added textile-mode and mmm-mode. xpath stuff

[emacs-init.git] / mmm-mode-0.4.8 / mmm.texinfo

\input texinfo
@c %**start of header
@setfilename mmm.info
@settitle MMM Mode Manual
@c %**end of header
@syncodeindex vr fn
@set MASON_VERSION 0.896

@dircategory GNU Emacs Lisp
@direntry
* MMM-Mode: (mmm).                 Multiple Major Modes for Emacs
@end direntry

@include version.texi

@ifinfo
This is edition @value{EDITION} of the MMM Mode Manual, last updated
@value{UPDATED}. It documents version @value{VERSION} of MMM Mode.

Copyright 2000 Michael Abraham Shulman.

Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
     
@ignore
Permission is granted to process this file through TeX and print the
results, provided the printed document carries a copying permission
notice identical to this one except for the removal of this paragraph
(this paragraph not being relevant to the printed manual).

@end ignore
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided also that the
sections entitled ``Copying'' and ``GNU General Public License'' are
included exactly as in the original, and provided that the entire
resulting derived work is distributed under the terms of a permission
notice identical to this one.
     
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation.

@end ifinfo

@titlepage
@title MMM Mode Manual
@subtitle Multiple Major Modes for Emacs
@subtitle Edition @value{EDITION}
@subtitle @value{UPDATED}
@author Michael Abraham Shulman
@page
@vskip 0pt plus 1filll
Copyright @copyright{} 2000 Michael Abraham Shulman.

Permission is granted to make and distribute verbatim copies of this
manual provided the copyright notice and this permission notice are
preserved on all copies.
     
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided also that the
sections entitled ``Copying'' and ``GNU General Public License'' are
included exactly as in the original, and provided that the entire
resulting derived work is distributed under the terms of a permission
notice identical to this one.
     
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation
approved by the Free Software Foundation.

@end titlepage

@ifinfo
@node Top, Overview, (dir), (dir)
@top MMM Mode

MMM Mode is a minor mode for Emacs which allows Multiple Major Modes to
coexist in a single buffer.

This is edition @value{EDITION} of the MMM Mode Manual, last updated
@value{UPDATED}, which documents version @value{VERSION} of MMM Mode.

@end ifinfo

@menu
* Overview::                    An overview and introduction to MMM Mode.
* Basics::                      The basics of how to use it.
* Customizing::                 Customizing how it works to your needs.
* Supplied Classes::            The supplied submode classes.
* Writing Classes::             Writing your own submode classes.
* Indices::                     Just that.

@detailmenu
 --- The Detailed Node Listing ---

Overview of MMM Mode

* Basic Concepts::              A simple explanation of how it works.
* Installation::                How to install MMM Mode.
* Quick Start::                 Getting started using MMM Mode quickly.

MMM Mode Basics

* MMM Minor Mode::              The Emacs minor mode that manages it all.
* Submode Classes::             What they are and how to use them.
* Selecting Classes::           How MMM Mode knows what classes to use.
* Insertion::                   Inserting new submode regions automatically.
* Re-parsing::                  Re-scanning for submode regions.
* Interactive::                 Adding submode regions manually.
* Global Mode::                 Turning MMM Mode on automatically.

The MMM Minor Mode

* Enabling MMM Mode::           Turning MMM Mode on and off.
* MMM Mode Keys::               Default key bindings in MMM Mode.

How MMM Mode selects submode classes

* File Classes::                Classes for a single file.
* Mode-Ext Classes::            Classes for a given mode or extension.
* Global Classes::              Classes for all MMM Mode buffers.

MMM Global Mode

* Major Mode Hook::             Using MMM's Major Mode Hook

Customizing MMM Mode

* Region Coloring::             Changing or removing background colors.
* Preferred Modes::             Choosing which major modes to use.
* Mode Line::                   What is displayed in the mode line.
* Key Bindings::                Customizing the MMM Mode key bindings.
* Local Variables::             What local variables are saved for submodes.
* Changing Classes::            Changing the supplied submode classes.
* Hooks::                       How to make MMM Mode run your code.

Supplied Submode Classes

* Mason::                       Mason server-side Perl in HTML.
* File Variables::              Elisp code in File Variables.
* Here-documents::              Code in shell and Perl here-documents.
* Javascript::                  Javascript embedded in HTML.
* Embedded CSS::                CSS Styles embedded in HTML.
* Embperl::                     Another syntax for Perl in HTML.
* ePerl::                       A general Perl-embedding syntax.
* JSP::                         Java code embedded in HTML.
* RPM::                         Shell scripts in RPM Spec Files.
* Noweb::                       Noweb literate programs.

Writing Submode Classes

* Basic Classes::               Writing a simple submode class.
* Paired Delimiters::           Matching paired delimiters.
* Region Placement::            Placing the region more accurately.
* Submode Groups::              Grouping several classes together.
* Calculated Submodes::         Deciding the submode at run-time.
* Calculated Faces::            Deciding the display face at run-time.
* Insertion Commands::          Inserting regions automatically.
* Region Names::                Naming regions for syntax grouping.
* Other Hooks::                 Running code at arbitrary points.
* Delimiters::                  Controlling delimiter overlays.
* Misc Keywords::               Other miscellaneous options.

Indices

* Concept Index::               Index of MMM Mode Concepts.
* Function Index::              Index of functions and variables.
* Keystroke Index::             Index of key bindings in MMM Mode.

@end detailmenu
@end menu

@node Overview, Basics, Top, Top
@comment  node-name,  next,  previous,  up
@chapter Overview of MMM Mode
@cindex overview of mmm-mode
@cindex mmm-mode, overview of

MMM Mode is a minor mode for Emacs which allows Multiple Major Modes to
coexist in a single buffer. The name is an abbreviation of `Multiple
Major Modes'@footnote{The name is derived from @file{mmm.el} for XEmacs
by Gongquan Chen <chen@@posc.org>, from which MMM Mode was adapted.}. A
major mode is a customization of Emacs for editing a certain type of
text, such as code for a specific programming language. @xref{Major
Modes, , , emacs, The Emacs Manual}, for details.

MMM Mode is a general extension to Emacs which is useful whenever one
file contains text in two or more programming languages, or that
should be in two or more different modes.  For example:

@itemize @bullet
@item
CGI scripts written in any language, from Perl to PL/SQL, may want to
output verbatim HTML, and the writer of such scripts may want to use
Emacs' html-mode or sgml-mode to edit this HTML code, while remaining
in the appropriate programming language mode for the rest of the
file.  @xref{Here-documents}, for example.

@item
There are now many ``content delivery systems'' which turn the CGI
script idea around and simply add extra commands to an HTML file,
often in some programming language, which are interpreted on the
server.  @xref{Mason}, @xref{Embperl}, @xref{ePerl}, @xref{JSP}.

@item
HTML itself can also contain embedded languages such as Javascript and
CSS styles, for which Emacs has different major modes.
@xref{Javascript}, and @xref{Embedded CSS}, for example.

@item
The idea of ``literate programming'' requires the same file to contain
documentation (written as text, html, latex, etc.) and code (in an
appropriate programming language).  @xref{Noweb}, for example.

@item
Emacs allows files of any type to contain `local variables', which can
include Emacs Lisp code to be evaluated. @xref{File Variables, , ,
emacs, The Emacs Manual}. It may be easier to edit this code in Emacs
Lisp mode than in whatever mode is used for the rest of the file.
@xref{File Variables}.

@item
There are many more possible uses for MMM Mode.  RPM spec files can
contain shell scripts (@pxref{RPM}).  Email or newsgroup messages may
contain sample code.  And so on.  We encourage you to experiment.
@end itemize

@menu
* Basic Concepts::              A simple explanation of how it works.
* Installation::                How to install MMM Mode.
* Quick Start::                 Getting started using MMM Mode quickly.
@end menu

@node Basic Concepts, Installation, Overview, Overview
@comment  node-name,  next,  previous,  up
@section Basic Concepts
@cindex dominant major mode
@cindex major mode, dominant
@cindex default major mode
@cindex major mode, default
@cindex submode regions
@cindex regions, submode
@cindex overlays, submode
@cindex submode overlays
@cindex mmm-ification

The way MMM Mode works is as follows. Each buffer has a @dfn{dominant}
or @dfn{default} major mode, which is chosen as major modes normally
are: the user can set it interactively, or it can be chosen
automatically with `auto-mode-alist' (@pxref{Choosing Modes, , , emacs,
The Emacs Manual}). Within the file, MMM Mode creates @dfn{submode
regions} within which other major modes are in effect. While the point
is in a submode region, the following changes occur:

@enumerate
@item
The local keymap is that of the submode. This means the key bindings for
the submode are available, while those of the dominant mode are not.
@item
The mode line (@pxref{Mode Line, , , emacs, The Emacs Manual}) changes
to show which submode region is active. This can be configured; see
@ref{Mode Line}.
@item
The major mode menu, both on the menu bar and the mouse popup, are that
of the submode.
@item
Some local variables of the submode shadow those of the default mode
(@pxref{Local Variables}). For the user, this serves to help make Emacs
behave as if the submode were the major mode.
@item
The syntax table and indentation are those of the submode.
@item
Font-lock (@pxref{Font Lock, , , emacs, The Emacs Manual}) fontifies
correctly for the submode.
@item
The submode regions are highlighted by a background color; see
@ref{Region Coloring}.

@end enumerate

The submode regions are represented internally by Emacs Lisp objects
known as @dfn{overlays}. Some of the above are implemented by overlay
properties, and others are updated by an MMM Mode function in
`post-command-hook'. You don't need to know this to use MMM Mode, but it
may make any error messages you come across more understandable.
@xref{Overlays, , , elisp, The GNU Emacs Lisp Reference Manual}, for
more information on overlays.

Because overlays are not saved with a file, every time a file is opened,
they must be created.  Creating submode regions is occasionally referred
to as @dfn{mmm-ification}.  (I've never had occasion to pronounce this,
but if I did I would probably say `mummification'. Like what they did in
ancient Egypt.)  You can mmm-ify a buffer interactively, but most often
MMM Mode will find and create submode regions automatically based on a
buffer's file extension, dominant mode, or local variables.


@node Installation, Quick Start, Basic Concepts, Overview
@comment  node-name,  next,  previous,  up
@section Installing MMM Mode

MMM Mode has a standard installation process.  See the file INSTALL for
generic information on this process.  To summarize, unpack the archive,
@command{cd} to the created MMM Mode directory, type @samp{./configure},
then @samp{make}, then @samp{make install}.  If all goes correctly, this
will compile the MMM Mode elisp files, install them in your local
site-lisp directory, and install the MMM Mode info file @file{mmm.info}
in your local info directory.

Now you need to configure your Emacs initialization file (usually
@file{~/.emacs}) to use MMM Mode.  First, Emacs has to know where to
find MMM Mode.  In other words, the MMM Mode directory has to be in
@code{load-path}.  This can be done in the parent directory's
@file{subdirs.el} file, or in the init file with a line such as:

@lisp
(add-to-list 'load-path "/path/to/site-lisp/mmm/")
@end lisp

Once @code{load-path} is configured, MMM Mode must be loaded.  You can
load all of MMM Mode with the line

@lisp
(require 'mmm-mode)
@end lisp

@noindent
but if you use MMM Mode only rarely, it may not be desirable to load all
of it at the beginning of every editing session.  You can load just
enough of MMM Mode so it will turn itself on when necessary and load the
rest of itself, by using instead the line

@lisp
(require 'mmm-auto)
@end lisp

@noindent
in your initialization file.

One more thing you may want to do right now is to set the variable
@code{mmm-global-mode}.  If this variable is @code{nil} (the default),
MMM Mode will never turn itself on.  If it is @code{t}, MMM Mode will
turn itself on in every buffer.  Probably the most useful value for it,
however, is the symbol @code{maybe} (actually, anything that is not
@code{nil} and not @code{t}), which causes MMM Mode to turn itself on in
precisely those buffers where it would be useful.  You can do this with
a line such as:

@lisp
(setq mmm-global-mode 'maybe)
@end lisp

@noindent
in your initialization file.  @xref{Global Mode}, for more detailed
information.


@node Quick Start,  , Installation, Overview
@comment  node-name,  next,  previous,  up
@section Getting Started Quickly

Perhaps the simplest way to create submode regions is to do it
interactively by specifying a region. First you must turn MMM Mode
on---say, with @kbd{M-x mmm-mode}---then place point and mark around the
area you want to make into a submode region, type @kbd{C-c % C-r}, and
enter the desired major mode. @xref{Interactive}, for more details.

A better way to add submode regions is by using submode classes, which
store a lot of useful information for MMM Mode about how to add and
manipulate the regions created.  @xref{Submode Classes}, for more
details.  There are several sample submode classes that come with MMM
Mode, which are documented later in this manual.  Look through these and
determine if one of them fits your needs.  If so, I suggest reading the
comments on that mode.  Then come back here to find out to use it.

To apply a submode class to a buffer interactively, turn MMM Mode on as
above, then type @kbd{C-c % C-c} and enter the name of the class.
Submode regions should be added automatically, if there are any regions
in the buffer appropriate to the submode class.

If you want a given file to always use a given submode class, you can
express this in a file variable: add a line containing the string
@samp{-*- mmm-classes: @var{class} -*-} at the top of the file.
@xref{File Variables, , , emacs, The Emacs Manual}, for more information
and other methods. Now whenever MMM Mode is turned on in that file, it
will be mmm-ified according to @var{class}. If @code{mmm-global-mode} is
non-nil, then MMM Mode will turn itself on whenever a file with a
@code{mmm-classes} local variable is opened. @xref{Global Mode}, for more
information.

If you want a submode class to apply to @emph{all} files in a certain
major mode or with a certain extension, add a line such as this to your
initialization file:

@lisp
(mmm-add-mode-ext-class @var{mode} @var{extension} @var{class})
@end lisp

@noindent
After this call, any file opened whose name matches the regular
expression @var{extension} @emph{and} whose default mode is @var{mode}
will be automatically mmm-ified according to @var{class} (assuming
@code{mmm-global-mode} is non-nil). If one of @var{extension} or
@var{mode} is @code{nil}, a file need only satisfy the other one to be
mmm-ified.

You can now read the rest of this manual to learn more about how MMM
Mode works and how to configure it to your preferences.  If none of the
supplied submode classes fit your needs, then you can try to write your
own.  @xref{Writing Classes}, for more information.

@node Basics, Customizing, Overview, Top
@comment  node-name,  next,  previous,  up
@chapter MMM Mode Basics

This chapter explains the most important parts of how to use MMM Mode.

@menu
* MMM Minor Mode::              The Emacs minor mode that manages it all.
* Submode Classes::             What they are and how to use them.
* Selecting Classes::           How MMM Mode knows what classes to use.
* Insertion::                   Inserting new submode regions automatically.
* Re-parsing::                  Re-scanning for submode regions.
* Interactive::                 Adding submode regions manually.
* Global Mode::                 Turning MMM Mode on automatically.
@end menu

@node MMM Minor Mode, Submode Classes, Basics, Basics
@comment  node-name,  next,  previous,  up
@section The MMM Minor Mode
@cindex mode, mmm minor
@cindex minor mode, mmm
@cindex mmm minor mode

An Emacs minor mode is an optional feature which can be turned on or off
in a given buffer, independently of the major mode. @xref{Minor Modes, ,
, emacs, The Emacs Manual}. MMM Mode is implemented as a minor mode
which manages the submode regions. This minor mode must be turned on in
a buffer for submode regions to be effective. When activated, the MMM
Minor mode is denoted by @samp{MMM} in the mode line (@pxref{Mode
Line}).

@menu
* Enabling MMM Mode::           Turning MMM Mode on and off.
* MMM Mode Keys::               Default key bindings in MMM Mode.
@end menu


@node Enabling MMM Mode, MMM Mode Keys, MMM Minor Mode, MMM Minor Mode
@comment  node-name,  next,  previous,  up
@subsection Enabling MMM Mode
@cindex mmm mode, turning on
@cindex mmm mode, turning off
@cindex turning on mmm mode
@cindex turning off mmm mode
@cindex mmm mode, enabling
@cindex mmm mode, disabling
@cindex enabling mmm mode
@cindex disabling mmm mode

If @code{mmm-global-mode} is non-@code{nil} (@pxref{Global Mode}),
then the MMM minor mode will be turned on automatically whenever a file
with associated submode classes is opened (@pxref{Selecting Classes}).
It is also turned on by interactive mmm-ification (@pxref{Interactive}),
although the interactive commands do not have key bindings when it is
not on and must be invoked via @kbd{M-x}. You can also turn it on (or
off) manually with @kbd{M-x mmm-mode}, in which case it applies all
submode classes associated with the buffer. Turning MMM Mode off
automatically removes all submode regions from the buffer.

@deffn Command mmm-mode @var{arg}
Toggle the state of MMM Mode in the current buffer. If @var{arg} is
supplied, turn MMM Mode on if and only if @var{arg} is positive.
@end deffn

@defun mmm-mode-on
Turn MMM Mode on unconditionally in the current buffer.
@end defun

@defun mmm-mode-off
Turn MMM Mode off unconditionally in the current buffer.
@end defun

@defvar mmm-mode
This variable represents whether MMM Mode is on in the current buffer.
Do not set this variable directly; use one of the above functions.
@end defvar


@node MMM Mode Keys,  , Enabling MMM Mode, MMM Minor Mode
@comment  node-name,  next,  previous,  up
@subsection Key Bindings in MMM Mode
@cindex mmm mode key bindings
@cindex key bindings in mmm mode
@findex mmm-insertion-help
@kindex C-c % h

When MMM Mode is on, it defines a number of key bindings. By default,
these are bound after the prefix sequence @kbd{C-c %}. Minor mode
keymaps are supposed to use @kbd{C-c @var{punctuation}} sequences, and I
find this one to be a good mnemonic because @samp{%} is used by Mason to
denote special tags. This prefix key can be customized; @ref{Key
Bindings}.

There are two types of key bindings in MMM Mode: @dfn{commands} and
@dfn{insertions}. Command bindings run MMM Mode interactive functions to
do things like re-parse the buffer or end the current submode region,
and are defined statically as normal Emacs key-bindings. Insertion
bindings insert submode region skeletons with delimiters into the
buffer, and are defined dynamically, according to which submode classes
(@pxref{Submode Classes}) are in effect, via a keymap default binding.

To distinguish between the two, MMM Mode uses distinct modifier keys for
each. By default, command bindings use the control key (e.g. @kbd{C-c %
C-b} re-parses the buffer), and insertion bindings do not (e.g. @kbd{C-c
% p}, when the Mason class is in effect, inserts a
@samp{<%perl>...</%perl>} region). This makes the command bindings
different from in previous versions, however, so the variable
@code{mmm-use-old-bindings} is provided. If this variable is set to `t'
before MMM Mode is loaded, the bindings will be reversed: insertion
bindings will use the control key and command bindings will not.

Normally, Emacs gives help on a prefix command if you type @kbd{C-h}
after that command (e.g. @kbd{C-x C-h} displays all key bindings
starting with @kbd{C-x}). Because of how insertion bindings are
implemented dynamically with a default binding, they do not show up when
you hit @kbd{C-c % C-h}. For this reason, MMM Mode defines the command
@kbd{C-c % h} which displays a list of all currently valid insertion key
sequences. If you use the defaults for command and insertion bindings,
the @kbd{C-h} and @kbd{h} should be mnemonic.

In the rest of this manual, I will assume you are using the defaults for
the mode prefix (@kbd{C-c %}) and the command and insertion modifiers.
You can customize them, however; @ref{Key Bindings}.


@node Submode Classes, Selecting Classes, MMM Minor Mode, Basics
@comment  node-name,  next,  previous,  up
@section Understanding Submode Classes
@cindex submode classes
@cindex classes, submode

A submode class represents a ``type'' of submode region. It specifies
how to find the regions, what their delimiters look like, what submode
they should be, how to insert them, and how they behave in other ways.
It is represented by a symbol, such as @code{mason} or
@code{eval-elisp}.

For example, in the Mason set of classes, there is one class
representing all @samp{<%...%>} inline Perl regions, and one
representing regions such as @samp{<%perl>...</%perl>},
@samp{<%init>...</%init>}, and so on. These are different to Mason, but
to Emacs they are all just Perl sections, so they are covered by the
same submode class.

But it would be tedious if whenever we wanted to use the Mason classes,
we had to specify both of these. (Actually, this is a simplification:
there are some half a dozen Mason submode classes.) So submode classes
can also ``group'' others together, and we can refer to the @code{mason}
class and mean all of them.

The way a submode class is used is to @dfn{apply} it to a buffer. This
scans the buffer for regions which should be submode regions according
to that class, and also remembers the class for later, so that new
submode regions can be inserted and scanned for later.


@node Selecting Classes, Insertion, Submode Classes, Basics
@comment  node-name,  next,  previous,  up
@section How MMM Mode selects submode classes

Submode classes that apply to a buffer come from three sources:
mode/extension-associated classes, file-local classes, and interactive
MMM-ification (@pxref{Interactive}). Whenever MMM Mode is turned on in a
buffer (@pxref{MMM Minor Mode}, and @ref{Global Mode}), it inspects the
value of two variables to determine which classes to automatically apply
to the buffer. This covers the first two sources; the latter is covered
in a later chapter.

@menu
* File Classes::                Classes for a single file.
* Mode-Ext Classes::            Classes for a given mode or extension.
* Global Classes::              Classes for all MMM Mode buffers.
@end menu


@node File Classes, Mode-Ext Classes, Selecting Classes, Selecting Classes
@comment  node-name,  next,  previous,  up
@subsection File-Local Submode Classes

@defvar mmm-classes
This variable is always buffer-local when set. Its value should be
either a single symbol or a list of symbols. Each symbol represents a
submode class that is applied to the buffer.
@end defvar

@code{mmm-classes} is usually set in a file local variables list.
@xref{File Variables, , , emacs, The Emacs Manual}. The easiest way to
do this is for the first line of the file to contain the string
@samp{-*- mmm-classes: @var{classes} -*-}, where @var{classes} is the
desired value of @code{mmm-classes} for the file in question. It can
also be done with a local variables list at the end of the file.


@node Mode-Ext Classes, Global Classes, File Classes, Selecting Classes
@comment  node-name,  next,  previous,  up
@subsection Submode Classes Associated with Modes and Extensions

@defopt mmm-mode-ext-classes-alist
This global variable associates certain submode classes with major modes
and/or file extensions. Its value is a list of elements of the form
@code{(@var{mode} @var{ext} @var{class})}. Any buffer whose major mode
is @var{mode} (a symbol) @emph{and} whose file name matches @var{ext} (a
regular expression) will automatically have the submode class
@var{class} applied to it.

If @var{mode} is @code{nil}, then only @var{ext} is considered to
determine if a buffer fits the criteria, and vice versa. Thus if both
@var{mode} and @var{ext} are nil, then @var{class} is applied to
@emph{all} buffers in which MMM Mode is on. Note that @var{ext} can be
any regular expression, although its name indicates that it most often
refers to the file extension.

If @var{class} is the symbol @code{t}, then no submode class is actually
applied for this association. However, if @code{mmm-global-mode} is
non-@code{nil} and non-@code{t}, MMM Mode will be turned on in matching
buffers even if there are no actual submode classes being applied.
@xref{Global Mode}.
@end defopt

@defun mmm-add-mode-ext-class @var{mode} @var{ext} @var{class}
This function adds an element to @code{mmm-mode-ext-classes-alist},
associating the submode class @var{class} with the major mode @var{mode}
and extension @var{ext}.

Older versions of MMM Mode required this function to be used to control
the value of @code{mmm-mode-ext-classes-alist}, rather than setting it
directly. In this version it is provided purely for convenience and
backward compatibility.
@end defun


@node Global Classes,  , Mode-Ext Classes, Selecting Classes
@comment  node-name,  next,  previous,  up
@subsection Globally Applied Classes and the Universal Class

In addition to file-local and mode-ext-associated submode classes, MMM
Mode also allows you to specify that certain submode classes apply to
@emph{all} buffers in which MMM Mode is enabled.

@defopt mmm-global-classes
This variable's value should be a list of submode classes that apply to
all buffers with MMM Mode on.  It can be overriden in a file local
variables list, such as to disable global class for a specific file.
Its default value is @code{(universal)}.
@end defopt

The default global class is the ``universal class'', which is defined in
the file @file{mmm-univ.el} (loaded automatically), and allows the
author of text to specify that a certain section of it be in a specific
major mode.  Thus, for example, when writing an email message that
includes sample code, the author can allow readers of the message (who
use emacs and MMM) to view the code in the appropriate major mode.  The
syntax used is @samp{@{%@var{mode}%@} ... @{%/@var{mode}%@}}, where
@var{mode} should be the name of the major mode, with or without the
customary @samp{-mode} suffix: for example, both @samp{cperl} and
@samp{cperl-mode} are acceptable.

The universal class also defines an insertion key, @samp{/}, which
prompts for the submode to use.  @xref{Insertion}.  The universal class
is most useful when @code{mmm-global-mode} is set to @code{t};
@ref{Global Mode}.


@node Insertion, Re-parsing, Selecting Classes, Basics
@comment  node-name,  next,  previous,  up
@section Inserting new submode regions

So much for noticing submode regions already present when you open a
file. When editing a file with MMM Mode on, you will often want to add a
new submode region. MMM Mode provides several facilities to help you.
The simplest is to just hit a few keys and have the region and its
delimiters inserted for you.

Each submode class can define an association of keystrokes with
``skeletons'' to insert a submode region. If there are several submode
classes enabled in a buffer, it is conceivable that the keys they use
for insertion might conflict, but unlikely as most buffers will not use
more than one or two submode classes groups.

As an example of how insertion works, consider the Mason classes. In a
buffer with MMM Mode enabled and Mason associated, the key sequence
@kbd{C-c % p} inserts the following perl section (the semicolon is to
prevent CPerl Mode from getting confused---@pxref{Mason}):

@example
<%perl>-<-;
-!-
->-</%perl>
@end example

In this schematic representation, the string @samp{-!-} represents the
position of point (the cursor), @samp{-<-} represents the beginning of
the submode region, and @samp{->-} its end.

All insertion keys come after the MMM Mode prefix keys (by default
@kbd{C-c %}; @pxref{Key Bindings}) and are by default single characters
such as @kbd{p}, @kbd{%}, and @kbd{i}. To avoid confusion, all the MMM
Mode commands are bound by default to control characters (after the same
prefix keys), such as @kbd{C-b}, @kbd{C-%} and @kbd{C-r}. This is a
change from earlier versions of MMM Mode, and can be customized; see
@ref{Key Bindings}.

To find out what insertion keys are available, consult the documentation
for the submode class you are using. If it is one of the classes
supplied with MMM Mode, you can find it in this Info file.

Because insertion keys are implemented with a ``default binding'' for
flexibility, they do not show up in the output of @kbd{C-h m} and cannot
be found with @kbd{C-h k}. For this reason, MMM Mode supplies the
command @kbd{C-c % h} (@code{mmm-insertion-help} to view the available
insertion keys.


@node Re-parsing, Interactive, Insertion, Basics
@comment  node-name,  next,  previous,  up
@section Re-Parsing Submode Regions
@cindex re-parsing submode regions
@cindex parsing submode regions
@cindex submode regions, re-parsing
@cindex regions, submode, re-parsing
@cindex submode regions, clearing
@cindex clearing submode regions
@cindex regions, submode, clearing
@kindex C-c % C-b
@kindex C-c % C-g
@kindex C-c % C-%
@kindex C-c % C-5
@kindex C-c % C-k

Describe @code{mmm-parse-buffer}, @code{mmm-parse-region},
@code{mmm-parse-block}, and @code{mmm-clear-current-region}.

@node Interactive, Global Mode, Re-parsing, Basics
@comment  node-name,  next,  previous,  up
@section Interactive MMM-ification Functions
@cindex interactive mmm-ification
@cindex mmm-ification, interactive
@cindex mmm-ification by region
@cindex mmm-ification by regexp
@cindex mmm-ification by class
@cindex region, mmm-ification by
@cindex regexp, mmm-ification by
@cindex class, mmm-ification by
@kindex C-c % C-r
@kindex C-c % C-c
@kindex C-c % C-x
@cindex mmm-ification, interactive history
@cindex history of interactive mmm-ification
@cindex interactive mmm-ification, history of

There are several commands you can use to create submode regions
interactively, rather than by applying a submode class to a buffer.
These commands (in particular, @code{mmm-ify-region}), can be useful
when editing a file or email message containing a snippet of code in
some other language.  Also see @ref{Global Classes}, for an alternate
approach to the same problem.

@table @kbd
@item C-c % C-r
Creates a submode region between point and mark. Prompts for the submode
to use, which must be a valid Emacs major mode name, such as
@code{emacs-lisp-mode} or @code{cperl-mode}. Adds markers to the
interactive history. (@code{mmm-ify-region})

@item C-c % C-c
Applies an already-defined submode class to the buffer, which it prompts
for. Adds this class to the interactive history.
(@code{mmm-ify-by-class})

@item C-c % C-x
Scans the buffer for submode regions (prompts for the submode) using
front and back regular expressions that it also prompts for. Briefly, it
starts at the beginning of the buffer and searches for the front regexp.
If it finds a match, it searches for the back regexp. If it finds a
match for that as well, it makes a submode region between the two
matches and continues searching until no more matches are found. Adds
the regexps to the interactive history. (@code{mmm-ify-by-regexp})

@end table

These commands are also useful when designing a new submode class
(@pxref{Submode Classes}). Working with the regexps interactively can
make it easier to debug and tune the class before starting to use it on
automatic. All these commands also add to value of the following
variable.

@defvar mmm-interactive-history
Stores a history of all interactive mmm-ification that has been
performed in the current buffer. This way, for example, the re-parsing
functions (@pxref{Re-parsing}) will respect interactively added regions,
and the insertion keys for classes that were added interactively are
available.
@end defvar

If for any reason you want to ``wipe the slate clean'', this command
should help you. By default, it has no key binding, so you must invoke
it with @kbd{M-x mmm-clear-history @key{RET}}.

@deffn Command mmm-clear-history
Clears all history of interactive mmm-ification in the current buffer.
This command does not affect existing submode regions; to remove them,
you may want to re-parse the buffer with @kbd{C-c % C-b}
(@code{mmm-parse-buffer}).
@end deffn


@node Global Mode,  , Interactive, Basics
@comment  node-name,  next,  previous,  up
@section MMM Global Mode
@cindex mode, mmm global
@cindex global mmm mode
@cindex mmm global mode
@vindex mmm-never-modes

When a file has associated submode classes (@pxref{Selecting Classes}),
you may want MMM Mode to turn itself on and parse that file for submode
regions automatically whenever it is opened in an Emacs buffer. The
value of the following variable controls when MMM Mode turns itself on
automatically.

@defopt mmm-global-mode
Do not be misled by the fact that this variable's name ends in
@samp{-mode}: it is not a simple on/off switch. There are three possible
(meanings of) values for it: @code{t}, @code{nil}, and anything else.

When this variable is @code{nil}, MMM Mode is never enabled
automatically. If it is enabled manually, such as by typing @kbd{M-x
mmm-mode}, any submode classes associated with the buffer will still be
used, however.

When this variable is @code{t}, MMM Mode is enabled automatically in
@emph{all} buffers, including those not visiting files, except those
whose major mode is an element of @code{mmm-never-modes}. The default
value of this variable contains modes such as @code{help-mode} and
@code{dired-mode} in which most users would never want MMM Mode, and
in which MMM might cause problems.

When this variable is neither @code{nil} nor @code{t}, MMM Mode is
enabled automatically in all buffers that would have associated submode
classes; i.e. only if there would be something for it to do. The value
of @code{mmm-never-modes} is still respected, however. Note that this
can include buffers not visiting files, if that buffer's major mode is
present in @code{mmm-mode-ext-classes-alist} with a @code{nil} value for
@var{ext} (@pxref{Mode-Ext Classes}). Submode class values of @code{t}
in @code{mmm-mode-ext-classes-alist} cause MMM Mode to be enabled in
matching buffers, but supply no submode classes to be applied.
@end defopt

@menu
* Major Mode Hook::             Using MMM's Major Mode Hook
@end menu


@node Major Mode Hook,  , Global Mode, Global Mode
@comment  node-name,  next,  previous,  up
@subsection The Major Mode Hook
@cindex hook, major mode
@cindex major mode hook
@vindex mmm-major-mode-hook

This section is intended for users who understand Emacs Lisp and want to
know how MMM Global Mode is implemented, and perhaps use the same
technique. In fact, MMM Mode exports a hook variable that you can use
easily, without understanding any of the details---see below.

In order to enable itself in @emph{all} buffers, however, MMM Mode has
to hook itself into all major modes.  Global Font Lock Mode from the
standard Emacs distribution (@pxref{Font Lock, , , emacs, The Emacs
Manual}) has a similar problem, and solves it by adding a function to
@code{change-major-mode-hook}, which is run by
@code{kill-all-local-variables}, which is run in turn by all major mode
functions at the @emph{beginning}.  This function stores a list of which
buffers need fontification.  It then adds a different function to
@code{post-command-hook}, which checks if the current buffer needs
fontification, and if so performs it.  MMM Global Mode uses the same
technique.

In the interests of generality, and for your use, the function that MMM
Mode runs in @code{post-command-hook} (@code{mmm-run-major-mode-hook})
is not specific to MMM Mode, but rather runs the hook variable
@code{mmm-major-mode-hook}, which by default contains a function
(@code{mmm-mode-on-maybe}) which possibly turns MMM Mode on, depending
on the value of @code{mmm-global-mode}.  Thus, to run another function
in all major modes, all you need to do is add it to this hook.  For
example, the following line in an initialization file will turn on Auto
Fill Mode (@pxref{Auto Fill, , , emacs, The Emacs Manual}) in all
buffers:

@lisp
(add-hook 'mmm-major-mode-hook 'turn-on-auto-fill)
@end lisp

@node Customizing, Supplied Classes, Basics, Top
@comment  node-name,  next,  previous,  up
@chapter Customizing MMM Mode

This chapter explains how to customize the appearance and functioning of
MMM Mode however you want.

@menu
* Region Coloring::             Changing or removing background colors.
* Preferred Modes::             Choosing which major modes to use.
* Mode Line::                   What is displayed in the mode line.
* Key Bindings::                Customizing the MMM Mode key bindings.
* Local Variables::             What local variables are saved for submodes.
* Changing Classes::            Changing the supplied submode classes.
* Hooks::                       How to make MMM Mode run your code.
@end menu

@node Region Coloring, Preferred Modes, Customizing, Customizing
@comment  node-name,  next,  previous,  up
@section Customizing Region Coloring
@cindex faces, submode
@cindex submode faces
@cindex customizing submode faces
@cindex default submode face

By default, MMM Mode highlights all submode regions with a background
color.  There are three levels of this decoration, controlled by the
following variable:

@defopt mmm-submode-decoration-level
This variable controls the level of coloring of submode regions.  It
should be one of the integers 0, 1, or 2, representing (respectively)
none, low, and high coloring.
@end defopt

No coloring means exactly that.  Submode regions have the same
background as the rest of the text.  This produces the minimal
interference with font-lock coloration.  In particular, if you want to
use background colors for font-lock, this may be a good idea, because
the submode highlight, if present, overrides any font-lock background
coloring.

Low coloring uses the same background color for all submode regions.
This color is specified with the face @code{mmm-default-submode-face}
(@pxref{Faces, , , emacs, The Emacs Manual}) which can be customized,
either through the Emacs ``customize'' interface or using direct Lisp
commands such as @code{set-face-background}.  Of course, other aspects
of the face can also be set, such as the foreground color, bold,
underline, etc.  These are more likely to conflict with font-lock,
however, so only a background color is recommended.

High coloring uses multiple background colors, depending on the function
of the submode region.  The recognized functions and their meanings are
as follows:

@table @samp
@item init
Code that is executed at the beginning of (something), as initialization
of some sort.

@item cleanup
Code that is executed at the end of (something), as some sort of clean
up facility.

@item declaration
Code that provides declarations of some sort, perhaps global or local
arguments, variables, or methods.

@item comment
Text that is not executed as code, but instead serves to document the
code around it.  Submode regions of this function often use a mode such
as Text Mode rather than a programming language mode.

@item output
An expression that is evaluated and its value interpolated into the
output produced.

@item code
Executed code not falling under any other category.

@item special
Submode regions not falling under any other category, such as component
calls.

@end table

The different background colors are provided by the faces
@code{mmm-@var{function}-submode-face}, which can be customized in the
same way as @code{mmm-default-submode-face}.


@node Preferred Modes, Mode Line, Region Coloring, Customizing
@comment  node-name,  next,  previous,  up
@section Preferred Major Modes

Certain of the supplied submode classes know only the language that
certain sections are written in, but not what major mode you prefer to
use to edit such code.  For example, many people prefer CPerl mode over
Perl mode; you may have a special mode for Javascript or just use C++
mode.  This variable allows you to tell submodes such as Mason
(@pxref{Mason}) and Embedded Javascript (@pxref{Javascript}) what major
mode to use for the submodes:

@defopt mmm-major-mode-preferences
The elements of this list are cons cells of the form
@code{(@var{language} . @var{mode})}.  @var{language} should be a symbol
such as @code{perl}, @code{html-js}, or @code{java}, while @var{mode}
should be the name of a major mode such as @code{perl-mode},
@code{cperl-mode}, @code{javascript-mode}, or @code{c++-mode}.

You probably won't have to set this variable at all; MMM tries to make
intelligent guesses about what modes you prefer.  For example, if a
function called @code{javascript-mode} exists, it is chosen, otherwise
@code{c++-mode} is used.  Similarly for @code{jde-mode} and
@code{java-mode}.
@end defopt

If you do need to change the defaults, you may find the following
function convenient.

@defun mmm-set-major-mode-preferences @var{language} @var{mode} &optional @var{default}
Set the preferred major mode for LANGUAGE to MODE.  If there is already
a mode specified for LANGUAGE, and DEFAULT is nil or unsupplied, then it
is changed.  If DEFAULT is non-nil, then any existing mode is unchanged.
This is used by packages to ensure that some mode is present, but not
override any user-specified mode.  If you are not writing a submode
class, you should ignore the third argument.
@end defun

Thus, for example, to use @code{my-java-mode} for Java code, you would
use the following line:

@lisp
(mmm-set-major-mode-preferences 'java 'my-java-mode)
@end lisp


@node Mode Line, Key Bindings, Preferred Modes, Customizing
@comment  node-name,  next,  previous,  up
@section Customizing the Mode Line Display

By default, when in a submode region, MMM Mode changes the section of
the mode line (@pxref{Mode Line, , , emacs, The Emacs Manual}) that
normally displays the major mode name---for example, @samp{HTML}---to
instead show both the dominant major mode and the currently active
submode---for example, @samp{HTML[CPerl]}.  You can change this format,
however.

@defopt mmm-submode-mode-line-format
The value of this variable should be a string containing one or both of
the escape sequences @samp{~M} and @samp{~m}.  The string displayed in
the major mode section of the mode line when in a submode is obtained by
replacing all occurrences of @samp{~M} with the dominant major mode name
and @samp{~m} with the currently active submode name.  For example, to
display only the currently active submode, set this variable to
@samp{~m}.  The default value is @samp{~M[~m]}.
@end defopt

The MMM minor mode also normally displays the string @samp{MMM} in the
minor mode section of the mode line to indicate when it is active.  You
can customize or disable this as well.

@defopt mmm-mode-string
This string is displayed in the minor mode section of the mode line when
the MMM minor mode is active.  If nonempty, it should begin with a space
to separate the MMM indicator from that of other minor modes.  To
eliminate the indicator entirely, set this variable to the empty string.
@end defopt


@node Key Bindings, Local Variables, Mode Line, Customizing
@comment  node-name,  next,  previous,  up
@section Customizing the MMM Mode Key Bindings

The default MMM Mode key bindings are explained in @ref{MMM Mode Keys},
and in @ref{Insertion}.  There are a couple of ways to customize these
bindings.

@defopt mmm-mode-prefix-key
The value of this variable (default is @kbd{C-c %}) should be a key
sequence to use as the prefix for the MMM Mode keymap.  Minor modes
typically use @kbd{C-c} followed by a punctuation character, but you can
change it to any user-available key sequence.  To have an effect, this
variable should be set before MMM Mode is loaded.
@end defopt

@defopt mmm-use-old-command-keys
When this variable is @code{nil}, MMM Mode commands use the control
modifier and insertion keys no modifier.  Any other value switches the
two, so that @code{mmm-parse-buffer}, for example, is bound to @kbd{C-c
% b}, while perl-section insertion in the Mason class is bound to
@kbd{C-c % C-p}.  This variable should be set before MMM Mode is loaded
to have an effect.
@end defopt

When MMM is loaded, it uses the value of @code{mmm-use-old-command-keys}
to set the values of the variables @code{mmm-command-modifiers} and
@code{mmm-insert-modifiers}, so if you prefer you can set these
variables instead.  They should each be a list of key modifiers, such as
@code{(control)} or @code{()}.  The Meta modifier is used in some of the
command and insertion keys, so it should not be used, and the Shift
modifier is not particularly portable between Emacsen---if it works for
you, feel free to use it.  Other modifiers, such as Hyper and Super, are
not universally available, but are valid when present.


@node Local Variables, Changing Classes, Key Bindings, Customizing
@comment  node-name,  next,  previous,  up
@section Changing Saved Local Variables

A lot of the functionality of MMM Mode---that which makes the major mode
appear to change---is implemented by saving and restoring the values of
local variables, or pseudo-variables.  You can customize what variables
are saved, and how, with the following variable.

@defvar mmm-save-local-variables
At its simplest, this is a list each of whose elements is a buffer-local
variable whose value is saved and restored for each major mode.  Each
elements can also, however, be a list whose first element is the
variable symbol and whose subsequent elements specify how and where the
variable is to be saved.  The second element of the list, if present,
should be one of the symbols @code{global}, @code{buffer}, or
@code{region}.  If not present, the default value is @code{global}.  The
third element, if present, should be a list of major mode symbols in
which to save the variable.  In the list form, the variable symbol
itself can be replaced with a cons cell of two functions, one to get the
value and one to set the value.  This is called a ``pseudo-variable''.
@end defvar

Globally saved variables are the same in all (MMM-controlled) buffers
and submode regions of each major mode listed in the third argument, or
all major modes if it is @code{t} or not present.  Buffer-saved
variables are the same in all submode regions of a given major mode in
each buffer, and region-saved variables can be different for each
submode region.

Pseudo-variables are used, for example, to save and restore the syntax
table (@pxref{Syntax, , , emacs, The Emacs Manual}) and mode keymaps
(@pxref{Keymaps, , , emacs, The Emacs Manual}).


@node Changing Classes, Hooks, Local Variables, Customizing
@comment  node-name,  next,  previous,  up
@section Changing the Supplied Submode Classes

If you need to use MMM with a syntax for which a submode class is not
supplied, and you have some facility with Emacs Lisp, you can write your
own; see @ref{Writing Classes}.  However, sometimes you will only want
to make a slight change to one of the supplied submode classes.  You can
do this, after that class is loaded, with the following functions.

@defun mmm-set-class-parameter @var{class} @var{param} @var{value}
Set the value of the keyword parameter @var{param} of the submode class
@var{class} to @var{value}.  @xref{Writing Classes}, for an explanation
of the meaning of each keyword parameter.  This creates a new parameter
if one is not already present in the class.
@end defun

@defun mmm-get-class-parameter @var{class} @var{param}
Get the value of the keyword parameter @var{param} for the submode class
@var{class}.  Returns @code{nil} if there is no such parameter.
@end defun



@node Hooks,  , Changing Classes, Customizing
@comment  node-name,  next,  previous,  up
@section Hooks Provided by MMM Mode

MMM Mode defines several hook variables (@pxref{Hooks, , , emacs, The
Emacs Manual}) which are run at different times.  The most often used is
@code{mmm-major-mode-hook} which is described in @ref{Major Mode Hook},
but there are a couple others.

@defvar mmm-mode-hook
This normal hook is run whenever MMM Mode is enabled in a buffer.
@end defvar

@defvar mmm-@var{major-mode}-hook
This is actually a whole set of hook variables, a different one for
every major mode.  Whenever MMM Mode is enabled in a buffer, the
corresponding hook variable for the dominant major mode is run.
@end defvar

@defvar mmm-@var{submode}-submode-hook
Again, this is a set of one hook variable per major mode.  These hooks
are run whenever a submode region of the corresponding major mode is
created in any buffer, with point at the start of the new submode
region.
@end defvar

@defvar mmm-@var{class}-class-hook
This is a set of one hook variable per submode class.  These hooks are
run when a submode class is first applied to a given buffer.
@end defvar

Submode classes also have a @code{:creation-hook} parameter which should
be a function to run whenever a submode region is created with that
class, with point at the beginning of the submode region.  This can be
set for supplied submode classes with @code{mmm-set-class-parameter};
@ref{Changing Classes}.


@node Supplied Classes, Writing Classes, Customizing, Top
@comment  node-name,  next,  previous,  up
@chapter Supplied Submode Classes

This chapter describes the submode classes that are supplied with MMM
Mode.

@menu
* Mason::                       Mason server-side Perl in HTML.
* File Variables::              Elisp code in File Variables.
* Here-documents::              Code in shell and Perl here-documents.
* Javascript::                  Javascript embedded in HTML.
* Embedded CSS::                CSS Styles embedded in HTML.
* Embperl::                     Another syntax for Perl in HTML.
* ePerl::                       A general Perl-embedding syntax.
* JSP::                         Java code embedded in HTML.
* RPM::                         Shell scripts in RPM Spec Files.
* Noweb::                       Noweb literate programs.
@end menu

@node Mason, File Variables, Supplied Classes, Supplied Classes
@comment  node-name,  next,  previous,  up
@section Mason: Perl in HTML

Mason is a syntax to embed Perl code in HTML and other documents.  See
@uref{http://www.masonhq.com} for more information.  The submode class
for Mason components is called `mason' and is loaded on demand from
`mmm-mason.el'.  The current Mason class is intended to correctly
recognize all syntax valid in Mason @value{MASON_VERSION}.  There are
insertion keys for most of the available syntax; use
@code{mmm-insertion-help} (@kbd{C-c % h} by default) with Mason on to
get a list.

If you want to have mason submodes automatically in all Mason files, you
can use automatic mode and filename associations; the details depend on
what you call your Mason components and what major mode you use.
@xref{Mode-Ext Classes}.  If you use an extension for your Mason files
that emacs does not automatically place in your preferred HTML Mode, you
will probably want to associate that extension with your HTML Mode as
well; @ref{Choosing Modes, , , emacs, The Emacs Manual}.  This also goes
for ``special'' Mason files such as autohandlers and dhandlers.

The Perl mode used is controlled by the user: @xref{Preferred Modes}.
The default is to use CPerl mode, if present.  Unfortunately, there are
also certain problems with CPerl mode in submode regions.  (Not to say
that the original perl-mode would do any better---it hasn't been much
tried.)  First of all, the first line of a Perl section is usually
indented as if it were a continuation line.  A fix for this is to start
with a semicolon on the first line.  The insertion key commands do this
whenever the Mason syntax allows it.

@example
<%perl>;
print $var;
</%perl>
@end example

In addition, some users have reported that the CPerl indentation
sometimes does not work. This problem has not yet been tracked down,
however, and more data about when it happens would be helpful.

Some people have reported problems using PSGML with Mason.  Adding the
following line to a @file{.emacs} file should suffice to turn PSGML off
and cause emacs to use a simpler HTML mode:

@lisp
(autoload 'html-mode "sgml-mode" "HTML Mode" t)
@end lisp

Earlier versions of PSGML may require instead the following fix:

@lisp
(delete '("\\.html$" . sgml-html-mode) auto-mode-alist)
(delete '("\\.shtml$" . sgml-html-mode) auto-mode-alist)
@end lisp

Other users report using PSGML with Mason and MMM Mode without
difficulty.  If you don't have problems and want to use PSGML, you may
need to replace @code{html-mode} in the suggested code with
@code{sgml-html-mode}.  (Depending on your version of PSGML, this may
not be necessary.)  Similarly, if you are using XEmacs and want to use
the alternate HTML mode @code{hm--html-mode}, replace @code{html-mode}
with that symbol.

One problem that crops up when using PSGML with Mason is that even
ignoring the special tags and Perl code (which, as I've said, haven't
caused me any problems), Mason components often are not a complete SGML
document.  For instance, my autohandlers often say

@example
<body>
  <% $m->call_next %>
</body>
@end example

in which case the actual components contain no doctype declaration,
@code{<html>}, @code{<head>}, or @code{<body>}, confusing PSGML.  One
solution I've found is to use the variable @code{sgml-parent-document}
in such incomplete components; try, for example, these lines at the end
of a component.

@example
%# Local Variables:
%# sgml-parent-document: ("autohandler" "body" nil ("body"))
%# sgml-doctype: "/top/level/autohandler"
%# End:
@end example

This tells PSGML that the current file is a sub-document of the file
@file{autohandler} and is included inside a @code{<body>} tag, thus
alleviating its confusion.


@node File Variables, Here-documents, Mason, Supplied Classes
@comment  node-name,  next,  previous,  up
@section Elisp in a Local Variables List

Emacs allows the author of a file to specify major and minor modes to be
used while editing that file, as well as specifying values for other
local Elisp variables, with a File Variables list.  @xref{File
Variables, , , emacs, The Emacs Manual}.  Since file variables values
are Elisp objects (and with the @code{eval} special ``variable'', they
are forms to be evaluated), one might want to edit them in
@code{emacs-lisp-mode}.  The submode class @code{file-variables} allows
this, and is suitable for turning on in a given file with
@code{mmm-classes}, or in all files with @code{mmm-global-classes}.


@node Here-documents, Javascript, File Variables, Supplied Classes
@comment  node-name,  next,  previous,  up
@section Here-documents

One of the long-time standard syntaxes for outputting large amounts of
code (or text, or HTML, or whatever) from a script (notably shell
scripts and Perl scripts) is the here-document syntax:

@example
print <<END_HTML;
<html>
  <head>
    <title>Test Page</title>
  </head>
  <body>
END_HTML
@end example

The @code{here-doc} submode class recognizes this syntax, and can even
guess the correct submode to use in many cases.  For instance, it would
put the above example in @code{html-mode}, noticing the string
@samp{HTML} in the name of the here-document.  If you use less than
evocative here-document names, or if the submode is recognized
incorrectly for any other reason, you can tell it explicitly what
submode to use.

@defopt mmm-here-doc-mode-alist
The value of this variable should be an alist, each element a cons pair
associating a regular expression to a submode symbol.  Whenever a
here-document name matches one of these regexps, the corresponding
submode is applied.  For example, if this variable contains the element
@code{("CODE" . cc-mode)}, then any here-document whose name contains
the string @samp{CODE} will be put in @code{cc-mode}.  The value of this
variable overrides any guessing that the @code{here-doc} submode class
would do otherwise.
@end defopt


@node Javascript, Embedded CSS, Here-documents, Supplied Classes
@comment  node-name,  next,  previous,  up
@section Javascript in HTML

The submode class @code{html-js} allows for embedding Javascript code in
HTML documents.  It recognizes both this syntax:

@example
<script language="Javascript">
function foo(...) @{
   ...
@}
</script>
@end example

and this syntax:

@example
<input type="button" onClick="validate();">
@end example

The mode used for Javascript regions is controlled by the user;
@xref{Preferred Modes}.


@node Embedded CSS, Embperl, Javascript, Supplied Classes
@comment  node-name,  next,  previous,  up
@section CSS embedded in HTML

CSS (Cascading Style Sheets) can also be embedded in HTML.  The
@code{embedded-css} submode class recognizes this syntax:

@example
<style>
h1 @{
   ...
@}
</style>
@end example

It uses @code{css-mode} if present, @code{c++-mode} otherwise.  This can
be customized: @xref{Preferred Modes}.


@node Embperl, ePerl, Embedded CSS, Supplied Classes
@comment  node-name,  next,  previous,  up
@section Embperl: More Perl in HTML

Embperl is another syntax for embedding Perl in HTML.  See
@uref{http://perl.apache.org/embperl} for more information.  The
@code{embperl} submode class recognizes most if not all of the Embperl
embedding syntax.  Its Perl mode is also controllable by the user;
@xref{Preferred Modes}.


@node ePerl, JSP, Embperl, Supplied Classes
@comment  node-name,  next,  previous,  up
@section ePerl: General Perl Embedding

Yet another syntax for embedding Perl is called ePerl.  See
@uref{http://www.engelschall.com/sw/eperl/} for more information.  The
@code{eperl} submode class handles this syntax, using the Perl mode
specified by the user; @xref{Preferred Modes}.


@node JSP, RPM, ePerl, Supplied Classes
@comment  node-name,  next,  previous,  up
@section JSP: Java Embedded in HTML

JSP (Java Server Pages) is a syntax for embedding Java code in HTML.
The submode class @code{jsp} handles this syntax, using a Java mode
specified by the user; @xref{Preferred Modes}.  The default is
@code{jde-mode} if present, otherwise @code{java-mode}.


@node RPM, Noweb, JSP, Supplied Classes
@comment  node-name,  next,  previous,  up
@section RPM Spec Files

@file{mmm-rpm.el} contains the definition of an MMM Mode submode class
for editing shell script sections within RPM (Redhat Package Manager)
spec files.  It is recommended for use in combination with
@file{rpm-spec-mode.el} by Stig Bjørlykke <stigb@@tihlde.hist.no> and
Steve Sanbeg <sanbeg@@dset.com>
(@uref{http://www.xemacs.org/~stigb/rpm-spec-mode.el}).

Suggested setup code:

@lisp
(add-to-list 'mmm-mode-ext-classes-alist
             '(rpm-spec-mode "\\.spec\\'" rpm-sh))
@end lisp

Thanks to Marcus Harnisch <Marcus.Harnisch@@gmx.net> for contributing
this submode class.

@node Noweb,  , RPM, Supplied Classes
@comment  node-name,  next,  previous,  up
@section Noweb literate programming

@file{mmm-noweb.el} contains the definition of an MMM Mode submode
class for editing Noweb documents.  Most Noweb documents use \LaTeX
for the documentation chunks.  Code chunks in Noweb are
document-specific, and the mode may be set with a local variable
setting in the document.  The variable @var{mmm-noweb-code-mode}
controls the global code chunk mode. Since Noweb files may have many
languages in their code chunks, this mode also allows setting the mode
by specifying a mode in the first line or two of a code chunk, using
the normal Emacs first-line mode setting syntax.  Note that this
first-line mode setting only matches a single word for the mode name,
and does not support the variable name setting of the generalized
first file line syntax.

@verbatim
% -*- mode: latex; mmm-noweb-code-mode: c++; -*-
% First chunk delimiter!
@
\noweboptions{smallcode}

\title{Sample Noweb File}
\author{Joe Kelsey\\
\nwanchorto{mailto:bozo@bozo.bozo}{\tt bozo@bozo.bozo}}
\maketitle

@
\section{Introduction}
Normal noweb documentation for the required [[*]] chunk.
<<*>>=
// C++ mode here!
// We might list the program here, or simply included chunks.
<<myfile.cc>>
@ %def myfile.cc

@
\section{[[myfile.cc]]}
This is [[myfile.cc]].  MMM noweb-mode understands code quotes in
documentation.
<<myfile.cc>>=
// This section is indented separately from previous.
@ 

@
\section{A Perl Chunk}
We need a Perl chunk.
<<myfile.pl>>=
#!/usr/bin/perl
# -*- perl -*-
# Each differently named chunk is flowed separately.
@ 

\section{Finish [[myfile.cc]]}
When we resume a previously defined chunk, they are indented together.
<<myfile.cc>>=
// Pick up where we left off...
@ 

@end verbatim

The quoted code chunks inside documentation chunks are given the mode
found in the variable @var{mmm-noweb-quote-mode}, if set, or the value
in @var{mmm-noweb-code-mode} otherwise.  Also, each quoted chunk is
set to have a unique name to prevent them from being indented as a
unit.

Suggested setup code:
@lisp
(mmm-add-mode-ext-class 'latex-mode "\\.nw\\'" 'noweb)
(add-to-list 'auto-mode-alist '("\\.nw\\'" . latex-mode))
@end lisp

In mmm-noweb buffers, each differently-named code chunk has a
different @code{:name}, allowing all chunks with the same name to get
indented together.

This mode also supplies special paragraph filling operations for use
in documentation areas of the buffer.  From a primary-mode
(@code{latex-mode, , emacs}) region, pressing @kbd{C-c % C-q} will mark all
submode regions with word syntax (@code{mmm-word-other-regions}), fill
the current paragraph (@code{(fill-paragraph justify)}), and remove the
syntax markings (@code{mmm-undo-syntax-other-regions}).

Thanks to Joe Kelsey <joe@@zircon.seattle.wa.us> for contributing this
class.


@node Writing Classes, Indices, Supplied Classes, Top
@comment  node-name,  next,  previous,  up
@chapter Writing Submode Classes

Sometimes (perhaps often) you may want to use MMM with a syntax for
which it is suited, but for which no submode is supplied.  In such cases
you may have to write your own submode class.  This chapter briefly
describes how to write a submode class, from the basic to the advanced,
with examples.

@menu
* Basic Classes::               Writing a simple submode class.
* Paired Delimiters::           Matching paired delimiters.
* Region Placement::            Placing the region more accurately.
* Submode Groups::              Grouping several classes together.
* Calculated Submodes::         Deciding the submode at run-time.
* Calculated Faces::            Deciding the display face at run-time.
* Insertion Commands::          Inserting regions automatically.
* Region Names::                Naming regions for syntax grouping.
* Other Hooks::                 Running code at arbitrary points.
* Delimiters::                  Controlling delimiter overlays.
* Misc Keywords::               Other miscellaneous options.
@end menu

@node Basic Classes, Paired Delimiters, Writing Classes, Writing Classes
@comment  node-name,  next,  previous,  up
@section Writing Basic Submode Classes
@cindex simple submode classes
@cindex submode classes, simple

Writing a submode class can become rather complex, if the syntax to
match is complicated and you want to take advantage of some of MMM
Mode's extra features.  But a simple submode class is not particularly
difficult to write.  This section describes the basics of writing
submode classes.

Submode classes are stored in the variable @code{mmm-classes-alist}.
Each element of this list represents a single submode class.  For
convenience, the function @code{mmm-add-classes} takes a list of submode
classes and adds them all to this alist.  Each class is represented by a
list containing the class name---a symbol such as @code{mason} or
@code{html-js}---followed by pairs of keywords and arguments called a
@dfn{class specifier}.  For example, consider the specifier for the
submode class @code{embedded-css}:

@lisp
(mmm-add-classes
 '((embedded-css
    :submode css
    :face mmm-declaration-submode-face
    :front "<style[^>]*>"
    :back "</style>")))
@end lisp

The name of the submode is @code{embedded-css}, the first element of the
list.  The rest of the list consists of pairs of keywords (symbols
beginning with a colon) such as @code{:submode} and @code{:front}, and
arguments, such as @code{css} and @code{"<style[^>]*>"}.  It is the
keywords and arguments that specify how the submode works.  The order of
keywords is not important; all that matters is the arguments that follow
them.

The three most important keywords are @code{:submode}, @code{:front},
and @code{:back}.  The argument following @code{:submode} names the
major mode to use in submode regions.  It can be either a symbol naming
a major mode, such as @code{text-mode} or @code{c++-mode}, or a symbol
to look up in @code{mmm-major-mode-preferences} (@pxref{Preferred
Modes}) such as @code{css}, as in this case.

The arguments following @code{:front} and @code{:back} are regular
expressions (@pxref{Regexps, , , emacs, The Emacs Manual}) that should
match the delimiter strings which begin and end the submode regions.  In
our example, CSS regions begin with a @samp{<style>} tag, possibly with
parameters, and end with a @samp{</style>} tag.

The argument following @code{:face} specifies the face (background
color) to use when @code{mmm-submode-decoration-level} is 2 (high
coloring).  @xref{Region Coloring}, for a list of canonical available
faces.

There are many more possible keywords arguments.  In the following
sections, we will examine each of them and their uses in writing submode
classes.


@node Paired Delimiters, Region Placement, Basic Classes, Writing Classes
@comment  node-name,  next,  previous,  up
@section Matching Paired Delimiters

A simple pair of regular expressions does not always suffice to exactly
specify the beginning and end of submode regions correctly.  For this
reason, there are several other possible keyword/argument pairs which
influence the matching process.

Many submode regions are marked by paired delimiters.  For example, the
tags used by Mason (@pxref{Mason}) include @samp{<%init>...</%init>} and
@samp{<%args>...</%args>}.  It would be possible to write a separate
submode class for each type of region, but there is an easier way: the
keyword argument @code{:save-matches}.  If supplied and non-nil, it
causes the regular expression @code{:back}, before being searched for,
to be formatted by replacing all strings of the form @samp{~@var{N}}
(where @var{N} is an integer) with the corresponding numbered
subexpression of the match for @code{:front}.  As an example, here is an
excerpt from the @code{here-doc} submode class.  @xref{Here-documents},
for more information about this submode.

@lisp
:front "<<\\([a-zA-Z0-9_-]+\\)"
:back "^~1$"
:save-matches 1
@end lisp

The regular expression for @code{:front} matches @samp{<<} followed by a
string of one or more alphanumeric characters, underscores, and dashes.
The latter string, which happens to be the name of the here-document, is
saved as the first subexpression, since it is surrounded by
@samp{\(...\)}.  Then, because the value of @code{:save-matches} is
present and non-nil, the string @samp{~1} is replaced in the value of
@code{:back} by the name of the here-document, thus creating a regular
expression to match the correct ending delimiter.


@node Region Placement, Submode Groups, Paired Delimiters, Writing Classes
@comment  node-name,  next,  previous,  up
@section Placing Submode Regions Precisely

Normally, a submode region begins immediately after the end of the
string matching the @code{:front} regular expression and ends
immediately before the beginning of the string matching the @code{:back}
regular expression.  This can be changed with the keywords
@code{:include-front} and @code{:include-back}.  If their arguments are
@code{nil}, or they do not appear, the default behavior is unchanged.
But if the argument of @code{:include-front} (respectively,
@code{:include-back}) is non-nil, the submode region will begin
(respectively, end) immediately before (respectively, after) the string
matching the @code{:front} (respectively, @code{:back}) regular
expression.  In other words, these keywords specify whether or not the
delimiter strings are @emph{included} in the submode region.

When @code{:front} and @code{:back} are regexps, the delimiter is
normally considered to be the entire matched region.  This can be
changed using the @code{:front-match} and @code{:back-match}
keywords.  The values of the keywords is a number specifying the
submatch.  This defaults to zero (specifying the whole regexp).

Two more keywords which affect the placement of the region
@code{:front-offset} and @code{:back-offset}, which both take integers
as arguments.  The argument of @code{:front-offset} (respectively,
@code{:back-offset}) gives the distance in characters from the beginning
(respectively, ending) location specified so far, to the actual point
where the submode region begins (respectively, ends).  For example, if
@code{:include-front} is nil or unsupplied and @code{:front-offset} is
2, the submode region will begin two characters after the end of the
match for @code{:front}, and if @code{:include-back} is non-nil and
@code{:back-offset} is -1, the region will end one character before the
end of the match for @code{:back}.

In addition to integers, the arguments of @code{:front-offset} and
@code{:back-offset} can be functions which are invoked to move the point
from the position specified by the matches and inclusions to the correct
beginning or end of the submode region, or lists whose elements are
either functions or numbers and whose effects are applied in sequence.
To help disentangle these options, here is another excerpt from the
@code{here-doc} submode class:

@lisp
:front "<<\\([a-zA-Z0-9_-]+\\)"
:front-offset (end-of-line 1)
:back "^~1$"
:save-matches 1
@end lisp

Here the value of @code{:front-offset} is the list @code{(end-of-line
1)}, meaning that from the end of the match for @code{:front}, go to the
end of the line, and then one more character forward (thus to the
beginning of the next line), and begin the submode region there.  This
coincides with the normal behavior of here-documents: they begin on the
following line and go until the ending flag.

If the @code{:back} should not be able to start a new submode region,
set the @code{:end-not-begin} keyword to non-nil.

@node Submode Groups, Calculated Submodes, Region Placement, Writing Classes
@comment  node-name,  next,  previous,  up
@section Defining Groups of Submodes

Sometimes more than one submode class is required to accurately reflect
the behavior of a single type of syntax.  For example, Mason has three
very different types of Perl regions: blocks bounded by matched tags
such as @samp{<%perl>...</%perl>}, inline output expressions bounded by
@samp{<%...%>}, and single lines of code which simply begin with a
@samp{%} character.  In cases like these, it is possible to specify an
``umbrella'' class, to turn all these classes on or off together.

@defun mmm-add-group @var{group} @var{classes}
The submode classes @var{classes}, which should be a list of lists,
similar to what might be passed to @code{mmm-add-classes}, are added
just as by that function.  Furthermore, another class named
@var{group} is added, which encompasses all the classes in
@var{classes}.
@end defun

Technically, an group class is specified with a @code{:classes} keyword
argument, and the subsidiary classes are given a non-nil @code{:private}
keyword argument to make them invisible.  But in general, all you should
ever need to know is how to invoke the function above.

@defun mmm-add-to-group @var{group} @var{classes}
Adds a list of classes to an already existing group.  This can be
used, for instance, to add a new quoting definition to @var{html-js}
using this example to add the quote characters ``%=%'':

@lisp
(mmm-add-to-group 'html-js '((js-html
                             :submode javascript
                             :face mmm-code-submode-face
                             :front "%=%"
                             :back "%=%"
                             :end-not-begin t)))
@end lisp
@end defun


@node Calculated Submodes, Calculated Faces, Submode Groups, Writing Classes
@comment  node-name,  next,  previous,  up
@section Calculating the Correct Submode

In most cases, the author of a submode class will know in advance what
major mode to use, such as @code{text-mode} or @code{c++-mode}.  If
there are multiple possible modes that the user might desire, then
@code{mmm-major-mode-preferences} should be used (@pxref{Preferred
Modes}).  The function @code{mmm-set-major-mode-preferences} can be
used, with a third argument, to ensure than the mode is present.

In some cases, however, the author has no way of knowing in advance even
what language the submode region will be in.  The @code{here-doc} class
is one of these.  In such cases, instead of the @code{:submode} keyword,
the @code{:match-submode} keyword must be used.  Its argument should be
a function, probably written by the author of the submode class, which 
calculates what major mode each region should use.

It is invoked immediately after a match is found for @code{:front}, and
is passed one argument: a string representing the front delimiter.
Normally this string is simply whatever was matched by @code{:front},
but this can be changed with the keyword @code{:front-form}
(@pxref{Delimiters}).  The function should then return a symbol
that would be a valid argument to @code{:submode}: either the name of a
mode, or that of a language to look up a preferred mode.  If it detects
an invalid match---for example, the user has specified a mode which is
not available---it should @code{(signal 'mmm-no-matching-submode nil)}.

Since here-documents can contain code in any language, the
@code{here-doc} submode class uses @code{:match-submode} rather than
@code{:submode}.  The function it uses is @code{mmm-here-doc-get-mode},
defined in @file{mmm-sample.el}, which inspects the name of the
here-document for flags indicating the proper mode.  For example, this
code should probably be in @code{perl-mode} (or @code{cperl-mode}):

@example
print <<PERL;
s/foo/bar/g;
PERL
@end example

This function is also a good example of proper elisp hygiene: when
writing accessory functions for a submode class, they should usually be
prefixed with @samp{mmm-} followed by the name of the submode class, to
avoid namespace conflicts.


@node Calculated Faces, Insertion Commands, Calculated Submodes, Writing Classes
@comment  node-name,  next,  previous,  up
@section Calculating the Correct Highlight Face

As explained in @ref{Basic Classes}, the keyword @code{:face} should be
used to specify which of the standard submode faces (@pxref{Region
Coloring}) a submode region should be highlighted with under high
decoration.  However, sometimes the function of a region can depend on
the form of the delimiters as well.  In this case, a more flexible
alternative to @code{:face} is @code{:match-face}.  Its value can be a
function, which is called with one argument---the form of the front
delimiter, as with @code{:match-submode}---and should return the face to
use.  A more common value for @code{:match-face} is an association list,
a list of pairs @code{(@var{delim} . @var{face})}, each specifying that
if the delimiter is @var{delim}, the corresponding region should be
highlighted with @var{face}.  For example, here is an excerpt from the
@code{embperl} submode class:

@lisp
:submode perl
:front "\\[\\([-\\+!\\*\\$]\\)"
:back "~1\\]"
:save-matches 1
:match-face (("[+" . mmm-output-submode-face)
             ("[-" . mmm-code-submode-face)
             ("[!" . mmm-init-submode-face)
             ("[*" . mmm-code-submode-face)
             ("[$" . mmm-special-submode-face))
@end lisp

Thus, regions beginning with @samp{[+} are highlighted as output
expressions, which they are, while @samp{[-} and @samp{[*} regions are
highlighted as simple executed code, and so on.  Note that
@var{mmm-submode-decoration-level} must be set to 2 (high decoration)
for different faces to be displayed.


@node Insertion Commands, Region Names, Calculated Faces, Writing Classes
@comment  node-name,  next,  previous,  up
@section Specifying Insertion Commands

As described in @ref{Insertion}, submode classes can specify key
sequences which automatically insert submode regions, with delimiters
already in place.  This is done by the keyword argument @code{:insert}.
Its value should be a list, each element of which specifies a single
insertion key sequence.  As an example, consider the following insertion
key sequence specifier, from the @code{embperl} submode class:

@lisp
(?p embperl "Region Type (Character): "
    @@ "[" str @@ " " _ " " @@ str "]" @@)
@end lisp

As you can see, the specifier is a list.  The first element of the list
is the character @samp{p}.  (The question mark tells Emacs that this is
a character object, not a one-character symbol.)  In general, the first
element can be any key, including both characters such as @samp{?p} and
function keys such as @samp{return}.  It can also be a dotted pair in
which the first element is a modifier symbol such as @code{meta}, and
the second is a character or function key.  The use of any other
modifier than meta is discouraged, as `mmm-insert-modifiers' is
sometimes set to \(control), and other modifiers are not very portable.
The second element is a symbol identifying this key sequence.  The third
element is a prompt string which is used to ask the user for input when
this key sequence is invoked.  If it is nil, the user is not prompted.

The rest of the list specifies the actual text to be inserted, where the
submode region and delimiters should be, and where the point should end
up.  (Actually, this string is simply passed to @code{skeleton-insert};
see the documentation string of that function for more details on the
permissible elements of such a skeleton.)  Strings and variable names
are inserted and interpolated.  The value entered by the user when
prompted, if any, is available in the variable @code{str}.  The final
location of the point (or the text around which the region is to be
wrapped) is marked with a single underscore @samp{_}.  Finally, the
@@-signs mark the delimiters and submode regions.  There should be four
@@-signs: one at the beginning of the front delimiter, one at the
beginning of the submode region, one at the end of the submode region,
and one at the end of the back delimiter.

The above key sequence, bound by default to @kbd{C-c % p}, always
prompts the user for the type of region to insert.  It can also be
convenient to have separate key sequences for each type of region to be
inserted, such as @kbd{C-c % +} for @samp{[+...+]} regions, @kbd{C-c %
-} for @samp{[-...-]} regions, and so on.  So that the whole skeleton
doesn't have to be written out half a dozen times, there is a shortcut
syntax, as follows:

@lisp
(?+ embperl+ ?p . "+")             
@end lisp

If the key sequence specification is a dotted list with four elements,
as this example is, it means to use the skeleton defined for the key
sequence given as the third element (@code{?p}), but to pass it the
fourth (dotted) element (@code{"+"}) as the `str' variable; the user is
not prompted.


@node Region Names, Other Hooks, Insertion Commands, Writing Classes
@comment  node-name,  next,  previous,  up
@section Giving Names to Submode Regions for Grouping

Submode regions can be given ``names'' which are used for grouping.
Names are always strings and are compared as strings.  Regions with
the same name are considered part of the same chunk of code.  This is
used by the syntax and fontification functions.  Unnamed regions are
not grouped with any others.

By default, regions are nameless, but with the @code{:match-name}
keyword argument a name can be supplied.  This argument must be a
string or a function.  If it is a function, it is passed a string
representing the front delimiter found, and must return the name to
use.  If it is a string, it is used as-is for the name, unless
@code{:save-name} has a non-nil value, in which case expressions such
as @samp{~1} are substituted with the corresponding matched
subexpression from @code{:front}.  This is the same as how
@code{:back} is interpreted when @code{:save-matches} is non-nil.

As a special optimization for region insertion (@pxref{Insertion
Commands}), the argument @code{:skel-name} can be set to a non-nil
value, in which case the insertion code will use the user-prompted
string value as the region name, instead of going through the normal
matching procedure.


@node Other Hooks, Delimiters, Region Names, Writing Classes
@comment  node-name,  next,  previous,  up
@section Other Hooks into the Scanning Process

Sometimes, even the flexibility allowed by all the keyword arguments
discussed so far is insufficient to correctly match submode regions.
There are several other keyword arguments which accept custom functions
to be invoked at various points in the MMM-ification process.

First of all, the arguments of @code{:front} and @code{:back}, in
addition to regular expressions, can be themselves functions.  Such
functions should ``act like'' a regular expression search: they should
start searching at point, take one argument as a limit for the search,
and return its result by setting the match data (presumably by calling
some regexp matching function).

This is rarely necessary, however, because often all that is needed is a
simple regexp search, followed by some sort of verification.  The
keyword arguments @code{:front-verify} and @code{:back-verify}, if
supplied, may be functions which are invoked after a match is found for
@code{:front} or @code{:back}, respectively, and should inspect the
match data (such as with @code{match-string}) and return non-nil if a
submode region should be begun at this match, nil if this match should
be ignored and the search continue after it.

The keyword argument @code{:creation-hook}, if supplied, should be a
function that is invoked whenever a submode region of this class is
created, with point at the beginning of the new region.  This can be
used, for example, to set local variables appropriately.

Finally, the entire MMM-ification process has a ``back door'' which
allows class authors to take control of the entire thing.  If the
keyword argument @code{:handler} is supplied, it overrides any other
processing and is called, and passed all other class keyword arguments,
instead of @code{mmm-ify} to create submode regions.  If you need to
write a handler function, I suggest looking at the source for
@code{mmm-ify} to get an idea of what must be done.


@node Delimiters, Misc Keywords, Other Hooks, Writing Classes
@comment  node-name,  next,  previous,  up
@section Controlling the Delimiter Regions and Forms

MMM also makes overlays for the delimiter regions, to keep track of
their position and form.  Normally, the front delimiter overlay starts
at the beginning of the match for @code{:front} and ends at the
beginning of the submode region overlay, while the back delimiter
overlay starts at the end of the submode region overlay and ends at
the end of the match for @code{:back}.  You can supply offsets from
these positions using the keyword arguments @code{:front-delim} and
@code{:back-delim}, which take values of the same sort as
@code{:front-offset} and @code{:back-offset}.

In addition, the delimiter regions can be in a major mode of their
own.  There are usually only two meaningful modes to use: the primary
mode or a non-mode like fundamental-mode.  These correspond to the
following two situations:

@itemize
@item
If the delimiter syntax which specifies the submode regions is
something @emph{added to} the syntax of the primary mode by a
pre-interpreter, then the delimiter regions should be in a non-mode.
This is the case, for example, with all server-side HTML script
extensions, such as @xref{Mason}, @xref{Embperl}, and @xref{ePerl}.
It is also the case for literate programming such as @xref{Noweb}.
This is the default behavior.  The non-mode used is controlled by the
variable @code{mmm-delimiter-mode}, which defaults to
fundamental-mode.

@item
If, on the other hand, the delimiter syntax and inclusion of different
modes is an @emph{intrinsic part} of the primary mode, then the
delimiter regions should remain in the primary mode.  This is the
case, for example, with @xref{Embedded CSS}, and @xref{Javascript},
since the @code{<style>} and @code{<script>} tags are perfectly valid
HTML.  In this case, you should give the keyword parameter
@code{:delimiter-mode} with a value of @code{nil}, meaning to use the
primary mode.
@end itemize

The keyword parameter @code{:delimiter-mode} can be given any major
mode as an argument, but the above two situations should cover the
vast majority of cases.

The delimiter regions can also be highlighted, if you wish.  The
keyword parameters @code{:front-face} and @code{:back-face} may be
faces specifying how to highlight these regions under high
decoration.  Under low decoration, the value of the variable
@code{mmm-delimiter-face} is used (by default, nothing), and of course
under no decoration there is no coloring.

Finally, for each submode region overlay, MMM Mode stores the ``form''
of the front and back delimiters, which are regular expressions that
match the delimiters.  At present these are not used for much, but in
the future they may be used to help with automatic updating of regions
as you type.  Normally, the form stored is the result of evaluating
the expression @code{(regexp-quote (match-string 0))} after each match
is found.

You can customize this with the keyword argument @code{:front-form}
(respectively, @code{:back-form}).  If it is a string, it is used
verbatim for the front (respectively, back) form.  If it is a function,
that function is called and should inspect the match data and return the
regular expression to use as the form.

In addition, the form itself can be set to a function, by giving a
one-element list containing only that function as the argument to
@code{:front-form} or @code{:back-form}.  Such a function should take
1-2 arguments.  The first argument is the overlay to match the delimiter
for.  If the second is non-nil, it means to insert the delimiter and
adjust the overlay; if nil it means to match the delimiter and return
the result in the match data.


@node Misc Keywords,  , Delimiters, Writing Classes
@comment  node-name,  next,  previous,  up
@section Miscellaneous Other Keyword Arguments

You can specify whether delimiter searches should be case-sensitive with
the keyword argument @code{:case-fold-search}.  It defaults to @code{t},
meaning that case should be ignored.  See the documentation for the
variable @code{case-fold-search}.

@node Indices,  , Writing Classes, Top
@comment  node-name,  next,  previous,  up
@chapter Indices

@menu
* Concept Index::               Index of MMM Mode Concepts.
* Function Index::              Index of functions and variables.
* Keystroke Index::             Index of key bindings in MMM Mode.
@end menu

@node Concept Index, Function Index, Indices, Indices
@comment  node-name,  next,  previous,  up
@section Concept Index

@printindex cp


@node Function Index, Keystroke Index, Concept Index, Indices
@comment  node-name,  next,  previous,  up
@section Function and Variable Index

@printindex fn


@node Keystroke Index,  , Function Index, Indices
@comment  node-name,  next,  previous,  up
@section Keystroke Index

@printindex ky


@bye

@c Local Variables:
@c mode: texinfo
@c mode: font-lock
@c mode: outline-minor
@c End: