This is the documentation for SET's Editor.

Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004 Salvador
Eduardo Tropea

This documentation may be freely distributed, provided this copyright notice
is left intact on all copies.

 EDITOR
*******

  Table of Contents
  *****************

1 Introduction
  1.1 Copying
  1.2 What is SETs Editor?
  1.3 About the Author
2 Documentation guide
  2.1 What can be configured to make easier the operation
  2.2 What can be configured to change the look of the editor
  2.3 What can be configured to change the syntax highlighting
  2.4 Things to get help or find some definition
  2.5 Running external programs (compilers/filters/others)
  2.6 Configuration files
3 Available commands and keys assignments
  3.1 Conventions
  3.2 Cursor movement
  3.3 Insert and Delete
  3.4 Blocks
  3.4.1 Block modes
  3.4.2 Selecting with the mouse or Shift
  3.4.2.1 Using the mouse
  3.4.2.2 Using the Shift key
  3.4.3 Indentation
  3.4.4 Rectangular Blocks
  3.5 Miscellaneous keyboard commands
4 Keyboard configuration
  4.1 How to configure the keyboard
  4.1.1 Assigning a sequence of commands
  4.1.2 Assigning a sLisp macro
  4.1.3 Assigning sLisp code
  4.2 Alt key configuration
  4.3 Restoring the default keyboard assignments
  4.4 Consulting scan codes
5 Pull-down menues
  5.1 File
  5.1.1 Open
  5.1.2 New
  5.1.3 Open Read-only copy
  5.1.4 Save
  5.1.5 Save as
  5.1.6 Save as UNIX or DOS
  5.1.7 Save as DOS or UNIX
  5.1.8 Save with same time
  5.1.9 Save all
  5.1.10 Print
  5.1.11 Print Setup
  5.1.12 Shell
  5.1.13 Quit
  5.1.14 Exit
  5.2 Edit
  5.2.1 Undo
  5.2.2 Redo
  5.2.3 Cut
  5.2.4 Copy
  5.2.5 Paste
  5.2.6 Show clipboard
  5.2.7 Clear
  5.2.8 Set Local
  5.2.9 Set Global
  5.2.10 Expand all tabs
  5.2.11 Compact text
  5.2.12 Copy to Windows Clipboard
  5.2.13 Paste from Windows Clipboard
  5.2.14 Copy to file Clipboard
  5.2.15 Paste from file Clipboard
  5.2.16 Push cursor position
  5.2.17 Pop cursor position
  5.2.18 Case (Menu)
  5.2.18.1 Block to upper
  5.2.18.2 Block to lower
  5.2.18.3 Character toggle
  5.2.18.4 Block invert
  5.2.18.5 Block alternate
  5.2.19 Insert new line (do not move)
  5.3 Search
  5.3.1 Find
  5.3.1.1 Regular Expressions Options
  5.3.2 Replace
  5.3.3 Search again
  5.3.4 Name current function
  5.3.5 Jump to function
  5.3.6 Jump to prototype
  5.3.7 Jump to symbol
  5.3.8 Go to line
  5.3.9 Class browser
  5.3.10 Word completion
  5.3.11 Jump to last cursor position
  5.3.12 Jump to last undo position
  5.4 Macro
  5.4.1 Record (Macro)
  5.4.2 Stop (Macro)
  5.4.3 Play (Macro)
  5.4.4 Choose (Macro)
  5.4.5 Repeat (Macro)
  5.4.6 Generate Code
  5.4.7 Run selected code
  5.4.8 Enter code to run
  5.4.9 Pseudo Macro (menu)
  5.5 Rectangle
  5.5.1 Start (Rectangle)
  5.5.2 End (Rectangle)
  5.5.3 Hide (Rectangle)
  5.5.4 Copy (Rectangle)
  5.5.5 Paste (Rectangle)
  5.5.6 Cut (Rectangle)
  5.5.7 Clear (Rectangle)
  5.5.8 Move (Rectangle)
  5.5.9 To upper (Rectangle)
  5.5.10 To lower (Rectangle)
  5.6 Windows
  5.6.1 Size/move
  5.6.2 Zoom
  5.6.3 Tile
  5.6.4 Cascade
  5.6.5 Next (Window)
  5.6.6 Previous (Window)
  5.6.7 Close
  5.6.8 List
  5.6.9 User Screen
  5.7 Tool&Ops
  5.7.1 Options
  5.7.1.1 Customize Colors
  5.7.1.2 Color Palette
  5.7.1.3 Color Theme
  5.7.1.4 Editor General
  5.7.1.5 Check for modified files
  5.7.1.6 Screen Saver
  5.7.1.7 SDG Options
  5.7.1.8 Run program (which one)
  5.7.1.9 Keyboard
  5.7.1.10 Key assignment
  5.7.1.11 Setup Alt keys
  5.7.1.12 Key pad behavior
  5.7.1.13 Back to defaults
  5.7.1.14 Consult scan codes
  5.7.1.15 Screen Options
  5.7.1.16 Encodings
  5.7.1.17 Fonts
  5.7.1.18 User Words
  5.7.1.19 Default global edition
  5.7.1.20 File open dialog
  5.7.1.21 Do not create backups for
  5.7.1.22 Search files under cursor in
  5.7.1.23 List of tag files
  5.7.1.24 Tag files options
  5.7.1.25 Regenerate central file
  5.7.1.26 Calendar options
  5.7.1.27 Advice dialogs
  5.7.2 Calculator (command/menu)
  5.7.3 SDG
  5.7.4 Run program
  5.7.5 Grep
  5.7.6 HTML Accents
  5.7.6.1 Convert accents to tags
  5.7.6.2 Convert tags to accents
  5.7.7 Export as HTML
  5.7.8 Insert key name
  5.7.9 Remap code page
  5.7.10 Profile Editor
  5.7.11 Redraw screen
  5.7.12 Paste Emacs mode
  5.7.13 Block quoted printable decode
  5.7.14 Un/Indent block
  5.7.14.1 Indent one space
  5.7.14.2 Unindent one character
  5.7.14.3 Indent one tab or gap
  5.7.14.4 Unindent one tab or gap
  5.7.14.5 Comment indent
  5.7.14.6 Comment unindent
  5.7.14.7 Arbitrary indent
  5.7.15 Delete memorized backups
  5.8 Project
  5.8.1 Open (Project)
  5.8.2 Close (Project)
  5.8.3 Save (Project)
  5.8.4 Save desktop here
  5.8.5 Export project
  5.8.6 Import project items
  5.9 Help
  5.9.1 InfView
  5.9.2 Another InfView
  5.9.3 Tip of the day
  5.9.4 Syntax help
  5.9.4.1 Options (Syntax help)
  5.9.4.2 Files to search (Syntax help)
  5.9.4.3 Search (Syntax help)
6 Menues configuration
  6.1 SubMenu and EndSubMenu
  6.2 Key names
  6.3 MenuItem entries
  6.4 Commands in menu items
  6.5 MenuItemC entries
  6.6 MenuSeparator
  6.7 The special submenu used for the right click
  6.8 Status bar ranges
  6.9 Visible status bar entries
  6.10 Invisible status bar entries
  6.11 Conditional menu entries
  6.12 Comments in menu files
7 Editing Modes
  7.1 Overwrite
  7.2 Autoindent
  7.3 Real Tabs
  7.4 Persistent Blocks
  7.5 Intelligent C indent
  7.5.1 Can you explain more about the behavior of this mode?
  7.5.2 Do you have more examples?
  7.6 Column cursor
  7.7 Row cursor
  7.8 Match pair highlight
  7.9 Match pair on the fly
  7.10 Do not wait to search the pair
  7.11 Transparent Blocks
  7.12 Optimal Fill
  7.13 Wrap Words
  7.14 Do not move the cursor on Paste
  7.15 Scroll Lock centers
  7.16 See Tabs
  7.17 Do not move inside tabs
  7.18 Tab indents
  7.19 Use indent size
  7.20 Keep trailing whitespace
  7.21 Backspace unindents
  7.22 Column Markers
  7.23 Syntax Highlighting
8 Syntax Highlighting File
  8.1 AllowedInsideNames
  8.2 CanStartAName
  8.3 Case
  8.4 CloseComment1
  8.5 EmacsModes
  8.6 EOLCInFirstCol
  8.7 EOLCInFirstCol1
  8.8 EOLCInFirstCol2
  8.9 EOLCInFirstUse1
  8.10 EOLCInFirstUse2
  8.11 EOLComment1
  8.12 Escape
  8.13 EscapeAnywhere
  8.14 Files
  8.15 FullNameMatch
  8.16 HexMarker
  8.17 Keywords
  8.18 Name
  8.19 NameMatch
  8.20 NoCheckNumbers
  8.21 OpenComment1
  8.21.1 Format of short syntax highlighting definitions
  8.22 PartialKeywords
  8.23 PMacros
  8.24 Preprocessor
  8.25 RelaxNumberCheck
  8.26 ShellScript
  8.27 ShortString
  8.28 SpecialSymbol
  8.29 String1
  8.30 Symbols1
  8.31 Symbols2
  8.32 UseInternal
9 Pseudo Macros
  9.1 Please enlighten me - what is that?
  9.2 How can I customize that?
10 sLisp macros
  10.1 How to write a sLisp macro
  10.2 How strings are parsed
  10.3 Running programs with a macro
  10.4 Editor specific commands
  10.4.1 AskString
  10.4.2 BindKey
  10.4.3 ComplChoose
  10.4.4 defmacro
  10.4.5 EvalString
  10.4.6 FindAgain
  10.4.7 FindString
  10.4.8 ForceUpdate
  10.4.9 GetCursorX
  10.4.10 GetCursorY
  10.4.11 GetCurWindowNumber
  10.4.12 GetMaxWindowNumber
  10.4.13 GetSystemInfo
  10.4.14 getenv
  10.4.15 GetSelection
  10.4.16 GetSyntaxAtCursor
  10.4.17 GetSyntaxLang
  10.4.18 InsertText
  10.4.19 KeyBindings
  10.4.20 MessageBox
  10.4.21 OpenFile
  10.4.22 ReplaceAgain
  10.4.23 ReplaceString
  10.4.24 RunProgram
  10.4.25 RunProgramRedir
  10.4.26 SelectionExists
  10.4.27 SelectWindowNumber
  10.4.28 SendCommands
  10.4.29 SetCursorXY
  10.4.30 ShowInMessageWindow
  10.4.31 ShowInStatusLine
  10.4.32 WhichEditor
  10.4.33 WordUnderCursor
  10.5 General sLisp commands
  10.5.1 and
  10.5.2 Operator = (assign)
  10.5.3 Operator & (bitwise and)
  10.5.4 cond
  10.5.5 Operator - (decrement)
  10.5.6 Operator != (different)
  10.5.7 Operator == (equal)
  10.5.8 eval
  10.5.9 exitloop
  10.5.10 gstr
  10.5.11 if
  10.5.12 Operator ++ (increment)
  10.5.13 left
  10.5.14 length
  10.5.15 loop
  10.5.16 Operator ~ (negated)
  10.5.17 not
  10.5.18 or
  10.5.19 Operator | (bitwise or)
  10.5.20 Operator + (plus)
  10.5.21 prex
  10.5.22 repeat
  10.5.23 right
  10.5.24 setv
  10.5.25 ShortFileName
  10.5.26 sstr
  10.5.27 strcasecmp
  10.5.28 strcmp
  10.5.29 strstr
  10.5.30 strxlt
  10.5.31 substr
  10.5.32 Operator - (substraction)
  10.5.33 tostr
  10.6 Writing macros that use text filters
  10.6.1 How to use Setedit for something it was not meant to
  10.6.1.1 Step 1 - Building your macro
  10.6.1.2 Step 2 - The filter program
  10.6.1.3 Step 3 - Binding your script to a keystroke
  10.6.1.4 Examples
11 Calculator
12 How to contact me
  12.1 Bugs
13 TAGS files
  13.1 What are tags?
  13.2 Which program is used to generate TAGS files?
  13.3 How should I generate the tags?
  13.4 Can these files be created and updated by the editor?
  13.5 What is the easiest way to use tags?
  13.6 What can I do with tags?
  13.7 Technical details about tags
14 Debugging
  14.1 Supported platforms for debugging
  14.2 Mechanism used to debug
  14.3 Quick start to debugging
  14.4 Supported debug targets
  14.5 Available debug options
  14.5.1 Debug options
  14.5.2 Path for sources
  14.5.3 Messages displayed
  14.5.4 Advanced debug options
  14.6 Debugging states
  14.6.1 Going to the connected debug state
  14.6.2 Going to the ready to run debug state
  14.7 Running the program to debug
  14.8 Stopping the program you are debugging
  14.8.1 Breakpoints
  14.8.2 Advanced breakpoint options
  14.8.3 Watchpoints
  14.9 Examining data
  14.9.1 Evaluate or Modify expression
  14.9.2 Watch an expression
  14.9.3 Watch an expression with scope
  14.9.4 Inspectors
  14.9.5 Data Window
  14.9.5.1 File - Read block (DW)
  14.9.5.2 File - Write block (DW)
  14.9.5.3 Move - Up (DW)
  14.9.5.4 Move - Down (DW)
  14.9.5.5 Move - Right (DW)
  14.9.5.6 Move - Left (DW)
  14.9.5.7 Move - Page down (DW)
  14.9.5.8 Move - Page up (DW)
  14.9.5.9 Move - First column (DW)
  14.9.5.10 Move - Last column (DW)
  14.9.5.11 Move - First row (DW)
  14.9.5.12 Move - Last row (DW)
  14.9.5.13 Move - First addr increment (DW)
  14.9.5.14 Move - First addr decrement (DW)
  14.9.5.15 Address - Change base address (DW)
  14.9.5.16 Address - Go to new address (DW)
  14.9.5.17 Address - Follow pointer (DW)
  14.9.5.18 Address - Follow pointer in new window (DW)
  14.9.5.19 Address - Recompute address (DW)
  14.9.5.20 Mode - Toggle auto follow (DW)
  14.9.5.21 Mode - Change display mode (DW)
  14.9.5.22 Mode - Toggle endian mode (DW)
  14.9.5.23 Mode - Change radix (DW)
  14.9.5.24 Block - Fill (DW)
  14.9.5.25 Block - Clear (DW)
  14.9.5.26 Block - Move (DW)
  14.9.5.27 Various - Less bytes per line (DW)
  14.9.5.28 Various - More bytes per line (DW)
  14.9.5.29 Various - Update memory (DW)
  14.9.5.30 Data Window Limitations and Details
  14.9.6 Stack window
  14.10 Resuming the execution of the program
  14.10.1 Continue execution
  14.10.2 Step over
  14.10.3 Trace into
  14.10.4 Executing until cursor position is reached
  14.10.5 Executing until return
  14.10.6 Returning immediatly
  14.11 Finishing and restarting a debug session
  14.11.1 Killing the program you are debugging
  14.11.2 Closing the debug session
  14.11.3 Destroying the debug session
  14.12 Examining the calling stack
  14.13 Cleaning the debug session
  14.14 Selecting the thread to debug
  14.15 Disassembler Window
  14.16 Debugging already running processes
  14.17 Debug Messages Window
  14.18 Editing a debug expression
  14.19 Debugging in the Linux console
15 Miscellaneous
  15.1 Configuration files location
  15.1.1 Configuration directories for DOS and Windows
  15.1.2 Configuration directories for POSIX systems
  15.2 Clipboard
  15.3 Time and date modifiers formats
  15.4 Regular Expressions
  15.5 Desktop Files
  15.6 Text mode attributes
  15.7 File Open
  15.7.1 Sorting of the files and directories in the dialog
  15.7.2 Files and directories excluded in the dialog
  15.8 Message Window
  15.9 Error messages from an external application
  15.10 Mouse under Linux
  15.11 Passing extra command line options
  15.12 How to run setedit remotely without root installation (UNIX)
16 Index
17 Index of key commands

1 Introduction
**************

   This document describes the use of the SET's editor; this editor was
designed for programmers and to be used alone or inside of the RHIDE.

   This documentation may be freely distributed with the editor or the RHIDE
package or any part thereof, provided this copyright notice is left intact on
all copies.

   Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the entire
resulting derived work is distributed under the terms of a permission notice
identical to this one.

  People who helped me develop the editor (alphabetically sorted):

   * ANDRIS PAVENIS <pavenis@lanet.lv>
     (with a lot of patches and reports)

   * BERND BECKER <munin@munin.inka.de>
     (with a lot of corrections to this text)

   * BURTON RADONS <loth@pacificcoast.net>
     (with the new calculator, bug reports and ideas)

   * FRANK DONAHOE <fdonahoe@wilkes1.wilkes.edu>
     (with a lot of corrections to this text)

   * GRZEGORZ ADAM HANKIEWICZ <gradha@users.sourceforge.net>
     (with tests and a lot of ideas)

   * IVAN BALDO <ibaldo@adinet.com.uy>
     (Debian packages, tests and a lot of ideas)

   * MAREK HABERSACK <grendel@ananke.amu.edu.pl>
     (with tests and a lot of ideas)

   * MOLNAR LASZLO <molnarl@cdata.tvnet.hu>
     (with the old calculator, tests and a lot of ideas)

   * ROBERT HHNE <robert.hoehne@gmx.net>
     (with some base routines, a lot of patches and reports)

  The editor is distributed under the GPL license. Please read the files
included in the source distribution for more information.

  This editor is included in the Robert Hhne's RHIDE as a replacement for the
original Turbo Vision's editor class.

1.1 Copying
===========

  The editor is distributed under the GPL license:

  This program is free software; you can redistribute it and/or modify
  it under the terms of the GNU General Public License as published by
  the Free Software Foundation; either version 2 of the License, or
  (at your option) any later version.

  This program is distributed in the hope that it will be useful,
  but WITHOUT ANY WARRANTY; without even the implied warranty of
  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  GNU General Public License for more details.

  You should have received a copy of the GNU General Public License
  along with this program; if not, write to the Free Software
  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.

  A copy of the license should be in the package, if not please tell me.

1.2 What is SETs Editor?
========================

  SET's editor is an editor designed to be used by programmers; the main target
of the editor is C and C++ code but Pascal and Clipper are supported too.
Currently I'm trying to make it more general so it's useful not only for
programming

  The editor was designed to be very similar to the DOS standard editors for C,
especially to Borland's IDE editor. The editor supports a lot of WordStar
style commands plus some CUA commands, so if you have used any editor that
uses these kinds of commands you'll find my editor very familiar. On the
other hand if you have never used a DOS editor, especially if you use VI on
UNIX machines you'll feel lost. In this case you may want to configure the
keyboard.

 ("How to configure the keyboard" Section 4.1).

1.3 About the Author
====================

The editor was created by Salvador Eduardo Tropea with some code contributed
by Robert Hhne

     E-Mail: Salvador Eduardo Tropea <salvador@inti.gov.ar>

     Telephone: (+5411) 4759-0013
     Postal Address:
     Salvador E. Tropea
     Curapalige 2124
     (1678) Caseros - 3 de Febrero
     Prov: Buenos Aires
     Argentina


2 Documentation guide
*********************

  This document is very extensive and was created while developing the editor
starting in 1996. For this reason sometimes is complex to find what you want
or to even know that something is possible. This chapter was created to try
to alleviate this problem. If you have any suggestions about what to include
here don't hesitate and contact me.

2.1 What can be configured to make easier the operation
=======================================================

  Here is a list of what can be configured to customize the editor input.

   * The keyboard can be configured  ("Keyboard configuration" Chapter 4).

   * The keyboard can be also configured using sLisp command.   ("sLisp
     macros" Chapter 10).

   * The menues be also configured.  ("Menues configuration" Chapter 6).

   * The shortcuts displayed at the bottom of the screen are called `status
     bar' and can be configured in the same way menues are configured.
     ("Menues configuration" Chapter 6).

   * The small menu that pops up when you right click on editor windows is
     also configurable using the menu configuration.  ("Menues configuration"
     Chapter 6).

   * You can define some abbreviations that can then expanded using `^Space'.
     They are language sensitive and they are called pseudo-macros or
     pmacros.  ("Pseudo Macros" Chapter 9).

   * You can ask the editor to record a group of operations and then
     reproduce them. This is what this documentation calls a macro. The
     operations to record and play macros can be found in the `Macro' menu (
     "Macro" Section 5.4). You can also generate the sLisp code for a macro
     and store it in the `macros.slp' file to use it from the menues or just
     bind it to a key.

   * If you want to automate more complex operations you can use sLisp
     macros. They use a very simple language that is quite similar to Lisp or
     Scheme languages.  ("sLisp macros" Chapter 10). sLisp macros can be
     called using the `Macro|Choose...' or `Macro|Repeat' menu options. You
     can also add menu entries that start certain sLisp macro or bind a sLisp
     macro to a key.  Additionally you can bind a small piece of sLisp code
     to a key or menu entry.

2.2 What can be configured to change the look of the editor
===========================================================

  Here are some things you can change to ajust how the editor looks.

   * The colors used for different elements of the editor can be customized
     using the `Tool&Ops | Options | Colors | Customize...' menu option.
     ("Customize Colors" Section 5.7.1.1).

   * Some predefined color schemes are available in the `Tool&Ops | Options |
     Colors | Theme...' menu option. You can contribute your own color scheme
     sending me you desktop file (desktop files have .dst as extension).
     ("Color Theme" Section 5.7.1.3).

   * The color palette can be customized using the `Tool&Ops | Options |
     Colors | Palette...' menu. This is available for most OSs and terminals,
     but not for all.  ("Color Palette" Section 5.7.1.2).

   * The character encodings used by the editor can be also configured. This
     is in the `Tool&Ops | Options | Encodings...' menu.  ("Encodings"
     Section 5.7.1.16).

   * Some terminals supports changing the screen or window size. This is in
     the `Tool&Ops | Options | Screen Options...' menu.  ("Screen Options"
     Section 5.7.1.15).

   * Some terminals supports changing the font and/or the font size. The
     editor includes various fonts suitable for it. This is in the `Tool&Ops
     | Options | Fonts...' menu.  ("Fonts" Section 5.7.1.17).

2.3 What can be configured to change the syntax highlighting
============================================================

  The editor highlights different elements in the code using different colors
here are things you can do to modify it.

   * The colors used for it can be configured.   ("What can be configured to
     change the look of the editor" Section 2.2).

   * The syntax highlighting can be configured and/or extended editing the
     `syntaxhl.shl' file.  ("Syntax Highlighting File" Chapter 8).

   * You can define some words as special words. An example of use is to
     highlight your own data types. They are called `user words' and you can
     define them for each supported syntax highlighting.  ("User Words"
     Section 5.7.1.18).

2.4 Things to get help or find some definition
==============================================

  When coding you usually need to access to help files describing the language
reference or you need to search in your files. This is a list of the most
important tools for it.

   * If the language and/or library you are using have documentation in the
     `info' format (the format used by the GNU project for on-line
     documentation) you can configure the editor to search the syntax of a
     command when pressing `^F1'.  ("Syntax help" Section 5.9.4).

   * You can browse info files using the InfView info browser. The editor
     also have a man page viewer. Both can be accessed from the `Help' menu.
     ("Help" Section 5.9).

   * The editor have a search and replace dialog that supports regular
     expressions and also supports replacement using `templates'. The `Search
     | Find' ( "Find" Section 5.3.1) and `Search | Replace' ( "Replace"
     Section 5.3.2) menu entries are used for it.

   * You can search for text using regular expressions. The `Powered Grep'
     dialog allows searching in project items and directories, also
     recursivelly.   ("Grep" Section 5.7.5).

   * Using `tag files' the editor can search for definitions, do word
     completion and browse classes.  ("TAGS files" Chapter 13).

   * For some languages the editor can parse your code and extract the
     function definitions.  ("Jump to function" Section 5.3.5).

2.5 Running external programs (compilers/filters/others)
========================================================

  You can run external programs to compile, apply filters or any other thing.
Here are some related options.

   * You run a external program using `Tool&Ops | Run program' menu option (
     "Run program" Section 5.7.4). To configure it you can use the `Tool&Ops
     | Options | Run program (which one)' menu option ( "Run program (which
     one)" Section 5.7.1.8).

   * The editor collects the messages from the external program and displays
     them in the `message window'.  ("Message Window" Section 15.8).

   * The editor parses the messages from the external program looking for
     errors. The parser can be selected using the `Tool&Ops | Options | Run
     program (which one)' menu option. You can add your own parsers editing
     the `errors.cle' file.  ("Error messages from an external application"
     Section 15.9).

   * You can call external programs binding a sLisp macro to a key or menu
     entry.  ("sLisp macros" Chapter 10),  ("Keyboard configuration" Chapter
     4),  ("Menues configuration" Chapter 6).

   * You can also call external filters using sLisp macros.   ("Writing
     macros that use text filters" Section 10.6).

2.6 Configuration files
=======================

  Here is a list of the most important configuration files.

   * The editor can read configuration files from various directories.
     ("Configuration files location" Section 15.1).

   * `syntaxhl.shl' configures the syntax highlghting.   ("Syntax
     Highlighting File" Chapter 8).

   * `pmacros.pmc' and other `.pmc' files are used to configure the pmacros
     (pseudo-macros).  ("Pseudo Macros" Chapter 9).

   * `errors.cle' configures the error parsers. They are used to collect
     error messages from external programs.   ("Error messages from an
     external application" Section 15.9).

   * `macros.slp' defines the available sLisp macros.  ("sLisp macros"
     Chapter 10).

   * `userword.txt' defines the user reserved words for syntax highlighting.
     ("User Words" Section 5.7.1.18).

   * `deflopts.txt' configures the default edition options for each syntax
     highlighting style.  ("Default global edition" Section 5.7.1.19).

   * `nobkp.txt' configures for which files the editor won't create back-ups.
     ("Do not create backups for" Section 5.7.1.21).

   * `tcedit.dst' and other `.dst' files. They are desktop files and contains
     information about opened windows and configuration options. They can be
     local (one per directory) or global (one for all the system). Projects
     also have its own desktop file. The format is binary.

   * `.epr' files. They are project files. The format is binary.

   * `keybind.dat' contains keyboard binding information. The format is
     binary.

   * `setenvir.dat' contains some global settings in binary form.

3 Available commands and keys assignments
*****************************************

  In this section I will explain the features of the editor and the default
configuration for the keyboard. If you want to change an assignment of a key
consult "configure the keyboard".   ("How to configure the keyboard" Section
4.1).

  Read the conventions topic first to understand my way of indicating
keystrokes.

  In the description of each command I'll include the internal name used by the
editor because this name is needed to configure the keyboard.

3.1 Conventions
===============

  I'll use some conventions when talking about the keystrokes needed to trigger
some command. So here is what I use:

  The key named <Ctrl> or Control is represented as `^'; this key doesn't have
any effect used alone inside of the editor so the `^' symbol will be used only
in conjunction with the name of a key indicating that you must press the two
keys at the same time. For example, `^A' is <Ctrl> and <A> at the same time.
When I say "at the same time" that means: press `<Ctrl>', hold it, and press
the other key; that's the reason to put <A> after <Ctrl>.

  To indicate a sequence of keystrokes I'll use a dash to separate the keys. For
example, `^K-B' is <Ctrl> and <K> at the same time, and then press <B>, of
course release `^K' first.

  To indicate keys pressed at the same time other than `^x' I'll use a plus.
For example, `Shift+^Insert' is the three keys at the same time!

  I don't think that you are stupid; the editor is written for programmers, but
I wanted to make that clear to avoid problems ;-).

3.2 Cursor movement
===================

   * Character left
     - Command: cmcCharLeft
     - Key: Left arrow
     - Alternate: ^S

   * Character right
     - Command: cmcCharRight
     - Key: Right arrow
     - Alternate: ^D

   * Word left
     - Command: cmcWordLeft
     - Key: ^Left arrow
     - Alternate: ^A

   * Word right
     - Command: cmcWordRight
     - Key: ^Right arrow
     - Alternate: ^F

   * End of the word
     - Command: cmcGoEndOfWord
     - Key:
     - Alternate:

   * Line up
     - Command: cmcLineUp
     - Key: Up arrow
     - Alternate: ^E

   * Line down
     - Command: cmcLineDown
     - Key: Down arrow
     - Alternate: ^X

   * Scroll the screen one line up
     - Command: cmcScrollUp
     - Key: ^W
     - Alternate:

   * Scroll the screen one line down
     - Command: cmcScrollDown
     - Key: ^Z
     - Alternate:

   * Page up
     - Command: cmcPageUp
     - Key: PgUp
     - Alternate: ^R

   * Page down
     - Command: cmcPageDown
     - Key: PgDn
     - Alternate: ^C

   * Beginning of line
     - Command: cmcLineStart
     - Key: Home
     - Alternate: ^Q-S

   * End of line
     - Command: cmcLineEnd
     - Key: End
     - Alternate: ^Q-D

   * Top of window
     - Command: cmcFirstLineInScreen
     - Key: ^Q-E
     - Alternate: ^Home

   * Bottom of window
     - Command: cmcLastLineInScreen
     - Key: ^Q-X
     - Alternate: ^End

   * Top of file
     - Command: cmcTextStart
     - Key: ^Q-R
     - Alternate: ^PgUp

   * Bottom of file
     - Command: cmcTextEnd
     - Key: ^Q-C
     - Alternate: ^PgDn



3.3 Insert and Delete
=====================

   * Delete the character under cursor
     - Command: cmcDelChar
     - Key: Del
     - Alternate: ^G

   * Delete character to left
     - Command: cmcBackSpace
     - Key: Backspace
     - Alternate: ^H

   * Delete line
     - Command: cmcDelLine
     - Key: ^Y
     - Alternate:

   * Delete to end of line
     - Command: cmcDelEnd
     - Key: ^Q-Y
     - Alternate: Shift+^Y

   * Delete to start of line
     - Command: cmcDelStart
     - Key: ^Q-H
     - Alternate:

   * Delete word at right
     - Command: cmcDelWord
     - Key: ^T
     - Alternate:

   * Delete word at left
     - Command: cmcDelPrevWord
     - Key: ^Backspace
     - Alternate:

   * Insert line
     - Command: cmcNewLine
     - Key: Enter
     - Alternate: ^N

   * Insert mode on/off
     - Command: cmcInsMode
     - Key: Ins
     - Alternate: ^V



  When you are in insert mode all the typed characters are inserted in the
text, but when the insert mode is off the typed characters replace the old
text. The editor starts with insert mode on. You can quickly know the mode by
the cursor shape. When the insert mode is on, the cursor is only a line, but
when it is off, the cursor is block shaped.

3.4 Blocks
==========

  A block is a selected portion of the text. You can copy, delete, `etc.' blocks
of text. The associated commands are:

   * Move to beginning of block
     - Command: cmcGoBeginBlock
     - Key: ^Q-B
     - Alternate:

   * Move to end of block
     - Command: cmcGoEndBlock
     - Key: ^Q-K
     - Alternate:

   * Set beginning of block
     - Command: cmcStartSelect
     - Key: ^K-B
     - Alternate:

   * Set end of block
     - Command: cmcEndSelect
     - Key: ^K-K
     - Alternate:

   * Hide/Show block
     - Command: cmcHideSelect
     - Key: ^K-H
     - Alternate:

   * Mark line
     - Command: cmcMarkLine
     - Key: ^K-L
     - Alternate:

   * Mark word
     - Command: cmcMarkWord
     - Key: ^K-T
     - Alternate:

   * Delete block and copy it to the Clipboard
     - Command: cmcCut
     - Key: ^K-Y
     - Alternate: Shift+Del

   * Copy the selected block
     - Command: cmcCopyBlock
     - Key: ^K-C
     - Alternate:

   * Move block
     - Command: cmcMoveBlock
     - Key: ^K+V
     - Alternate:

   * Copy to Clipboard
     - Command: cmcCopy
     - Key: ^Ins
     - Alternate:

   * Delete block
     - Command: cmcClear
     - Key: ^Del
     - Alternate:

   * Paste from Clipboard
     - Command: cmcPaste
     - Key: Shift+Ins
     - Alternate:

   * Read block from disk
     - Command: cmcReadBlock
     - Key: ^K-R
     - Alternate: Shift+^R

   * Write block to disk
     - Command: cmcWriteBlock
     - Key: ^K-W
     - Alternate: Shift+^W

   * Replace the block by the Clipboard block
     - Command: cmcReplaceSelect
     - Key: Shift+^Ins
     - Alternate:

   * Convert to Uppercase
     - Command: cmcToUpper
     - Key: ^K-M
     - Alternate:

   * Convert to Lowercase
     - Command: cmcToLower
     - Key: ^K-O
     - Alternate:

   * Invert case
     - Command: cmcInvertCase
     - Key: none
     - Alternate:

   * Alternate case
     - Command: cmcAltCase
     - Key: none
     - Alternate:

   * Report the length of the block
     - Command: cmcSelLength
     - Key: ^Q-L
     - Alternate:



3.4.1 Block modes
-----------------

  There are two block modes. One is the mode that the old editor of RHIDE used.
This mode is used in CUA programs. The other is called Persistent Blocks.

  In the normal mode each time you select a block and then insert anything in
it (with `cmcPaste' or by typing anything) the selected block is deleted and
is replaced by the new text.

  In persistent blocks the selection is not replaced and is not lost when you
move the cursor. From this comes the name "Persistent." In this mode you can
use `cmcMoveBlock' and `cmcCopyBlock' without using the Clipboard. In
addition you can apply indentations to the block ( "Indentation" Section
3.4.3), search only inside it, `etc.' That's what makes this mode much more
powerful than the former. If you really like to replace the selected text by
the selection of the Clipboard, that's the default behaviour of the first
mode. You can use the `cmcReplaceSelect' command to achieve the same in the
Persistent Blocks mode.

3.4.2 Selecting with the mouse or Shift
---------------------------------------

  The described commands for selecting a block, `cmcStartSelect' and
`cmcEndSelect', are good but not so quick. There are other ways to do this.

3.4.2.1 Using the mouse
.......................

  Using the mouse you need only point to the start place, and while pressing the
left button moving the mouse to the end point of your block.

  To select a word with the mouse just double click on it.

3.4.2.2 Using the Shift key
...........................

  Using the <Shift> key you only need to move the cursor to the start point,
keep `<Shift>' pressed and move the cursor to the end point with any of the
available cursor commands.  ("Cursor movement" Section 3.2).

3.4.3 Indentation
-----------------

  You can indent or unindent a block of text using various commands, but you
must keep in mind that for now the editor is limited in the following way: If
you are using tabs to indent your text, don't mix tabs with spaces and, if
you are using spaces to indent, don't mix them with real tabs.  ("Real Tabs"
Section 7.3).

   * Indent block one position adding a space
     - Command: cmcIndentBlkOne
     - Key: ^K-I
     - Alternate: Shift+^I

   * Unindent block one character - not an x position
     - Command: cmcUnIndentBlkOne
     - Key: ^K-U
     - Alternate: Shift+^U

   * Indent block
     - Command: cmcIndentBlk
     - Key: ^K-Tab
     - Alternate:

   * Unindent block
     - Command: cmcUnIndentBlk
     - Key: ^K-Shift+Tab
     - Alternate:



`cmcUnIndentBlkOne' unindents deleting one char at the start of the line so,
if the line is indented with tabs, the line will retract one tab.

`cmcIndentBlk' acts according to the mode. If you are using tabs, the editor
will put one tab at the beginning of each line. If you aren't using tabs the
editor will operate the Tab command on the first line and then will use this
amount of indentation on the entire block.  ("Real Tabs" Section 7.3).

`cmcUnIndentBlk' acts according to the mode too.  ("Real Tabs" Section 7.3).
Just like `cmcUnIndentBlkOne' it deletes one tab but if you don't use tabs
the editor uses Backspace on the first used column of the first line of the
block and unindents by the resulting amount the rest of the block.

The following commands aren't applied to the whole block, they apply only to
the line where the cursor is positioned.

   * Smart Indent block
     - Command: cmcSmartIndent
     - Key: ^Tab

   * Smart Unindent block
     - Command: cmcSmartUnIndent
     - Key: Shift+^Tab



`cmcSmartIndent' and `cmcSmartUnIndent' indents take as reference the { } pair
where the cursor is, for example:

       {
     line1
          line2
        line3
       }

After indenting line1 with `cmcSmartIndent' and line2 with `cmcSmartUnIndent'
you get:

       {
        line1
        line2
        line3
       }

The indentation is made with spaces and you must put the cursor on the first
letter of the line, the l in this example.

3.4.4 Rectangular Blocks
------------------------

  The editor includes a mode where you can select a rectangular portion of the
text and copy, cut, clear, paste, move, `etc.' this region.  This tool is
very useful for modifications on columns.

  Attention! The selected area is based on the X,Y coordinates. For this reason
if you insert lines before the bottom of the rectangle the area won't be
moved. I don't plan to move the area by now because that takes some CPU and I
think that this selection is made just before using it. So don't report that
like a bug. That is the way it works!

   * Set beginning of block
     - Command: cmcSelRectStart
     - Key: ^K-Shift+B

   * Set end of block
     - Command: cmcSelRectEnd
     - Key: ^K-Shift+K

   * Hide/Show block
     - Command: cmcSelRectHide
     - Key: ^K-Shift+H

   * Delete block and copy it to an special Clipboard
     - Command: cmcSelRectCut
     - Key: ^K-ShiftT

   * Move block
     - Command: cmcSelRectMove
     - Key: ^K+Shift+V

   * Copy to special Clipboard
     - Command: cmcSelRectCopy
     - Key: ^K-Shift+C

   * Delete block
     - Command: cmcSelRectDel
     - Key: ^K-Shift+L

   * Paste from special Clipboard
     - Command: cmcSelRectPaste
     - Key: ^K-Shift+P

   * Convert to uppercase
     - Command: cmcSelRectToUpper
     - Key:

   * Convert to lowercase
     - Command: cmcSelRectToLower
     - Key:



3.5 Miscellaneous keyboard commands
===================================

   * Autoindent mode on/off
     - Command: cmcIndentMode
     - Key: ^O

   * Find place marker
     - Command: cmcGotoMarkn
     - Key: ^Q n*

   * Set marker
     - Command: cmcPutMarkn
     - Key: ^K n*

   * Search for open curly bracket from where the cursor is
     - Command: cmcSearchStart
     - Key: ^[

   * Search for close curly bracket from where the cursor is
     - Command: cmcSearchEnd
     - Key: ^]

   * Search for ( from where the cursor is
     - Command: cmcSearchOpPar
     - Key: Shift+^9

   * Search for ) from where the cursor is
     - Command: cmcSearchClPar
     - Key: Shift+^0

   * Search for [ from where the cursor is
     - Command: cmcSearchOpCor
     - Key: Shift+^[

   * Search for ] from where the cursor is
     - Command: cmcSearchClCor
     - Key: Shift+^]

   * Search for the complementary pair
     - Command: cmcSearchComplement
     - Key: ^Q ESC

   * Undo
     - Command: cmcUndo
     - Key: Alt+Backspace

   * PMacro's Trigger
     - Command: cmcExpandCode
     - Key: ^Space

   * Goto Line
     - Command: cmcGotoEditorLine
     - Key: ^J

   * Set the options for the current window (Not in RHIDE)
     - Command: cmcSetLocalOptions
     - Key: Alt+L

   * Set the default options (Not in RHIDE)
     - Command: cmcSetGlobalOptions
     - Key: Alt+G

   * Convert all tabs to spaces
     - Command: cmcExpandAllTabs
     - Key: From menu

   * Compact the text using tabs
     - Command: cmcCompactBuffer
     - Key: From menu

   * Start recording a macro
     - Command: cmcRecordMacro
     - Key: Shift+F10

   * Stop recording a macro
     - Command: cmcStopMacro
     - Key: Alt+F10

   * Play a macro
     - Command: cmcPlayMacro
     - Key: ^F10



4 Keyboard configuration
************************

  The editor can be configured to trigger one or more commands pressing one key
or an arbitrary sequence of keys. Unlike old versions, now the sequence of
keys isn't limited.  Additionally you no longer need to configure the editor
for non-US keyboards.

  You can also assing sLisp defined macros to keys or even small portions of
sLisp code to keys. In addition you can change the key assignments using
sLisp macros.

  If you are looking for information about the names used for the keys in the
editor consult:  ("Key names" Section 6.2).

4.1 How to configure the keyboard
=================================

 These options are located under the menu option called Tool&Ops, submenu
Options, submenu Keyboard, submenu Key assignment; yes is a little deep in
the menu.

  After selecting this option you'll get a window with the keyboard
assignments. This window shows entries of the type `Key sequence -> Command
sequence'. You can delete an assignment by selecting it and choosing the
`Delete' button.

  The list is sorted by a internal criteria. The keys with <Shift> have an `S'
in front, for <Ctrl> you'll see a `C' and for <Alt> an `A'. The editor can
distinguish the left and right <Alt> keys; if you enable it the right <Alt>
will be represented by an `a'.  ("Alt key configuration" Section 4.2).

  To add a new assignment press the `Add' button. A window called `Sequence of
keys' will appear. This window is used to choose the sequence of keys that
will trigger an action in the editor. The sequence can be as large as you
want, so if you want to assign a sequence like this: `^A-Shift+^Insert-Alt+Z'
you can, but I doubt you really want to use such a complex combination even
if the editor is flexible enough to allow it.

  To add a new key to the sequence use the `Add' button, to delete a key use the
`Delete' button. The `Add' button always adds a key to the end of the list; to
insert a key in the sequence use the `Insert' button, it will insert the key
before the selected key. Finally you can choose if you want to assign a
sequence of commands or a sLisp macro to this key sequence.  ("sLisp macros"
Chapter 10).

4.1.1 Assigning a sequence of commands
--------------------------------------

  A window called `Commands' will appear. The mechanism to add, insert and
delete commands is the same as used for a keyboard sequence. This time when
you add or insert a new command to the sequence a window with all the
available commands will pop-up.  The meaning of each command can be found in
the indices of this help.

  To make a selection with the commands, like when holding <Shift>, you must use
the `SelectOn' and `SelectOff' commands. For an example you can take a look
at the assignments for the `Shift+Left' or `Shift+Right' keys.

4.1.2 Assigning a sLisp macro
-----------------------------

  A window called `Macros' will appear. This window shows all the macros
defined in the `macros.slp' file.  ("sLisp macros" Chapter 10).

  The main advantage of using macros instead of command sequence is that macros
can insert text in your code.

4.1.3 Assigning sLisp code
--------------------------

  A window called `sLisp code' will appear. This window shows allows entering a
small portion of sLisp code. For long code use define a macro.   ("sLisp
macros" Chapter 10).

  The sLisp code must start with `(' and end with `)' or you'll get parser
errors.

4.2 Alt key configuration
=========================

  The editor can distinguish the left and right <Alt> keys. As old versions
didn't allowed that and as different users use different <Alt> keys the
editor doesn't make any difference between them anymore by default.

  The `Setup Alt keys' menu option (under Tool&Ops | Options | Keyboard) allows
to enable it. Three options are offered:

   * Left Alt
     - Meaning: The menues are tiggered by the left Alt and you can use the
     right Alt for commands

   * Right Alt
     - Meaning: The reverse as for Left Alt - right Alt activates menues,
     left Alt commands

   * Both Alt
     - Meaning: Both keys can be used for menues


4.3 Restoring the default keyboard assignments
==============================================

  If you need to restore the original keyboard assignment because you did
something very wrong you can use this option for that. The option is located
under Tool&Ops | Options | Keyboard.

4.4 Consulting scan codes
=========================

  If you need to know the scan code of a key for your program you don't need to
use a table or another program. The editor has a tool built in that shows
scancodes for keypresses under Tool&Ops | Options | Keyboard.

5 Pull-down menues
******************

The menues are configurable; for this reason the following structure is just
one of many possible arrangements.

If you need or want to configure the shortcuts activating the menues or menue
options look at the `menubind.smn' file. The format is self explanatory and
the editor supports syntax highlighting for these files.   ("Menues
configuration" Chapter 6).

5.1 File
========

This menu contains the file operations (save, load, print, `etc.') and the
program exit functions.

5.1.1 Open
----------

This option brings the file open dialog ( "File Open" Section 15.7). From
this dialog you can select a file to load and edit.

Choose the Open button or select a file with `<Enter>' to open the file. Use
`<Esc>' to abort.

If the file is read-only a dialog will ask you if you want to make the file
writable, in this case the editor will try to change the file attributes. Not
always is it possible to change these attributes, as for example, the CDs are
read-only and obviously you can't change files on them.

Name of the command: cmeOpen.
Assigned key: `F3'

5.1.2 New
---------

Use this option to create a new and empty editor window. The window's name
will be `Untitled'.

An alternative way to do it is just using the `Open' option and give a new
name instead of selecting an existing file.  ("Open" Section 5.1.1). The
advantage of this method is that the new window already has a meaningful
name, instead of having to save the file first.

Name of the command: cmeNew.

5.1.3 Open Read-only copy
-------------------------

Use this option to open a copy of the file you are editing in another window.

Be careful, the new copy is loaded from disk so it may be unsychronized. The
copy will automatically become read-only so you won't be able to modify it
and accidentally overwrite the first copy.

For more information:  ("Editor General" Section 5.7.1.4).

Name of the command: cmeOpenROCopy.

5.1.4 Save
----------

This option saves the contents of the current window. Only the contents of
this window are saved; not anything else. Additionally it only saves if the
window was modified.

If the window is named `Untitled' this command acts like  ("Save as" Section
5.1.5).

Name of the command: cmcSave.
Assigned key: `F2'

5.1.5 Save as
-------------

This option allows to save the contents of the current window specifying a
new name for the file. For this purpose the file open dialog is used ( "File
Open" Section 15.7).

If the file already exists a dialog will pop-up asking for confirmation to
overwrite the existing file.

The title of the window is changed to reflect the new file name.

Name of the command: cmcSaveAs.

5.1.6 Save as UNIX or DOS
-------------------------

This menu entry says `Save as UNIX' for DOS or Win32 systems and `Save as
DOS' for UNIX systems.

This option is very similar to the `Save as' option ( "Save as" Section
5.1.5). The only difference is that the editor will change the end of line
character.

Additionally there is an option to save UNIX files as UNIX files without
converting it to DOS style. Tool&Ops|Options|Editor General ("Editor General"
Section 5.7.1.4).

Name of the command: cmcSaveAsConvertEOL.

5.1.7 Save as DOS or UNIX
-------------------------

This menu entry says `Save as DOS' for DOS or Win32 systems and `Save as
UNIX' for UNIX systems.

This option is useful when the editor is configured to save files without
converting the end of line character and you want to force a convertion.

Name of the command: cmcSaveAsNoConvertEOL.

5.1.8 Save with same time
-------------------------

This option is very similar to the `Save' option ( "Save" Section 5.1.4). The
only difference is that the editor won't modify the creation time. This
option is very useful to modify header files avoiding the recompilation of
the whole project. A common case is when you only add constants to a header
that is included by various files but only one will use the new constants.

Name of the command: cmcSaveSameTime.

5.1.9 Save all
--------------

All modified editing windows will be saved.

Name of the command: cmcSaveAll.

5.1.10 Print
------------

This option prints the current editor window. Don't use this option without
configuring the printer module.  ("Print Setup" Section 5.1.11).

This option was designed to print source files, to print plain text files or
avoid all the formating features. Select the portion of text to be printed
and save the block (`^K-W') to a file with the name of the device to which
your printer is connected. As an example to print to the DOS LPT 1 device
just write the block to the `lpt1' file.

The editor will report the number of lines processed and printed in the
message window.

To learn more about the message window  ("Message Window" Section 15.8).

Name of the command: cmePrintEditor.

5.1.11 Print Setup
------------------

This option brings up a dialog to configure the printing module. After
configuring it you can print using the `Print' option.  ("Print" Section
5.1.10).  If you want to print a plain text (without formating) consult the
`Print' option too.

The dialog asks for the following parameters:

   * Total lines per page: The total number of lines that fits on one page
     including the footer and header lines.

   * Columns w/o margin: The number of colums that fits on the page excluding
     the desired margin.

   * Left margin: The number of columns left blank at the left side of the
     page.

   * Print line numbers: Select this option to get the line numbers printed

   * Time format: The time format specified in the C style.  ("Time and date
     modifiers formats" Section 15.3).

   * Date format: The date format specified in the C style.  ("Time and date
     modifiers formats" Section 15.3).

   * Title: only used in the header.

   * Author: only used in the header.

   * Output file: Specify the printer device here. For example: the DOS LPT 1
     is the lpt1 file.

   * Printer initialization: The sequence of character used to initialize the
     printer. Normally it includes a reset and a font selection.  Specify the
     values separated by commas.

   * Before heading: Commands to send before the header.

   * After heading: Commands to send after the header.

   * Before footer: Commands to send before the footer.

   * After footer: Commands to send after the footer.

Use the `Ok' button to confirm or the `Cancel' button to reject.

The `Epson' button fills the values with the default settings for Epson
printers (ESCP2 language). The `HP' button fills the values with the default
settings for Hewlett Packard ink-jet printers.

The printer module was designed by me some years ago when I needed to present
a program at my University (Universidad Tecnologica Nacional) and I wanted to
add to the source code listing a header, footer, lines number, page number,
date/time of the printing, project and author. So that's what the routines
does. I know they are limited but they generate a very good listing, much
better than just printing the text without any formating.

Name of the command: cmeSetUpPrinter.

5.1.12 Shell
------------

Calls the default command interpreter indicated by the `COMSPEC' enviroment
variable. I guess you know that by typing exit you'll get back to the editor.

Name of the command: cmeDosShell.

5.1.13 Quit
-----------

This option exits the editor deleting all the back-up, desktop and project
files located in the current directory. That's useful if you want to keep the
directory clean.

When you use this option a dialog will appear asking for confirmation, you
can avoid this dialog in the future by checking the "don't show again" option.

Name of the command: cmeQuitDelete.
Assigned key: `Alt+Q'

5.1.14 Exit
-----------

This option just exits the program. The editor asks for each modified editor
window if you want to save the changes. All the settings are stored in the
desktop file automatically.

Name of the command: cmeQuit.
Assigned key: `Alt+X'

5.2 Edit
========

This menu contains all the editing operations that have a menu shortcut.

5.2.1 Undo
----------

This option reverts the last edit operation. Upto 32 operations can be
reverted.

Name of the command: cmcUndo.
Assigned key: `Alt+BackSpace'

5.2.2 Redo
----------

This option takes back the last `Undo'.  ("Undo" Section 5.2.1).

Name of the command: cmcRedo.

5.2.3 Cut
---------

The selected text is copied to the clipboard and deleted from the text.
("Clipboard" Section 15.2).

Name of the command: cmcCut.
Assigned key: `Shift+Del'

5.2.4 Copy
----------

The selected text is copied to the clipboard.  ("Clipboard" Section 15.2).

Name of the command: cmcCopy.
Assigned key: `^Ins'

5.2.5 Paste
-----------

The text that's currently selected in the clipboard is inserted at the cursor
position.  ("Clipboard" Section 15.2).

Name of the command: cmcPaste.
Assigned key: `Shift+Ins'

5.2.6 Show clipboard
--------------------

This option shows the clipboard window.  ("Clipboard" Section 15.2).

Name of the command: cmeShowClip.

5.2.7 Clear
-----------

Deletes the selected text. It isn't copied to the clipboard.

Name of the command: cmcClear.
Assigned key: `^Del'

5.2.8 Set Local
---------------

This option pops up the local configuration dialog. The values in this dialog
are valid only for the current editor window.

The first group of options are related to the editing modes.   ("Editing
Modes" Chapter 7).

The syntax highlight group of options allows to choose the type of highlight.
("Syntax Highlighting" Section 7.23).

Additionally the window allows to indicate the tab size and the column where
the wrap cuts the lines.

Name of the command: cmcSetLocalOptions.
Assigned key: `Alt+L'

5.2.9 Set Global
----------------

This option pops up the global configuration dialog. The values in this
dialog are used as default values. If you exit the dialog using the `Ok'
button these values aren't applied to any of the editor windows, they just
act as default values for newly opened and created files. If you exit the
dialog using the `To all' button these values are applied to all the editor
windows. To set the options for just one window  ("Set Local" Section 5.2.8).

The options are related to the editing modes.  ("Editing Modes" Chapter 7).

Additionally the window allows to indicate the tab size and the column where
the wrap cuts the lines.

Name of the command: cmcSetGlobalOptions.
Assigned key: `Alt+G'

5.2.10 Expand all tabs
----------------------

This option converts all the tabs to spaces.  ("Real Tabs" Section 7.3).

The tabs are expanded to the current tab size, check this value before using
this option.

Name of the command: cmcExpandAllTabs.

5.2.11 Compact text
-------------------

This option converts spaces to tabs where possible.  ("Real Tabs" Section
7.3).

Be careful, if the tab size is too small the editor will generate tons of
tabs, even in places where you may not want a tab.

Name of the command: cmcCompactBuffer.

5.2.12 Copy to Windows Clipboard
--------------------------------

Copies the selected text to Windows' clipboard. Of course you must be running
Windows to be able to use this ;-)). Don't expect this feature to work always
though, as Windows has bugs in the WinOldAp module; which is used to provide
this function.

There is a command called `cmcCutClipWin' you can use to cut a portion of
text copying it to Windows' clipboard.

Name of the command: cmcCopyClipWin.

Name of the command: cmcCutClipWin.

5.2.13 Paste from Windows Clipboard
-----------------------------------

Pastes the Windows' clipboard content at the cursor's position. Of course you
must be running Windows and have some text in the clipboard.  Don't expect
this feature to work always though, as Windows has bugs in the WinOldAp
module; which is used to provide this function.

Name of the command: cmcPasteClipWin.

5.2.14 Copy to file Clipboard
-----------------------------

Copies the selected text to a special file used as clipboard. This option is
usually available on UNIX as a replacement for the Windows clipboard
functions. The file is created in a user specific directory, for this reason
the information is only available to one user.

Name of the command: cmcCopyClipFile.

5.2.15 Paste from file Clipboard
--------------------------------

Pastes text from a special file used as clipboard. This option is usually
available on UNIX as a replacement for the Windows clipboard functions. The
file is created in a user specific directory, for this reason the information
is only available to one user.

Name of the command: cmcPasteClipFile.

5.2.16 Push cursor position
---------------------------

It stores the cursor position (x,y) in a stack. You can restore it later
using  ("Pop cursor position" Section 5.2.17). The stack can hold up to
eleven nested positions; if you try to push a 12th value the oldest is
discarded.

Name of the command: cmcPushCursorPos.

5.2.17 Pop cursor position
--------------------------

It restores the cursor position (x,y) from a stack. You can store the
position using  ("Push cursor position" Section 5.2.16). The stack can hold
up to eleven nested positions; if you try to push a 12th value the oldest is
discarded.

Name of the command: cmcPopCursorPos.

5.2.18 Case (Menu)
------------------

This submenu contains operations to convert blocks or single characters to
uppercase or lowercase and similar operations

5.2.18.1 Block to upper
.......................

Converts all the selected characters to uppercase.

Name of the command: cmcToUpper.

5.2.18.2 Block to lower
.......................

Converts all the selected characters to lowercase.

Name of the command: cmcToLower.

5.2.18.3 Character toggle
.........................

Converts the character under the cursor to lowercase if it was uppercase and
`vice versa'.

Name of the command: cmcToggleCharCase.

5.2.18.4 Block invert
.....................

Converts all the selected uppercase characters to lowercase and vice versa.

Name of the command: cmcInvertCase.

5.2.18.5 Block alternate
........................

It will convert the first character to uppercase, the second to lowercase and
so on. If you want the reverse you can use the command to invert a block.
("Block invert" Section 5.2.18.4).

Name of the command: cmcAltCase.

5.2.19 Insert new line (do not move)
------------------------------------

It inserts a new line character at cursor position but without modifying the
cursor position. Note it doesn't apply indentation or other special things to
the inserted line.

Name of the command: cmcInsertNewLine.
Assigned key: `Ctrl+N'

5.3 Search
==========

This menu contains the search and replace commands of the editor. The search
function also includes advanced tools specifically designed for programmers,
like the options to jump to a function definition or a symbol.

5.3.1 Find
----------

This command searches a string in the current editor. The dialog contains the
following fields:

   * Text to find: The text you want to find.

   * Case sensitive: Use this option if the case of the text to search is
     important.

   * Whole words only: When this option is set the editor searches for whole
     words. If not the text to search can be only part of a word.

   * Regular expressions: Use this option if the text to search is a regular
     expression. A description of the syntax used by regular expressions can
     be found in the libc.  ("Regular Expressions" Section 15.4). Three types
     of regular expression are supported.  ("Regular Expressions Options"
     Section 5.3.1.1).

   * Only inside comments: Matches will be returned only if the first
     character is inside a comment. Every type of comment supported by the
     syntax highlighting is supported. It doesn't have any sense for plain
     text.

   * Only outside comments: Matches will be returned only if the first
     character is outside a comment. If the previous option is also enabled
     this option takes precedence.

   * Scope - Global: The search will ignore the selected text.

   * Scope - Selected text: The search will be done only inside the selected
     text.

   * Origin - From cursor: The search will start from the position of the
     cursor.

   * Origin - Entire scope: The search will start from the beginning of the
     text.

The regular expressions search is slow and this can be noticed in large
files. Don't use it if the search can be done without it. The editor will
automatically disable the regular expressions search if the `Text to find'
contains only alphanumeric characters. This can be disabled  ("Regular
Expressions Options" Section 5.3.1.1).

To repeat the search use the `Search again' command.  ("Search again" Section
5.3.3).

Name of the command: cmcFind.
Assigned key: `^Q-F'

5.3.1.1 Regular Expressions Options
...................................

This dialog is accessed from the `RegEx Ops' button in the Find or Replace
dialogs. Here you can indicate some important options about the regular
expressions:

RegEx style:

   * Basic POSIX: old style POSIX regular expressions.

   * Extended POSIX: new style POSIX regular expressions.

   * Perl Compatible: Perl like regex, using the PCRE library.

Replace text:

   * Normal text: the match is replaced by the text in the replace text area.

   * Dollar tags: the text to replace is parsed searching for `$n' tags.
     These markers are replaced by the correspondent subexpression. If you
     know Perl that's just like using the $0, $1, `etc.' after a search.  If
     you never used Perl and don't know what a subexpression is, just play a
     little and you'll get the idea; use $n where n is the desired
     subexpression.

Optimize:

   * Try to use normal search: if the editor detects that the search string
     contains plain text will use a simple search even when regular
     expressions are enabled. That's about ten times faster.

   * Ever use RegEx: the editor doesn't try to be smart and always uses what
     you selected.

For more information about regular expressions  ("Regular Expressions"
Section 15.4).

5.3.2 Replace
-------------

This command searches portions of text and replaces it with another text.
The options for the search are the same as in the `Find' command.   ("Find"
Section 5.3.1). The additional replace options are:

   * New text: The text used to replace the matching text.

   * Prompt on replace: When this option is enabled the editor asks for
     confirmation before doing the replace.

   * Replace all: Use it to replace all the possible ocurrences.

To repeat the search use the `Search again' command.  ("Search again" Section
5.3.3).

Name of the command: cmcReplace.
Assigned key: `^Q-A'

5.3.3 Search again
------------------

This option repeats the last search.

Name of the command: cmcSearchAgain.
Assigned key: `^L'

5.3.4 Name current function
---------------------------

It shows the name of the function at the position of the cursor in the status
line of the window. It has the same limitations of the `Jump to function'
option.  ("Jump to function" Section 5.3.5).

Name of the command: cmcWhichFunctionIs.
Assigned key: `'

5.3.5 Jump to function
----------------------

his option opens a dialog with the list of functions in your source file you
can choose to jump to. The functions are searched in the source file so you
don't need to compile it. As the parser uses heuristics it may fail. For an
option that uses a more reliable but also slower and complicated search see
the `Jump to symbol' option.   ("Jump to symbol" Section 5.3.7).

The `Browse' button sends the list of functions to the message window so you
can browse the list and visit each function.  ("Message Window" Section 15.8).

Currently only a few languages are supported, they are:

   * C/C++

   * Clipper 5.x

   * Syntax Highlight Files

   * Texinfo sources (chapters, sections, `etc.')

   * Assembler files (labels)

   * PHP

In C sources fails are common if your code has an unbalanced number of curly
brackets. Here is a piece of code that made the heuristic get lost:

     #ifndef MSI_USE_GETDELIM
        if (readPipe(textMsgPipe, id, buf, maxLen))
        {
           *buflen = strlen(buf);
     #else
        if ((bytes = readPipe(textMsgPipe, id, buf)) > 0)
        {
           *buflen = bytes - 2;
     #endif
         ....
        }

Here the editor saw two open curly brackets but only one close bracket.  To
avoid this you should add:

         ....
     #ifndef MSI_USE_GETDELIM
        }
     #else
        }
     #endif

Currently the parser tries to be smart and only check half of the sequence,
just like if all `#if' conditions were true. It means the parser won't get
lost with this specific problem, but very complex situations may break it
nonetheless.

Name of the command: cmcJumpToFunction.
Assigned key: `Alt+F2'

5.3.6 Jump to prototype
-----------------------

This option pops up a dialog with the list of C function prototypes in your
source file. You can then choose one to jump to. The prototypes are searched
in the source file so you don't need to compile it. As the parser uses an
heuristic approach it may fail, especially if your code has an odd number of
curly brackets.

Name of the command: cmcJumpToPrototype.

5.3.7 Jump to symbol
--------------------

This option uses `TAGS' files to search for symbols. In order to make it work
you should read about `TAGS' files first. The editor will try to help you to
generate the `TAGS' file if you don't have one.   ("TAGS files" Chapter 13).

The word to the left of the cursor will be used to start a search in the list
of available symbols. You can make incremental searches in the list and
pressing <Enter> the editor will jump to the symbol.

More about tags?  ("TAGS files" Chapter 13).

Name of the command: cmeSearchTag.
Assigned key: `Ctrl+F2'

5.3.8 Go to line
----------------

Allows you to indicate to which line you want to jump to.

Name of the command: cmcGotoEditorLine.
Assigned key: `^J'

5.3.9 Class browser
-------------------

This option uses `TAGS' files to search for symbols. In order to make it work
you should read about `TAGS' files first. The editor will try to help you to
generate the `TAGS' file if you don't have one.   ("TAGS files" Chapter 13).^

First the editor will create an index of available classes from the `TAGS'
files and then will do a search using the word to the left of the cursor. If
an exact match was found the editor will go directly to browse this class,
otherwise you'll get a list of available classes to choose one.  You can make
incremental searches in this list typing the first letters of the class name.

The main browser dialog contains three buttons offering different information
about the selected class. They are:

   * This class: shows a list of the members for this class. You can select
     any of them to jump to the member code or go back to the main dialog.

   * This and parents: shows a list of the members for this class and all the
     parent classes. The members for this class are first, then come the
     members of the parents, then the members of the grandparents and so on.
     The information is more compact and you'll only see the name of the
     class where the member was found and the level (one for the parent, two
     for grandparents, `etc.').

   * Sorted: This option is the same as the previous but the list is
     alphabetically sorted.

In addition to these buttons you'll see a list of parent and child classes.
Selecting a class from those lists and pressing the corresponding button you
can browse the selected class.

More about tags?  ("TAGS files" Chapter 13).

Name of the command: cmeClassBrowser.

5.3.10 Word completion
----------------------

This option uses `TAGS' files to search for symbols. In order to make it work
you should read about `TAGS' files first. The editor will try to help you to
generate the `TAGS' file if you don't have one.   ("TAGS files" Chapter 13).

The word to the left of the cursor is used to perform a search in the list of
available symbols. If a partial match is found a list of available symbols
will pop-up. To select the symbol you want you can use the arrow cursors or
continue typing until you reach the desired symbol. When you type <Enter>,
<Space> or any symbol the select word will be inserted completing the word.
Pressing <Esc> or <Tab> will abort the process.

More about tags?  ("TAGS files" Chapter 13).

Name of the command: cmeWordCompletion.
Assigned key: `Alt+Right'

5.3.11 Jump to last cursor position
-----------------------------------

This command moves the cursor to the position it was before the last
movement. You can use it to go back after a search.

Name of the command: cmcJumpLastCursorPos.
Assigned key: `Ctrl+Q-P'

5.3.12 Jump to last undo position
---------------------------------

This command moves the cursor to the position it was before the last movement
sequence. This is equivalent to undo the last group of movements.

Name of the command: cmcLastPosCur.

5.4 Macro
=========

This submenu has the options to record and replay macros.

5.4.1 Record (Macro)
--------------------

Starts recording all the operations you do in the current editor window.

Name of the command: cmcRecordMacro.
Assigned key: `Shift+F10'

5.4.2 Stop (Macro)
------------------

Stops the macro recording.

Name of the command: cmcStopMacro.
Assigned key: `Alt+F10'

5.4.3 Play (Macro)
------------------

Replays the saved macro. A macro saved in one window can be used in another.

Name of the command: cmcPlayMacro.
Assigned key: `^F10'

5.4.4 Choose (Macro)
--------------------

This option lists all the available `sLisp macros' so you can choose one to
execute.

To learn more about these macros:  ("sLisp macros" Chapter 10).

Name of the command: cmcChooseMacro.

5.4.5 Repeat (Macro)
--------------------

It re-runs the last selected macro.  ("Choose (Macro)" Section 5.4.4). That's
useful if the macro isn't assigned to a key and you need to use it several
times.

To learn more about these macros:  ("sLisp macros" Chapter 10).

Name of the command: cmcRepeatMacro.
Assigned key: `Shift+F3'

5.4.6 Generate Code
-------------------

This option translates the recorded macro into an `sLisp macro'. The
generated code is inserted at the cursor's position.

To learn how to record a macro:  ("Record (Macro)" Section 5.4.1).

To learn more about these macros:  ("sLisp macros" Chapter 10).

Name of the command: cmcGenCodeForMacro.

5.4.7 Run selected code
-----------------------

It interpretes the selected text as an sLisp macro.

Name of the command: cmcRunSel_sLisp.

5.4.8 Enter code to run
-----------------------

It asks for a text and interpretes it as an sLisp macro. The maximum length is
one kilobyte.

Name of the command: cmcRunEnter_sLisp.

5.4.9 Pseudo Macro (menu)
-------------------------

It shows a list of all the pseudo macros available for the current syntax
highlighting mode. The purpose of this option is just to show you what's
available; of course you can choose any from the list but that's much slower
than using them directly.

The letters enclosed in brackets are the triggers for the pseudo macro. To
learn more about pseudo macros:  ("Pseudo Macros" Chapter 9).

Name of the command: cmcChoosePMacrosList.

5.5 Rectangle
=============

This menu contains the rectangular block operations. As they are a bit hard
to code and these functions are normally ignored by the users, I put them in
a very visible menu.

5.5.1 Start (Rectangle)
-----------------------

Selects the top-left corner of the rectangular area.

To learn more about rectangular blocks:  ("Rectangular Blocks" Section 3.4.4).

Name of the command: cmcSelRectStart.
Assigned key: `^K-Shift+B'

5.5.2 End (Rectangle)
---------------------

Selects the bottom-right corner of the rectangular area.

To learn more about rectangular blocks:  ("Rectangular Blocks" Section 3.4.4).

Name of the command: cmcSelRectEnd.
Assigned key: `^K-Shift+K'

5.5.3 Hide (Rectangle)
----------------------

Hides the rectangular selection.

To learn more about rectangular blocks:  ("Rectangular Blocks" Section 3.4.4).

Name of the command: cmcSelRectHide.
Assigned key: `^K-Shift+H'

5.5.4 Copy (Rectangle)
----------------------

Copies the rectangular selection into the clipboard. This clipboard isn't the
same clipboard used by the normal selections and is overwritten each time you
copy to it.

To learn more about rectangular blocks:  ("Rectangular Blocks" Section 3.4.4).

Name of the command: cmcSelRectCopy.
Assigned key: `^K-Shift+C'

5.5.5 Paste (Rectangle)
-----------------------

Inserts the contents of the rectangular clipboard at the cursor position.

To learn more about rectangular blocks:  ("Rectangular Blocks" Section 3.4.4).

Name of the command: cmcSelRectPaste.
Assigned key: `^K-Shift+P'

5.5.6 Cut (Rectangle)
---------------------

Copies the rectangular selection into the clipboard and then deletes the
selected text.

To learn more about rectangular blocks:  ("Rectangular Blocks" Section 3.4.4).

Name of the command: cmcSelRectCut.
Assigned key: `^K-Shift+T'

5.5.7 Clear (Rectangle)
-----------------------

Deletes the selected text.

To learn more about rectangular blocks:  ("Rectangular Blocks" Section 3.4.4).

Name of the command: cmcSelRectDel.
Assigned key: `^K-Shift+L'

5.5.8 Move (Rectangle)
----------------------

Moves the selected text to the cursor's position.

To learn more about rectangular blocks:  ("Rectangular Blocks" Section 3.4.4).

Name of the command: cmcSelRectMove.
Assigned key: `^K-Shift+M'

5.5.9 To upper (Rectangle)
--------------------------

Converts all the characters inside the rectangle to uppercase. This operation
basically does a cut of the block, then processes all the characters and
finally makes a paste of the modified block. As a side effect of this
operation tabs inside or crossing the rectangle boundaries are converted to
spaces.

To learn more about rectangular blocks:  ("Rectangular Blocks" Section 3.4.4).

Name of the command: cmcSelRectToUpper.

5.5.10 To lower (Rectangle)
---------------------------

Converts all the characters inside the rectangle to lowercase. This operation
basically does a cut of the block, then processes all the characters and
finally makes a paste of the modified block. As a side effect of this
operation tabs inside or crossing the rectangle boundaries are converted to
spaces.

To learn more about rectangular blocks:  ("Rectangular Blocks" Section 3.4.4).

Name of the command: cmcSelRectToLower.

5.6 Windows
===========

This menu contains the options to handle the windows in the editor.

5.6.1 Size/move
---------------

You can resize the windows by dragging the bottom right corner of the window.
You can move the windows by dragging the title line of the window.

Additionally this command allows the same operations to be done without the
mouse. Once you entered in this mode the border of the window changes it's
color and you can move the window using the arrow keys. To resize the window
use the arrow keys holding the <Shift>. Furthermore <Home>, <End>, <PgUp> and
<PgDown> can be used to move the window to one end of the desktop. When
you're done you can end this mode by pressing <ESC> or <ENTER>.

Name of the command: cmeResize.
Assigned key: `^F5'

5.6.2 Zoom
----------

Changes the size of the window to be as large as the whole desktop. The
second time you use this command the window is resized to it's previous size.
It's the equivalent of the maximize and restore options of other editors.

Name of the command: cmeZoom.
Assigned key: `F5'

5.6.3 Tile
----------

Arranges all the windows in a way that you can see all at the same time.

Name of the command: cmeTile.

5.6.4 Cascade
-------------

Arranges all the windows in such a way that they overlap.

Name of the command: cmeCascade.

5.6.5 Next (Window)
-------------------

Selects the next window. The windows are linked in a circular list, this
command selects the next window in the list. To change the order of the
windows you can directly select a window with the mouse or <ALT> plus a
number. That window will become the current one, and the other will be the
`previous' window.

Use it to select the most recently used windows.

Name of the command: cmeNext.
Assigned key: `F6'

5.6.6 Previous (Window)
-----------------------

Selects the previous window. The windows are linked in a circular list, this
command selects the previous window in the list. To change the order of the
windows you can directly select a window with the mouse or <ALT> plus a
number. That window will become the current one, and the other will be the
`previous' window.

Use it to select the most recently used windows.

Name of the command: cmePrev.
Assigned key: `Shift+F6'

5.6.7 Close
-----------

Closes the current window. If the content wasn't saved the editor will ask if
you want to.

Name of the command: cmeClose.
Assigned key: `Alt+F3'

5.6.8 List
----------

Pops up the List of Windows dialog. This dialog contains the list of all the
editor windows, the special windows and the closed windows.

The editor windows are numbered starting from two and the list is sorted by
number. If a window isn't saved an asterisk is placed between the number and
the name.

The special windows are: the message window (the number of lines is
indicated), the project window, the clipboard window (the bytes used by it is
indicated) and the InfView windows.

The closed windows list is sorted alphabetically and holds the last closed
windows. The editor stores important information about these windows so if
you close and re-open one of them the size of the window and other values are
restored.

You can jump to any of the windows by selecting it with the mouse, with the
arrow keys and <ENTER>, double clicking, or by using the `Go' button.

A press of <Delete> or using the `Delete' button will delete the closed
windows or close any open editor window.

Using the `ReEnumerate' button you can force to compute the window numbers so
all become continuous.

Name of the command: cmeListWin.
Assigned key: `Alt+0'

5.6.9 User Screen
-----------------

Shows the DOS screen. Press any key to go back to the editor.

Name of the command: cmeUserScreen.
Assigned key: `Alt+F5'

5.7 Tool&Ops
============

This menu contains all the configuration submenues and some useful tools.

5.7.1 Options
-------------

This submenu contains the configuration submenues. You can find the local and
global edition options in the Edit menu.  ("Edit" Section 5.2).

Some of the deeper submenues are explained here because the documentation
tools limit how deep subsections can be nested.

5.7.1.1 Customize Colors
........................

This command allows to customize the colors used by the editor. Almost all
the colors are configurable.

The first list, called `Group', is the list of things used by the editor.
Each entry in the group list has one or more colors in the `Item' list.
First select the group you want to customize and then press <Tab> to move the
cursor to the items list. To customize a color just select it in the list of
items; the dialog will show a text example in the bottom right corner and the
`Foreground' and `Background' colors will be indicated. Using the mouse or
moving with <Tab> and using the arrow keys you can select any of the
available colors; the sample text area will show the resulting combination.

Exiting the dialog with <ENTER> or with the `Ok' button the new colors will
be applied. The colors are stored in the desktop file. If you are using one
centralized desktop file these colors will be used each time you run the
editor, if not the colors will be used only when you run the editor in that
very same directory.

To learn more about the scope of the desktop files and how to indicate
default values:  ("Desktop Files" Section 15.5).

If you want to use other colors not listed in this dialog you must customize
the palette.  ("Color Palette" Section 5.7.1.2).

Name of the command: cmeSetColors.

5.7.1.2 Color Palette
.....................

This option allows to configure the palette of colors used by the editor. I
think you know about palettes but I wrote a little explanation, just in case.
("Text mode attributes" Section 15.6).

The `Color' radio buttons are used to select the index you want to customize.
The `Red', `Green' and `Blue' scroll bars can be used to customize the color.
To modify one of the components use the mouse or use the <R>, <G> and <B> to
increase the values and <Shift>+<R>, <Shift>+<G> and <Shift>+<B> to decrease
the values.

Exiting the dialog with <ESC> restores the values you had before selecting
this option. Choosing the `Default' button the colors are set to the default
values of VGA cards.

These settings are stored in the desktop file.  ("Desktop Files" Section
15.5).

Name of the command: cmeEditPalette.

5.7.1.3 Color Theme
...................

With this option you can choose from a list of predefined groups of colors.

Currently only a couple of color themes are available, users are encouraged to
contribute their own themes.

Name of the command: cmeColorTheme.

5.7.1.4 Editor General
......................

This will open the general configuration dialog containing several options.

The `Save options' groups control various settings about which and how files
are created by the editor.

   * Make backups: When enabled the editor keeps a backup of your files using
     the `.bkp' extension. It saved me many times.

   * UNIX style backups: When enabled the backups are created appending a
     tilde symbol to the file extension. This is recommended only for systems
     that support long file names, not pure DOS.

   * Hidden backups: The backups are created as hidden files, it can be used
     to keep backups, but at the same time make them less annoying.

   * Remember bkps to delete: When you exit with `Quit' the editor deletes
     backup files. To do it the editor deletes files ending with `bkp', but
     if you use UNIX style backups or just edit files in a directory other
     than the current it's hard to know where these files are located. For
     this reason the editor keeps a list of created backup files.  If you
     exit normally this list is lost, so the next time you use the editor and
     exit with `Quit' only the backups created during this session will be
     deleted. This behavior can be modified choosing this option, when enabled
     the editor will store the list in the desktop file and retrieves it the
     next time you start the editor. By default this option is disabled
     because users that don't know about this mechanism and use a centralized
     desktop file could end up with a very long list of backup files, wasting
     memory and disk space.

   * Don't create desktop files: When enabled the editor creates only one
     desktop file and not one per directory.  ("Desktop Files" Section 15.5).

   * Save desktop files hidden: Just what the name says. That's useful if you
     want to create desktop files in each directory but don't want to see
     them in directory listings.

   * Tile windows vertically first: It affects the `Windows|Tile' option.
     Normally this option starts splitting the screen by dividing the height
     of windows. When this option is selected the width is divided first.

   * Save UNIX files as UNIX: When enabled the editor saves to disk UNIX
     files in UNIX format. That means that the conversion is done only
     internally and the format of the file on disk isn't altered when you
     save.

   * Do not remember cursor position: If you enable it the editor won't
     remember the last cursor position of each editor window.

   * Do not warn about read-only files: It disables the dialog that warns
     about opening read-only files. Disabling it you won't be asked about
     reverting the read-only attribute until you try to save the file.

   * Open read-only files as R.O. buffers: When enabled the editor won't
     allow you to edit files with the read-only attribute. Note: it doesn't
     disable the warning, use the above mentioned option for this.

In the open, save, `etc.' dialogs you can sort backup files in a special way
so they don't interfere with the rest of the files.   ("File open dialog"
Section 5.7.1.20).
You can also configure the editor to avoid creating backup files for some
particular filenames or directories.  ("Do not create backups for" Section
5.7.1.21).

The `Clock' group allows to turn on/off the clock and to choose 24 hs or
AM/PM style. The clock is displayed in the top right corner of the screen.

The `Max. editor copies' controls how many copies of the same file can be
opened at the same time. By default the value is one, so when you try to open
a file twice the editor will just show you the first copy. Specifying a value
different than one will allow you to open more copies of the same file. Only
the first copy can be modified and the rest are read-only snap-shots of the
file. Note it only affects the number of read-only copies automatically
created by the editor, you can open as many as you want using the `File |
Open Read-only copy' menu option. For more information:  ("Open Read-only
copy" Section 5.1.3).

The `Max. closed to remember' value specifies how many closed files are
remembered in the list of windows. This value can't be less than three and
can't be greater than two hundred. If you reduce this value and there are
already more files remembered the editor won't reduce the number immediately,
instead you must choose which files to remove by hand.

The `+ Desktop' button opens a second dialog containing options about what is
stored in the desktop files. Each section selects if the option will be
remembered always (ever), only when no files are specified in the command
line or never.

   * Remember editor windows: it affects the opened files.

   * Remember other windows: it affects other windows, like the help windows.

   * Remember closed windows: it affects the list of closed files.

The second dialog contains a button to return to the first dialog.

The `+ Others' button opens a third dialog containing options to configure
the behavior of the message window when the last or first error in the list
is reached.

   * Just stop: nothing particular is done, the editor just stops.

   * Indicate with a message: a message indicating you reached the first or
     last pops up.

   * Wrap (circular list): the list becomes a circular list. So when you
     reach the last message the editor jumps to the first and `vice versa'.

   * Make a beep: a sound is generated.

Additionally you can choose if you want project and message windows to be
vertically or horizontally oriented in this dialog.

   * Use the vertical direction: message and project windows are vertically
     oriented. By default they are positioned at the left side. When this
     option is selected the project window has only one column.

   * Use the right side: use the right side of the screen instead of the left
     side. That's used only when the direction is vertical.

One important detail is that this options affects the size of newly opened
files. If you already opened windows in this project the editor will have
them memorized. So you should set up this options before creating a project.

To control the size of the windows when you open a new file you have two
options.

   * Use reserved width or 7 (hz dir): In this case the editor will make
     windows smaller than the desktop by substracting the value entered in
     reserved width, for vertical direction projects, or a fixed ammount of
     seven for horizontal direction projects.

   * Avoid message and project windows: When selected the editor will try to
     avoid using the desktop surface used by the message and project windows.
     The size computed is the size of these windows without zooming. If the
     resulting size is too small the window will overlap.

This dialog also has a button to go back to the main dialog.

These settings are stored in the desktop file.  ("Desktop Files" Section
15.5).

Name of the command: cmeEdGralOptions.

5.7.1.5 Check for modified files
................................

When you edit a file the editor uses a copy of this file stored on memory.
The copy on disk can be changed by another user or by a program. If it
happends you can be working on a different file and you'll destroy those
changes when saving. For this reason the editor must check if the
modification time of the file changed. The editor does this check when:

  1. You run an external program. This program could generate some output
     file and it could be one of the opened files.

  2. You select an editor window. This is done in the `idle loop', so this
     check could be delayed a fraction of a second.

  3. Periodically for the focused file.

The configuration dialog can be used to disable or fine tune these checks.
The `Seconds between checks' option refers to the periodical checks.  The
other options are:

   * Don't check after executing an external program. It disables the checks
     done after running an external program.

   * Don't check while idle. It disables the the checks done periodically and
     also the checks done when you select a an editor window. That's because
     both are done when the editor have some free time.

When the editor detects a file on disk is newer than the file on memory a
dialog asking if you want to load the new file pops up. This question will be
done only once and no matters what you answer the editor will asume the
problem is solved. This is to avoid an endless storm of questions.

If you choose to load the file from disk the editor will check if the memory
copy is also modified. In this case a collision between your changes and the
external changes can lose data. For this reason the editor offers four
options:

  1. Load file from disk (discard changes): This is the default action, use
     it if you want to get a fresh copy discarding any changes done in memory.

  2. Abort operation: The editor does nothing and you must solve the problem
     manually.

  3. Load and show differences: This is like the first option, but the editor
     will invoke `diff' and load it's output. You can analize the differences
     between your changes and the external changes to solve the problem.

  4. Don't load and show differences: This is like the second option, but the
     editor invokes `diff' like in the third.

The last two options are offered only if GNU diff is installed in your
system. You can get GNU diff from the GNU site (http://www.gnu.org/).

Note that when the editor offers a diff output this is loaded from a temporal
file that is discarded as soon as loaded. For this reason the editor marks
this window as read only. If you want to use this file to as input for
`patch' just use the `Save As' option.

Name of the command: cmeSetModiCkOps.

5.7.1.6 Screen Saver
....................

This dialog customizes the screen saver. Note the question mark at the end of
the term "screen saver" I did it because the plasma screen saver isn't too
good to be used as a real saver for your screen.  You can enable or disable,
choose the time the editor will wait before starting the screen saver and the
screen saver style. The Test button can be used to see how the screen saver
looks like.

If you leave the mouse pointer in the upper right corner of the screen for a
few seconds the screen saver is activated. This time is three seconds by
default and can be configured entering the amount of seconds in the second
box labeled `Time'.

Two types of screen savers are supported: internal and external. Internal
screen savers are hardcoded in the editor. External screen savers are
external programs started by the editor. If you select an external screen
saver from the list the Info and Help buttons are enabled. Pressing these
buttons you can get more information about the screen saver. You can pass
additional parameters to the external screen saver filling the `External
Saver' box.

If you want to write your own screen saver please download the sources of the
editor and read the explanations found in the `scrnsave' directory. An
external screen saver is basically a simple program that supports some
special command line options and returns some specified return values.

Name of the command: cmeScreenSaverOpts.

5.7.1.7 SDG Options
...................

These options customize the SDG module (SET's Documentation Generator).  See
Section 'Top' in documentation for 'SDG'.

   * Format file: Indicates the name of the format file used to generate the
     documentation.

   * Intermediate file: The name of the temporal file used in the process.

   * Base output: The base name of the output file. Doesn't include the
     extension.

   * Directory of formats: The place where the editor will search for the
     format file.

   * Keep intermediate: When enabled the temporal file isn't deleted, so you
     can see possible errors in it.

The SDG module uses the files listed in the project to collect the
documentation from the comments.  ("Project" Section 5.8).

These settings are stored in the desktop file.  ("Desktop Files" Section
15.5).

Name of the command: cmeSDGDialog.

5.7.1.8 Run program (which one)
...............................

The editor can run an external program and collect the errors reported by it
if you press the shortcut (`^F9'). A good example is the make program. Here
you can select the name of the program. If you need to run more than one
command separate it with `;'.

The editor will redirect the stderr (standard error output) and stdout
(standard output) of the program and then will analize it looking for errors.
The dialog includes a list of parsing algorithms to analize the errors from
the external program. To learn how to configure the editor for other formats
or just fine tune any of them  ("Error messages from an external application"
Section 15.9).

The other options found in this dialog are a little bit complex. Here I'll
try to explain each of these options but I recommend to just try them to see
how they work.

Option `Use OS screen to run the program': The editor will try to restore the
contents of the screen. So it will look like as it was before running the
editor. Meaning the program will be executed and the editor will redraw all
the contents of the desktop and windows afterwards. This mechanism is useful
when the program you want to run is interactive or doesn't use the standard
output. Is important to understand that on some platforms and terminals the
editor can't restore the contents of the screen and will just clean it to the
grey over black color. It's also important to understand that when this
option is enabled the program can't be executed in multitasking mode; it
means the option `Don't try to run in background' will be implicitly selected.

Option `Don't try to run in background': On some platforms (currently only
Linux) the editor can execute the external program as a child process. It
means the program will execute in parallel with the editor (in background).
When the editor does this you'll see the message window indicate the program
is running but won't say when you are back in the editor. Instead the output
of the external program will start to fill the message window. You can select
any other window and continue working while the external program runs.  When
the external program ends the editor will also collect the rest of the
messages and errors in the background. While the editor is running the
external program and/or parsing the remaining messages the status bar will
show an option `Ctrl+C Stop'; clicking on it or pressing the indicated key the
editor will stop the background process and will also stop collecting
messages. The purpose of this option: When this option is enabled the editor
won't try to run the external program in parallel even if the platform
supports it. That's faster, but if the external program is slow you'll be
forced to wait until it finishes and you won't be able to stop the program
from the editor.

Option `Always parse in background': If enabled the editor will collect the
messages and errors in the background even if the platform doesn't support the
execution of the external program in the background. On platforms that don't
support the execution of the external program as a thread the editor will
block until the external program ends and then will parse the messages and
errors in the background. This is useful when the amount of messages and
errors is big and the parsing will take a long time. In this way you can
continue working while the editor does this job. You must understand that's
even slower but you can at least use the time for reading or editing text.

Option `Jump to the first error': If enabled the editor jumps to the file and
line of the first error reported by the external program.

Option `Don't redirect stdout': Complements the `Use OS screen to run the
program' option. When enabled the editor collects errors only from the
standard error and not from the standard output. This is useful for
interactive applications that sends its output to the standard output.

The `Message window scroll' group of options are mutually exclusive and give
some control over the behavior of the message window. When you start
executing the external program the message window will automatically get the
focus. As messages and errors are added to this window the window scrolls and
shows the last message. When the external program finishes and the editor
parses all the messages and errors the message window will get the focus
again. Finally, if the editor found errors, the message window will scroll to
the first line. This is the default behavior and you'll be able to see each
of this steps only if the editor is running the external program in the
background and/or parsing the messages and errors in the background. This
behavior corresponds to the `Ever' (always) option. If you select the `Never'
option the editor won't scroll the message window. In this case you can
browse the messages even while the editor is collecting them. Finally you can
choose `Only if not focused'. In this case you'll be able to browse the
messages when the message window is selected, but if you select an other
window the editor will start to scroll the message window. The fastest option
is `Never' but then you won't see if the external program finished executing,
unless you have the message window selected all the time.

The `Lines per pass' option is associated with the speed of parsing. This
option takes effect only when the messages and errors are parsed in the
background. What this option indicates is how many lines of messages and
errors will be parsed before releasing the CPU. A bigger value will make the
parsing faster but will make the editor slower and you'll start having
problems to select windows and write text. You must experiment with this
parameter. On my machine a value of 20 is acceptable.

All of these settings are stored in the desktop file.  ("Desktop Files"
Section 15.5).

For more information about the behavior of the message window  ("Message
Window" Section 15.8).

Name of the command: cmeConfRunCommand.

5.7.1.9 Keyboard
................

This is a submenu but due to limitations in the documentation tools I was
forced to put it with the rest of the options listed in the `Options' submenu.

It contains all the options to customize the keyboard.

5.7.1.10 Key assignment
.......................

With this command you can fully customize the keys used by the editor windows.
It doesn't include the menues, for that you must edit the `menubind.smn' file.

To learn how to use this command consult:  ("How to configure the keyboard"
Section 4.1).

Name of the command: cmeEditKeyBind.

5.7.1.11 Setup Alt keys
.......................

It allows you to select how the editor interpretes the left and right alt
keys. For more information:  ("Alt key configuration" Section 4.2).

Name of the command: cmeSetUpAltKeys.

5.7.1.12 Key pad behavior
.........................

[DOS]

Here you can choose how the keypad is interpreted by the editor. Two options
are provided. One is the BIOS default, in this mode the <NumLock> changes
between arrows and numbers. In the other mode the behavior is similar, but
holding shift and pressing a number will behave like an arrow key shifted,
that's very common in DOS applications so that's the default.

Name of the command: cmeKeyPadBehavior.

5.7.1.13 Back to defaults
.........................

This option restores the default key assignments of the editor. Use it if you
messed up the keyboard configuration and you want to get back the original
values.

Name of the command: cmeKbBackDefault.

5.7.1.14 Consult scan codes
...........................

Used to consult the keyboard scan codes:  ("Consulting scan codes" Section
4.4).  This dialog must be closed using the mouse to press the `OK' button
because it displays the scan codes for the keys, even for <ESC> and <ENTER>.

Name of the command: cmeSeeScanCodes.

5.7.1.15 Screen Options
.......................

This command opens the screen options configuration dialog. Here you can
customize the video mode or window size. This dialog is available only when
you are running in a terminal that can control the screen size. If you are
looking for the screen saver:  ("Screen Saver" Section 5.7.1.6).

Not all hardware drivers support it and not all the options are usable for
all the drivers. With some you can't control the screen size, like in Linux,
in other cases you have total control, like when you run the editor in X11,
and in others you have limited control, like in DOS.

The first group of options defines how the editor will try to set the video
mode. The available options are:

   * Don't force: The editor won't try to change the screen size.

   * Same as last run: The editor will remember the last size used and will
     try to restore the closest possible size.

   * External program: It currently makes only sense in DOS. It can be used
     when you have an external program that can set the video mode for you.
     In this case you must also indicate the command line for it in the
     External program box.

   * Closest to specified size: The editor will try to select a screen size
     that's as close to the values indicated in Width, Height, Chars width
     and Chars height boxes as possible. Note that some drivers support only
     a small set of fixed values and others have silly limitations imposed by
     the operating system.

   * Specified mode number: This is mainly for DOS. Use it if you know about
     a video mode supported by your video card that isn't known by the
     editor. In this case you must indicate the video mode number in the Mode
     number box.

Even when the video mode mechanism is for DOS you can use one of the video
modes known by the editor even when running on X11 or other terminals where
the screen size can be changed at will. The following is a list of known
video modes:

   * 0x0003
     - Width x Height: 80 x 25
     - Char Cell: 9 x 16

   * 0x0103
     - Width x Height: 80 x 28
     - Char Cell: 9 x 14

   * 0x0703
     - Width x Height: 80 x 30
     - Char Cell: 9 x 16

   * 0x0803
     - Width x Height: 80 x 34
     - Char Cell: 9 x 14

   * 0x0203
     - Width x Height: 80 x 35
     - Char Cell: 9 x 10

   * 0x0303
     - Width x Height: 80 x 40
     - Char Cell: 9 x 10

   * 0x0403
     - Width x Height: 80 x 43
     - Char Cell: 9 x  8

   * 0x0503
     - Width x Height: 80 x 50
     - Char Cell: 9 x  8

   * 0x0108
     - Width x Height: 80 x 60
     - Char Cell: 9 x  8

   * 0x0D03
     - Width x Height: 82 x 25
     - Char Cell: 8 x 16

   * 0x0903
     - Width x Height: 90 x 30
     - Char Cell: 9 x 16

   * 0x0A03
     - Width x Height: 90 x 34
     - Char Cell: 9 x 14

   * 0x0B03
     - Width x Height: 94 x 30
     - Char Cell: 9 x 16

   * 0x0C03
     - Width x Height: 94 x 34
     - Char Cell: 9 x 14

   * 0x0109
     - Width x Height: 132 x 25
     - Char Cell: 9 x 14

   * 0x010A
     - Width x Height: 132 x 43
     - Char Cell: 9 x 11

   * 0x010B
     - Width x Height: 132 x 50
     - Char Cell: 9 x 10

   * 0x010C
     - Width x Height: 132 x 60
     - Char Cell: 9 x  8



Name of the command: cmeSetScreenOps.

5.7.1.16 Encodings
..................

This dialog is used to select the encodings used by the editor. This is a
complex topic and you'll have to play a little bit with these options before
you can get all the funtionality. Usually you won't need to use it unless you
deal with more than one encoding.

Currently the editor can only use files where one letter corresponds to just
one character. It limits the editor to 256 different symbols. This is the
common case for most operating systems and environments. Which symbols are
represented by these 256 values is called an encoding.

The encodings are also known as code pages. An example of a code page is the
DOS code page 437 used by PC VGA boards or the ISO 8859-1 encoding used by
most POSIX systems as a default encoding. Currently more than 47 encodings
are supported. These encodings cover the latin, cyrillic and greek alphabets.
If your alphabet or favorite encoding isn't supported please consider
contributing.

In most cases the editor will detect the encoding currently used by the
operating system without problems. A special case is the GNU/Linux OS where
the concept of code page doesn't really exist and you can find all kinds of
errors in the fonts and unicode maps.

This dialog can have two, three or four encodings to select, they are:

   * Application: which encoding the application uses, ie. used in your
     documents.

   * Input: how the data from the keyboard is encoded. It is usually the same
     encoding used by the screen. If the application and input code pages
     don't match the editor will translate the data from the keyboard to the
     application encoding.

   * Screen: the encoding if the screen. Not all hardware drivers support it
     because some of them work with a fixed code page and this value is known.
     Here you indicate how the font used by your hardware or OS is encoded. If
     this value doesn't match the application code page the editor will
     translate the data before sending it to the screen.

   * Second font: how the secondary font is encoded. This is available only
     when the driver supports more than one font at the same time.

If any of these settings doesn't have the Force encoding option selected then
the editor will use what was detected.

If you force the screen encoding the selected value must match the OS
encoding. The exception is when you select a font in the Fonts dialog. In
this case the editor will recode the selected font to match the encoding.
Don't forget this interaction with the Fonts dialog.  ("Fonts" Section
5.7.1.17).

Note that you can easily edit documents encoded in a code page different than
the one used by the OS without doing any specific recoding. Example: if you
are using X11 and the ISO 8859-1 encoding but you want to edit a DOS file
encoded using the 850 code page you just need to select the application
encoding as PC 850 and force it. Your keyboard will generate ISO 8859-1
values, they will be translated into 850 encoding and put in your document.
At the same time the text encoded using 850 code page will be translated to
ISO 8859-1 before displaying it.

Name of the command: cmeEncodings.

5.7.1.17 Fonts
..............

This dialog is used to select the fonts used by the editor. Not all hardware
drivers support custom fonts and some only support only one font. The editor
can handle up to two fonts at the same time. However, the use of two fonts is
complex.

Only the fonts marked with Load font are used, don't forget to enable it
before leaving this dialog.

The size of the fonts must be the same, that's why even on systems that
supports two fonts only one size can be selected.

The offered fonts are the ones provided by the editor and not the OS. This is
because some systems doesn't provide fonts and the systems that provide fonts
usually provides only a few fixed width fonts. The font format used is
documented in the Turbo Vision library.

The fonts marked with `limited' don't cover all the encodings supported by
the editor.  ("Encodings" Section 5.7.1.16).

When you load a font the encoding for this font will be the value indicated
by the screen encoding option selected in the Encodings dialog. If you change
this encoding the editor will automatically change the encoding of the font.

Name of the command: cmeFonts.

5.7.1.18 User Words
...................

This option is used to define reserved words defined by the user. A very
common use is to define `typedef's you normally use in your programs.  The
user words are language dependant. They are highlighted with a special color
different than the color used for reserved words.

The first dialog is used to select the language, the names are the ones
defined in the syntax highlighting file ( "Syntax Highlighting" Section
7.23). Selecting one of the names and pressing <ENTER> pops up the second
dialog.

The next dialog is used to add or remove words to the list. Pressing <ENTER>
you confirm the changes and they are saved to disk. If you exit with <ESC>
the old list is preserved.

The user words are stored in a file called `userword.txt', in the same
directory all other configuration files are stored. Such a file isn't
included in the distribution because those words must be defined by the user.
You can edit the file by hand if you want, the format is very simple. The
start of a list is marked with `.' followed by the name of the language. The
items of the list are marked with `+'. Any line starting with a different
character will be ignored during the parsing.

Note: The menu option is "Tool&Ops|Options|User Words".

Name of the command: cmeEditUserWords.

5.7.1.19 Default global edition
...............................

To understand how this menu option works you must know some details about the
global options of the editor.  ("Set Global" Section 5.2.9).

The global options are good, but sometimes you want to make some small
differences depending on the kind of file you are editing. For example: I
want the `Intelligent C indent' mode enabled for C, but not for most of the
files, I also want to wrap lines for Texinfo files and files without syntax
highlighting. That's impossible to achieve just using the global options,
that's when this option comes into play.

The mechanism is like this: each time you open a new file the editor will
copy the default global options to it and select the syntax highlighting
according to various things, mainly the extension. Once the editor selects
the syntax highlighting the next step is to transfer some options that apply
only to the selected syntax highlighting. By default the list of options to
apply is empty but you can add some using this menu option.

The list of options indicates which settings will be enabled, or disabled, in
addition to the global options. That's something very important to keep in
mind, the list does not replace the global options, it modifies them. So the
list will say things like "also enable the intelligent C indent", "disable
the wrap lines", `etc.'

The first dialog shows the list of syntax highlighting defined in the
`syntaxhl.shl' and has three buttons. The `Edit' button is the default
button, so pressing <ENTER> you'll edit the settings for the selected syntax
highlighting. The `No SHL' button is used to edit the options that will be
applied for files with no syntax highlighting associated to.

Once you selected a syntax highlighting a dialog with the list of the settings
for it will appear. The first time it will be empty because those are defined
by the user, and therefore no defaults are provided. In this list the
settings that will be enabled in addition to the global options are marked
with a `+' before the name of the option. The options that will be disabled
are marked with a `-'. The options are the same described in the editing
modes section.  ("Editing Modes" Chapter 7).

The dialog contains an `Add' and a `Delete' button like other dialogs. You
can also use the <Insert> and <Delete> keys. When adding a new setting to the
list a dialog containing the list of available settings is displayed. Note
that this list contains all the settings that are available, once you add one
of them to the previous dialog it is removed from this list.

After selecting one setting to add a new dialog will be displayed. This
dialog will ask information related to this setting. Most settings are flags
that can be `added' or `removed' from the global options, but some of them
are just values that will overwrite the global options. Examples of the last
type are the tab size and the wrap column; in this case the dialog will ask
for the value.

The values are stored in a very simple format in the `deflopts.txt' file. You
can edit this file by hand, but in this case you'll need to know the names of
the settings. The format is very simple, a line starting with `.' starts a
section, the full stop is followed by the name of the highlighting affected.
Flags that will be added (enabled or ored) are marked with a `+' as in the
dialog, flags that will be substracted (disabled or anded) are marked with a
`-' again like in the dialog, in fact the dialog shows the same string as
those in the file. If the setting is numeric it will ever start with `+' and
after the name a `=' and the asigned value will follow. Just play a little
bit and see the resulting file.

Name of the command: cmeEditDeflOpts.

5.7.1.20 File open dialog
.........................

This command is used to configure some details of the file open dialog.
("File Open" Section 15.7).

Name of the command: cmeFileOpenOptions.

5.7.1.21 Do not create backups for
..................................

Sometimes you don't want to create backups for some particular files. In my
case I use a tool called `cvs'. It generates some temporal files and calls
the editor so I can write some information. Those files are short and
temporal, creating backups for them doesn't make any sense.

This command shows a configuration dialog where you can enter a list of
regular expressions. If any of these regular expressions match with the file
name of the file you are about to save then the editor won't create a backup
file for it. The regular expressions are Perl style because I think they are
much more intuitive than POSIX regex.

The list shown in this dialog is stored in a file called `nobkp.txt'.  The
exact place of the file depends on your system like other files created by
the editor.

Note that files listed here aren't remembered in the list of closed editors.
That's because the editor assumes those files are temporal and you won't need
to use them again.

Name of the command: cmeEditNoBkp.

5.7.1.22 Search files under cursor in
.....................................

When you press `Ctrl+Enter' the editor tries to load the file that's in the
text at the cursor position. If the cursor is in an include line the editor
will extract the name of the header. In any other case the editor will try to
find where the name starts and ends.

If the file isn't located in the current directory the editor will try to
find the file in the list of directories indicated by this option.

You can use environment variables for this list. Here is an example that
shows how to use the content of the `ALLEGRO' environment variable:

     $(ALLEGRO)/include

If the editor failed to find the file and a project is loaded a search in the
project is performed. If this also fails the editor looks for the file in the
same directory of the file you are editing.

This list is stored in the desktop file.

Name of the command: cmeIncludeList.

5.7.1.23 List of tag files
..........................

This dialog controls which tag files are loaded to search for symbols. To
learn more about tags please read the corresponding chapter.  ("TAGS files"
Chapter 13).

When you add a file the editor tries to load it and shows how many symbols it
contains.

After selecting a new tag file a dialog will ask if you want to add it using
a relative path. This is a very important question. If you are using one
centralized desktop file and you have just a few projects you can add a list
of tag files in this desktop file, but the list must contain absolute
entries. If you are entering a list of tag files for a project or a local
desktop file I recommend using relative paths, but it depends on your needs.

Name of the command: cmeTagFiles.

5.7.1.24 Tag files options
..........................

To maintain the tag files you should use a makefile. This makefile will
update the tag files when any of your sources changes.

To make things easier the editor provides an option to automatically keep the
tag files updated. This option can be enabled from this dialog and only works
if you are using a project.

When this option is enabled and you perform a search that involves tags the
editor will automatically check if any of the files included in your project
is newer than the tag files. If newer files were found the editor will invoke
ctags to update the tags. If you delete the tag files the editor will create
a new one.

This option works with a tags file called `tags' located in the same
directory as the project. This file will be listed in the list of tag files
to search but marked with `automatic'. You won't be able to remove this entry
from the list manually.

When new items are added to the project they are marked in a way that their
symbols will be added to the tags file, but when you remove items symbols
from this file aren't removed. If this is annoying just remove the `tags'
file and let the editor create a new one.

Name of the command: cmeTagsOps.

5.7.1.25 Regenerate central file
................................

The central file is maintained using the incremental features of ctags. Some
times ctags fails to remove obsolete entries in the tags file. This bug can
become really annoying. In this case you can use this command to regenerate
the file from scratch.

Name of the command: cmeTagsAutoRegen.

5.7.1.26 Calendar options
.........................

The editor can mark holidays in the calendar using a different color. As
holidays depends on the country this feature needs to know:

  1. What's your country

  2. About the holidays celebrated in your country.

The first is achieved using the contents of the LANG environment variable and
a file called `holidays.conf'. This dialog can be used to force the country
when the LANG isn't enough or you don't want to set this variable.

The second is achieved using plug-ins. Currently the editor supports it only
for systems that implements the `dlopen' family of functions. If you can't
find a plug-in for your country and you know how to write small C programs
consider wrtiting a plug-in for your country. The information about how to do
it is contained in the sources distribution. Look for a directory called
`holidays', it contains a `README' file. If you need assistance contact the
author. Please consider contributing the plug-in so other people from your
country can share the benefits.

Name of the command: cmeHolidaysConf.

5.7.1.27 Advice dialogs
.......................

In some situations the editor offers some advice to guide new users. These
are called `advice dialogs'. They can be disabled using a check box in the
dialog.

This option is used to enable or disable any of the advice dialogs. Texts
marked with an `X' indicates the corresponding dialog is enabled. Use the
<space> key to change the desired dialog.

Name of the command: cmeAdviceDiagConf.

5.7.2 Calculator (command/menu)
-------------------------------

This command shows the calculator:  ("Calculator" Chapter 11).

Name of the command: cmeCalculator.
Assigned key: `Alt+F4'

5.7.3 SDG
---------

Runs the documentation module: See Section 'Top' in documentation for 'SDG'.

To configure the SDG module:  ("SDG Options" Section 5.7.1.7).

Name of the command: cmeSDG.
Assigned key: `F9'

5.7.4 Run program
-----------------

Runs the desired program. To customize which program to run:  ("Run program
(which one)" Section 5.7.1.8).

Name of the command: cmeRunCommand.
Assigned key: `Ctrl+F9'

5.7.5 Grep
----------

This command pops up the `Powered Grep' dialog. Grep is a very powerful tool
to search text in files. To be able to use it you must have the grep tool
installed and in the path. It isn't shipped with the editor.

   * Pattern Box: The text you want to search for. You can use regular
     expressions here.  ("Regular Expressions" Section 15.4).

   * Files to search: The mask used to select the files grep will search
     through. Wildcards and some limited basic regular expressions are
     supported here.

   * Directories to search: The list of directories where the search will be
     performed.

The `Source of pattern' options are used to define the text that will be
searched for:

   * Pattern box is the pattern: grep will search the text indicated in the
     pattern box.

   * Pattern box is a file name: to search words contained in a text file.

   * Use clipboard selection: use the clipboard selection instead of the
     pattern box text.  ("Clipboard" Section 15.2).

The `Type of pattern' option selects how the pattern is interpreted. The
options are directly related to the grep switches `-G', `-E' and `-F'. You
can select basic regular expressions, extended regular expressions or just a
list of matching values separated by carriage returns.   ("Regular
Expressions" Section 15.4).

The `Place to search' group is used to select which files are examined in the
search:

   * Use files to search: The files to search content is the mask.

   * Search in opened windows: The search is performed in all the opened text
     files.

   * Search in project: The search is performed in all the project files.
     ("Project" Section 5.8).

   * Recurse in subdirs: When enabled the editor will run grep not only in
     the indicated directories but in any subdirectory contained therein as
     well. That's the main reason because I call it Powered Grep.

The `Options' group contains various options that are self-explanatory.  They
include: case sensitive search, whole word and whole line matching and
inverse matching. The last reports the lines that don't match, so be careful.

After the search the matched lines are displayed in the message window and by
pressing `Alt+F7' and `Alt+F8' you can examine them.

To learn more about the message window  ("Message Window" Section 15.8).

Name of the command: cmeGrepDialog.

5.7.6 HTML Accents
------------------

The following options are useful for people using ISO Latin 1 accents in your
HTML code. Even when the current code page is different than ISO Latin 1.

5.7.6.1 Convert accents to tags
...............................

It converts all the accents in the text to ISO-Latin-1 HTML tags. That's
useful when editing html files because you can type accents naturally and can
simply use this option to generate the right tags. It works for any code page
selected.

Name of the command: cmeHTMLAccents.

5.7.6.2 Convert tags to accents
...............................

It converts all the ISO-Latin-1 HTML tags into accents. That's useful when
reading html files because you can convert the tags to symbols. It works for
any code page selected.

Name of the command: cmeHTMLTag2Accent.

5.7.7 Export as HTML
--------------------

This option is used to export the current text file as HTML. The default
options generates a very good WYSIWYG result. This option works for any
syntax highlighting mode and for any color configuration you want.

Under DOS you can customize the editor's palette, this feature will export
the customized colors too.

Note that due to limitations in the HTML language the editor can change the
background color on a word-by-word basis only.

The available options are:

   * File name as title: uses the full path and name of the file as the title
     for the generated HTML.

   * Same background color as the editor: defines the background of the HTML
     file to be equal to the background of the editor's window.

   * Monospaced font: sets the font for the HTML to `Courier New'.

   * Bold attribute: sets the font for the HTML to `bold'.

You can also choose between colorized or simple output. Using colors the size
of the file is increased a lot but the result is very nice.

Name of the command: cmeExportAsHTML.

5.7.8 Insert key name
---------------------

This command brings up a dialog asking you to press a key. When you press a
key the dialog is closed and the name of the key is inserted at the cursor
position. You can use it to configure the menues (`menubind.smn').   ("How to
configure the keyboard" Section 4.1).

Name of the command: cmcInsertKeyName.

5.7.9 Remap code page
---------------------

With this command you can change the code page encoding of the current
document. This operation will translate all the characters from the current
encoding to a new one. Characters that don't have an equivalent in the new
code page are converted to spaces.

The dialog asks for the original code page ('from' list) and the new code page
('to' list). Additionally you can allow the editor to also translate the first
32 values. Translating the first 32 values can be dangerous; for this reason
the editor won't translate carriage return, line feed and tabs even if you
choose to remap the first values.

This operation affects the buffer globally so you can't use undo. Therefore
you should keep a copy of the file and don't save it if the results aren't
what you expected.

This option is very useful to exchange texts between different operating
systems.

Name of the command: cmeRemapCodePage.

5.7.10 Profile Editor
---------------------

This option is used to measure the speed of the editor. Use large files and
to make results comparable always use the same one.

Name of the command: cmcProfileEditor.

5.7.11 Redraw screen
--------------------

This command forces a redraw of the screen. It might be needed if some
application running in the background messed up your console.

Name of the command: cmeReDraw.

5.7.12 Paste Emacs mode
-----------------------

Pastes a comment at the start of the file indicating the Emacs mode and the
tab size used for this file. That's very useful if the file doesn't have an
extension or the extension is ambiguous. It is also a good idea to do it if
you'll send the file to another person and want to indicate which tab size
you used.

The editor understands this comment and sets the syntax highlighting and tab
size to the value indicated.

Name of the command: cmcPasteEmacsMode.

5.7.13 Block quoted printable decode
------------------------------------

This option decodes the selected text assuming it is encoded with the quoted
printable MIME spec. That's useful if you have an e-mail with non-ASCII
characters and it was encoded with this method. That's very useful for
spanish accents.

Name of the command: cmcQuotedPrintDecode.

5.7.14 Un/Indent block
----------------------

This submenu contains the block indentation operations.

5.7.14.1 Indent one space
.........................

Indents a block one space. For more information  ("Indentation" Section
3.4.3).

Name of the command: cmcIndentBlkOne.

5.7.14.2 Unindent one character
...............................

Unindents a block one character. For more information  ("Indentation" Section
3.4.3).

Name of the command: cmcUnIndentBlkOne.

5.7.14.3 Indent one tab or gap
..............................

Indents a block as if you used the <Tab> key in each line.  For more
information  ("Indentation" Section 3.4.3).

Name of the command: cmcIndentBlk.

5.7.14.4 Unindent one tab or gap
................................

Unindents a block as if you used the <Backspace> key in all of the lines. For
more information  ("Indentation" Section 3.4.3).

Name of the command: cmcUnIndentBlk.

5.7.14.5 Comment indent
.......................

This command inserts a comment at the start of each line of the selected
block. The comment used is the one defined in the syntax highlighting file as
`EOLComment1'; if none is defined or the file doesn't have any syntax
highlighting this command does nothing.  ("EOLComment1" Section 8.11).

If no text is selected this command first selects the current line and then
applies the indentation.

Name of the command: cmcCommentIndent.

5.7.14.6 Comment unindent
.........................

This command removes as many chars from each selected line as the length of a
comment sequence. The comment used is the one defined in the syntax
highlighting file as `EOLComment1'; if none is defined or the file doesn't
have any syntax highlighting this command does nothing.   ("EOLComment1"
Section 8.11).

This command doesn't check if each line you selected starts with the defined
comment, so be careful.

If no text is selected this command first selects the current line and then
applies the indentation.

Name of the command: cmcCommentUnIndent.

5.7.14.7 Arbitrary indent
.........................

This command pops up a dialog asking for a text to be used as indentation.
The text will be inserted at the start of each line of the selected block.

Name of the command: cmcArbitraryIndent.

5.7.15 Delete memorized backups
-------------------------------

Deletes all the memorized backups. This includes all the backups created while
the current project/desktop file was opened. If you want to also delete
backup files created during previous sessions you must enable a special
option that makes the editor keep a list across sessions.   ("Editor General"
Section 5.7.1.4).

Name of the command: cmeDeleteBkps.

5.8 Project
===========

The project files are used to indicate groups of files. Each project has its
own desktop file so you can have different settings for different groups of
files.  ("Desktop Files" Section 15.5).

There are several reasons to use projects:

   * If you want to work on a group of files and you will be editing these
     files for a long time use a project. Then you will be able to select
     which file to edit from the project window, as this window is sorted
     alphabetically it's easy to make incremental searches (typing the first
     letters) to find the file. Additionally the editor saves the window
     position and other stuff for all the files listed in the project, even
     if they aren't listed in the closed windows list. I use it for my web
     site files; they are more than 44 and the list of windows (`Alt+0')
     doesn't help.

   * If you are using the SDG module to collect the comments of the files you
     need to use a project to specify them.

   * You can use a project to list a set of files to search with grep.  Each
     time you want to search in these files you just open this project and
     perform the search.

By default the project window is located at the bottom of the desktop, but
you can change this.  ("Editor General" Section 5.7.1.4).

This window shows only the names of the files and the relative path only when
names are repeated. You can change it to display the relative path for all
files pressing `Alt+V' when the project window is selected.

5.8.1 Open (Project)
--------------------

Opens a project file. To create a new file just enter a new name in the
dialog.

Name of the command: cmeOpenPrj.

5.8.2 Close (Project)
---------------------

Close the project file.

Name of the command: cmeClosePrj.

5.8.3 Save (Project)
--------------------

Saves the current project. It's a good idea to save the project after adding
or removing a lot of files.

Name of the command: cmeSavePrj.

5.8.4 Save desktop here
-----------------------

Saves a desktop file in the working directory. A desktop file is created even
if the editor is configured to use only one central desktop file. This is
useful to locally store settings without creating a project.

Name of the command: cmeSaveDesktop.

5.8.5 Export project
--------------------

Creates a text file containing the names of the project items. The file
contains one file per line.

Name of the command: cmeExportPrj.

5.8.6 Import project items
--------------------------

Imports the names given in the file as project items. The file with the file
names to import must contain one file per line. Relative paths are
recommended. The editor will check if the file exists and if that's a regular
file, if not the item will be rejected.

At the end of the import process the editor informs how many items were
added, how many were already in the project and how many file names didn't
correspond to existing files.

Name of the command: cmeImportPrj.

5.9 Help
========

5.9.1 InfView
-------------

Well, I think you figured it out, that's the help.

Name of the command: cmeInfView.
Assigned key: `F1'

5.9.2 Another InfView
---------------------

It opens another InfView window. The editor always opens one window that's
used by the help system. When you close this window the editor just hides it
and when you press <F1> the window is un-hided and the help is displayed.
That allows the existence of the `Previous help' command. But some times you
could want to brise one or more help files without losing the help window, in
this case you need more than one InfView opened.

Name of the command: cmeAnotherInfView.

5.9.3 Tip of the day
--------------------

Once a day the editor shows a usage tip when you start it. Each tip talks
about one interesting feature that most of the people overlook. Reading them
you'll discover a lot of interesting things about the editor.

Each tip has one or more buttons at the right, which represent a link to a
help topic related to this tip. If you want to learn more about the tip's
topic you can browse the help using the buttons.

There are three options at the bottom of the window, they are self
explanatory. The first disables the annoying tips ;-), the second shows the
tips in a dialog box once a day and the third shows the same text in the
message window (once a day of course). The third option is less annoying than
the second but you lose the link buttons.

The text displayed by the tips can be found in the `editor.tip' file. You can
edit it to show anything but be careful because the parser isn't very
tolerant to typos.

To learn more about the message window  ("Message Window" Section 15.8).

Name of the command: cmeTipOfTheDay.

5.9.4 Syntax help
-----------------

When programming in a language like C you probably can't remember the exact
name of all the library functions; the djgpp libc help contains around 650
nodes and the Allegro help around 400. Placing the cursor over the name of a
library function and pressing <^F1> you'll get help about this function.  If
the name isn't exactly typed you'll get a list of the closest matches. The
following topics explain how to configure it.

5.9.4.1 Options (Syntax help)
.............................

This dialog box allows the configuration of the syntax help.  See Syntax help.

The search method used can be:

   * Exact: Only exact matches are reported.

   * Substring: Partial matches are reported.

   * Fuzzy: It uses a special algorithm that reports words similar to the one
     you are searching.

The available options are:

   * Case sensitive: The search interpretes lower case characters as different
     than uppercase characters.

   * Sort by score: The matches are sorted by score when reported. When
     disabled the sorting criteria is alphabetical. A greater score means the
     match is more similar to the word you are searching. A score of 1000
     means exact match.

   * Fuzzy value: That's used only when the selected mode is the fuzzy mode.
     It indicates the minimal score a word must have to be displayed as a
     possible match. Experiment with different values and see the scores
     reported.

Name of the command: cmeSyntaxHelpOps.

5.9.4.2 Files to search (Syntax help)
.....................................

Here you can indicate in which info files the editor will search the name of
the function; the default is OS dependent.  ("Syntax help" Section 5.9.4).

You can specify an info node or just the name of the file. In the first case
the editor will read all the cross references found in this node, in the
second the editor will use all the nodes of the file. Normally the node that
contains all the relevant references is called Index, but there are
exceptions; one interesting case is libc.

In the dialog the editor shows which nodes are searched in. If the name has a
question mark to the left it means the editor didn't read the file yet,
pressing <^F1> over any word the editor will search it and hence will read
the help files. If the name has an asterisk it means some error was
encoutered when trying to read this file. Finally if no mark is indicated it
means the editor successfully read the file, additionally the number of nodes
found is indicated to the right. You can add or remove nodes from the list.

Name of the command: cmeSyntaxHelpFiles.

5.9.4.3 Search (Syntax help)
............................

It makes the syntax search and report the matches found. If only one match
was found the editor jumps to this node.  ("Syntax help" Section 5.9.4).

Name of the command: cmeSyntaxHelp.
Assigned key: `^F1'

6 Menues configuration
**********************

  If you need or want to configure the shortcuts activating the menues or menue
options look at the `menubind.smn' file. The format is self explanatory and
the editor supports syntax highlighting for these files.

  This file also configures the `status bar', that's the line at the bottom of
the screen. This line is context sensitive and also contains shortcuts.  Some
of the shortcuts can be invisible.

  Finally you can also configure the menu displayed when you click over an
editor window using the right mouse button.

  The menues and status bar can be used to complement the keyboard assignments.
You can assing editor commands, sLisp macros or small portions of sLisp code
to them.

  This chapter is a small reference about the commands you can use in the
`menubind.smn' file.

6.1 SubMenu and EndSubMenu
==========================

  The `SubMenu' command starts a menu definition. It can be nested. When used
at top level defines a new entry in the menu bar. When used inside another
`SubMenu' section defines a child menu.

  The `EndSubMenu' marks the end of a menu section.

  The syntax for `SubMenu' is:

       SubMenu: "Visible name", key
       ...
       EndSubMenu

 The visible name is what you'll see in the menu bar or the entry in the
parent menu. Usually you assign a key to submenues so they can be easily
selected. That's what you put in the `key' field. To highlight the key you
can use the `~' character. This character indicates the beggining and the end
of a highlighted text. The following example shows a submenu called File
that's triggered when `Alt+F' is pressed and that marks the `F' with a
different color:

       SubMenu: "~F~ile", kbAlF

6.2 Key names
=============

  When defining a menu entry you must indicate the name of the key or key
combination. The editor have a command specially designed to help you with
it, look in the `Tool&Ops' menu ( "Insert key name" Section 5.7.8). Here is a
small explanation about how those names are created.

  In menues all key combinations start with the `kb' preffix. Then follows the
key modifiers. They can be:

   * Sh
     - key: Shift

   * Ct
     - key: Control

   * al
     - key: Right Alt

   * Al
     - key: Left Alt


  Avoid using the right alt definitions unless you know very well the side
effects. You can specify more than one modifier but they must be in the order
shown in the table. Suppose you want to indicate a complex key combination
where shift, control and the left alt key must be pressed together the result
is `kbShCtAl'. Note that is not valid if you change the order. As an example
`kbShAlCt' is invalid.

  After the modifier comes the key name. The key name is usually the character
you obtain by pressing the key or the name of the character you obtain. For
the alphabet characters is quite simple, `kbA' is just the key that generates
the `a'. The same is for numbers. Function keys are also easy, they are named
F and the number of the function key.

  Things becomes more complex and even confusing for other characters. At first
you must know the name I gave to each symbol and then you must understand a
small complexity created by the fact that some shifted keys generates
non-related symbols. Here is an example: `kbSh1' is valid but you won't get
this combination because when you press <1> plus shift what you obtain is
`!'. The editor will interpret it as `kbAdmid'.

  Here is a list of the names, but the best is to use the `Insert key name'
option: OpenBrace, BackSlash, CloseBrace, Pause, Esc, BackSpace, Tab, Enter,
Colon, Quote, Grave, Comma, Stop, Slash, Asterisk, Space, Minus, Plus, PrnScr,
Equal, Home, Up, PgUp, Left, Right, End, Down, PgDn, Insert, Delete, Caret,
Admid, DobleQuote, Numeral, Dolar, Percent, Amper, OpenPar, ClosePar,
DoubleDot, LessThan, GreaterThan, Question, A_Roba, Or, UnderLine, OpenCurly,
CloseCurly, Tilde, Macro, WinLeft, WinRight and WinSel.

  When you have to indicate that no key the `kbUnknown' name can be used.

6.3 MenuItem entries
====================

  To add a menu entry you can use the `MenuItem' command. The syntax is:

     MenuItem: "Name", Command[, Key [, Context [,"KeyName"]]]

  The name is the visible entry in the menu. The command is the command
assigned to this menu entry. If this menu entry have a shortcut it must be
indicated in the key field. The context field is used to indicate a help
context. The keyname field is just an string that is displayed at the right
of the name of the menu entry to show the shortcut. This syntax is not very
common because you usually don't want to specify a particular help context.

6.4 Commands in menu items
==========================

  You can use any of the editor commands in the commands field. They are listed
in the commands index ( "Index of key commands" Chapter 17). Additionally you
can specify the name of a sLisp macro or a small piece of sLisp code.

  To indicate the name of a sLisp macro just use the following syntax:
`cm(NAME)'. To enter a small portion of sLisp code use it: `cm((CODE))'. Here
is an example that just prints a message in the message window:
`cm((ShowInMessageWindow 'Hi! ;-)'))'.

6.5 MenuItemC entries
=====================

  This is the most common way to add a menu entry. The syntax is:

     MenuItemC: "Name", Command[, Key [,"KeyName"]]

  The only difference with `MenuItem' ( "MenuItem entries" Section 6.3) is that
you don't have to specify a help context. Help contexts are numbers that
associates a context in the editor with some text in the on-line help.
Usually the number is the same used for the command. The `MenuItemC' assumes
the help context is the same as the command number.

6.6 MenuSeparator
=================

  It adds a division line to the menu. Use it to visually group related menu
entries.

6.7 The special submenu used for the right click
================================================

  To define the menu associated with the right mouse button you must define a
top-level submenu called `Editor Right Click'. This submenu will pop up when
you right click over an edition window and is just like any other submenu.

6.8 Status bar ranges
=====================

  The content of the status bar is context sensitive. For this reason the
content of the status bar is defined for a help context or for a range of
help context values. The ranges can overlap and the first match is used. For
this reason the order is important.

  Ranges are started using the `StatusRange' command and closed using the
`EndStatusRange' command. The syntax for the first is:

       StatusRange: first_context, number_of_contexts

6.9 Visible status bar entries
==============================

  The syntax for visible entries is:

       StatusEntry: Label, Command, Key

  The label is the visible text and the rest of the values are just like the
ones used by the `MenuItem' entries.  ("MenuItem entries" Section 6.3).

6.10 Invisible status bar entries
=================================

  The syntax is almost the same used by the visible ones, the most important
difference is that they doesn't need a label.  ("Visible status bar entries"
Section 6.9).

       StatusHiddenEntry: Command, Key

6.11 Conditional menu entries
=============================

  The `menubind.smn' files support some conditional statements. They are quite
similar to the preprocessor conditionals used in C language and can be used
to exclude or include some menu entries according to the operating system and
other configuration details.

  The most important structure is:

       $if CONDITION
       ...
       $else
       ...
       $endif

  The `else' section is optional. In this case if the `CONDITION' is true the
entries in the first section are used, otherwise the entries in the `else'
section are used.

  Contitions can use and, or and not boolean operations. The `and' and `or'
operators are used for the first two and the `!' operator for not. Currently
you can use parentheses in conditionals but this is not fully tested.

  The available variables to evaluate are:

   * Alpha
     - Meaning: The binary was compiled for Alpha (DEC) CPUs

   * BCPP
     - Meaning: The binary was compiled using Borland C++

   * BZIP2
     - Meaning: Support for bzip2 compressed files included

   * CALCULATOR
     - Meaning: Built-in calculator is available

   * CALENDAR
     - Meaning: Built-in calendar is available

   * Cygwin
     - Meaning: The binary was compiled using Cygwin

   * djgpp
     - Meaning: The binary was compiled using djgpp

   * DOS
     - Meaning: The binary was compiled for DOS

   * DrvDOS
     - Meaning: The editor using the DOS TV driver

   * DrvQNX
     - Meaning: The editor using the QNX RtP TV driver

   * DrvQNX4
     - Meaning: The editor using the QNX 4 TV driver

   * DrvUNIX
     - Meaning: The editor using the generic UNIX TV driver

   * DrvWin32
     - Meaning: The editor using the Win32 TV driver

   * DrvWinGr
     - Meaning: The editor using the WinGr TV driver

   * DrvWinNT
     - Meaning: The editor using the WinNT TV driver

   * DrvX11
     - Meaning: The editor using the X11 TV driver

   * DrvXTerm
     - Meaning: The editor using the XTerm TV driver

   * FreeBSD
     - Meaning: The binary was compiled for FreeBSD

   * GCC
     - Meaning: The binary was compiled using gcc

   * HPPA
     - Meaning: The binary was compiled for HP-PA Risc CPUs

   * Itanium
     - Meaning: The binary was compiled for IA64 CPUs

   * Linux
     - Meaning: The binary was compiled for Linux

   * MinGW
     - Meaning: The binary was compiled using MinGW

   * MIPS
     - Meaning: The binary was compiled for MIPS CPUs

   * MIXER
     - Meaning: Support for the audio mixer included

   * MP3
     - Meaning: Support to play MP3 included

   * MSC
     - Meaning: The binary was compiled using Microsoft C

   * PCRE
     - Meaning: Support for Perl RegEx included

   * PPC
     - Meaning: The binary was compiled for Power PC CPUs

   * QNXRtP
     - Meaning: The binary was compiled for QNX

   * Solaris
     - Meaning: The binary was compiled for Solaris

   * SPARC
     - Meaning: The binary was compiled for 32 bits SPARC CPUs

   * SPARC64
     - Meaning: The binary was compiled for 64 bits SPARC CPUs

   * UNIX
     - Meaning: The binary was compiled for a UNIX style OS (Linux included)

   * Unknown
     - Meaning: The binary was compiled for an unknown CPUs

   * Win32
     - Meaning: The binary was compiled for Windows

   * x86
     - Meaning: The binary was compiled for IA32 CPUs


  In addition to the above mentioned methode you can also define small macros
using the `$define' directive. You can also undefine them using `$undef'
directive. These directives doesn't support multiline definitions, or worst,
they do but this won't work as you should spect.

6.12 Comments in menu files
===========================

  You can add comments to menu files, they must start with the `#' character
and it must be in the first column. Lines starting with `#' are ignored when
parsing menu files.

7 Editing Modes
***************

  The editor has various settings that control the function and aspect of the
editor.

  The settings are:

7.1 Overwrite
=============

  This setting controls if the typed characters are inserted in the buffer or
replace the original ones. For a detailed explanation  ("Insert and Delete"
Section 3.3).

7.2 Autoindent
==============

  This setting controls what happens when you press <ENTER>. If this setting is
off the cursor goes to the column 1 of a new line. If the mode is on the
editor will try to mimic the indentation of the code by inserting spaces or
tabs.  ("Optimal Fill" Section 7.12).

7.3 Real Tabs
=============

  This setting controls what happens when you press `<TAB>'. If this setting is
on the editor will insert an ASCII 9 at this position.

  An ASCII 9 is a TAB, that means that the width of this char is enough to move
the cursor to the next tabulator column. In the editor the tabulator columns
are equidistant and the positions are controled by the Tab Size value.

  If this setting is off the editor won't put any ASCII 9 in your text.  The
behavior is configured by the `Tab indents' option. Read the section about it
for more information.  ("Tab indents" Section 7.18).

  You can also indent using spaces when this option is disabled. For more
information consult the `Use indent size' option.   ("Use indent size"
Section 7.19).

  In the past (versions older than 0.4.44) another thing controlled by this
setting was the behaviour of the <Backspace> key, but now that's controlled
by the `Backspace unindents' option.  ("Backspace unindents" Section 7.21).

  The editor is much more coherent when you choose to use TABs or not use TABs.
If you mix the two modes you'll get some unexpected results, especially in
the indentation of the blocks.

  If you never use tabs, it's better you work with this setting off.  Then
you'll get much from the editor.

  Now you can say: `But I really need tabs because I'm editing a make file!' or
`because I will send the file using an ultra slow link' `and I want the
compression granted by the tabs.' In these cases you can first expand all
tabs, then work without real tabs and at the finish of your work compact all
possible spaces and generate a file with tabs.  ("Miscellaneous" Chapter 15).

  Most tabs users also like to enable the `Optimal Fill' option.  ("Optimal
Fill" Section 7.12).
I also suggest using the following options to complement the indentation when
using tabs: `Autoindent' ON, `Intelligent C indent' OFF, `Optimal Fill' ON,
`Do not move inside tabs' ON, `Tab smart indents' OFF, `Use indent size' OFF
and `Backspace unindents' OFF.

7.4 Persistent Blocks
=====================

  This setting controls the behaviour of the selected area.

  The "Block modes" chapter for a detailed explanation.  ("Block modes" Section
3.4.1).

7.5 Intelligent C indent
========================

  This mode was designed to be used jointly with the Pseudo Macros ( "Pseudo
Macros" Chapter 9) and the Real Tabs mode off to achieve an easy way to
indent the code that does a better job than the Autoindent mode.

  In this mode the spaces inserted after pressing <ENTER> depend on the first
word in the last line. For example, if you have:

     if (a==b)_

  With the cursor in the '_' position and press <ENTER> you'll get:

     if (a==b)
       _

  Now you can do either of two things, 1) press space and write the code that
will be executed by the if, or 2) if this a multiline code press { and
<ENTER>.  In the last case you'll get:

     if (a==b)
       {
        _

  Now type your first line of code. Press <ENTER>. Write your next line.  Press
<ENTER> again and then <Backspace>:

     if (a==b)
       {
        1st line;
        2nd line;
       _

  Now type } and press <ENTER> one more time:

     if (a==b)
       {
        1st line;
        2nd line;
       }
     _

  As you can see the code is perfectly aligned without a significant work on
your part.

  I tried to make this mode as smart as possible, but it still needs more work.
If you have suggestions contact me.

  Another important thing is that this indentation has a personal style, my
style ;-), so it may be you don't like it. If that's the case you can do the
following things:

   * Use another way of indentation offered by the editor. For example, turn
     on the Real Tabs mode and indent with tabs.

   * Customize the `cpmacros.pmc' file ( "Pseudo Macros" Chapter 9), actually
     this file is coherent with this mode.

   * The editor isn't configurable like Emacs or Brief using a language, but
     is written in C++. If you know C++ contact me and I'll help you to write
     the routines that you need to get an indentation in your own style :-).

7.5.1 Can you explain more about the behavior of this mode?
-----------------------------------------------------------

  I'll try to describe the behaviour of the mode:

  Each time you press <ENTER> the editor inserts a `\r\n' string in your text,
after that the editor searches one line located above the new line that has
at least one character on it. This line is taken as reference. The editor
analyzes this line searching for:

   * The first non-blank character on this line.

   * The first word on this line.

   * The first parenthesis.

   * The balance of parentheses on the line.

   * The last non-blank and non-comment character in the line.

  Now, if the line contains { at the beginning the editor goes to the first
column after the {.

  If the line contains } at the start the editor will go to the same column of
the } and then will perform a <Backspace>. If the Real Tabs mode is off,
that's an unindent.  ("Real Tabs" Section 7.3).

  If the line starts with a C++ comment the effect is the same as in
Autoindent.  ("Autoindent" Section 7.2).

  If the line starts with a C comment the editor will try to skip the comment
and analyze the rest of the line, but if the comment doesn't end on this line
the editor will go to the column where the `/' is.

  If the line starts with `/' the editor goes to this column.

  If the line has more `(' than `)' the editor will go to the column of the
first non-blank after the first `('.

  If the line has more `)' than `(' the editor will search the line where the
number of parentheses is balanced, then will analyze this line. If the whole
line still generates an unbalanced situation the editor will go to the first
used column in the line that was found the first time. But if this line will
leave all brackets balanced the editor will take the first word on the line
and will use it as reference.

  At last, and according to the word found, the editor will use this word as
reference. The editor recognises the following keywords:

   * `do'
     - Action: +2 but not if ... ;

   * `if'
     - Action: +2 but not if ... ;

   * `for'
     - Action: +3 but not if ... ;

   * `else'
     - Action: +2

   * `case'
     - Action: +5

   * `while'
     - Action: +2 but not if ... ;

   * `switch'
     - Action: +2

   * `break'
     - Action: unindent

   * `return'
     - Action: unindent

   * `default'
     - Action: +5


  The numbers represent how many spaces are added with reference to the first
letter of the word.  'not if ... ;' means that, if the line ends with a
semicolon, the editor will do the same as for Autoindent.  ("Autoindent"
Section 7.2). The unindent is performed with <Backspace>.

  Seasick?  ("Do you have more examples?" Section 7.5.2).

  Note: Some of these features were added in v0.2.14 of the editor based on a
suggestion of stud73@nortel.ca <Bradford L. Spencer> about the behaviour of
the mode on a line like this `printf("Num: %d",' with the rest of the
parameters on the next line.

7.5.2 Do you have more examples?
--------------------------------

  Well here are some other examples. I used a strange convention, like this: if
I say type `a[ENTER]{', type the letter `a', then press `<ENTER>' and finally
press the `{' key.

Example 1: (similar to one explained before but is to show the convention)

     Type:
     if (a==1)[ENTER]{[ENTER]a=2;[ENTER]b=3;[ENTER][BACKSPACE]}[ENTER]

     You'll get:

     if (a==1)
       {
        a=2;
        b=3;
       }
     [<--- cursor here]

Example 2: A switch/case example

     Type:
     switch(a)[ENTER]{[ENTER]case 1:[ENTER]a=2;[ENTER]b=3;[ENTER]break;
     [ENTER]case 2:[ENTER]b=5;[ENTER]break;[ENTER][BACKSPACE]}[ENTER]

     You'll get:

     switch(a)
       {
        case 1:
             a=2;
             b=3;
             break;
        case 2:
             b=5;
             break;
       }
     [<--- cursor here]

Example 3: A call to a function that takes a lot of parameters

     Type:
     printf([SPACE]"Num: %d",[ENTER]a[SPACE]);[ENTER]

     You'll get:

     printf( "Num: %d",
             a );
     [<--- cursor here]

Example 4: A lot of parentheses

     Type:
     if[SPACE]([SPACE](a==1)[SPACE]||[ENTER](b==2)[SPACE]||[ENTER]
     c[SPACE])[ENTER]

     You'll get:

     if ( (a==1) ||
          (b==2) ||
          c )
       [<--- cursor here]

Example 5: Comment trying to interfere part 1

     Type:
     /*-a-*/for[SPACE](x=1;x;--x)[ENTER]

     You'll get:

     /*-a-*/for (x=1;x;--x)
               [<--- cursor here]

Example 6: Comment trying to interfere part 2

     Type:
     for[SPACE](x=1;x;--x);[SPACE]//-b[ENTER]

     You'll get:

     for (x=1;x;--x); //-b
     [<--- cursor here]

 Note: Of course you can fool the editor but as you can see it is relatively
smart ;-).

7.6 Column cursor
=================

  This setting enables a strange feature of the editor. When on the column
where the cursor is is highlighted. This feature is very good to check if
parts of your code are aligned correctly.

  If you like this mode but think it's very uncomfortable to use all the time
contact me and if I get enough feedback I'll create a hotkey for it to be
turned on/off quickly.

7.7 Row cursor
==============

  This setting is similar to the column cursor but acts on the row where the
cursor is. If you enable the two modes you'll get a cross on the screen
showing where the cursor is.

7.8 Match pair highlight
========================

  This mode acts showing the pairs of (/), [/] and {/} on the fly. Each time
you type one of these symbols the editor will search the matching pair, if
the editor finds it and the match is on the screen both will be highlighted,
if the match is outside the screen the editor will inform the position on the
status line, and if there is no match the editor will inform the situation in
the status line too.

  That's very useful when you are typing complex parenthetical expressions or a
very nested code. You can use it together with the `cmcSearchStart',
`cmcSearchEnd', `cmcSearchOpPar', `cmcSearchClPar', `cmcSearchOpCor' and
`cmcSearchClCor' commands.  ("Miscellaneous" Chapter 15).

  If you want to get highlight not only after typing but also when moving the
cursors you'll need to enable the `Match pair on the fly' option.   ("Match
pair on the fly" Section 7.9).

7.9 Match pair on the fly
=========================

  This mode is very similar to the `Match pair highlight' mode. If you don't
know how it works please read the `Match pair highlight' section first.
("Match pair highlight" Section 7.8).

  The main difference is that this mode highlights the pair when the cursor is
over the character to search.

  The highlight is done half a second after you stop typing to avoid interfering
with your typing. But if you have a fast machine, not just a 386, you can
configure the editor to do the search without waiting.   ("Do not wait to
search the pair" Section 7.10).

7.10 Do not wait to search the pair
===================================

  This option works only when `Match pair on the fly' is enabled. When enabled
the editor doesn't wait to do the search of the complementary pair.  I think
this could impact the performance of scrolling on very slow machines and
that's why it's optional.  ("Match pair on the fly" Section 7.9).

7.11 Transparent Blocks
=======================

  When this mode is on you can see the syntax highlighting of selected blocks.
Normally the selection affects the background and foreground colors, but when
using transparent blocks only the background is affected.

7.12 Optimal Fill
=================

  This mode was added for the people that use ASCII 9 tabulators in their code
( "Real Tabs" Section 7.3). Normally the editor uses spaces to indent the
code or, in general, to fill any gap in the text. When you enable this mode
the editor will use as many tabs as possible to fill these gaps. That's what
the tab users normally expect.

7.13 Wrap Words
===============

  Even when the editor is mainly intended for programmers it became apparent
that some other user groups want to use it too.

  The word wrap added to the editor is a very simple one, it just inserts a new
line if you type a word beyond the wrap column, that's all. You won't get
automatic reformat functions like in text editors intended for love letters
(like the one from the Bill Gates company).

  The wrap column box is used to enter the column that triggers the wrap.

7.14 Do not move the cursor on Paste
====================================

  That's a global setting. When it's on the cursor isn't moved after pasting.
Normally the cursor is moved to the end of the pasted block, but sometimes
it's better if the cursor isn't moved.

7.15 Scroll Lock centers
========================

  When this mode is on the Scroll Lock key has a special meaning. If the Scroll
Lock LED of your keyboard is on then the editor centers the current line in
the window. The effect is very strange but the advantage is that you don't
need to follow the movement of the line with your eyes because it's always in
the same place.

7.16 See Tabs
=============

  In this mode the tabs are highlighted, two colors are used for this purpose,
one for even and the other for odd numbered tabs. In this way you can clearly
see where a tab is located and the size of the tab. The colors can be
customized from the Colors menu option.  ("Customize Colors" Section 5.7.1.1).

  This mode was introduced in v0.4.23 and is globally enabled by default.

7.17 Do not move inside tabs
============================

  In this mode the cursor can't be placed inside a tab character. This
definition is fuzzy and confusing so here I'll try to explain it better. One
tab character can be expanded to one or more characters when displayed on the
screen. Normally you can place the cursor in any of the spaces that belong to
a tab character. I think this behavior is the best because you are free to
move the cursor to any place you want, but tab users get confused when they
type and discover they had the cursor in the middle of a tab. It produces a
cursor jump. To avoid this kind of surprise a lot of editors don't allow to
position the cursor in these spaces, only in the first space.

  This mode was introduced in v0.4.23 and is globally enabled by default.

7.18 Tab indents
================

  When `Use real tabs' option is disabled the editor will insert enough spaces
to move the cursor to the next tab-stop or indent position. But if this
option is enabled the editor will insert enough spaces to move the cursor to
the beggining of the next word in the previous line.  Confused? Sorry for my
English, an example will clarify that:

This is a line above the line you are

That's the line where the cursor is, at column one.

  After pressing TAB you'll get:

This is a line above the line you are

     That's the line where the cursor is, at column one.

  Now you'll ask, Why this? Is that useful? The answer is that's very useful to
keep your code indented. Experiment using that under a line with if, for,
`etc.'

  Another option is to indent like the Tab key but using a size different than
the tab size. This can be done disabling this option and enabling `Use indent
size'.  ("Use indent size" Section 7.19).

7.19 Use indent size
====================

  When the `Real Tabs' and `Tab indents' options are disabled you can indent
with the tab key but using spaces. Sometimes people want to use tabs of eight
spaces because this is the most common value for consoles and printers but,
at the same time, they want to indent by a different amount of spaces
pressing tab. In this situation you must enable this option and configure the
indentation amount in the `Indent size' box.

7.20 Keep trailing whitespace
=============================

  Normally the editor purges any space after the last visible character in a
line. That's very useful to save disk space and to avoid problems with end of
line continuation sequences, like in C language. But sometimes you may want
to avoid it for some reason. Enabling this option the editor won't try to
remove extra spaces at the end of lines.

7.21 Backspace unindents
========================

  When this option is enabled the <Backspace> deletes as many spaces as
necessary to move the cursor to the first used column of the previous line.
Basically it keeps the indentation. However, this is true only when all the
characters located to the left of the cursor are spaces or tabs.

  In versions older than 0.4.44 this option was implicitly enabled when the
`Real Tabs' option was disabled. When loading old desktop files the editor
enables/disables this option based in the `Real Tabs' option.

7.22 Column Markers
===================

  Column markers highlight a column of text. This option is useful for
programming languages where the column is important or when you just need to
have some column positions as references.

  You can set various column markers at the same time, just enable this option
and enter a list of columns in the associated text box. The columns should be
separated by spaces and sorted incrementally. The editor will format the list
in this way.

7.23 Syntax Highlighting
========================

  The editor can highlight the syntax of your code. The available modes are:

   * No highlighting, all the code is with the same color.

   * C/C++ Highlighting.

   * Pascal Highlighting.

   * Clipper Highlighting.

   * User defined. Including already defined for:
        * 4DOS batch files

        * 80x86 assembler (AT&T syntax)

        * 80x86 assembler (Intel syntax)

        * 8x51 assembler

        * Ada

        * BASIC

        * C/C++

        * Cascading Style Sheets version 2

        * Clipper 5.x

        * Command Line Errors File

        * Environment files

        * Flat assembler

        * Fortran

        * HTML

        * Internationalization files (.po)

        * Java

        * Java Script

        * Lua

        * Makefiles

        * Menu files

        * Modula 2

        * Netwide Assembler (NASM)

        * Objetive C

        * Pascal

        * PDP11 assembler

        * Perl

        * PHP

        * PIC assembler (Microchip)

        * PLM/51

        * PMacros files

        * PostScript

        * POV-ray

        * Python

        * Ruby

        * SDG format files

        * sLisp macros

        * SQL

        * The syntax highlight file itself

        * TCL/Tk

        * TeX

        * Texinfo

        * Turbo Vision configuration files

        * UNIX shell scripts

        * VHDL

        * XML

        * WML

  The kind of highlighting is chosen automatically using the extension of the
file. Additionally the editor supports Emacs-like mode selection. Emacs
searches the name of the editing mode in the first kilobyte and last three
kilobytes of text. The editing mode is delimited by `-*-' and the editor will
use it instead of the file extension to select the highlighting mode.  That's
very useful for files without extensions like the new C++ headers.  To add
even more flexibility I'm supporting another technique used by some C++
headers from Silicon Graphics and Hewlett Packard, in these headers the mode
is located in the last lines using some special keywords.

  The editor also supports the convention used for UNIX script files. If a file
starts with `#!' this line indicates the program that must be executed to
interpret the script. The editor reads this line and extracts the name of
this program and searches it in the `ShellScript' definition.

  The C/C++ highlighting was designed for the GNU C compiler.

  The Pascal highlighting was designed for the GNU Pascal compiler.

  The Clipper highlighting was designed for the CA-Clipper 5 compiler.

  The highlighting can be customized,  ("Syntax Highlighting File" Chapter 8).

8 Syntax Highlighting File
**************************

  The highlighting can be customized editing the `syntaxhl.shl' file. You can
add syntax highlighting to almost any kind of files, a good example is the
fact that the files used to configure the editor have their own syntax
highlighting.   ("Configuration files location" Section 15.1).

  The file declares the settings for each kind of files. Each declaration ends
with an `End' marker. The `#' acts as a start of command if it appears in the
first column.

  Important things you have to know to add a new syntax highlighting:

   * In previous versions when adding a new highlight mode it ought to have
     been added at the end of the file, now this isn't mandatory.

   * The editor remembers various settings of loaded files even after closing
     the file. This includes the syntax highlighting, so if you add a new one
     or added a new extension to the list and you open a file that you
     already opened the editor will remember the last settings and doesn't
     use the new setting. In this case you must force it by hand.

   * If you feel that this new highlight mode can be used by other users send
     it to me and I'll include it in the next release. Many of the currently
     included syntax highlightings were contributed by users.


  If you want to add some keywords to a language for personal use don't do it
in the `Keywords' section because you will need to edit this file each time
you install a new release of the editor. The editor provides another file for
it and also a nice user interface to add and delete words defined by the
user.  ("User Words" Section 5.7.1.18).

  The next sections covers the supported settings.

8.1 AllowedInsideNames
======================

  Some languages include symbols inside names, you can list these symbols here.
Normally the editor allows letters, digits and underscore. These characters
are allowed inside the names and not at the start of the name.
("CanStartAName" Section 8.2).

8.2 CanStartAName
=================

  Some languages includes symbols at the start of names, you can list these
symbols here. Normally the editor allows letters and underscore. These
characters are allowed at the start of the name and not inside.
("AllowedInsideNames" Section 8.1).

8.3 Case
========

  When this setting is 1 all the keywords become case sensitive. If the
keywords aren't case sensitive don't use it.

  This should be declared as one of the first settings because it affects how
other definitions are loaded in memory. I suggest using it before `Name' and
`Files' and after the others.

8.4 CloseComment1
=================

  Used to indicate the end of a multiline comment. The lenght is limited to four
characters. If the file format has two different terminations use
`CloseComment2' for the second.

  For more information  ("Format of short syntax highlighting definitions"
Section 8.21.1).

8.5 EmacsModes
==============

  It indicates which Emacs modes will use this highlighting. The modes must be
separated by commas. The modes are't case sensitive.

  As in Emacs the editor looks at the first kilobyte of text and at the last
three kilobytes for the mode delimited by `-*-'.

  The priority is as follows: `EmacsModes', `ShellScript' and finally the
extension (`Files').  ("ShellScript" Section 8.26).  ("Files" Section 8.14).

8.6 EOLCInFirstCol
==================

  When this setting is on the one line comments starts only if the sequence of
characters is present in the first column.

8.7 EOLCInFirstCol1
===================

  That's like EOLCInFirstCol but affects only the EOLComment1 and not both.
("EOLCInFirstCol" Section 8.6).

8.8 EOLCInFirstCol2
===================

  That's like EOLCInFirstCol but affects only the EOLComment2 and not both.
("EOLCInFirstCol" Section 8.6).

8.9 EOLCInFirstUse1
===================

  That's like EOLCInFirstCol1 but the starting sequence may be the first
non-blank character and does not need to be located in the first column.
("EOLCInFirstCol1" Section 8.7).

8.10 EOLCInFirstUse2
====================

  That's like EOLCInFirstCol2 but the starting sequence may be the first
non-blank character and does not need to be located in the first column.
("EOLCInFirstCol2" Section 8.8).

8.11 EOLComment1
================

  Used to indicate the start of a comment that ends at the end of the line. The
lenght is limited to four characters. If the file format has two different
terminations use `EOLComment2' for the second.

  For more information  ("Format of short syntax highlighting definitions"
Section 8.21.1).

8.12 Escape
===========

  Indicates which character acts as escape character inside strings or to
concatenate lines.

8.13 EscapeAnywhere
===================

  Indicates that escape characters can be found anywhere, not just inside
strings or at the end of lines like in C.

8.14 Files
==========

  It indicates which extensions will use this highlighting mode. The extensions
must be separated by commas. The extensions are case sensitive, be careful.

  The editor can also choose the highlighting using the Emacs mode or the
program used to execute the file if that's an UNIX script.

  The priority is as follows: `EmacsModes', `ShellScript', `FullNameMatch',
`NameMatch' and finally the extension (`Files').  ("ShellScript" Section
8.26).  ("EmacsModes" Section 8.5).  ("FullNameMatch" Section 8.15).
("NameMatch" Section 8.19).

8.15 FullNameMatch
==================

  It can be used to indicate a regular expression to match the full path and
name of the file. If the expression matches the editor will use this syntax
highlighting.

  The regular expression must be in Perl format. You can read about it in any
book about Perl.

8.16 HexMarker
==============

  This setting indicates what prefix is used for hexadecimal numbers. No postfix
is supported yet. The lenght is limited to four characters.

  For more information  ("Format of short syntax highlighting definitions"
Section 8.21.1).

8.17 Keywords
=============

  It can be used as many times as needed and is used to indicate the reserved
keywords of the language. The separator is the comma.

8.18 Name
=========

  It sets the name of the syntax highlighting. This name is used in the local
settings dialog (cmcSetLocalOptions).

8.19 NameMatch
==============

  It can be used to indicate a regular expression to match the name of the
file. If the expression matches the editor will use this syntax highlighting
mode.

  The regular expression must be in Perl format. You can read about it in any
book about Perl.

8.20 NoCheckNumbers
===================

  When this setting is on the numbers aren't highlighted.

8.21 OpenComment1
=================

  Used to indicate the start of a multiline comment. The lenght is limited to
four characters. If the file format has two different starts use
`OpenComment2' for the second.

  The maximum length is four characters, for more information about the format:
("Format of short syntax highlighting definitions" Section 8.21.1).

8.21.1 Format of short syntax highlighting definitions
------------------------------------------------------

  All the characters after the equal sign are taken as part of the field. Blank
spaces after the equal sign are ignored, so `Field= //' is equivalent to
`Field=//'.

  As the field may start with spaces and spaces at the end of line are invisible
you can quote the text using the double quote letter. If you do it the first
character after the equal sign must be the double quote, if you left an space
like this: `Field= "' the double quote will be interpreted as part of the
field.

  When quoting text the back slash is the escape character, so this:
`Filed="\""' will be interpreted as asigning `"' to `Field'.

8.22 PartialKeywords
====================

  When this setting is enabled the list of the editor will highlight partial
matches of the keywords. For example, suppose a keyword is defined as `key'
and you type `keytable', in this case the editor will highlight the `key'
part of the word. This feature is experimental and makes the highlighting
quite slower.

8.23 PMacros
============

  Indicates the name of the pseudo macros file used for this kind of files.
Using different files for different formats you can assign common triggers to
different actions. For example, you can use i( to trigger an if() {} else {}
in C and an if then else in BASIC.  ("Pseudo Macros" Chapter 9).

8.24 Preprocessor
=================

  Indicates which symbol starts a preprocessor line.

8.25 RelaxNumberCheck
=====================

  This is used when a number can start a name and that isn't a error.  Normally
the editor takes it as a wrong number but when this option is enabled the
check for numbers is relaxed and they aren't highlighted as wrong values.

8.26 ShellScript
================

  It indicates which scripts will be highlighted. You must indicate the names
of the programs associated with this script. The names are case sensitive.

  The editor looks for the `#!' characters at the start of the file, if found
the name of the program is extracted and searched in this list.

  The priority is as follows: `EmacsModes', `ShellScript' and finally the
extension (`Files').  ("EmacsModes" Section 8.5).  ("Files" Section 8.14).

8.27 ShortString
================

  Used to indicate the start and end of strings, used for small strings or
characters.  The string ends at the end of line if not explicitly closed.

8.28 SpecialSymbol
==================

  That's used to mark pairs of characters that form a particular symbol that
will generate problems if the editor sees them as separated symbols. It was
created to avoid problems with `$#' in Perl and bash scripts and with `@@' in
Texinfo files. You can define more than one character here.

  The characters that can be combined with it are specified using the
`SpecialSymbolCont' definition.

8.29 String1
============

  Used to indicate the start and end of strings, used for long strings. You can
specify more than one character, in this case any of them can open or close
the string.

  The editor supports up to three diferent strings declared as `String1',
`String2' and `String3'.

8.30 Symbols1
=============

  Used to indicate which characters are allowed to be symbols, normally used
for boolean and arithmetic symbols.

8.31 Symbols2
=============

  Used to indicate what characters are allowed to be symbols, normally used for
flow control and subindex symbols.

8.32 UseInternal
================

  That's optional. When a highlighting mode has this setting the editor will
use the internal routines and will ignore almost all the settings. Only the
extensions ( "Files" Section 8.14) and pseudo macros ( "PMacros" Section
8.23) settings are used.

  The value assigned can be: 1 for GNU C highlighting, 2 for standard Pascal
and 3 for Clipper.

9 Pseudo Macros
***************

  This feature is very useful to save some keystrokes. With this feature you
can create a lot of shortcuts to make your life easier.

9.1 Please enlighten me - what is that?
=======================================

  So now, What's a Pseudo Macro? It's like a macro but it's triggered by two
things: one the text before the cursor, and two the `^Space' combination.

  And what's the result? The result is customizable, but by default there are
some predefined behaviours. For example, in a C source window type the
following two characters: `#i' and then press `^Space' ... (to create an empty
file with .c extension go to File|Open and type a new name, foo.c for
example).

  Surprised? I bet! You got: #include <.h> and the cursor just at the right
place to write the name of the header.

Now write the famous stdio word, press `<End>', then `<Enter>' twice and now
write the following two letters: `ma' and then `^Space' ...

  I bet this time you had an inkling of what will happen so you aren't that
surprised ;-). Anyways, Surprise! Now type `pr' and the magic keys and ...
`printf("");' appears. To end the happy history now type "Hello world!" That's
all. You wrote the hello world program at a very good speed.

  Now you know what I mean when I say pseudo-macros.

9.2 How can I customize that?
=============================

  It's very easy. Each syntax highlighting has its own pseudo macros file. Which
file belongs to which syntax highlighting is indicated in the `syntaxhl.shl'
file.  ("Syntax Highlighting File" Chapter 8). In the case of C/C++ the file
is called `cpmacros.pmc' and is located in the same directory where the rest
of the configuration files are installed. This file contains the definitions
for each pseudo macro that can be triggered in C/C++ files. You can define as
many pseudo macros as you want.  ("Configuration files location" Section
15.1).

  Important: If you are using RHIDE consult the RHIDE documentation to know
where the `cpmacros.pmc' file is stored.

  The `cpmacros.pmc' file is a very good example and is self-explanatory but
I'll include one example here.

Trigger: "i("
Mode: 0,1,0,1,0
 "if (@0)\n"
 "  {@1\n"
   "}\n"
 "\belse\n"
 "  {@2\n"
   "}"

  The Trigger keyword defines the two letters before the cursor that will
trigger the pseudo macro.

  The Mode keyword indicates the mode that the editor will use when inserting
the text. The modes are:

   * Overwrite                ("Overwrite" Section 7.1).

   * Autoindent               ("Autoindent" Section 7.2).

   * Use Tabs                 ("Real Tabs" Section 7.3).

   * Persistent blocks        ("Block modes" Section 3.4.1).

   * Intelligent C indent     ("Intelligent C indent" Section 7.5).

  Each mode can be 0 or 1. The editor will return to the original mode after
expanding the pseudo macro. In this case the macro is expanded like this: No
overwrite, autoindent, don't use tabs, the blocks are persistent and don't be
smart when indenting.

  The rest is the code to insert enclosed by ". You can use \b to indicate
backspace, \n for newline and \\ to indicate a simple \. If you want to
insert a @ you have to type it twice @@, because this character has a special
meaning (see below).

  After the insertion the cursor is positioned in the place marked with @0.
Don't forget to signal this point or the cursor will be positioned at the
beginning of the file. The places marked with @1, @2 and @3 are saved in the
markers 7, 8 and 9.

  Currently I defined pseudo macros only for C/C++ and Perl. I did it in a way
that you can use the same pseudo macro for both languages. For example: `#i'
is expanded to `#include <.h>' for C/C++ files and to `request "";' for Perl
files. If you write similar pseudo macros for other languages please send me
the file and I'll add it to the next release.

10 sLisp macros
***************

  The editor uses a lisp like language to store macros. The macros are stored
in a file called `macros.slp'.  ("Configuration files location" Section 15.1).

  The macros can be assigned to keys ( "How to configure the keyboard" Section
4.1), selected from the `Macro' menu or from the menu. From the `Macro' menu
you can choose a macro from the list. If you want to repeat it you can do so
with one keystroke.

  To add a new macro to the list you must edit the `macros.slp' file. You can
write the new macro by hand or you can use the `Macro' menu to record a macro
and then generate the sLisp code for this macro.

  To assign a macro to a menu entry you must edit the `menubind.smn'. The macros
are called using `cm(name)' where `name' is the name of the macro you want to
trigger.

  To assign a piece of sLisp code to a menu entry you must use `cm((code))'.
Here is an example:

      MenuItemC: "Test macro from menu", cm((ShowInMessageWindow "Hi! ;-)"))

  Don't insert spaces between the starting and ending parentheses pairs.

  You can also assing a piece of sLisp code to a key.   ("How to configure the
keyboard" Section 4.1).

10.1 How to write a sLisp macro
===============================

  I'll show you an example and then I'll explain each part of the example:

     (defmacro 'Testing 1 2 3 ;-)'
      (eval
       (SendCommands cmcLineEnd)
       (InsertText (+ 'Hi!' CR 'How are you?'))
       (SendCommands cmcLineUp cmcLineEnd)
      )
     )

  That defines a macro called "Testing 1 2 3 ;-)" that will be expanded to all
the code inside of the `eval' sentence. The `SendCommands' command sends one
or more commands to the editor. The `InsertText' command inserts one string
in the code.  To concatenate strings use the `+' operator. Currently you can
use the `\n' escape sequence inside a string to indicate a carriage return,
but in the example above `CR' - a constant - is used to concatenate strings.

10.2 How strings are parsed
===========================

  sLisp strings can be delimited by single or double quotes (`'' or `"').
Currently both produce the same result but in the future it could change to
be similar to Perl.

  A few C escape sequences are supported inside strings. This means the `\ '
character has a special meaning. Here are the values supported:

   * n
     - Interpreted as: an OS dependant carriage return (rn for DOS and n for
     UNIX)

   * l
     - Interpreted as: simple line feed (ASCII 10 for DOS and UNIX)

   * r
     - Interpreted as: simple carriage return (ASCII 13 for DOS and UNIX)

   * t
     - Interpreted as: tab (ASCII 9 for DOS and UNIX)


  In strings that will be inserted in the text you should use the `\\n' escape
sequence to maintain compatibility across different OSs. You can also use the
built-in constant called `CR'.

10.3 Running programs with a macro
==================================

  There is a sLisp command called `RunProgram'. It takes a string as parameter
which can contain one or more programs separated by `;' or the commands
separator used by your command shell.

  The standard output and error are redirected and the results are sent to the
Message Window. If the program is a GNU tool and reported errors the editor
will parse these errors and will allow you to directly jump to the file and
line where the error was reported.

To learn more about the message window  ("Message Window" Section 15.8).

10.4 Editor specific commands
=============================

  This section describes commands related to files being edited.

10.4.1 AskString
----------------

  (AskString TITLE MESSAGE)

  Pops up a dialog asking for user input. The text is returned as an sLisp
string variable. If the user canceled the input, the string has zero length.

10.4.2 BindKey
--------------

  (BindKey KEY_SEQ MACRO)
  (BindKey KEY_SEQ COMMAND ...)

  This command creates a key binding data type. It doesn't change anything
alone and must be used with the `KeyBindings' command.

  You can assign a defined sLisp macro or a piece of sLisp code using the first
syntax. When assigning a sLisp code it must begin with `(' and end with `)'.

  You can also assign a one command or a sequence of commands using the second
syntax. The commands are specified using the same constants used for
`SendCommands'.  ("SendCommands" Section 10.4.28).

  The KEY_SEQ argument must be a string containing the key sequence used to
trigger this definition.

  The following example defines two keyboard sequences. The first is triggered
when <Ctrl> and <K> are pressed together, then released and then <Z> is
pressed (all in lower case mode). This sequence moves the cursor one line up
and then one character to the left. The second sequence (`^K-X') triggers a
small sLisp sentence that prints a message in the message window.

     (KeyBindings
      (BindKey "CtK,Z" cmcLineUp cmcCharLeft)
      (BindKey "CtK,X" "(ShowInMessageWindow 'Hi! ;-)')")
     )

  When writing a key sequence you can use the `Insert key name' menu option if
you don't know the name of a key combination.  ("Insert key name" Section
5.7.8).

  An important detail is that this command can replace already defined keys and
can even replace a group of assignments by the new definition. You can assign
a command to `CtK' removing all the `^K' assigments during this operation.

10.4.3 ComplChoose
------------------

  (ComplChoose OPTIONS DELIMITER [FLAGS])

  Brings up a floating drop-down list with the options. The OPTIONS parameter
is a delimited string, the delimiter is specified by the DELIMITER parameter.

  The function returns the option selected by the user or an empty string if
the user aborted.

  The list is sorted alphabetically, the user can choose the item using the
cursors or doing an incremental search. Only characters allowed for reserved
words are accepted, any symbol will choose the currently selected item, other
characters will abort.

  The FLAGS parameter fine tunes the behavior of the routine. Currently only
one thing can be changed. By default the string returned contains the
character that produced the selection appended at the end of the string.
Passing 1 for FLAGS the character isn't concatenated.

10.4.4 defmacro
---------------

  (defmacro NAME EXPRESSION)

  Defines a new editor macro called NAME. When this macro is called the
EXPRESSION is evaluated.  ("How to write a sLisp macro" Section 10.1).

10.4.5 EvalString
-----------------

  (EvalString STRING)

  Executes the sLisp code contained in the STRING variable.

10.4.6 FindAgain
----------------

  (FindAgain)

  Repeats the last search. It should be used when you didn't gave control to
the user since the last search because the user could perform a different
search changing the compiled search. Returns the text found or an empty string
if no match could be found.  ("FindString" Section 10.4.7).

10.4.7 FindString
-----------------

  (FindString STRING [OPTIONS_FLAGS])

  Performs a search for the provided STRING. Returns the text found or an empty
string if no match could be found. This operation moves the cursor position
as in an interactive search. The match becomes temporally highlighted in the
text.

  Even when the OPTIONS_FLAGS argument is optional it is strongly recommended
to provide it. If nothing is provided the user options will be used instead.
The options are provided as flags combined with the `|' operand. Some of the
flags doesn't have any sense when combined with others, you should avoid
doing it. The available flags are:

Search style:

   * edfCaseSensitive: perform a case sensitive search.

   * edfWholeWordsOnly: don't match partial words.

   * edfRegularEx: the string is a regular expression. Note this flag must be
     indicated in adition to the type of regular expressions used.

Scope:

   * edfSearchInComm: search only inside comments.

   * edfSearchOutComm: search only outside comments.

   * edfOnlySelection: search only inside the selected text.

   * edfFromCursor: start searching from the cursor position. That's the
     default action.

   * edfFromBeggining: start searching from the beggining of the selected
     scope.

Regular expressions options:

   * edfBasicRegEx: use basic POSIX regular expressions. Used in combination
     with edfRegularEx.

   * edfExtendedRegEx: use extended POSIX regular expressions. Used in
     combination with edfRegularEx.

   * edfPerlRegEx: use Perl regular expressions. Used in combination with
     edfRegularEx.

   * edfOptimizeRegEx: if the string to search doesn't contain special
     characters and regular expressions are enabled perform a normal search.
     This makes the search much more faster. This is the default action so
     you don't really need to specify this flag.

   * edfNoOptimizeRegEx: it is the opposite of edfOptimizeRegEx flag. It
     disables the optimization. Use it when you are sure the string is a
     regular expression even when it doesn't look like a regular expression.
     This option was created just in case the optimization fails to detect a
     real regular expression.

Others:

   * edfShowFuncName: display the name of the function where the match was
     found in the status line.

  A very important detail about regular expressions: In sLisp the `\' character
is the escape character. It means that you have to use `\\' to get `\'.

  The following example searchs for the `Hello' word followed by a digit using
Perl regular expressions. Then prints the result of the search in the message
window and after it repeats the search.

     (eval
      (= "str" (FindString "Hello\\d" (| edfRegularEx edfPerlRegEx)))
      (if str
         (ShowInMessageWindow (+ "Found: " str))
         (ShowInMessageWindow "Not found.")
      )
      (= "str" (FindAgain))
      (if str
         (ShowInMessageWindow (+ "Found: " str))
         (ShowInMessageWindow "Not found.")
      )
     )

  To learn more about the search options consult the find command ( "Find"
Section 5.3.1) and the regular expression options ( "Regular Expressions
Options" Section 5.3.1.1). You could be also interested in learning more
about POSIX regular expressions ( "Regular Expressions" Section 15.4), for
Perl regular expressions consult some Perl reference. This command is
complemented by `FindAgain' ( "FindAgain" Section 10.4.6), `ReplaceString' (
"ReplaceString" Section 10.4.23) and `ReplaceAgain' ( "ReplaceAgain" Section
10.4.22) commands.

10.4.8 ForceUpdate
------------------

  (ForceUpdate)

  It forces an update at this point of the macro. That's useful if you want to
stop to ask the user something or the macro will take some time to complete
and don't want to give the impression the editor hangs.

10.4.9 GetCursorX
-----------------

  (GetCursorX)

  Returns the X cursor coordinate. Note the coordinates are computed from zero.

10.4.10 GetCursorY
------------------

  (GetCursorY)

  Returns the Y cursor coordinate. Note the coordinates are computed from zero.

10.4.11 GetCurWindowNumber
--------------------------

  (GetCurWindowNumber)

  Returns the window number for the currently selected edition window.

10.4.12 GetMaxWindowNumber
--------------------------

  (GetMaxWindowNumber)

  Returns the window number for the edition window with the biggest window
number. The lowest number is two.

10.4.13 GetSystemInfo
---------------------

  (GetSystemInfo WHICH)

  Returns information about the system and similar stuff. The WHICH argument
selects which kind of information is returned. Available values are:

   * edfInfTVDriver: The short name of the Turbo Vision driver in use.  Some
     of the possible drivers are: DOS, Linux, QNX, QNX4, UNIX, Win32, WinGr,
     WinNT, X11 and XTerm.

   * edfInfOS: The name of the operating system. Some of the possible names
     are: DOS, UNIX and Win32.

   * edfInfOSFlavor: Some of the possible flavors are: Linux, FreeBSD,
     Solaris and QNXRtP.

   * edfInfCPU: The CPU family used to compile the editor. Some of the
     possible names are: x86, Alpha, SPARC64, SPARC, PPC, HPPA, MIPS, Itanium
     and Unknown.

   * edfInfCompiler: The compiler used to compile the editor. Some of the
     possible compilers are: GCC, BCPP and MSVC.

   * edfInfCompilerFlavor: Some of the possible flavors are: djgpp, MinGW and
     Cygwin.

10.4.14 getenv
--------------

  (getenv NAME)

  This command returns the content of the environment variable called NAME.
Internal variables of the editor are also accessable. If the variable doesn't
exist an empty string is returned.

10.4.15 GetSelection
--------------------

  (GetSelection)

  This command returns the selected text in the editor as a sLisp string.
("How to write a sLisp macro" Section 10.1).

10.4.16 GetSyntaxAtCursor
-------------------------

  (GetSyntaxAtCursor)

  This command returns the syntax highlighting flags for the cursor position.
That's very specific and can be used to find out if the cursor is inside a
comment, string or preprocessor code.

  To understand how it works use the following macro and trigger it in many
places to see what you get:

     (defmacro 'Test edfWE'
       (eval
        (ShowInStatusLine (+ "" (GetSyntaxAtCursor)))
       )
     )

  The following table explains the flags returned:

   * edfComInside
     - Meaning: Inside an end-of-line comment

   * edfEndCom
     - Meaning: This line ends a multiline comment

   * edfEndCom2
     - Meaning: This line ends a multiline comment type 2

   * edfExtCom
     - Meaning: This line extends a multiline comment

   * edfExtCom2
     - Meaning: This line extends a multiline comment type 2

   * edfExtOneLineCom
     - Meaning: The end-of-line comment is extended to the next line

   * edfExtPrepro
     - Meaning: The preprocessor line follows in the next

   * edfExtString
     - Meaning: Inside a string

   * edfExtString2
     - Meaning: Inside a string type 2

   * edfExtString3
     - Meaning: Inside a string type 3

   * edfInsideCom
     - Meaning: Inside a multiline comment

   * edfInsideCom2
     - Meaning: Inside a multiline comment type 2

   * edfPrepro
     - Meaning: Inside preprocessor code

   * edfStartCom
     - Meaning: This line starts a multiline comment

   * edfStartCom2
     - Meaning: This line starts a multiline comment type 2

   * edfStartInCom
     - Meaning: This line is commented out by the previous

   * edfStartInCom2
     - Meaning: This line is commented out by the previous (type 2)

   * edfStartString
     - Meaning: This line startes inside a string

   * edfStartString2
     - Meaning: This line startes inside a string type 2

   * edfStartString3
     - Meaning: This line startes inside a string type 3


  Something very important is that the editor will only analyze the line up to
the cursor's position. So if the cursor is inside a string the editor will
think this line extends a string.

10.4.17 GetSyntaxLang
---------------------

  (GetSyntaxLang)

  Returns the language used for syntax highlight in the current editor. If none
syntax highlight is in use it returns an empty string.

10.4.18 InsertText
------------------

  (InsertText STRING [SELECTED [MOVE]])

  Inserts the STRING at the cursor's position. By default the cursor is moved
behind the insertion and the inserted text isn't selected. If SELECTED is 1
the text is selected. If the MOVE parameter is 0 the cursor isn't moved and if
the parameter is 1 the action depends on the current editor's setup.

10.4.19 KeyBindings
-------------------

  (KeyBindings KEY_BINDING ...)

  Adds the KEY_BINDING to the keyboard definitions. As the operation to insert
a key in the keyboard definitions is slow this command is used to group all
the operations in just one operation. Currently the only command that returns
a key binding is `BindKey'.  ("BindKey" Section 10.4.2).

  It returns zero if the operation failed and one if the operation succeed.  If
any of the specified key bindings fails the keyboard configuration isn't
altered.

10.4.20 MessageBox
------------------

  (MessageBox MESSAGE [OPTIONS])

  Pops up a small dialog showing the MESSAGE to the user. The OPTIONS can take
the following values (use 'or' to combine them):

   * edfMBCancelButton: has a `Cancel' button.

   * edfMBNoButton: has a `No' button.

   * edfMBOKButton: has an `Ok' button.

   * edfMBOKCancel: has an `Ok' and a `Cancel' button.

   * edfMBYesButton: has a `Yes' button.

   * edfMBYesNoCancel: has a `Yes', a `No' and a `Cancel' button.

   * edfMBConfirmation: use Confirmation as caption.

   * edfMBError: use Error as caption.

   * edfMBInformation: use Information as caption.

   * edfMBWarning: use Warning as caption.

10.4.21 OpenFile
----------------

  (OpenFile STRING)

  Opens the file STRING in a window and makes it the current one. If the file
doesn't exist it does nothing. Always returns 1.

10.4.22 ReplaceAgain
--------------------

  (ReplaceAgain)

  Repeats the last search and replace operation. It should be used when you
didn't gave control to the user since the last search because the user could
perform a different search changing the compiled search. Returns a boolean
value indicating the status of the last search.  ("ReplaceString" Section
10.4.23).

10.4.23 ReplaceString
---------------------

  (ReplaceString STRING REPLACEMENT [OPTIONS_FLAGS])

  Performs a search and replace operation. Returns a boolean value indicating
the status of the last search. Consult the `Find' command ( "FindString"
Section 10.4.7) and the replace options of the editor ( "Replace" Section
5.3.2) for more information.

  In addition to the `Find' options you have the following options:

   * edfPromptOnReplace: ask confirmation to the user.

   * edfReplaceAll: repeat the search and replace operation until no match
     can be found in the current scope.

   * edfNormalText: used to indicate that the REPLACEMENT string is a simple
     text and not a replacement template. This is the default behavior.

   * edfTagsText: used to indicate that the REPLACEMENT string contains `$n'
     tags. For more information consult the section for the regular
     expressions dialog of the editor ( "Regular Expressions Options" Section
     5.3.1.1).

  The following example searchs for the `Hello' word followed by a digit using
Perl regular expressions and replaces the text by the digit followed by an
space and the `Bye' word. Then prints the result of the operation in the
message window.

     (eval
      (if (ReplaceString "Hello(\\d)" "$1 Bye"
           (| edfRegularEx edfPerlRegEx edfTagsText))
       (ShowInMessageWindow "Replaced.")
       (ShowInMessageWindow "Not replaced.")
      )
     )

10.4.24 RunProgram
------------------

  (RunProgram PROGRAM_NAME [[FLAGS] PARSER])

  Calls the desired program. You can pass more than one using the `;' separator
or any separator supported by your shell. Under DOS `;' is ok even when
command.com doesn't support it. The stderr and stdout are redirected and
captured by the message window of the editor.

  Use the operator `+' to concatenate strings and pass arguments.

  Using `edfRunUserScreen' for the FLAGS argument the editor will use the user
screen when running the external program. Using `edfRunNoRedirect' the editor
won't redirect the standard output. For interactive applications that uses
standard input and output you should both flags combined using the or
operator.  ("Operator | (bitwise or)" Section 10.5.19).

  Another option you can use is `edfRunStopDebug'. This option will stop any
active debug session. That's a very delicated option and can be used when the
macro is calling a compiler to rebuild an executable. In some OSs you must
stop the debugger in order to be able to overwrite the executable.
Additionaly gdb have some bugs that needs it as workaround.

  The PARSER parameter tells the editor which error parser will be used to
parse the output of the external program. By default the GNU one is used.

  To learn more about the message window  ("Message Window" Section 15.8).

10.4.25 RunProgramRedir
-----------------------

  (RunProgramRedir PROGRAM_NAME [INPUT_TEXT])

  This function is similar to RunProgram, but the result isn't dumped to the
message window, instead the output of the program is returned as a string.

  Additionally you can specify a text to send to the standard input of the
program you are calling. You can easily call external filters with it. I
provide an example of how to replace a selected text with the output of an
external filter in the `macros.slp' file.  ("Configuration files location"
Section 15.1).

  Be careful, that's a very powerful command, but very dangerous too.

10.4.26 SelectionExists
-----------------------

  (SelectionExists)

  Returns 1 if some text is selected and the selection is visible.

10.4.27 SelectWindowNumber
--------------------------

  (SelectWindowNumber NUMBER)

  Selects the indicated window. Positive numbers are reserved for edition
windows, except window number one that's the project window. Negative numbers
are used for special windows. The function returns a boolean value indicating
if the desired window was selected.

  When you select an edition window it becomes the target of all sLisp
operations.

  The available constants for special windows are:

   * edfWinASCII

   * edfWinCalendar

   * edfWinClipboard

   * edfWinHelp

   * edfWinMP3

   * edfWinMan

   * edfWinMessage

   * edfWinPrj

10.4.28 SendCommands
--------------------

  (SendCommands COMMAND ...)

  Sends all the listed commands to the editor. The editor commands are sLisp
constants that start with `cmc'. These commands are the same explained in the
keyboard section as `cmcXXXXX'.

10.4.29 SetCursorXY
-------------------

  (SetCursorXY X [Y])

  Sets the cursor coordinates. Note the coordinates starts from zero. If the Y
argument is omitted only the X coordinate is changed. This function returns
the current Y coordinate, it doesn't have to be the same value you indicated
in the Y argument because the editor limits this value to the last line in
the current buffer.

10.4.30 ShowInMessageWindow
---------------------------

  (ShowInMessageWindow STRING [CLEAR])

  Prints the desired string in the message window. That's a very good way to
show a result to the user.

  If the message window isn't selected or visible it gets the focus and becomes
visible. The optional parameter allows you to clean the contents of the
window.  ("Message Window" Section 15.8).

10.4.31 ShowInStatusLine
------------------------

  (ShowInStatusLine STRING)

  Prints the passed string in the status line of the current editor. Tabs are
converted to one space and the message stops at the first carriage return or
line feed. That's a very good way to show a result to the user.

10.4.32 WhichEditor
-------------------

  (WhichEditor [OPTION])

  This command takes one optional parameter and returns the file name of the
current file under edition.

  The optional parameter can be:

   * edfWEFull (0)
     - Meaning: Full name and path
     - Example: `c:/temp/test.txt'

   * edfWEFullNoExt (1)
     - Meaning: Same as 0 but without extension
     - Example: `c:/temp/test'

   * edfWEPath (2)
     - Meaning: Path for the file
     - Example: `c:/temp/'

   * edfWEDisk (3)
     - Meaning: Disk drive
     - Example: `c:'

   * edfWEExtension (4)
     - Meaning: File extension
     - Example: `.txt'

   * edfWENameNoExt (5)
     - Meaning: Name without extension
     - Example: `test'


10.4.33 WordUnderCursor
-----------------------

  (WordUnderCursor [MAX_LENGTH [OPTIONS]])

  This command takes one optional parameter and returns the word that's located
under the cursor. That's very useful for things like searching help about the
function the user is typing or things like that.

  The maximum length is optional and if you don't specify it 256 is used.
Values under 4 or over 32768 are adjusted to fit this range for security
issues.

  The OPTIONS parameter is 0 by default. Passing 1 the editor will return the
word located to the left of the cursor, if the cursor is located at the first
character after this word.

10.5 General sLisp commands
===========================

  This section describes general sLisp commands that can be used from the
editor and the SDG configuration files.

10.5.1 and
----------

  (and EXP1 EXP2 ...)

  Returns the logical and of the provided expressions. The values are evaluated
from left to right. If at any point the result is zero then zero is returned
and the rest isn't evaluated.

  Each value is evaluated as one or zero.

  (and 20 "hello" 4)
  Returns 1.

  (and 20 4 2)
  Returns 1.

10.5.2 Operator = (assign)
--------------------------

  (= VARIABLE VALUE)

  This is a shortcut for the `setv' command.  ("setv" Section 10.5.24).

10.5.3 Operator & (bitwise and)
-------------------------------

  (& EXP1 EXP2 ...)

  Returns the bitwise and of the provided expressions. The values are evaluated
from left to right. If at any point the result is zero then zero is returned
and the rest isn't evaluated.

  (& 20 4 2)
  Returns 0.

  (& 20 4)
  Returns 4.

10.5.4 cond
-----------

  (cond CONDITION EXP [CONDITION2 EXP2 ...])

  Evaluates the conditions until one is true and returns the associated
expression. If all conditions are false it returns 0.

  (cond 0 1 2 3)
  Returns 3.

  (cond 1 "Hi!" 0 "Bye")
  Returns "Hi!".

10.5.5 Operator - (decrement)
-----------------------------

 (- EXPRESSION)

  Returns the integer expression decremented by one. Currently only integer
values are supported.

  If the EXPRESSION is an string and it corresponds to the name of a defined
integer variable the variable is decremented and the resulting value is
returned.

  (- 8)
  Returns 7.

  (setv "a_var" 4)   (++ "a_var")
  Returns 3.

10.5.6 Operator != (different)
------------------------------

  (!= EXP1 EXP2)

  Returns  zero if both expressions represent the same value. Currently both
expressions must be of the same type.

  (== 1 2)
  Returns 1

  (== 2 (+ 1 1))
  Returns 0

  (== "Hello" "Hello")
  Returns 0

10.5.7 Operator == (equal)
--------------------------

  (== EXP1 EXP2)

  Returns an integer different than zero if both expressions represent the same
value. Currently both expressions must be of the same type.

  (== 1 2)
  Returns 0

  (== 2 (+ 1 1))
  Returns 1

  (== "Hello" "Hello")
  Returns 1

10.5.8 eval
-----------

  (eval EXPRESSION ...)

  This command is used to specify more than one action where just one is
expected.  Additionally you could write sLisp code in a string and then
evaluate it. The expressions are evaluated from left to right, the last
evaluated result is returned.

10.5.9 exitloop
---------------

  (exitloop )

  Exits from a loop.  ("loop" Section 10.5.15).

10.5.10 gstr
------------

  (gstr STRING POSITION)

  Returns the POSITIONth character of the STRING. The first character is in the
0 position.

  (gstr "hello" 4)
  Returns "o".

10.5.11 if
----------

  (if CONDITION STATEMENT_1 [STATEMENT_2])

  It evaluates CONDITION, if the boolean value of it is true then STATEMENT_1
is evaluated. If the value is false and you provided a second statement then
STATEMENT_2 is evaluated. If the value is false and the second statement is
missing nothing is evaluated and the resulting value is 0.

  (if "hello" 4)
  Returns 4.

  (if "" 4 3)
  Returns 3.

  (if 0 4)
  Returns 0.

10.5.12 Operator ++ (increment)
-------------------------------

 (++ EXPRESSION)

  Returns the integer expression incremented by one. Currently only integer
values are supported.

  If the EXPRESSION is an string and it corresponds to the name of a defined
integer variable the variable is incremented and the resulting value is
returned.

  (++ 8)
  Returns 9.

  (setv "a_var" 4)   (++ "a_var")
  Returns 5.

10.5.13 left
------------

  (left STRING NUMBER)

  Returns the first NUMBER characters of the STRING.

  (left "hello" 4)
  Returns "hell".

10.5.14 length
--------------

  (length STRING)

  This function returns the number of characters in STRING.

  (length "hello")
  Returns 5.

10.5.15 loop
------------

  (loop EXPRESSION ...)

  Evaluates the expressions until an `exitloop' command is found.

  (= "i" 0)
  (loop
   (InsertText (tostr i))
   (if (== i 5) (exitloop))
   (++ "i")
  )
  Inserts "012345" text.

10.5.16 Operator ~ (negated)
----------------------------

  (~ VALUE)

  Returns the VALUE negated (at bit level).

  (~ "hello")
  Returns 0.

  (~ "")
  Returns 0xFFFFFFFF.

  (~ 1)
  Returns -2.

10.5.17 not
-----------

  (not VALUE)

  Returns the VALUE negated.

  (not "hello")
  Returns 0.

  (not "")
  Returns 1.

  (not 1)
  Returns 0.

10.5.18 or
----------

  (or EXP1 EXP2 ...)

  Returns the logical or of the provided expressions. The values are evaluated
from left to right. If any of them is non-zero the result is non-zero and the
rest isn't evaluated.

  (or 20 4 2)
  Returns 20.

10.5.19 Operator | (bitwise or)
-------------------------------

  (| EXP1 EXP2 ...)

  Returns the bitwise or of the provided expressions. The values are evaluated
from left to right. If at any point all bits are one the rest of the values
isn't evaluated.

  (| 20 4 2)
  Returns 22.

10.5.20 Operator + (plus)
-------------------------

  (+ VALUE1 VALUE2 ...)

  Returns the result of adding all the parameters. The type of the result is
determined by the first parameter. Currently you can't mix strings and
integers but if you bug me enough I'll add it. Adding strings means
concatenation.

  (+ "hel" "lo")
  Returns "hello".

  (+ 3 5)
  Returns 8.

  (+ "hello" 1)
  Syntax error.

10.5.21 prex
------------

  (prex TEXT REGEX)

  Performes a search using the REGEX Perl regular expression in the specified
TEXT.

  Returns the number of matchs. This value is 0 if the search failed and at
least one if succeed. To retreive the results of the matches you can use
internal variables called `_X', where X is the number of the match. The match
number 0 is the main result, the others correspond to the subexpressions
contained in REGEX. Note that variables with numbers above the bigest
subexpression number are undefined or could contain the result of a previous
search.

  A very important detail is that back slash characters are escape characters
in sLisp string. It means that `"\\d"' is equivalent to `\d'.

  The following example shows `123' in the status line of the current edition
window:

  (eval    (prex "Hi 123 bye 456 nope" "i (\\d*) b")    (ShowInStatusLine _1)
)

10.5.22 repeat
--------------

  (repeat TIMES EXPRESSION ...)

  Evaluates the expressions until an `exitloop' command is found or the TIMES
number of iterations is reached.

  (= "i" 0)
  (repeat 6
   (InsertText (tostr i))
   (++ "i")
  )
  Inserts "012345" text.

10.5.23 right
-------------

  (right STRING NUMBER)

  Returns the last NUMBER characters of the STRING.

  (right "hello" 3)
  Returns "llo".

10.5.24 setv
------------

  (setv VARIABLE VALUE)

  Assigns VALUE to the VARIABLE. If the VARIABLE wasn't yet defined the
variable is created. The VARIABLE parameter must be a string representing the
name of the variable, it may also be an expression. This command have a
shortcut.  ("Operator = (assign)" Section 10.5.2).

  Variable names can contain underscores, but the names starting with
underscores are reserver for internal use.

  (setv "counter" 1)
  Assigns 1 to the COUNTER variable, if COUNTER doesn't exist counter is
created.

  (setv (+ "counter" "1") 1)
  Assigns 1 to the COUNTER1 variable, if COUNTER1 doesn't exist counter1 is
created.

10.5.25 ShortFileName
---------------------

  (ShortFileName FILE_NAME)

  Returns the short name for FILE_NAME. That's useful only for Windows and if
long file names are supported. It can be used to pass a file name to an
external program that doesn't support long file names.

10.5.26 sstr
------------

  (sstr STRING POSITION VALUE)

  Changes the POSITIONth character of STRING to the first character of VALUE.
The first character is at 0.

  (sstr "hello" 1 "a")
  Returns "hallo".

10.5.27 strcasecmp
------------------

  (strcasecmp S1 S2)

  This function is the same as `strcmp' but this version isn't case sensitive.
("strcmp" Section 10.5.28).

10.5.28 strcmp
--------------

  (strcmp S1 S2)

  This function compares S1 and S2. The returned value is zero if the strings
are equal, a positive number if S1 comes after S2 in the ASCII table, else a
negative number.

10.5.29 strstr
--------------

  (strstr STRING SEARCH [START_POS])

  Searches the string SEARCH in the string STRING starting at START_POS
position. If START_POS is not provided the search is done from the beginning
of STRING. The returned value is the offset of SEARCH in STRING or -1 if the
value wasn't found.

  (strstr "hello" "el")
  Returns 1.

10.5.30 strxlt
--------------

  (strxlt STRING SEARCH REPLACE)

  Searches for the character listed in SEARCH in the passed STRING replacing
each ocurrence by the correponding value found in REPLACE.

  (strxlt "hello" "eo" "12")
  Returns "h1ll2".

10.5.31 substr
--------------

  (substr STRING POSITION [LEN])

  Returns LEN characters of STRING starting at the indicated POSITION.  If LEN
is omitted all the characters in the string starting at POSITION are
returned. The first character is at 0.

  (substr "hello" 1 3)
  Returns "ell".

  (substr "hello" 2)
  Returns "llo".

10.5.32 Operator - (substraction)
---------------------------------

  (- VALUE1 VALUE2)

  Returns the result of substracting VALUE2 from VALUE1. Both variables must be
integer.

  (- 20 3)
  Returns 17.

10.5.33 tostr
-------------

  (tostr VALUE)

  Returns VALUE converted into a string.

  (tostr 2)
  Returns "2".

10.6 Writing macros that use text filters
=========================================

  The sLisp language provides the commands needed to take a selected text, pass
it to an external program, and collect their output. You can finally replace
the original text by the new text. This kind of programs are usually called
filters. Some applications of this are complex recoding using GNU recode,
text formating, code indentation using GNU indent, `etc.'

  The following sections were contributed by Grzegorz Adam Hankiewicz and
explains how to use this feature.

10.6.1 How to use Setedit for something it was not meant to
-----------------------------------------------------------

SET has always said that setedit is not a word processing program. He's
right, it's best at doing things like programming. But once you get used to
it the road to building another emacs is clear: you start wanting to use it
even to cook your breakfast. This document, however, centers on how to
control text format easily under setedit, which is not that ambitious :)

As you will notice, in some version SET added the "word wrap" option. IMHO it
should be removed, if you try to use it, it will behave quite differently
from what you expect (ie: if you are writting tabbed lines, the editor will
ignore them and puts your cursor right away in the first column ignoring any
possible autoindentation options and tab uses you might have set for the
document). Also, it's setedit's choice of characters which will be used to
split the line. Bleah... it really sucks.

But setedit is customizable. You can write macros in a lisp like language.
The language still doesn't allow you to control tightly things like text
formatting, but there's one cool option: you can pipe a block selection to
another external program and the editor will substitute it with the output of
this external program. Certainly not the most optimized way of doing things,
but flexible enough to write a custom extension which treats your text.

And here we go... the purpose will be to create a macro, which feeds a text
block to an external program which we will write. Simple as that, all the
power you wanted at your finger tips.

10.6.1.1 Step 1 - Building your macro
.....................................

Macros are written in sLisp, and stored somewhere in a `macros.slp' file.
This file can be a global read-only file or it can be located at
`~/.setedit/macros.slp'.  It's pure text, and somewhere (ie: at the end of
the file) you will insert the following chunk of code:

     ;*******************************************************************
     ;
     ; MACRO:   Block reformat
     ;
     ; DESCRIPTION: This script gets the current selection and feeds it to
     ; an external program. If no selection is present a message box will
     ; be printed.
     ;
     ;*******************************************************************/

     (defmacro 'Block reformat'
      (eval
       (setv "input" (GetSelection))
       (if (length input)
        ; Call the filter
        (eval
         (setv "output"
          (RunProgramRedir
           ; filter line here
           "~/project/email-fmt.py/email-fmt.py "
           input
          )
         )
         (SendCommands cmcCut)
         (InsertText output 1)
        )
        ; Ask the user to select something first
        (MessageBox "Please select a text first")
       )
      )
     )

Ok, don't worry about the syntax of the macro or how it's written, if you are
curious you can take a look at setedit's documentation to find out what all
that mess means. The important line is the one which follows the comment `;
filter line here'. That line is actually the path to the external program we
will be using, which can be stored anywhere on your computer, it only needs to
have execution permission. I am the only user of my machine, so I don't need
to put things under /local/bin to feel comfortable :) Please change that to
point at your filter script.

10.6.1.2 Step 2 - The filter program
....................................

The filter program has to be simple: receives text lines as input and prints
the result to stdout. You can say: `Hey! that calls some weird Python script
which I don't have... why don't I call GNU's fmt command instead?'.  Good
point. Do it. If you are meaning to write simple texts that's ok. In fact,
after the command you can specify parameters to fmt like the width of the
lines, and there you go, you have a text formatting macro in Setedit. But if
you are meaning to answer emails (ie: setedit is your default editor and mail
programs like mutt call it to write messages) or do something more
sophisticated, you better use something like this (big chunk of code follows):

     #!/usr/bin/env python

     import sys, popen2, tempfile, os

     class LINE:
        pass

     indentation = " \t>:;#"

     def detect_indentation_characters(line):
        """Returns number of indentation chars minus trailing whitespace."""
        count = 0
        for f in line:
           if f in indentation: count = count + 1
           else: break

        return len(line[:count].rstrip()), count

     def initial_scan(line):
        """Add attributes to the lines."""
        l = LINE()
        l.line = line
        l.total_length = len(line)
        l.ident, l.soft_ident = detect_indentation_characters(line)
        l.length = l.total_length - l.soft_ident
        return l

     def same_indentation_lines(lines):
        """Returns number of lines with same indentation level."""
        f = 1
        while f < len(lines):
           if lines[f].ident != lines[0].ident:
              break
           else:
              f = f + 1

        return f

     def reformat_lines(lines, temp_filename):
        """Dumps lines in list to file, the reads fmt's output."""
        assert lines
        ident = lines[0].ident
        length = max (70 - ident, 20)

        # create temporary file with content to ident
        file = open(temp_filename, "wt")
        num = 0
        while num < len(lines):
           string = lines[num].line[ident:]
           stripped = string.lstrip()
           if stripped:   file.write(stripped)
           else:          file.write(string)
           num += 1
        file.close()

        # call external tool and read it's stdout
        stdout, stdin = popen2.popen2 (["fmt", "-w", "%d" % length,
           "-u", temp_filename])
        stdin.close()
        if ident:   padding = "%s " % lines[0].line[:ident]
        else:       padding = ""
        new_lines = stdout.readlines()
        stdout.close()
        os.unlink(temp_filename)

        # output lines, taking care of last line's trailing '\n'
        for f in range(len(new_lines)-1):
           sys.stdout.write("%s%s" % (padding, new_lines[f]))
        if lines[num-1].line.find("\n") >= 0:
           sys.stdout.write("%s%s" % (padding, new_lines[-1:][0]))
        else:
           sys.stdout.write("%s%s" % (padding, new_lines[-1:][0][:-1]))

     def main():
        lines = map(initial_scan, sys.stdin.readlines())
        temp_filename = tempfile.mktemp(".email-fmt.tmp")

        f = 0
        while f < len(lines):
           look = same_indentation_lines(lines[f:])
           reformat_lines(lines[f:f+look], temp_filename)
           f = f + look

     if __name__ == "__main__":
        main()
     # end of python code

Woah, that's more than the previous script. True, but this 86 line Python
script is a cool wrapper: fmt doesn't know exactly how you want to format
your text. If you are replying somebody's email, some text will have quotes.
Wouldn't it be nice if fmt could detect them? Yes, but it doesn't. Remember
that the unix style is doing one simple thing, but doing it extremely well.

So we solve this with the Python script. It isn't really long (try removing
blank lines and comments), and it's pretty clear (to any experienced
functional programmer) what it does. It scans the sent block of text for
existant quote characters like those defined in the indentation variable,
which are used frequently by most email programs. These lines will be kept
together, and the script will call fmt with them, but removing the quote
characters and reducing the line width. To the result, the script will add
the apropiate quote indentation level.

This means that you can select say 20 lines of text which contain three
different quote indentation levels.  If these use the defined indentation
characters the script will split them, format them separately and join them
before returning them to setedit. The result is obvious: a cool formatting
feature which is cumbersome to build into setedit, and which can be
customized any way you want. You can write your script in Python, ruby, perl,
C, etc, etc... you put the limit, not the program.

10.6.1.3 Step 3 - Binding your script to a keystroke
....................................................

Nice, we have the macro, we have the script, but how do we run it? In a
horrible world you would have to reproduce these steps to format a text block:

   * Select text block

   * Open macro menu

   * Select `Select...'

   * Choose the macro

If you did it once, you can repeat the macro with `Shift+F3', but this is not
nice, it prevents you from running other macros and you aren't saved from the
punishment of doing all that once each time you run setedit, and since we
want this for email replies, the editor will be run independently for each
answer. Ugh!

Luckily you can bind a macro to a key ( "Assigning a sLisp macro" Section
4.1.2).  The option is located under the menu option called Tool&Ops, submenu
Options, submenu Keyboard, submenu Key assignment. Once you reach that, you
can see all the key bindings for the editor. Notice that none seem to use Alt
for keystrokes, which is nice because you can assume they are free for
customizing. So you just have to type a nice combo. I use `Alt+F' for
"Formatting". Once you have the combo you assign the macro "Block reformat"
to it.

Say ok to all menus and try it yourself: write a few lines with only a word
in each of them. Select those lines and press `Alt+F'. If you did it right
you will see all those words formatted in a single paragraph. Now try that
with email indentation. Think of all the possibilities of using this trick
along with the "rectangle selection" feature of the editor.

10.6.1.4 Examples
.................

I prepared some examples so you could see the good things of all this work.
All the examples are the result of selecting all lines and using the macro.
You can easily reproduce them yourself.

Written in setedit:

      So I am just     a lazy boy typing this
      which I know will be
         formatted correctly with that cool
          python script                     written by a
            guy with funny name.

After the `Alt+F' combo:

     So I am just a lazy boy typing this which I know will be formatted
     correctly with that cool python script written by a guy with
     funny name.

Written in setedit:

     On  9 Jan 01 at 14:25, Grzegorz Adam Hankiewicz wrote:
     > On Sun, 7 Jan 2001, Petr Vandrovec wrote:
     > > (2) you can specify only xres 'x' yres '-' bpp '@' fv, you cannot
     > >     specify for example horizontal/vertical sync polarity, sync-on-green
     > >     mode, or even shift picture left-right (modify
     > >     left/right/upper/lower/hslen/vslen fields of var_screeninfo)
     > >
     > > First problem is probably unresolvable without linking modedb to each
     > > fbdev separately, as if you removing __init, others will come to you
     > > with big staff.
     > >
     > > Second problem is probably only 'invent how to write it' problem... You
     > > can look at matroxfb picture size/refresh related parameters - all of
     > > them should be parsed by modedb for benefit of all drivers...
     >
     > Aha.. do you mean I have to add new options and switches to modedb, taking
     > out the parsing code from the matroxfb module? For example, adding new
     > modedb functions which have more parameters, to remain backwards
     > compatible with the existant code?

     In my opinion it should still take one char* and...

After the Alt+F combo:

     On 9 Jan 01 at 14:25, Grzegorz Adam Hankiewicz wrote:
     > On Sun, 7 Jan 2001, Petr Vandrovec wrote:
     > > (2) you can specify only xres 'x' yres '-' bpp '@' fv, you
     > > cannot specify for example horizontal/vertical sync polarity,
     > > sync-on-green mode, or even shift picture left-right (modify
     > > left/right/upper/lower/hslen/vslen fields of var_screeninfo)
     > >
     > > First problem is probably unresolvable without linking modedb
     > > to each fbdev separately, as if you removing __init, others will
     > > come to you with big staff.
     > >
     > > Second problem is probably only 'invent how to write it'
     > > problem... You can look at matroxfb picture size/refresh related
     > > parameters - all of them should be parsed by modedb for benefit
     > > of all drivers...
     >
     > Aha.. do you mean I have to add new options and switches to modedb,
     > taking out the parsing code from the matroxfb module? For example,
     > adding new modedb functions which have more parameters, to remain
     > backwards compatible with the existant code?

     In my opinion it should still take one char* and...

11 Calculator
*************

  The calculator inside the editor was originally made by Laszlo Molnar.
Laszlo is a friend of mine from Hungary and is the author of the great DJP
progam (a djgpp exe compressor and now UPX, the best EXE compressor in all
the categories). Currently the editor is compiled with a new calculator with
some advanced features. This new calculator have the same features plus some
interesting additions and was developed by Burton Radons.

  The sources of the calculator are in the `parser.c' file. They are free and
you can use it for any purpose. There are three sources.

  Here is the documentation of the calculator written by Laszlo:

  The purpose of this program is to provide a simple but powerful 'calculator'
for programmers, to help with coding and debugging, where GDB's expression
evaluator is not enough.

  You may say "Hey, I can write a better one with `flex' and `bison'", and you
may be right. I can make a better one too. But it'll be 4-5 times longer!
This calculator is only 10 kbytes of C code. What I think? It's not that bad.

  The parser algorithm I use is called 'Operator Precedence Parsing' (I
translated this from Hungarian, so I may be wrong ;-). It works with
'operator precedence grammars' (a subset of LR(1) grammars), which means that
there can't be two non-terminating tokens next to each other on the right
side of your grammar rules. It's ideal for expression evaluation.

  With this parser you can use numbers, operators, parentheses and functions
like in C.

Here are the operators in decreasing precedence:

  1. `~' unary not `-' unary minus

  2. `**' power

  3. `*' multiplication `/' division `%' modulo

  4. `+' plus `-' binary minus

  5. `<<' shift left `>>' shift right

  6. `<' less than * `<=' less or equal than * `>' greater than * `>='
     greater or equal than *

  7. `==' equal to * `!=' different than *

  8. `&' bitwise and

  9. `^' bitwise xor

 10. `|' bitwise or

 11. `&&' logical and *

 12. `||' logical or *

 13. `?:' conditional *


  The operators marked with asterisk are available only in Burton's version.

  The calculator includes the following functions: `sin, cos, tan, sinh, cosh,
tanh, asin, acos, atan, log, log10, exp, abs, sqrt, ceil' and `floor'.  They
work as you expect.  Additionally the calculator provides some radix
conversion routines: `bin, oct, dec' and `hex'.

  The calculator uses doubles, but you can use numbers in the usual integer
formats also: 0x... for base 16, 0b... for base 2 and 0... for base 8.  The
result of the calculation is displayed as a double for base 10, and converted
to long long format for the other radixes.

  In addition the new calculator has the following features:

  You can define variables just by assigning a value to them. So if you enter
`x=5' you can use `x' in other calculation like this `x**2'. You can also use
the C/C++ assign plus operation. Post and pre increment and decrement are
also available.

  You can define functions like this `f(x)=x**2+2' so then entering `f(5)' will
give as result `27'.

  The C/C++ conditionals are available, so the following: `f(5)>=27 ? 6 : 2'
will give `6' assuming you defined `f(x)' as in the above example.

  You can separate operations with commas as in C. The result of the last
operation is the result of the compound.

12 How to contact me
********************

  If you have any suggestions or a bug report contact me at the address shown
in the author section.  ("About the Author" Section 1.3).

12.1 Bugs
=========

  If you find a bug please contact me, the Undo thing is the most complex one
and I know that needs some work on it.

  When reporting a bug please don't tell me: Some times some strange thing
happens ... Try to find a pattern to the problem. What situation triggers the
problem? ... with which file(s)? Then send me the file and the description.

  When sending me a file UUEncode it to avoid problems related with e-mail.

13 TAGS files
*************

This chapter describes some details about `TAGS' files. They are referreded
as tags in the rest of the sections.

I wrote a small ilustrated tutorial showing how to quickly use this. The
tutorial is available in the editor home page:
http://setedit.sf.net/tags.html. It is also included in some of the
distributions of the editor.

I strongly recommend reading the documentation of the program to generate the
tag files.

13.1 What are tags?
===================

TAGS files are plain text files containing a list of symbols from your source
code. For each symbol the file indicates in which file it's located and how to
find the symbol inside the file. Modern programs also include very important
information like which type of symbol is defined and if this symbol is part of
a bigger construct like a class.

This information is very useful to do searches. The editor implements
facilities to jump to any defined symbol, browse classes and do word
completion using information from tag files.

13.2 Which program is used to generate TAGS files?
==================================================

A lot of programs generate tag files but the editor uses special features
only found in the `Exuberant Ctags' program. For this reason I recommend
using this program to generate the tag files. You can find this program at
Source Forge (http://ctags.sourceforge.net).

13.3 How should I generate the tags?
====================================

It depends on your project and how you compile the project. It also depends
on how fast your system is. I currently use a tags file for the editor and
another for the Turbo Vision library. The best is to refresh the tags file
using the same makefile you use to compile your project. If your system isn't
very fast or your project is too big you could try using one tags file for
each directory in your project. The editor can be configured to collect
symbols from various tag files.

The command line options I recommend to use with `ctags' are:

   * -R recursive, to search in all subdirectories.

   * --fields=+i+l+m+z Inheritance information, language of source file that
     contains the tag, implementation information and type of tag.

Without the `+i' and `+z' option you won't be able to use the class browser.

13.4 Can these files be created and updated by the editor?
==========================================================

I recommend using makefiles to maintain tag files because you have all the
control over when and how they are updated. But if you are using a fast
machine and want to simplify the process you can let the editor create and
maintain a tags file for your project. You'll find options for it in the
Tool&Ops | Options | Tag files menu option.  ("Tag files options" Section
5.7.1.24).

13.5 What is the easiest way to use tags?
=========================================

Here is an explanation about how to quickly use tag files:

  1. Install `Exuberant Ctags' (http://ctags.sourceforge.net).

  2. Create a project (Project | Open).  ("Open (Project)" Section 5.8.1).

  3. Add all the files you want to include to the project. You can import the
     list of files from a text file.  ("Import project items" Section 5.8.6).

  4. Enable automatic tag generation using a central file (Tool&Ops | Options
     | Tag files | Options).  ("Tag files options" Section 5.7.1.24).

  5. Perform a search, the tags file will be generated during this process so
     be patient if your system is slow or your project is huge.

13.6 What can I do with tags?
=============================

With the tag files you can:

  1. Jump to any defined symbol.  ("Jump to symbol" Section 5.3.7).

  2. Browse C++ class hierarchy.  ("Class browser" Section 5.3.9).

  3. Ask the editor to complete a partially typed symbol.  ("Word completion"
     Section 5.3.10).

13.7 Technical details about tags
=================================

The editor loads the tags to memory to allow fast searches. For this reason
you must be aware that they consume lots of memory. In some tests I did using
GNU/Linux I found the amount needed to load a tags file is close to the size
of the file on disk. For this reason I recommend to avoid generating symbols
for files you don't need.

The time to parse and load tag files can be annoying on slow systems. For
this reason the editor never loads the tag files until they are really
needed. That's why you'll notice a delay for the first search, because the
files are loaded at this time and not when the editor is started.

Tag files don't catch all definitions, they aren't perfect, so please read
the man pages. They also lack some important information, like what type a
variable (int, float, `etc.') is.

14 Debugging
************

Important! 2004/08/17 version 0.5.4: Currently this chapter is incomplete and
the debugging features of the editor are considered unstable.

Only C and C++ programs can be debugged. If you use another language supported
by gdb and want to help supporting it please contact me. Languages not
supported by gdb needs more work to get supported, even if the debugger used
for them is similar to gdb.

14.1 Supported platforms for debugging
======================================

Currently only Linux for IA32 (x86) CPUs is fully supported. Support for
other OSs where gdb (The GNU debuger) is available and that provides a
complete POSIX interface should be easy to obtain. If you want to help adding
support for your platform please contact the author.   ("How to contact me"
Chapter 12).

Architectures other than IA32 are partially supported. If that's your case
you can help testing and providing information about it. The two most
important things you can help to fix are: the stack window and the syntax
highlight of the disassembler window. You just need to know a little bit
about the native assembler.

DOS and any other target where the fork/exec mechanism and pipes
communication isn't implemented aren't supported. To support them I need a
lot of help. In particular I need to make some code to handle the gdb/MI
protocol using libgdb.a. If you think you can help with it contact me.

The support for Linux is for X and for the real console. You can't debug
programs if you don't have access to the real console of the local machine.

Only programs that can be debugged by gdb can be debugged using the editor.

14.2 Mechanism used to debug
============================

The editor doesn't include a built-in debugger like RHIDE does. Instead you
have to install the GNU Debuger (gdb). Only versions of gdb that implements
the MI protocol are usable. The version 5 implements MI version 1 and the
version 6 implements MI version 2.

To debug the editor starts gdb and communicates with the debugger using a
POSIX mechanism called pipe. It means the editor can just send commands to
gdb like you could do manually. The consequence is that you all the power of
gdb, but also all the problems and bugs found in gdb. During the development
of the debug features I found more than two bugs in gdb each week.

The editor uses a special mode of gdb called MI. MI stands for Machine
Interface. In this mode the gdb responses are specially designed to be parsed
by another program instead of read by a human bean. This makes the
communication much more robust but this mode is less tested than the usual
mode (CLI Command Line Interface).

The MI mode allows to send CLI commands to gdb. So you can send your own
commands to gdb. This should be used only if the editor doesn't provide any
way to achieve the same result. If you need to do it very often please
consider commenting it in the editor's mailing list. We could add the proper
user interface for what you need.  ("Debug Messages Window" Section 14.17).

14.3 Quick start to debugging
=============================

The following is a very short explanation of how to start debugging your
program. It assumes you already know what's a debugger and you are
familiarized with the user interface. The user interface is quite similar to
the RHIDE's user interface and that's almost the same used by Borland
debuggers.

  1. Create a new project. All the debugging options will be stored in the
     project file.  ("Open (Project)" Section 5.8.1).

  2. Add your source files to the project. You can do it pressing the
     `Insert' key when the project window is selected or importing a list of
     files.  ("Import project items" Section 5.8.6).

  3. Compile your program with debug information. For `gcc' you should use
     the `-g' command line option and avoid using the `-s' option.  You'll
     get better integration if you compile your program using the internal
     features of the editor to run external programs.  ("Run program (which
     one)" Section 5.7.1.8).

  4. Select the `Debug|Step over' menu option. If you didn't set the debug
     options the editor will ask to do it. The most important option is the
     name of the program you want to debug.

  5. If the editor can't find the source file you'll be asked to indicate its
     location.

  6. If all goes as spected you'll see your source file and the first line of
     the main function highlighted. You are debugging your program. Now you'll
     want to consult the sections about breakpoints, stepping in your code and
     displaying data.

14.4 Supported debug targets
============================

Currently you can especify the following targets:

   * Local executables. That's the most common case where you debug a program
     you compiled locally.

   * Remote debugging. In this case you debug a program that runs on another
     machine connected through a serial communication or network.

   * Already running processes. You can also debug a process that's currently
     running in your local machine.

Other gdb targets aren't supported.

In all the cases you should have a local copy of the program with debug
information. When you debug a remote program the remote copy doesn't have to
include debug information.

14.5 Available debug options
============================

You can configure the following debug options. Remmember that most of them
are stored in project files.

14.5.1 Debug options
--------------------

This dialog is obtained using the `Debug|Options|Program and mode...' menu
entry.

Here you setup the most relevant debug options. These options are stored in
project files. For this reason you should use a project if you want to avoid
setting these options again. The available options are:

   * Program to debug: Indicates the local copy of the program to debug.
     This copy must be compiled with debug information and the debug
     information must be preserved during the linking stage. Using the
     `Browse' button you can choose a file using a dialog to navigate your
     filesystem.

   * Mode: It selects the kind of target you will debug.   ("Supported debug
     targets" Section 14.4).

   * Program arguments: The command line options for the program you want to
     debug. This option is used only for local targets.

   * Forced terminal: When the editor debugs your program the input and
     output of the program are redirected to a different terminal. This
     mechanism avoids interferences between the editor and the program you
     want to debug.  The terminal is selected or created automatically. If
     you need to force a particular terminal fill this option. Otherwise
     leave it blank.

   * Remote protocol type: This option is used for remote targets. Here you
     must specify one of the protocols supported by gdb. The most common
     protocol is `extended-remote'.

   * Remote location: Also used for remote debugging. Here you indicate the
     location of the remote machine or the serial device. For TCP/IP you must
     specify the host and the port number using `host:port'. For serial links
     you must specify the serial device. Consult gdb documentation for
     details.

Name of the command: cmeDebugOptions.

14.5.2 Path for sources
-----------------------

This dialog is obtained using the `Debug|Options|Path for sources...' menu
entry.

The most common way to compile your program is by indicating the relative
path for the source file. When you debug your program you can do it from any
directory. For this reason gdb can have problems to find the source files.
The editor will analyze your project and extract paths from it. Additionally
the editor will use the path for the executable as reference. The resulting
list is passed to gdb. But sometimes it isn't enough and you must indicate
the location for the files manually.

The easiest way to add a directory is using the `Browse' button. It will
bring a tree view of your filesystem. This dialog supports incremental search.

When gdb reports a source file the editor uses this list to find the exact
location of the file. If the file couldn't be found the editor will ask you
to select the file using a dialog like the one used to open files. For this
reason you usually don't need to add the directories manually. The exception
is when you need to set a breakpoint and you get an error from gdb indicating
the file couldn't be found. In this case the best recommendation is to add
the file to the project, but you can add the directory here.

Name of the command: cmeSourceList.

14.5.3 Messages displayed
-------------------------

This dialog is obtained using the `Debug|Options|Messages displayed...'  menu
entry.

This dialog controls which messages will be displayed in the Debugger Window.
("Debug Messages Window" Section 14.17). You can filter the following kind
of messages:

   * GDB console (CLI): Messages sent to the console. Usually they are human
     readable messages generated by gdb in response to console commands.

   * Target: The output of the target. This kind of messages are specified by
     gdb documentation but never generated. They are supposed to contain the
     tunneled output of your program.

   * Log: Messages sent to the gdb log. Usually they contain technical
     information about gdb details.

   * GDB/MI commands sent to gdb: Machine Interface commands sent by the
     editor to gdb.

   * GDB/MI responses from gdb: Machine Interface responses from gdb.

Usually you'll be interested only in the console and log messages. The target
messages aren't really implemented in gdb and the MI messages are only useful
to control if the editor is sending the proper information and gdb replying
to it.

Name of the command: cmeDbgOptsMsgs.

14.5.4 Advanced debug options
-----------------------------

This dialog is obtained using the `Debug|Options|Advanced...' menu entry.

The options contained in this dialog are used to fine tune the debug session.
Some of these options are dangerous and can produce problems if you don't set
them properly.

   * GDB executable: The binary executed as debugger. Use the browse button
     to select it. If you leave it blank the editor will choose the first
     binary called gdb found in your path.

   * X terminal executable: The binary used to create a terminal for the
     program to debug when the editor is connected to an X server.

   * Main function: The name of the first function executed in your program.
     For C and C++ programs that's usually `main'. Sometimes the name is
     different. One example is the Allegro library where `main' is replaced by
     `__mangled_main'.

   * GDB Time out [s]: The ammount of seconds to wait for a gdb response.
     After this time the editor will ask if you want to wait more time or
     assume the response will never arrive. You can't indicate less than two
     seconds.

   * Max. lines in Debug window: Maximum lines to hold in the Debugger
     Window. When this value is reached the older lines are discarded. You
     can't indicate less than 100 lines.

   * No gdb banner after connecting: Suppress the gdb banner printed in the
     Debugger Window when we start gdb.  ("Debug Messages Window" Section
     14.17).

   * Enable MI v2 features: Enable a few features only available in gdb MI
     version 2. Currently the editor can't determine the exact version
     implemented in gdb so you must indicate it. Versions newer than 6.0
     seems to implement MI version 2. But cvs snapshots of gdb doesn't report
     a useful version.

   * No symbols bug workaround: Actually the editor implements a workaround
     for a bug in gdb when searching for file names in the table of debug
     symbols.  This workaround produces a slower start-up. This bug seems to
     affect versions from 5.3 to 6.2 (inclusive).

   * No source code in disasm. window: Disables the use of mixed source code
     and assembler in the disassembler window. This is faster.

All the above options are stored globally and affects all projects. The only
exception is the name of the main function which is stored in the project
file.

Name of the command: cmeDbgOptionsAdv.

14.6 Debugging states
=====================

Not all operations can be performed at any time. GDB must be running in order
to run the program and the program running in order to stop it. For this
reason the editor keeps track of gdb state. The current state is displayed in
the Debugger Window.  ("Debug Messages Window" Section 14.17).The following
states are defined:

   * Disconnected: gdb isn't running.

   * Connected: gdb is running but no target was selected.

   * Ready to Run: gdb is running and we selected a target.

   * Running: owr program is running. Note that currently gdb doesn't
     implement full asynchronous operation. It means that you can't send
     commands to gdb while the state is `running'. GDB won't be able to reply
     commands in this state.

   * Stopped: the program was stopped and can resume its execution.

Some menu options can be used to force a particular state. So you can
directly jump to `connected' or `ready to run' states from the `disconnected'
state. This is useful if you want to send commands to gdb that only works in
these states.  ("Debug Messages Window" Section 14.17).

When you specify an action that is invalid for the current state the editor
will try to reach the proper state. As an example, if you use the `Step over'
option when you are in `disconnected' state the editor will first try to
start gdb and go to `connected', then will try to specify the target and go
to `ready to run' and finally will emulate the step action by using a
temporal breakpoint on the main function.

The menu entry called `Debug|Debug session|Close' is used to go to the
`disconnected' state from any of the other states.

The menu entry called `Debug|Stop' is used to go to the `stopped' state from
the `running' state. And the `Debug|Kill' option is used to go to the `ready
to run' state from the `stopped' or `running' states.

14.6.1 Going to the connected debug state
-----------------------------------------

This operation is obtained using the `Debug|Go to state|Connected' menu entry.

With this function the editor will try to reach the `Connected' debugging
state.  ("Debugging states" Section 14.6).

It can be used when the debugger isn't running and you want to start it
without specifying the target to debug. In this way you can send commands to
gdb that are useful only before you specify the target.

Name of the command: cmeDbgGoConnected.

14.6.2 Going to the ready to run debug state
--------------------------------------------

This operation is obtained using the `Debug|Go to state|Ready to run' menu
entry.

With this function the editor will try to reach the `Ready to run' debugging
state.  ("Debugging states" Section 14.6).

This operation will try to reach the `Connected' state first.   ("Going to
the connected debug state" Section 14.6.1). If this is achieved then the
editor will indicate to gdb which target do you want to debug. For local
debugging the program won't be started so you can send to gdb commands that
apply only before starting the program to debug.

Name of the command: cmeDbgGoReadyToRun.

14.7 Running the program to debug
=================================

This operation is obtained using the `Debug|Run/Continue/Atach' menu entry.

You must first configure the name of the program you want to debug. If you
don't do it the editor will ask if you want to do it.  ("Debug options"
Section 14.5.1).

Once your program is running you can stop it using the `Debug|Stop' option.
But usually is better to use breakpoints and watchpoints instead.

While the program is running you can't consult the value of variables.
That's because gdb gets blocked until your program is stopped.   ("Debugging
states" Section 14.6).

This command have other uses. If your program is already stopped it resumes
its execution.  ("Continue execution" Section 14.10.1). If you are debugging
a program that is already running this command will atach to the running
process and stop the execution.

Name of the command: cmeDbgRunContinue.
Assigned key: `Shift+F9'

14.8 Stopping the program you are debugging
===========================================

Once your program is running you can stop it using the `Debug|Stop' option.
But usually is better to use breakpoints and watchpoints instead.

Using breakpoints you can stop the execution at a desired point of your
program. Using watchpoints you can stop the execution when some data value
changes or is accessed. Note that gdb watchpoints aren't fully reliable.

Name of the command: cmeDbgStop.

14.8.1 Breakpoints
------------------

Breakpoints are stop points in your program. The easiest way to create a
breakpoint is by using the `Debug|Breakpoint' menu option. This will create a
breakpoint at the current file and line. To remove it just use the same menu
option. Breakpoint lines are highlighted. The default color is red.
("Customize Colors" Section 5.7.1.1).

Note that gdb accepts breakpoints even for source lines that didn't generate
code. In this case gdb will stop the execution when the nearest source line
is reached.

When you edit your program breakpoints are automatically moved. If you use
the editor options to call an external program in order to recompile your
program the editor will inform gdb the new place for the breakpoints.

To create more complex breakpoints you have to use the `Debug|Edit
breakpoints...' menu option.  ("Advanced breakpoint options" Section 14.8.2).

Name of the command: cmeBreakpoint.
Assigned key: `Ctrl+F8'

14.8.2 Advanced breakpoint options
----------------------------------

This dialog is obtained using the `Debug|Edit breakpoints...' menu entry.
Using this dialog you can:

   * Modify existing breakpoints. Modify button.

   * Add new breakpoints. New button.

   * Remove breakpoints. Delete button.

   * Enable or disable a breakpoint. Enable and Disable buttons. Note that
     currently this is not the same concept implemented in gdb. I.e. we
     remove the breakpoint from gdb but not from the editor list.

   * Jump to the source file where the breakpoint is located. Show button.

The list of breakpoints contains the following information:

   * E: an asterisk if enabled.

   * Where: breakpoint location.

   * Condition: the additional condition for conditional breakpoints.

   * Count: the number of times this breakpoint will be ignored when enabled.

   * Thre.: the thread affected by this breakpoint. All threads are affected
     by default.

When you add or modify a breakpoint a dialog offering the following options
is used:

   * Type: This is the kind of specification for the breakpoint location.
     GDB supports four ways to specify a breakpoint combining the source file
     name, the line, function and code address.

   * Condition: If this field contains a condition the breakpoint will stop
     the program execution when the location is reached and the condition is
     satisfied.

   * Count: When this value is specified the breakpoint will be ignored this
     ammount of times. Note that gdb keeps a counter that is reset every time
     you set the breakpoint. When the program is started the editor will
     force a refresh on this value. That isn't the normal gdb behavior. It
     means that a count of five will make the breakpoint to be ignored the
     first five times it was reached after starting your program.

   * Thread: You can specify a thread id here.

   * Enabled: Indicates if the breakpoint is enabled.

   * Hardware assisted: When you are debugging code that is located in a read
     only memory space you have to use hardware assisted breakpoints. Read
     the gdb documentation for more information.

When a breakpoint fails to apply the editor will disable it to avoid farther
gdb errors.

Name of the command: cmeDbgEditBreakPts.

14.8.3 Watchpoints
------------------

This dialog is obtained using the `Debug|Edit watchpoints...' menu entry.

Unlike breakpoints the watchpoints are data related. Some people calls them
`data breakpoints'. Watchpoints can stop the execution when a data value is:

   * Written

   * Read

   * Accessed (Written or Read)

Note that in order to get usable results this needs hardware assistance. It
means that not all platforms supports it. Additionally the number of
watchpoints is limited and gdb imposes more restrictions to what can be used
as a watchpoint. In some cases gdb tries to emulate it using software and the
result can be really bad. During development I found a case where gdb did it
and as a result it eated all the CPU time and was really hard to stop.

Using this dialog you can:

   * Modify existing watchpoints. Modify button.

   * Add new watchpoints. New button.

   * Remove watchpoints. Delete button.

   * Enable or disable a watchpoint. Enable and Disable buttons.

The list of watchpoints contains the following information:

   * E: an asterisk if enabled.

   * R/W: type of watchpoint (Read, Write or Read+Write)

   * Expression: The expression to monitor.

When you add or modify a watchpoint a dialog offering the following options
is used:

   * Expression: the expression that will be monitored.

   * Type: the action that will stop the execution.

   * Enabled: indicates if the watchpoint is enabled.

Name of the command: cmeDbgEditWatchPts.

14.9 Examining data
===================

There are many tools to examine data values of the program you are debugging.
You can:

   * Evaluate an expression to know the resulting value.   ("Evaluate or
     Modify expression" Section 14.9.1).

   * Monitor a simple value as the program flows.  ("Watch an expression"
     Section 14.9.2).

   * Monitor an expression with scope.  ("Watch an expression with scope"
     Section 14.9.3).

   * Monitor a complex data structure.  ("Inspectors" Section 14.9.4).

   * Examine the memory space.  ("Data Window" Section 14.9.5).

   * Examine the stack.  ("Stack window" Section 14.9.6).

14.9.1 Evaluate or Modify expression
------------------------------------

This dialog is obtained using the `Debug|Evaluate/Modify...' menu entry.

The first input line is used to enter the expression you want to evaluate.
The result is obtained using the `Evaluate' button. That's the default button
so you can just press `Enter'. The resulting value is displayed in the
`Result' input line.

To assign a different value to the expression you must use the `New value'
input line and press the `Change' button. Note that the value used in
`Expression' must be expressed in a way that can be used for an assignment
(`Expression' = `New Value').

The `Copy' button will copy the `Result' value to the editor's clipboard. The
`Paste' value will paste from the editor's clipboard to the `Expression'
input line.

Using the `Inspect' button you can create an Inspector Window based on the
`Expression'.  ("Inspectors" Section 14.9.4).

This dialog is modal and is used to evaluate a value just once. To see how a
value changes as yourt program flows use watches or inspectors.

Name of the command: cmeDbgEvalModify.
Assigned key: `Ctrl+F4'

14.9.2 Watch an expression
--------------------------

This functionality is obtained using the `Debug|Watch an expression|Normal
watch' menu entry.

Using this option you can add an expression to the Watches Window. This
expression will be evaluated every time the program stops and the result will
be updated in the Watches Window.

Adding too much watches could slow down the debug process. Also note that the
expression will be evaluated in the current scope. If you need to monitor an
expression in a fixed scope use a watch with scope.   ("Watch an expression
with scope" Section 14.9.3). For complex data structures you'll get better
results using Inspectors.  ("Inspectors" Section 14.9.4).

Name of the command: cmeDbgWatchExpNorm.
Assigned key: `Ctrl+F7'

14.9.3 Watch an expression with scope
-------------------------------------

This functionality is obtained using the `Debug|Watch an expression|With
scope' menu entry.

Using this option you can add an expression to the Watches Window. This
expression will be evaluated every time the program stops and the result will
be updated in the Watches Window.

Adding too much watches could slow down the debug process. These watches are
implemented using a gdb mechanism that is faster when only a few expressions
changes in each execution.

Unlike normal watches the scope of the expression is fixed. It means you'll
be monitoring the same value all the time and not different variables. Also
note that when their scope is finished they are deleted from the list.

For complex data structures you'll get better results using Inspectors.
("Inspectors" Section 14.9.4).

Name of the command: cmeDbgWatchExpScp.

14.9.4 Inspectors
-----------------

This functionality is obtained using the `Debug|Watch an expression|Using the
Inspector' menu entry.

Inspectors are used to monitor complex data structures as the program flows
or just navigate them to verify them. I.e. you can navigate a linked list
using an inspector.

Unlike watches each expression you evaluate have its own window. You can have
as many inspector windows as you like. The windows are resizable and have
scroll bars.

The representation of the data inside an inspector is a tree. You can expand
or collapse the branches using the `E' and `C' keys. On most systems you can
also use the `+' and `-' keys. If a line begins with a `+' it means that you
can expand the tree from this point. If it starts with a `-' you can collapse
it. After the name of the data member you'll find its data type enclosed in
brackets. At the end of the line the value of the data member is displayed.

Some lines will look a like this: `varXX.public'. That's just the way gdb
indicates the accesibility of the members. The `varXX' part is the name of
the internal gdb value. You don't have to worry about this name.

When you select a line containing some data type that looks like a pointer
the `Inspect' option of the status bar will be enabled. Using it you'll get a
new inspector for the data pointed by the selected pointer. The `Enter' key
can be used for this.

When the line contains a data value that can be modified the `Modify' option
of the status bar will be enabled. The `Ctrl'+`M' key can be used to modify
the selected value.

You can also use the `Format' option (key `Ctrl'+`F') to change the format
used to represent the selected member.

Inspectors are evaluated in the scope they were created. When its scope is
finished they stop working. If you enter to the same scope, or another where
the expression is valid, you can use the `Recycle' option. The keyboard
shortcut for it is `Ctrl'+`R'.

Name of the command: cmeDbgInspector.
Assigned key: `Ctrl+F6'

14.9.5 Data Window
------------------

This functionality is obtained using the `Debug|Watch an expression|Data
window' menu entry.

The Data Window can be used to examine and/or modify the memory. Its
functionality can be compared with an hexadecimal editor. The number of
operations and options related to the Data Window are really big. For this
reason the Data Window have its own menu. When you select a Data Window the
menu will change to show the Data Window commands.

The original code for the Data Window was contributed by Molnar Laszlo. He
designed it for RHIDE, but RHIDE doesn't document all the available
functionality of the Data Window.

When you modify the memory in the Data Window the changes aren't reflected to
the real memory space. In order to reflect the changes you must use the
`Enter' key or the `Various|Update memory' menu option. Every time the
program stops the content of the Data Window is updated and the changes
highlighted. You can force the editor to recompute the original expression
used to get the memory address used in the Data Window in each update. To
achieve it use the `Mode|Toggle auto follow' menu option.

The following is a description of the special menu used for the Data Window.
Only the commands related to the Data Window are explained.

Name of the command: cmeDbgDataWindow.

14.9.5.1 File - Read block (DW)
...............................

Used to read data from a file and store it in the memory space of the program
you are debugging. You have to choose a file and then indicate the
destination address and the maximum ammounts of bytes to read from the file.

Some general limits and functionality common to the Data Window operations
applies.  ("Data Window Limitations and Details" Section 14.9.5.30).

Name of the command: cmDWRead.
Assigned key: `Ctrl+R'

14.9.5.2 File - Write block (DW)
................................

Used to write data from the memory space of the program you are debugging to
a file. You have to indicate the source address and the how much data to
write and then choose a file name.

Some general limits and functionality common to the Data Window operations
applies.  ("Data Window Limitations and Details" Section 14.9.5.30).

Name of the command: cmDWWrite.
Assigned key: `Ctrl+W'

14.9.5.3 Move - Up (DW)
.......................

Moves the cursor position one row up. If the cursor is at the top row the
starting address will be decremented by the bytes contained in a row and the
whole block will be fetched.

Changes not tranferred to the target will be lost if a new block of data is
fetched.

Name of the command: cmDWUp.
Assigned key: `Cursor up'

14.9.5.4 Move - Down (DW)
.........................

Moves the cursor position one row down. If the cursor is at the bottom row the
starting address will be incremented by the bytes contained in a row and the
whole block will be fetched.

Changes not tranferred to the target will be lost if a new block of data is
fetched.

Name of the command: cmDWDown.
Assigned key: `Cursor down'

14.9.5.5 Move - Right (DW)
..........................

Moves the cursor to the right. If the cursor is in the rightmost column it
will remain there.

Name of the command: cmDWRight.
Assigned key: `->'

14.9.5.6 Move - Left (DW)
.........................

Moves the cursor to the left. If the cursor is in the leftmost column it will
remain there.

Name of the command: cmDWLeft.
Assigned key: `<-'

14.9.5.7 Move - Page down (DW)
..............................

Increments the starting memory address by the number of bytes contained in a
row multiplied by the number of rows.

Changes not tranferred to the target will be lost.

Name of the command: cmDWPgDn.
Assigned key: `Page Down'

14.9.5.8 Move - Page up (DW)
............................

Decrements the starting memory address by the number of bytes contained in a
row multiplied by the number of rows.

Changes not tranferred to the target will be lost.

Name of the command: cmDWPgUp.
Assigned key: `Page Up'

14.9.5.9 Move - First column (DW)
.................................

Moves the cursor to the leftmost column.

Name of the command: cmDWFirstColumn.
Assigned key: `Home'

14.9.5.10 Move - Last column (DW)
.................................

Moves the cursor to the rightmost column.

Name of the command: cmDWLastColumn.
Assigned key: `End'

14.9.5.11 Move - First row (DW)
...............................

Moves the cursor to the first row in the same column.

Name of the command: cmDWFirstRow.
Assigned key: `Ctrl+Home'

14.9.5.12 Move - Last row (DW)
..............................

Moves the cursor to the last row in the same column.

Name of the command: cmDWLastRow.
Assigned key: `Ctrl+End'

14.9.5.13 Move - First addr increment (DW)
..........................................

Increments the starting memory address by one.

Changes not tranferred to the target will be lost.

Name of the command: cmDWBaseIncrement.
Assigned key: `Ctrl+->'

14.9.5.14 Move - First addr decrement (DW)
..........................................

Decrements the starting memory address by one.

Changes not tranferred to the target will be lost.

Name of the command: cmDWBaseDecrement.
Assigned key: `Ctrl+<-'

14.9.5.15 Address - Change base address (DW)
............................................

When you open a new Data Window the addresses displayed are the same used by
the program you are debugging. If you need to convert them into relative you
can change the base address. When you specify a new base address it is used
as reference.

Name of the command: cmDWBaseAddress.
Assigned key: `Ctrl+B'

14.9.5.16 Address - Go to new address (DW)
..........................................

Used to change the expression used to determine the starting address of the
Data Window. The address corresponding to this expression is always
highlighted used a different color.

Name of the command: cmDWGotoAddress.
Assigned key: `Ctrl+G'

14.9.5.17 Address - Follow pointer (DW)
.......................................

Interprets the memory at the cursor position as a pointer to another block of
memory. This pointer is used as the new expression for the Data Window.

Name of the command: cmDWFollowPointer.
Assigned key: `Ctrl+F'

14.9.5.18 Address - Follow pointer in new window (DW)
.....................................................

Interprets the memory at the cursor position as a pointer to another block of
memory. This pointer is used as the expression for a new Data Window. The
extra asterisk is to cancel an `&' added by the editor.

Name of the command: cmDWFollowPtnNew.
Assigned key: `Ctrl+O'

14.9.5.19 Address - Recompute address (DW)
..........................................

Evaluates the original expression to determine the new starting address. You
can do it automatically every time the program stops using the auto follow
mode.  ("Mode - Toggle auto follow (DW)" Section 14.9.5.20).

Name of the command: cmDWRecompute.
Assigned key: `Ctrl+H'

14.9.5.20 Mode - Toggle auto follow (DW)
........................................

Toggles the auto follow mode. When the auto follow mode is enabled the
expression supplied to determine the memory displayed by the Data Window is
automatically recomputed every time the program stops.

The auto follow mode is indicated with an `A' in the indicator widget located
at the bottom right of the Data Window.

Name of the command: cmDWTogAutoF.
Assigned key: `Ctrl+A'

14.9.5.21 Mode - Change display mode (DW)
.........................................

Changes the display mode to the next in the list of available modes. The
available modes are:

   * Bytes: each cell represents one byte in memory. The right side of the
     window contains its ASCII representations. Note that ASCII
     representation is suppressed when you select decimal radix.

   * Word: each cell represents two bytes.

   * Double Word: each cell represents four bytes.

   * ASCII: Only the ASCII representation is displayed. In this mode you can
     simply type characters to change the memory.

Name of the command: cmDWDispMode.
Assigned key: `Ctrl+D'

14.9.5.22 Mode - Toggle endian mode (DW)
........................................

Toggles the endian mode. When a new Data Window is created the data is
displayed using the endian type of the program you are debugging. You can
change the endian mode at any time.

Little endian mode is indicated in the indicator widget located at the bottom
right of the Data Window using an `e' and big endian using `E'.

Name of the command: cmDWTogEndian.
Assigned key: `Ctrl+E'

14.9.5.23 Mode - Change radix (DW)
..................................

Changes the radix of the represented numbers. The available radices are
hexadecimal and decimal.

Hexadecimal radix is indicated in the indicator widget located at the bottom
right of the Data Window using an `X' and decimal using `D'.

Name of the command: cmDWRadix.
Assigned key: `Ctrl+X'

14.9.5.24 Block - Fill (DW)
...........................

Fills a memory block with an indicated data value. You have to indicate the
destination address, how many bytes will be filled and the value used to fill
the block.

Some general limits and functionality common to the Data Window operations
applies.  ("Data Window Limitations and Details" Section 14.9.5.30).

Name of the command: cmDWFill.
Assigned key: `Ctrl+I'

14.9.5.25 Block - Clear (DW)
............................

Fills a memory block with zeroes. You have to indicate the destination
address and how many bytes will be filled.

Some general limits and functionality common to the Data Window operations
applies.  ("Data Window Limitations and Details" Section 14.9.5.30).

Name of the command: cmDWClear.
Assigned key: `Ctrl+L'

14.9.5.26 Block - Move (DW)
...........................

Copies a block of memory to another memory location. You have to indicate the
source address, the destination address and how many data will be copied.

Some general limits and functionality common to the Data Window operations
applies.  ("Data Window Limitations and Details" Section 14.9.5.30).

Name of the command: cmDWMove.
Assigned key: `Ctrl+M'

14.9.5.27 Various - Less bytes per line (DW)
............................................

Decrements the ammount of data cells displayed in each row.

Changes not tranferred to the target will be lost.

Name of the command: cmDWLessLines.
Assigned key: `-'

14.9.5.28 Various - More bytes per line (DW)
............................................

Increments the ammount of data cells displayed in each row.

Changes not tranferred to the target will be lost.

Name of the command: cmDWMoreLines.
Assigned key: `+'

14.9.5.29 Various - Update memory (DW)
......................................

Transfers changes to the memory space of the program you are debugging.

Name of the command: cmDWUpdateMemory.
Assigned key: `Enter'

14.9.5.30 Data Window Limitations and Details
.............................................

Most Data Window operations uses a common dialog that adapts its fields to
the operation you want to perform. For this reason the operations shares the
same functionality and limitations.

All the fields (New Value, From, To, Expression, Length and Value) accepts
any kind of expression supported by gdb. You can include arithmetic
operations, variables, `etc.'

The Length and Value fields are evaluated as an unsigned value. The other
values are evaluated as addresses. In the last case the editor tries to
determine if you typed an address (something starting with a digit), an
explicit reference (`&' C/C++ operator) or a CPU register (they start with
`$' in gdb). If the editor thinks you didn't specify an address the `&'
operator will be automatically added. If you have a local variable called
`var' and you specify it then the editor will use `&var'.

In most cases the transferring big memory blocks between the target program
and the editor is an slow process. For this reason the editor will show a
warning if the size of the block is bigger than 128 KB and will refuse to
operate with more than 1 MB. This is also a security meassurement. If you
think this is a wrong idea please tell me.

14.9.6 Stack window
-------------------

The stack window is a specialized version of the Data Window. It starts using
the `esp' register as starting address, one cell in each line, four bytes
cells and auto follow mode enabled.  ("Data Window" Section 14.9.5).

Currently this is implemented only for the IA32 architecture. If you know the
name of the stack pointer register for your architecture please contact me so
I can add support for it. Also note that this window uses the `esp' register
and not the stack frame register (`ebp'). I don't know if users will be
interested on having a similar window for the stack frame.

Name of the command: cmeDbgStackWindow.

14.10 Resuming the execution of the program
===========================================

Once you stopped the execution of your program and you examined its data
you'll most probably want to resume its execution. This section describes
various operations to do it controlling how much code will be executed.

You can resume execution until a breakpoint, watchpoint or the end of program
is reached.  ("Continue execution" Section 14.10.1).

You can execute until the next line of code is reached. You can achieve it
using two different commands. The first won't stop inside a function called
during by the execution ( "Step over" Section 14.10.2) and the second will do
it ( "Trace into" Section 14.10.3).

You can also execute until a line of code is reached.   ("Executing until
cursor position is reached" Section 14.10.4).

Another option is to run the code until the current function is finished.
("Executing until return" Section 14.10.5). It's also possible to return
immediatly without executing the rest of the function.  ("Returning
immediatly" Section 14.10.6).

14.10.1 Continue execution
--------------------------

When the program is stopped the menu entry named `Debug|Run/Continue/Atach'
can be used to resume the execution.

If the program isn't started this command is used to start the execution
(Run) or atach to an already running process (Atach).

Name of the command: cmeDbgRunContinue.
Assigned key: `Shift+F9'

14.10.2 Step over
-----------------

This functionality is obtained using the `Debug|Step over' menu entry.

This command resumes the execution of a stopped program until the next line
of code is reached. Function calls that appear within the line of code are
executed without stopping.

When applied to the Assembler Window it works at instruction level.

Name of the command: cmeDbgStepOver.
Assigned key: `F8'

14.10.3 Trace into
------------------

This functionality is obtained using the `Debug|Trace into' menu entry.

Continue running your program until control reaches a different source line,
then stops it. If a function call is found this command will enter the
function unless it doesn't have debug information.

When applied to the Assembler Window it works at instruction level and no
debug information is needed to enter functio calls.

Name of the command: cmeDbgTraceInto.
Assigned key: `F7'

14.10.4 Executing until cursor position is reached
--------------------------------------------------

This functionality is obtained using the `Debug|Go to cursor' menu entry.

Runs your program until control reaches the source line where the cursor is
located.

Name of the command: cmeDbgGoToCursor.
Assigned key: `F4'

14.10.5 Executing until return
------------------------------

This functionality is obtained using the `Debug|Until return' menu entry.

Resumes the execution until the current function is finished. This command
can't be used in the main function. If you want to return from the current
function without executing the remaining code use the `Return now' option.
("Returning immediatly" Section 14.10.6).

Name of the command: cmeDbgFinishFun.

14.10.6 Returning immediatly
----------------------------

This functionality is obtained using the `Debug|Return now' menu entry.

This command doesn't realy resume the execution. Instead it simulates a
function return and the control is passed to the calling function. If you
want to execute the rest of the code use the `Until return' option.
("Executing until return" Section 14.10.5).

Name of the command: cmeDbgReturnNow.

14.11 Finishing and restarting a debug session
==============================================

Sometimes you'll want to finish the process you are currently debugging
without farther execution or even finish the whole debug session. This
section describes the available options related to it.

If you want to finish the current process without wainting until it finishes
you can `kill' it.  ("Killing the program you are debugging" Section
14.11.1). This operation is also called restart.

You can `kill' the process and finish the debug session. In this case the
debugger will also finish execution.  ("Closing the debug session" Section
14.11.2).

Finally you can `kill' the process, `kill' the debugger and instruct the
editor to forget about breakpoints, watchpoints, `etc.' That's a dangerous
operation used only when no other operation helps.   ("Destroying the debug
session" Section 14.11.3).

14.11.1 Killing the program you are debugging
---------------------------------------------

This functionality is obtained using the `Debug|Restart (Kill)' menu entry.

This operation will finish the execution of the program you are debugging and
go to the `Ready to Run' state.  ("Debugging states" Section 14.6).

After it you can start executing your program from the beginning.

Name of the command: cmeDbgKill.

14.11.2 Closing the debug session
---------------------------------

This functionality is obtained using the `Debug|Debug session|Close' menu
entry.

This operation will finish the execution of the program you are debugging and
then will finish the debugger. After it the state will be `Disconnected'.
("Debugging states" Section 14.6).

After it you can start a new debug session.

Name of the command: cmeDbgCloseSession.

14.11.3 Destroying the debug session
------------------------------------

This functionality is obtained using the `Debug|Debug session|Destroy' menu
entry.

This operation isn't recommended for normal use. In addition to kill the
program you are debugging and finishing the debugger it also makes the editor
to forget about breakpoints, watchpoints and other settings related to the
debug session.

After it you can start a new debug session.

Name of the command: cmeDbgEndSession.

14.12 Examining the calling stack
=================================

This functionality is obtained using the `Debug|Calling stack' menu entry.

When your program stopped you may want to know which functions were called in
order to reach the current source line. Using this command the editor will
send the list of functions to the Message Window.  ("Message Window" Section
15.8).

After sending the list of functions to the Message Window the editor will
jump to the first one. Using `Alt'+`F7' and `Alt'+`F8' you can navigate this
list.

In addition to the source file, function name, line of code and memory
address you'll also find the arguments of the function.

Name of the command: cmeDbgCallStack.
Assigned key: `Ctrl+F3'

14.13 Cleaning the debug session
================================

This dialog is obtained using the `Debug|Debug session|Clear debug elements'
menu entry.

Usually you'll want the editor to remmeber the breakpoints, watchpoints,
watches, `etc.' used in your last debug session. For this reason this data is
stored in the project file. But sometimes you could want to clean all and do
a fresh start. This command allows you to delete all of this or just a
portion. You erase items of the following types:

   * Breakpoints.  ("Advanced breakpoint options" Section 14.8.2).

   * Watchpoints.  ("Watchpoints" Section 14.8.3).

   * Watches.  ("Watch an expression" Section 14.9.2).

   * Inspectors.  ("Inspectors" Section 14.9.4).

   * Data windows.  ("Data Window" Section 14.9.5).

Name of the command: cmeDbgCleanElem.

14.14 Selecting the thread to debug
===================================

This dialog is obtained using the `Debug|Select thread' menu entry.

GDB can debug multithread programs that uses POSIX threads. The starting
thread is the one you'll stop, step, trace, `etc.' If you want to select
another thread to debug you can use this option. Note that gdb uses its own
thread ids. The editor will offer a list of known threads and indicate which
one is the current thread. GDB doesn't have any MI command to know which one
is the current thread, for this reason the editor assumes that's the last
thread id informed during a stop.

Debugging multithread programs is complicated and I don't have good test
examples. If you need to enhance the current options and want to cooperate
with me please contact me.

Name of the command: cmeDbgThreadSel.

14.15 Disassembler Window
=========================

This functionality is obtained using the `Debug|Disassembler Window' menu
entry.

If you need to debug your program at the instruction level or debug some code
that lacks debug information that's the tool you need.

The Disassembler Window is divided in two sections. The left side contains
the assembler code and the right side the CPU registers. The code side
displays the assembler using a read-only edition window. If the code have
debug information you'll see the source code mixed as comments.

The `Step over', `Trace into' and `Go to cursor' operations works at
instruction level when applied to this window.

You can manually open this window. Additionlly the editor will open this
window every time the execution stops in a function without debug
information. When the editor can determine the function name and the source
file that contains it the whole function will be disassembled. If only the
function name can be determined the editor will disassemble from the
beginning of the function upto 500 bytes after the current location. If
nothing can be determined the editor will disassemble 500 bytes from the
current location.

Sometimes gdb informs bogus function locations. What's worst is that
disassembling from the informed place upto the current location can be too
slow and even make gdb to SIGSEGV. To avoid it the editor won't disassemble
from the beginning of the function if the distance is bigger than 20000 bytes.

When this window is opened the code section is selected. You can switch to
the registers section using the `Tab' key. Selecting a register and using the
`Enter' key you can modify its value.

If you get an error from gdb stating `bad register number' when you open this
window please try upgrading gdb. That's a bug in the MI code. I submitted a
patch to fix this problem to the proper list. It seems to affect gdb 6.1 and
6.2.

Name of the command: cmeDbgDisAsmWin.

14.16 Debugging already running processes
=========================================

You can debug a process that's already running in memory. This is really
useful when your program exposes an uncommon bug that's hard to reproduce.

The steps to do it are almost the same used for regular programs. Only a few
details changes:

   * As the program is already running you have to stop it instead of run it.

   * When you finish debugging you could want to let it run without farther
     intrusion.

The first action is called `atach' and the second `detach'.

To start debugging an already running process you must select this option in
the `Debug Options' dialog.  ("Debug options" Section 14.5.1). Then you must
know the process id for the process you want to debug. This can be found
using the `ps' POSIX command.

To atach gdb to the running process you must use the
`Debug|Run/Continue/Atach' menu option. It will ask you the process id (pid).

Once gdb is atached to the running process it is stopped and the state is
changed to `Stopped'.  ("Debugging states" Section 14.6).

When you no longer need to debug the process you can detach from it using the
`Debug|Debug session|Detach' menu option. The process will resume its normal
execution.

Name of the command: cmeDbgDetach.

14.17 Debug Messages Window
===========================

This window is used to:

   * Show gdb messages.  ("Messages displayed" Section 14.5.3).

   * Show the current debugging state.  ("Debugging states" Section 14.6).

   * Show extra information about the current state. It includes errors from
     gdb or in the communication with gdb.

   * Send commands to gdb.

When you start a new debug session this window becomes visible. When the
session is closed the window is hided. When the window is hided you can
access it using the list of windows dialog.  ("List" Section 5.6.8).

The maximum ammount of messages displayed in this window is limited to avoid
wasting memory. You can configure this value using the advanced options
dialog.  ("Advanced debug options" Section 14.5.4).

To send commands to gdb use the `Insert' key.

14.18 Editing a debug expression
================================

When a dialog asks for an expression to watch, modify, `etc.' you can enter
any expression that gdb can analize. This includes arithmetic operations and
function calls.

14.19 Debugging in the Linux console
====================================

When you debug in a Linux console the editor uses a separated virtual
terminal for the program you want to debug. The terminal is automatically
selected but you can specify it using the debug options dialog.   ("Debug
options" Section 14.5.1).

One reason to specify a fixed terminal is when you have more than one video
card. In this case you can allocate some virtual terminals in the secondary
video board and use them to run the program to debug.

When using only one video card the debug window should be in some visible part
of the editor's desktop. This window indicates the current status and hence
you can know if your program is stopped or running.  ("Debug Messages Window"
Section 14.17).

Remmember that when the editor is running in a Linux console you must use the
`Ctrl' + `Alt' + `Fn' key combination to switch to another virtual terminal.

15 Miscellaneous
****************

15.1 Configuration files location
=================================

  The configuration files are read from various directories allowing to use
system wide defaults and also user defined settings. The mechanism used to
determine which file to load depends on the OS. For DOS and Windows the
editor doesn't know much about users and uses an approach that's different to
the one used for POSIX systems (Linux, FreeBSD, QNX, Solaris, `etc.').  The
Cygwin version of the editor is considered as a POSIX version even when
that's a Windows binary.

15.1.1 Configuration directories for DOS and Windows
----------------------------------------------------

  Here is the list of directories where the editor looks for a configuration
file. Note that Cygwin binaries are an exception and they behave as binaries
compiled for POSIX systems.  ("Configuration directories for POSIX systems"
Section 15.1.2).

  One concept used during the search is the `home directory'. It is determined
using the `HOME' environment variable. If this variable isn't defined the
editor tries with `HOMEDIR'. We represent it using `~'.

  To represent the content of an environment variable the examples use
`%VARIABLE%'. In the examples we will represent the configuration file naming
it `FILE'.

  1. `~/FILE'

  2. `%SET_FILES%/FILE'

  3. The directory where you started the editor.

  The `SET_FILES' environment variable should point to the directory where
system wide configuration files are stored. Note that if the `SET_FILES'
environment variable isn't defined the editor tries to figure out it using
the following sequence:

  1. The installation directory detected or specified at compilation time
     adding the `share/setedit' directory.

  2. The directory where the binary is located eliminating the `bin' part and
     adding `share/setedit' directory.

15.1.2 Configuration directories for POSIX systems
--------------------------------------------------

  Here is the list of directories where the editor looks for a configuration
file. Note it includes Cygwin binaries.

  One concept used during the search is the `home directory'. It is determined
using the `HOME' environment variable. If this variable isn't defined the
editor tries with `HOMEDIR'. We represent it using `~'.

  To represent the content of an environment variable the examples use
`$VARIABLE'. In the examples we will represent the configuration file naming
it `FILE'. In POSIX systems a hidden file is created using a dot a preffix.

  1. `~/.setedit/FILE' (Note: this directory is created if it doesn't exist).

  2. `~/.setedit/.FILE'

  3. `~/FILE'

  4. `~/.FILE'

  5. `$SET_FILES/FILE'

  The `SET_FILES' environment variable should point to the directory where
system wide configuration files are stored. Note that if the `SET_FILES'
environment variable isn't defined the editor tries to figure out it using
the following sequence:

  1. The installation directory detected or specified at compilation time
     adding the `share/setedit' directory.

  2. `/usr/share/setedit'

  3. `/usr/local/share/setedit'

15.2 Clipboard
==============

The clipboard is just another editor window, which holds the text you used the
`Cut' and `Copy' commands on, and retrieve it with the `Paste' command.

Unlike the windows clipboard the one provided with the editor doesn't lose
the old contents when you copy to it. This approach has the important
advantage that you can copy text from various parts to the clipboard and then
paste all the text in one place with just one operation. To do it select the
clipboard window, select the text to paste and the editor pastes the text
selected in the clipboard window.

The disadvantage is that all the text copied to the clipboard remains there
and if you are running on a machine with low memory and handling huge files
you can fill all the memory quickly. To avoid this check the clipboard size
(with `Alt+0') and exit the program if it gets too big. That normally isn't
needed on machines with enough free disk space that can be used as swap.

15.3 Time and date modifiers formats
====================================

This values are the same used by the `strftime' function of the standard
`libc'. The editor uses it in the printing module.  ("Print Setup" Section
5.1.11).

   * `%A'
     - Meaning: The full weekday name (`Friday')

   * `%a'
     - Meaning: The abbreviated weekday name (`Fri')

   * `%B'
     - Meaning: The full month name (`October')

   * `%b, %h'
     - Meaning: The abbreviated month name (`Oct')

   * `%C'
     - Meaning: Short for `%a %b %e %H:%M:%S %Y' (`Fri Oct  1 15:30:34 1993')

   * `%c'
     - Meaning: Short for `%m/%d/%y %H:%M:%S' (`10/01/93 15:30:34')

   * `%e'
     - Meaning: The day of the month

   * `%D'
     - Meaning: Short for `%m/%d/%y' (`10/01/93')

   * `%d'
     - Meaning: The day of the month

   * `%H'
     - Meaning: The hour (0-24)

   * `%I'
     - Meaning: The hour (1-12)

   * `%j'
     - Meaning: The Julian day

   * `%k'
     - Meaning: The hour (0-24)

   * `%l'
     - Meaning: The hour (1-12)

   * `%M'
     - Meaning: The minutes

   * `%m'
     - Meaning: The month (1-12)

   * `%n'
     - Meaning: A newline (`n')

   * `%p'
     - Meaning: AM or PM (`PM')

   * `%R'
     - Meaning: Short for `%H:%M' (`15:30')

   * `%r'
     - Meaning: Short for `%I:%M:%S %p' (`03:30:35 PM')

   * `%S'
     - Meaning: The seconds

   * `%T, %X'
     - Meaning: Short for `%H:%M:%S' (`15:30:35')

   * `%t'
     - Meaning: A tab (`t')

   * `%U'
     - Meaning: The week of the year

   * `%W'
     - Meaning: The week of the year

   * `%w'
     - Meaning: The day of the week (0-6) (`5')

   * `%x'
     - Meaning: Short for `%m/%d/%y' (`10/01/93')

   * `%y'
     - Meaning: The year (00-99) of the century (`93')

   * `%Y'
     - Meaning: The year

   * `%Z'
     - Meaning: The timezone abbreviation (`EDT')

   * `%%'
     - Meaning: A percent symbol (`%')


15.4 Regular Expressions
========================

The editor supports regular expressions in the search and replace commands,
here is a description of the syntax.

Regular expressions ("RE"s), as defined in POSIX 1003.2, come in two forms:
modern REs (roughly those of `egrep'; 1003.2 calls these _extended_ REs) and
obsolete REs (roughly those of `ed'; 1003.2 _basic_ REs).  Obsolete REs
mostly exist for backward compatibility in some old programs; they will be
discussed at the end.  1003.2 leaves some aspects of RE syntax and semantics
open; `(*)' marks decisions on these aspects that may not be fully portable
to other 1003.2 implementations.

A (modern) RE is one(*) or more non-empty(*) _branches_, separated by `|'.
It matches anything that matches one of the branches.

A branch is one(*) or more _pieces_, concatenated.  It will try to find a
match for the first, then for the second, etc.

A piece is an _atom_ possibly followed by a single(*) `*', `+', `?', or
_bound_.  An atom followed by `*' matches a sequence of 0 or more occurences
of the atom.  An atom followed by `+' matches a sequence of 1 or more
occurences of the atom.  An atom followed by `?' matches a sequence of 0 or 1
occurences of the atom.

A _bound_ is a `{' followed by an unsigned decimal integer, possibly followed
by `,' possibly followed by another unsigned decimal integer, always followed
by `}'.  The integers must lie between 0 and `RE_DUP_MAX' (255(*)) inclusive,
and if there are two of them, the first may not exceed the second.  An atom
followed by a bound containing one integer `i' and no comma matches a
sequence of exactly `i' occurences of the atom.  An atom followed by a bound
containing one integer `i' and a comma matches a sequence of `i' or more
occurences of the atom.  An atom followed by a bound containing two integers
`i' and `j' matches a sequence of `i' through `j' (inclusive) occurences of
the atom.

An atom is a regular expression enclosed in `()' (one occurence for the
regular expression), an empty set of `()' (matches the null string(*)), a
_bracket expression_ (see below), `.' (matching any single character), `^'
(matching the null string at the beginning of a line), `$' (matching the null
string at the end of a line), a `\\' followed by one of the characters
`^.[$()|*+?{\\' (matches that character taken as an ordinary character), a
`\\' followed by any other character(*) (matches that character taken as an
ordinary character, as if the `\\' had not been present(*)), or a single
character with no other significance (matching that character).  A `{'
followed by a character other than a digit is an ordinary character, not the
beginning of a bound(*).  It is illegal to end an RE with `\\'.

A _bracket expression_ is a list of characters enclosed in `[]'.  It normally
matches any single character from the list (but see below).  If the list
begins with `^', it matches any single character (but see below) _not_ from
the rest of the list.  If two characters in the list are separated by `-',
this is shorthand for the full _range_ of characters between those two
(inclusive) in the used character set, e.g. `[0-9]' matches any decimal digit
in ASCII.  It is illegal(*) for two ranges to share an endpoint, e.g.
`a-c-e'.  Ranges are very character set-dependent, and portable programs
should avoid relying on them.

To include a literal `]' in the list, make it the first character (following
a possible `^').  To include a literal `-', make it the first or last
character, or the second endpoint of a range.  To use a literal `-' as the
first endpoint of a range, enclose it in `[.' and `.]' to make it a collating
element (see below).  With the exception of these and some combinations using
`[' (see next paragraphs), all other special characters, including `\\', lose
their special significance within a bracket expression.

Within a bracket expression, a collating element (a character, a
multi-character sequence that collates as if it were a single character, or a
collating-sequence name for either) enclosed in `[.' and `.]'  stands for the
sequence of characters of that collating element.  The sequence is a single
element of the bracket expression's list.  A bracket expression containing a
multi-character collating element can thus match more than one character,
e.g. if the collating sequence includes a `ch' collating element, then the RE
`[[.ch.]]*c' matches the first five characters of "chchcc".

Within a bracket expression, a collating element enclosed in `[=' and `=]' is
an equivalence class, standing for the sequences of characters of all
collating elements equivalent to that one, including itself.  (If there are
no other equivalent collating elements, the treatment is as if the enclosing
delimiters were `[.' and `.]'.)  For example, if o and \o'o^' are the members
of an equivalence class, then `[[=o=]]', `[[=\o'o^'=]]', and `[o\o'o^']' are
all synonymous.  An equivalence class may not be an endpoint of a range.

Within a bracket expression, the name of a _character class_ enclosed in `[:'
and `:]' stands for the list of all characters belonging to that class.
Standard character class names are:

     alnum	digit	punct
     alpha	graph	space
     blank	lower	upper
     cntrl	print	xdigit

These stand for the character classes defined by `isalnum', `isdigit',
`ispunct', `isalpha', `isgraph' , `isspace' (`blank' is the same as `space'),
`islower', `isupper', `iscntrl', `isprint', and `isxdigit', respectively.  A
locale may provide others.  A character class may not be used as an endpoint
of a range.

There are two special cases(*) of bracket expressions: the bracket
expressions `[[:<:]]' and `[[:>:]]' match the null string at the beginning
and end of a word respectively.  A word is defined as a sequence of word
characters which is neither preceded nor followed by word characters.  A word
character is an `alnum' character (as defined by the `isalnum' library
function) or an underscore.  This is an extension, compatible with but not
specified by POSIX 1003.2, and should be used with caution in software
intended to be portable to other systems.

In the event that an RE would match more than one substring of a given
string, the RE matches the one starting earliest in the string.  If the RE
could match more than one substring starting at that point, it matches the
longest.  Subexpressions also match the longest possible substrings, subject
to the constraint that the whole match be as long as possible, with
subexpressions starting earlier in the RE taking priority over ones starting
later.  Note that higher-level subexpressions thus take priority over their
lower-level component subexpressions.

Match lengths are measured in characters, not collating elements.  A null
string is considered longer than no match at all.  For example, `bb*' matches
the three middle characters of `abbbc', `(wee|week)(knights|nights)' matches
all ten characters of `weeknights', when `(.*).*' is matched against `abc' the
parenthesized subexpression matches all three characters, and when `(a*)*' is
matched against `bc' both the whole RE and the parenthesized subexpression
match the null string.

If case-independent matching is specified, the effect is much as if all case
distinctions had vanished from the alphabet.  When a letter that exists in
multiple cases appears as an ordinary character outside a bracket expression,
it is effectively transformed into a bracket expression containing both
cases, e.g. `x' becomes `[xX]'.  When it appears inside a bracket expression,
all case counterparts of it are added to the bracket expression, so that
(e.g.) `[x]' becomes `[xX]' and `[^x]' becomes `[^xX]'.

No particular limit is imposed on the length of REs(*).  Programs intended to
be portable should not employ REs longer than 256 bytes, as an implementation
can refuse to accept such REs and remain POSIX-compliant.

Obsolete (_basic_) regular expressions differ in several respects.  `|', `+',
and `?' are ordinary characters and there is no equivalent for their
functionality.  The delimiters for bounds are `\\{' and `\\}', with `{' and
`}' themselves being ordinary characters.  The parentheses for nested
subexpressions are `\(' and `\)', with `(' and `)' themselves being ordinary
characters.  `^' is an ordinary character except at the beginning of the RE
or(*) the beginning of a parenthesized subexpression, `$' is an ordinary
character except at the end of the RE or(*) the end of a parenthesized
subexpression, and `*' is an ordinary character if it appears at the
beginning of the RE or the beginning of a parenthesized subexpression (after
a possible leading `^').  Finally, there is one new type of atom, a _back
reference_: `\\' followed by a non-zero decimal digit _d_ matches the same
sequence of characters matched by the _d_th parenthesized subexpression
(numbering subexpressions by the positions of their opening parentheses, left
to right), so that (e.g.) `\\([bc]\\)\\1' matches `bb' or `cc' but not `bc'.

15.5 Desktop Files
==================

Each time you run the editor it searches for a desktop file in the current
directory, if the editor can't find any desktop file it searches in the
directory indicated by the enviroment variable `SET_FILES'. If no desktop
files are found the editor uses internal default values. By creating a
desktop file in the `%SET_FILES%' directory you'll indicate default values.
To create a desktop file in this directory just run the editor in the
directory, customize it and leave the editor. This feature is very useful to
customize things like: Colors, Palette, Global editor options, `etc.'

When you exit from the editor it saves all the settings, windows positions,
`etc.' to a desktop file stored in the current directory. Some people don't
like this and therefore a special mode in which the editor stores just one
desktop file in the `%SET_FILES%' directory and not in each directory exists.
This approach has the advantage of saving disk space, but you won't have
local settings. If you use this approarch and want to keep a local
configuration in a directory use a project.  ("Project" Section 5.8). To
store only one centralized desktop file:  ("Editor General" Section 5.7.1.4).

The desktop files have `.dst' as extension.

15.6 Text mode attributes
=========================

Text modes uses up to 16 colors for forground and 16 colors for background,
that's because the attributes of each character are stored in a 1 byte
allowing just 256 combinations. So four bits are dedicated to the background
color giving 16 combinations and the other four bits for the foreground.

As we have only 16 colors there is a big chance that our preferred colors
aren't included. For this reason the VGA chip uses a palette of colors. That
means that these 16 colors aren't fixed and you can indicate what colors to
use. The values are just index values and you can assign to it any color.

The VGA chip supports 18(6:6:6) bits per color giving 262,144 combinations.
The colors are created using the RGB (Red Green and Blue) method. That's
simply because the CRT (Cathode Rays Tube) of the monitor uses this method to
create the colors. You have 6 bits per component giving 64 different levels.

There is a little more of complexity added to the VGA card. The most
significant bit of the background is used to create blinking text. The editor
avoids this mode because it restricts the background combinations to 8 and
the editor doesn't need blinking text. The other funny thing of the attribute
is the most significant bit of the foreground color. It has a special meaning
but even when using it, it doesn't reduce the number of combinations.  That
means that this bit can select two things at the same time. By now you'll be
asking: What? This bit can be used to select a second font, in this way you
can have upto 512 different characters on the screen. The editor can exploit
it but you must understand that this bit is used for the foreground color
too. You can customize the palette to reduce the number of foreground
combinations to 8 and defining the 9th color with the same value as the 0th
color and so on. Then you can assign different colors and different fonts to
the things in the editor. You can for example use a font for the menues and
other for the text, or a font for the code and other for the comments.  The
editor goes even further allowing different code page encodings for each
font. I really never saw it in any editor. You can use it for example to
write your code using a font with the ISO Latin 1 encoding (used by Linux and
Windows in the USA and part of Europe) and have your comments in russian
(using cyrillic characters).

To learn how to customize the palette:  ("Color Palette" Section 5.7.1.2).
To learn how to assign colors to the things used in the editor:  ("Customize
Colors" Section 5.7.1.1).
To learn how to select a font and an encoding:  ("Screen Options" Section
5.7.1.15).
15.7 File Open
==============

This dialog is used to select a file in various places. Even when it has
different names and purposes the dialog is always the same.

The dialog is similar to the one used by most of the T/GUI programs so I'll
focus on special details and particularities.

All the files and directories are shown together, the directories have a
slash at the end. The current directory and selected file information is
shown at the bottom of the dialog.

To select a different disk, at least in DOS, you can simply type the drive
letter followed by a colon in the text input area and press <Enter>. You can
specify a mask in the text area too. Even though under DOS a single asterisk
matches any filename, the *.* mask won't match with a filename that lacks
extension. You can use some asterisks and question marks in the mask, they
work like in most of the shells (DOS command.com included). Additionally the
mask can contain brackets, if your shell doesn't use it and you never used it
here is an example: `*.[ch]*' will match with test.c, test.h, test.cc,
test.cpp, `etc.' the brackets indicates the possible characters that will
match, in this case the first character after the point should be c or h.
Pressing the down arrow while you are in the text input area you'll get a
list of the paths in which you selected files in the past. That's very useful
when you load files from two or more very different directories.

The `Home' changes the directory to the one from where you started the
editor, that's very useful when you navigate a lot and you want to quickly
return to the directory from where you started.

The `Options' button brings a dialog to configure the sorting options and
which files to exclude. These options are described in the following
subsections.

The editor remembers the last place where you selected a file the last time
you used it. This information is stored in the desktop file. This information
is unique for the following selections: open a file to edit, save a file,
open a help file, save a block of text, open/save a project, open an MP3 file
and save an MP3 file. In this way you can be opening files from a directory
and saving newly created files to another at the same time without indicating
the directory each time you open/save a file.

15.7.1 Sorting of the files and directories in the dialog
---------------------------------------------------------

By default the sorting criteria is the following:

   * Names sorting is case insensitive.

   * Directories are listed after files.

   * The parent directory (`..') is the last entry.

   * Files starting with a dot are alphabetically listed.

But this can be customized with the `File open dialog' menu option or the
`Options' button of the dialog. It brings a dialog with the following options:

   * Sort type: controls how the names are sorted.
        * Alphabetical: both, directories and file names, are mixed.

        * Directories first: the directories are listed first.

        * Files first: file names are listed first.


   * Case style: controls if the dialog differentiates between capital and
     lower case ones.
        * Capital letters goes first: the names are sorted according to the
          ASCII table

        * Case insensitive: the names are sorted lexicographically.


   * Parent directory: this option controls the location of the parent
     directory link (`..'). It doesn't affect the `Alphabetical' sorting.
        * First in the list: the parent is the first entry in the list.

        * At the end of the list: the parent is the last entry in the list.


   * Files starting with a dot: it controls how names starting with a dot are
     handled.
        * Normally sorted: they will be sorted alphabetically. As the `ASCII'
          code of a dot is less than the code of any letter they will be
          first in the list. That's quite annoying when you use UNIX style
          backups and they are also hidden files.

        * After the rest: they will be put at the end of the list.

The other options you'll find in this configuration dialog are described in
another subsection.  ("Files and directories excluded in the dialog" Section
15.7.2).

Something very important that you must know is how the <Shift> key is
interpreted.

When the sorting is `Alphabetical' and the list is case sensitive (`Capital
letters are first') the shift key affects the case of the typed letters so
you must be careful; check the state of <Caps Lock> now and then as well.

When the sorting isn't `Alphabetical' the <Shift> has a special meaning. If
you press shift while typing the first letter of the name then the search
will be done in the list of directories instead of the list of file names.
Once you are in one of the lists the rest of the search is done in this list.
If the list is case sensitive it creates an interesting conflict because you
could need to press shift for some file name and then the search will be done
in the directories list. Don't forget this.

15.7.2 Files and directories excluded in the dialog
---------------------------------------------------

By default all names found are listed but this can be customized with the
`File open dialog' menu option or the `Options' button of the dialog. The
related options are:

   * Exclude files: controls which names are excluded.
        * Ending with tilde: names ending with a tilde are excluded. They
          usually belong to UNIX-style backups. The editor generates these
          files when you configure it to generate UNIX style backups.

        * Ending with .bkp: names with the bkp extension are excluded. By
          default the editor generates backup files with this extension.

        * Starting with dot: names starting with a dot are excluded. In UNIX
          they are hidden files. The editor can be configured to create
          backup files as hidden files.

Personally I only exclude files with the bkp extension and sort files
starting with a dot so they appear at the end of the list.

15.8 Message Window
===================

This window is used to show important information that you may want to have
at hand. Some examples are: the results of a printing operation, the errors
collected from an external program, the hits of the powered grep, the tip of
the day, the output of an external program, etc.

The message window doesn't have a number but can be accessed from the list of
windows (default `Alt+0') or from the `Windows | Message Window' menu option.

If the messages are errors or hits from grep you can use `Alt+F7' and
`Alt+F8' to quickly jump to the next/previous line. You can also just select
any of the error/hits. The errors are parsed by the editor according to the
user selection. Many formats are supported by the editor and the mechanism is
configurable. See the next section for more information.

When the last or first error in the list is reached the editor indicates it.
This behavior can be configured.  ("Editor General" Section 5.7.1.4).

You can also delete entries in the message window pressing <Delete>. To add
more flexibility the content of the message window can be stored in a file or
copied to the clipboard, to do this just use the menu.

For more information about the behavior of the message window when you run
external programs  ("Run program (which one)" Section 5.7.1.8).

15.9 Error messages from an external application
================================================

Originally the editor only supported the GNU style. After many users asked
for support of other error formats Grzegorz Adam Hankiewicz suggested doing
it in a configurable way. So starting with v0.4.41 the editor can be
configured to parse the errors from an external application.

The configuration is stored in a file called `errors.cle'. The syntax is very
similar to the one used in the `syntaxhl.shl' file. All definitions start
with a `Name' declaration that indicates the name of the parsing options and
ends with an `End' marker.  ("Configuration files location" Section 15.1).

The `Pattern' entry tells the editor how to parse a line containing an error
from the external program. The pattern is actually a Perl regular expression.
I used it because they are much more easy to learn than POSIX regular
expressions and they are a lot more logical to me.

The `File', `Line', `Severity' and `Description' entries indicate - with
subexpressions - the file name, line number, degree of severity and
description of the error. Note that actually only the first two are used. If
you don't know what a subexpression is here is a hint: look at the
parentheses.

The `EnterDirPat' is another Perl regex to indicate how make informs about a
directory change. `EnterDirDir' is the related subexpression.  That's needed
for GNU make, I guess other make tools have a similar mechanism. `LeaveDir'
is the pattern generated by make to indicate a change to the parent directory.

15.10 Mouse under Linux
=======================

When you are running in a console the mouse is captured by the editor and you
can't use it to copy/paste between consoles. If you want to do it you must
hold down the right <Alt> key and use the mouse as usually. Note this can be
altered if you changed the alt keys configuration.   ("Alt key configuration"
Section 4.2). Additionally note that when you paste the keys are interpreted
as <Alt> plus the key you pressed so the carriage returns aren't interpreted
very well.

If you are using an X terminal the mouse isn't interactive because X
terminals only report when the button is pressed and released but not when
it's moved. You can do all the normal operations but you won't get the usual
feedback.

15.11 Passing extra command line options
========================================

You can pass extra command line options defining an environment variable.
It's called SET_CMDLINE. The editor will parse this variable and add it to
the options passed in the command line. These options will be interpreted
before the options passed in the command line.

Any space (including tabs) are interpreted as separators. If you need to pass
a file name containing spaces you must enclose the name with double quotes.
If you need to pass a name containing a double quote you must escape it using
a back slash. Here are some examples:

   * ops 1
     - The editor interpretes: two options "ops" and "1"

   * "ops 1"
     - The editor interpretes: one option "ops 1"

   * ops"1
     - The editor interpretes: one option containing a double quote inside


This mechanism can be used to set default options and avoid typing them all
the time. Another use is when you need to pass options to the editor and you
are using Eterm; which can't pass command line options, so you must use this
mechanism.

15.12 How to run setedit remotely without root installation (UNIX)
==================================================================

This text was contributed by Grzegorz Adam Hankiewicz:

I'm a lucky guy and I've been hired by eFaber (http://www.efaber.net) to do
some open source hacking. The first day I started I was given a computer and
received strong orders: `Install Suse Linux 8.1 and customize your
environment'. Obviously I chose to install setedit, and it worked great.
However, from time to time I have to go to another workstation or login
remotely. We export `/home' through NFS, so apparently we have the same
configuration everywhere. But setedit doesn't work that way.

First of all, setedit depends on the TVision library, and only my PC has it
installed. setedit also needs a global directory where shared files reside,
usually `/usr/share/setedit', or something else pointed by `SET_FILES'.  One
solution would be to install setedit everywhere, but this is not scalable,
and it's error prone. If I wanted to upgrade setedit or TVision I would spend
much time copying and installing on every machine. Besides, even if my
coworkers let me use root, or told me the password, I would never be given
access to the main server, and it's the only one with email I/O. What's the
solution?

The solution is to put the files needed by setedit on the exported directory,
in other words, `$HOME'. After you compile setedit, put the binary in
`~/bin'. Find where your TVision is (librhtv*) and copy the library to
`~/lib'. If you run `ldd' on `~/bin/setedit' you will see that the librhtv
dependency is resolved to your local file system. If you logged now to
another computer, the library wouldn't be found.

To avoid this, add something similar to this to your .bashrc:

        export PATH=$PATH:~/bin
        export LD_LIBRARY_PATH=~/lib

After you log in again, try to use ldd on the binary, and now you should see
the linker resolve it to your `~/lib' everywhere.  First step complete.  The
second step is to move the shared files.  I created the directory in
`~/bin/setedit_shared_files', and made `/usr/share/setedit' on my local
machine a symlink to it, so newer installs of setedit overwrite the correct
shared files.  Now make `SET_FILES' point to it:

        export SET_FILES=~/bin/setedit_shared_files

Nice. Now, if you run setedit it will be able to load, but you will still
find two annoying facts: setedit always looks for the `setedit.info' file
(which is usually found in `/usr/info/setedit.info.gz'), and the language
translations won't be found because GNU's gettext doesn't know anything about
our exportation trick.

To solve the info file warning, copy setedit.info.gz to
`~/bin/setedit_shared_files' and add that directory to the appropriate
environment variable:

        export INFOPATH=$INFOPATH:$SET_FILES

Finally, to correct the gettext problem, first make `SET_LOCELEDIR' point to
`SET_FILES':

        export SET_LOCALEDIR=$SET_FILES

And now, in the `SET_FILES' directory create the `xxxx/LC_MESSAGES/'
directory structure, where `xxxx' is a two letter code representing a
language.  I would then create `~/bin/setedit_shared_files/es/LC_MESSAGES',
and put inside the file `setedit.mo' I found typing `locate setedit.mo' on my
workstation.  After copying the file there, GNU's gettext will find the
translation file always, from any directory, and you will be able to hack
from every computer as if you were at home.

Again, what you have to add to your `.bashrc' file is:

        export PATH=$PATH:~/bin
        export LD_LIBRARY_PATH=~/lib
        export SET_FILES=~/bin/setedit_shared_files
        export INFOPATH=$INFOPATH:$SET_FILES
        export SET_LOCALEDIR=$SET_FILES

Happy hacking!

16 Index
********

* About the Author:                      Section 1.3
* Address - Change base address (DW):    Section 14.9.5.15



17 Index of key commands
************************

* 0x0003 <9 x 16>:                       Section 5.7.1.15
* 0x0103 <9 x 14>:                       Section 5.7.1.15
* 0x0108 <9 x  8>:                       Section 5.7.1.15
* 0x0109 <9 x 14>:                       Section 5.7.1.15
* 0x010A <9 x 11>:                       Section 5.7.1.15
* 0x010B <9 x 10>:                       Section 5.7.1.15
* 0x010C <9 x  8>:                       Section 5.7.1.15
* 0x0203 <9 x 10>:                       Section 5.7.1.15
* 0x0303 <9 x 10>:                       Section 5.7.1.15
* 0x0403 <9 x  8>:                       Section 5.7.1.15
* 0x0503 <9 x  8>:                       Section 5.7.1.15
* 0x0703 <9 x 16>:                       Section 5.7.1.15
* 0x0803 <9 x 14>:                       Section 5.7.1.15
* 0x0903 <9 x 16>:                       Section 5.7.1.15
* 0x0A03 <9 x 14>:                       Section 5.7.1.15
* 0x0B03 <9 x 16>:                       Section 5.7.1.15
* 0x0C03 <9 x 14>:                       Section 5.7.1.15
* 0x0D03 <8 x 16>:                       Section 5.7.1.15
* 132 x 25 <9 x 14>:                     Section 5.7.1.15
* 132 x 43 <9 x 11>:                     Section 5.7.1.15
* 132 x 50 <9 x 10>:                     Section 5.7.1.15
* 132 x 60 <9 x  8>:                     Section 5.7.1.15
* 80 x 25 <9 x 16>:                      Section 5.7.1.15
* 80 x 28 <9 x 14>:                      Section 5.7.1.15
* 80 x 30 <9 x 16>:                      Section 5.7.1.15
* 80 x 34 <9 x 14>:                      Section 5.7.1.15
* 80 x 35 <9 x 10>:                      Section 5.7.1.15
* 80 x 40 <9 x 10>:                      Section 5.7.1.15
* 80 x 43 <9 x  8>:                      Section 5.7.1.15
* 80 x 50 <9 x  8>:                      Section 5.7.1.15
* 80 x 60 <9 x  8>:                      Section 5.7.1.15
* 82 x 25 <8 x 16>:                      Section 5.7.1.15
* 90 x 30 <9 x 16>:                      Section 5.7.1.15
* 90 x 34 <9 x 14>:                      Section 5.7.1.15
* 94 x 30 <9 x 16>:                      Section 5.7.1.15
* 94 x 34 <9 x 14>:                      Section 5.7.1.15
* Alternate case <none>:                 Section 3.4
* Autoindent mode on/off <^O>:           Section 3.5



