Move doc of backup-directory-alist to the Backup node

[emacs.git] / doc / misc / pcl-cvs.texi

\input texinfo  @c -*-texinfo-*-
@c %**start of header
@setfilename ../../info/pcl-cvs.info
@settitle PCL-CVS---Emacs Front-End to CVS
@include docstyle.texi
@syncodeindex vr fn
@c %**end of header

@copying
Copyright @copyright{} 1991--2016 Free Software Foundation, Inc.

@quotation
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.3 or
any later version published by the Free Software Foundation; with no
Invariant Sections, with the Front-Cover Texts being ``A GNU Manual'',
and with the Back-Cover Texts as in (a) below.  A copy of the license
is included in the section entitled ``GNU Free Documentation License''.

(a) The FSF's Back-Cover Text is: ``You have the freedom to copy and
modify this GNU manual.''
@end quotation
@end copying

@dircategory Emacs misc features
@direntry
* PCL-CVS: (pcl-cvs).           Emacs front-end to CVS.
@end direntry

@c The titlepage section does not appear in the Info file.
@titlepage
@sp 4
@c The title is printed in a large font.
@center @titlefont{User's Guide}
@sp 1
@center @titlefont{to}
@sp 1
@center @titlefont{PCL-CVS---The Emacs Front-End to CVS}
@ignore
@sp 2
@center release 2.9
@c -release-
@end ignore
@sp 3
@center Per Cederqvist
@center Stefan Monnier
@c -date-

@c  The following two commands start the copyright page
@c  for the printed manual.  This will not appear in the Info file.
@page
@vskip 0pt plus 1filll
@insertcopying
@end titlepage

@contents

@c ================================================================
@c                   The real text starts here
@c ================================================================

@ifnottex
@node Top
@top PCL-CVS

This manual describes PCL-CVS, the GNU Emacs front-end to CVS@.  It
is nowhere near complete, so you are advised to use @kbd{M-x
customize-group RET pcl-cvs @key{RET}} and to look at the documentation strings
of the various commands and major modes for further information.
@c This manual is updated to release 2.5 of PCL-CVS.

@insertcopying

@end ifnottex

@menu
* About PCL-CVS::               Credits, history, @dots{}

* Getting started::             An introduction with a walk-through example.
* Buffer contents::             An explanation of the buffer contents.
* Selected files::              To which files are commands applied.
* Commands::                    All commands, grouped by type.

* Log Edit Mode::               Major mode to edit log messages.
* Log View Mode::               Major mode to browse log changes.
@c * CVS Status Mode::             Major mode to view CVS' status output.
* Customization::               How you can tailor PCL-CVS to suit your needs.
* Bugs::                        Bugs (known and unknown).

* GNU Free Documentation License:: The license for this documentation.
* Function and Variable Index::  List of functions and variables.
* Concept Index::               List of concepts.
* Key Index::                   List of keystrokes.

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

About PCL-CVS

* Contributors::                Contributors to PCL-CVS.

Commands

* Entering PCL-CVS::            Commands to invoke PCL-CVS
* Setting flags::               Setting flags for CVS commands
* Updating the buffer::
* Movement commands::           How to move up and down in the buffer
* Marking files::               How to mark files that other commands
                                will later operate on.
* Committing changes::          Checking in your modifications to the
                                CVS repository.
* Editing files::               Loading files into Emacs.
* Getting info about files::    Display the log and status of files.
* Adding and removing files::   Adding and removing files
* Undoing changes::             Undoing changes
* Removing handled entries::    Uninteresting lines can easily be removed.
* Ignoring files::              Telling CVS to ignore generated files.
* Viewing differences::         Commands to @samp{diff} different versions.
* Invoking Ediff::              Running @samp{ediff} from @file{*cvs*} buffer.
* Updating files::              Updating files that Need-update.
* Tagging files::               Tagging files.
* Miscellaneous commands::      Miscellaneous commands.

Customization

* Customizing Faces::

@end detailmenu
@end menu

@node     About PCL-CVS
@chapter About PCL-CVS
@cindex About PCL-CVS

PCL-CVS is a front-end to CVS versions 1.9 and later.
It concisely shows the present status of a checked out module in an
Emacs buffer and provides single-key access to the most frequently used CVS
commands.  Note that the @code{vc-dir} command (@pxref{VC Directory
Mode, , , emacs, The GNU Emacs Manual}) provides similar
functionality, but for several version control systems, including CVS.

PCL-CVS was originally written many years ago by Per Cederqvist who
proudly maintained it until January 1996, at which point he released the
beta version 2.0b2 and passed on the maintainership to Greg A Woods.
Development stayed mostly dormant for a few years during which
version 2.0 never seemed to be able to leave the ``beta'' stage while a
separate XEmacs version was slowly splitting away.  In late 1998,
Stefan Monnier picked up development again, adding some major new
functionality and taking over the maintenance.

@menu
* Contributors::                Contributors to PCL-CVS.
@end menu

@node     Contributors
@section Contributors to PCL-CVS
@cindex Contributors
@cindex Authors

Contributions to the package are welcome.  I have limited time to work
on this project, but I will gladly add any code that you contribute to
me to this package (@pxref{Bugs}).

The following persons have made contributions to PCL-CVS.

@itemize @bullet
@item
Brian Berliner wrote CVS, together with some other contributors.
Without his work on CVS this package would be useless@dots{}

@item
Per Cederqvist wrote most of the otherwise unattributed functions in
PCL-CVS as well as all the documentation.

@item
@c inge@@lysator.liu.se
Inge Wallin wrote the skeleton of
@file{pcl-cvs.texi}, and gave useful comments on it.  He also wrote
the files @file{elib-node.el} and @file{compile-all.el}.  The file
@file{cookie.el} was inspired by Inge.

@item
@c linus@@lysator.liu.se
Linus Tolke contributed useful comments
on both the functionality and the documentation.

@item
@c jwz@@jwz.com
Jamie Zawinski contributed
@file{pcl-cvs-lucid.el}, which was later renamed to
@file{pcl-cvs-xemacs.el}.

@item
Leif Lonnblad contributed RCVS support (since superseded by the new
remote CVS support).

@item
@c jimb@@cyclic.com
Jim Blandy contributed hooks to automatically
guess CVS log entries from @file{ChangeLog} contents, and initial support of
the new Cygnus / Cyclic remote CVS, as well as various sundry bug fixes
and cleanups.

@item
@c kingdon@@cyclic.com
Jim Kingdon contributed lots of fixes to
the build and installation procedure.

@item
@c woods@@weird.com
Greg A. Woods contributed code to implement
the use of per-file diff buffers, and vendor join diffs with emerge and
ediff, as well as various and sundry bug fixes and cleanups.

@item
@c greg.klanderman@@alum.mit.edu
Greg Klanderman implemented
toggling of marked files, setting of CVS command flags via prefix
arguments, updated the XEmacs support, updated the manual, and fixed
numerous bugs.

@item
@c monnier@@gnu.org
Stefan Monnier added a slew of other
features and introduced even more new bugs.  If there's any bug left,
you can be sure it's his.

@item
@c wordy to avoid an underfull hbox
@c masata-y@@is.aist-nara.ac.jp
Masatake YAMATO made a gracious
contribution of his cvstree code to display a tree of tags which was later
superseded by the new @code{cvs-status-mode}.
@end itemize

Apart from these, a lot of people have sent us suggestions, ideas,
requests, bug reports and encouragement.  Thanks a lot!  Without you
there would be no new releases of PCL-CVS.


@node Getting started
@chapter Getting started
@cindex Introduction
@cindex Example run
@cindex Sample session

This document assumes that you know what CVS is, and that you at least
know the fundamental concepts of CVS@.  If that is not the case, you
should read the CVS documentation.  Type @kbd{info -f cvs} or @kbd{man
cvs}.

PCL-CVS is only useful once you have checked out a module.  So before
you invoke it, you must have a copy of a module somewhere in the file
system.

You can invoke PCL-CVS by typing @kbd{M-x cvs-examine @key{RET}}.
You can also invoke it via the menu bar, under @samp{Tools}.
Or, if you prefer, you can also invoke PCL-CVS by simply visiting the
CVS administrative subdirectory of your module, with a prefix argument.
For example, to invoke PCL-CVS in a separate frame, type @kbd{C-u C-x 5
f ~/my/project/CVS @key{RET}}.

The function @code{cvs-examine} will ask for a directory.  The command
@samp{cvs -n update} will be run in that directory.  (It should contain
files that have been checked out from a CVS archive.)  The output from
@code{cvs} will be parsed and presented in a table in a buffer called
@file{*cvs*}.  It might look something like this:

@example
Repository : /usr/CVSroot
Module     : test
Working dir: /users/ceder/FOO/test


In directory .:
           Need-Update            bar
           Need-Update            file.txt
           Modified               namechange
           Need-Update            newer
In directory sub:
           Modified               ChangeLog

--------------------- End ---------------------
-- last cmd: cvs -f -z6 -n update -d -P --
@end example

In this example, your repository is in @file{/usr/CVSroot} and CVS has
been run in the directory @file{/users/ceder/FOO/test}.  The three files
(@file{bar}, @file{file.txt} and
@file{newer}) that are marked with @samp{Need-Update} have been changed
by someone else in the CVS repository.  Two files (@file{namechange}
and @file{sub/ChangeLog}) have been modified locally, and need to be
checked in.

You can move the cursor up and down in the buffer with @kbd{C-n} and
@kbd{C-p} or @kbd{n} and @kbd{p}.  If you press @kbd{c} on one of the
@samp{Modified} files, that file will be checked in to the CVS
repository. @xref{Committing changes}.  You can also press @kbd{O} to
update any of the files that are marked @samp{Need-Update}.  You can
also run @kbd{M-x cvs-update @key{RET}} (bound to @kbd{M-u} in the
@file{*cvs*} buffer) to update all the files.

You can then press @kbd{=} to easily get a @samp{diff} between your
modified file and the base version that you started from, or you can
press @kbd{l} to get the output from @samp{cvs log}.  Many more such
commands are available simply by pressing a key (@pxref{Getting info
about files}).

@node Buffer contents
@chapter Buffer contents
@cindex Buffer contents
@cindex @file{*cvs*} buffer contents

The display contains several columns, some of which are optional.
These columns are, from left to right:

@itemize @bullet

@item
Optionally, the head revision of the file.  This is the latest version
found in the repository.  It might also contain (instead of the head
revision) a sub status which typically gives further information about
how we got to the current state, for example @samp{patched},
@samp{merged}, @dots{}

@item
An asterisk when the file is @dfn{marked} (@pxref{Selected
files}).

@item
The actual status of the file wrt the repository.  See below.

@item
Optionally, the base revision of the file.  This is the version
which the copy in your working directory is based upon.

@item
The file name.

@end itemize

The @samp{file status} field can have the following values:

@table @samp
@item Modified
The file is modified in your working directory, and there was no
modification to the same file in the repository.  This status can have
the following substatus:

@table @samp
@item merged
The file was modified in your working directory, and there were
modifications in the repository as well, but they were merged
successfully, without conflict, in your working directory.
@end table

@item Conflict
A conflict was detected while trying to merge your changes to @var{file}
with changes from the repository.  @var{file} (the copy in your
working directory) is now the output of the @code{rcsmerge} command on
the two versions; an unmodified copy of your file is also in your
working directory, with the name @file{.#@var{file}.@var{version}},
where @var{version} is the RCS revision that your modified file started
from.  @xref{Viewing differences}, for more details.

A conflict can also come from a disagreement on the existence of the file
rather than on its content.  This case is indicated by the following
possible substatus:

@table @samp
@item removed
The file is locally removed but a new revision has been committed to
the repository by someone else.

@item added
The file is locally added and has also been added to the repository
by someone else.

@item modified
The file is locally modified but someone else has removed it from the
repository.
@end table

@item Added
The file has been added by you, but it still needs to be checked in to
the repository.

@item Removed
The file has been removed by you, but it still needs to be checked in to
the repository.  You can resurrect it by typing @kbd{a} (@pxref{Adding
and removing files}).

@item Unknown
A file that was detected in your directory, but that neither appears in
the repository, nor is present on the list of files that CVS should
ignore.

@item Up-to-date
The file is up to date with respect to the version in the repository.
This status can have a substatus of:

@table @samp
@item added
You have just added the file to the repository.

@item updated
The file was brought up to date with respect to the repository.  This is
done for any file that exists in the repository but not in your source,
and for files that you haven't changed but are not the most recent
versions available in the repository.

@item patched
The file was brought up to date with respect to the remote repository by
way of fetching and applying a patch to the file in your source.  This
is equivalent to @samp{updated} except that CVS decided to use a hopefully
more efficient method.

@item committed
You just committed the file.
@end table

@item Need-Update
Either a newer version than the one in your source is available in the
repository and you have not modified your checked out version, or the
file exists in the repository but not in your source.  Use
@samp{cvs-mode-update} bound to @kbd{O} to update the file.

@item Need-Merge
You have modified the checked out version of the file, and a newer
version is available in the repository.  A merge will take place when
you run a @samp{cvs-update}.

@item Missing
The file has been unexpectedly removed from your working directory
although it has not been @samp{cvs remove}d.
@end table

@node Selected files
@chapter Selected files
@cindex Selected files
@cindex Marked files
@cindex File selection
@cindex Active files
@cindex Applicable

Many of the commands work on the current set of @dfn{selected} files
which can be either the set of marked files (if any file is marked and
marks are not ignored) or whichever file or directory the cursor is on.

If a directory is selected but the command cannot be applied to a
directory, then it will be applied to the set of files under this
directory which are in the @file{*cvs*} buffer.

@findex cvs-mode-force-command
@findex cvs-allow-dir-commit
Furthermore, each command only operates on a subset of the selected
files, depending on whether or not the command is @dfn{applicable} to
each file (based on the file's status).  For example,
@code{cvs-mode-commit} is not applicable to a file whose status is
@samp{Need-Update}.  If it should happen that PCL-CVS guesses the
applicability wrong, you can override it with the special prefix
@code{cvs-mode-force-command} normally bound to @kbd{M-f} (and file a
bug report).  The applicability rule can be slightly changed with
@code{cvs-allow-dir-commit} and @code{cvs-force-dir-tag}.

By default, marks are always in effect (you may change this, however, by
setting the variable @code{cvs-default-ignore-marks}) except for the
commands that @samp{tag} or @samp{diff} a file (which can be changed
with the variable @code{cvs-invert-ignore-marks}).

In addition, you may use the special prefix @code{cvs-mode-toggle-marks}
normally bound to @key{T} to toggle the use of marks for the following
command.

This scheme might seem a little complicated, but once one gets used to
it, it is quite powerful.

For commands to mark and unmark files, see @ref{Marking files}.

@node Commands
@chapter Commands

@iftex
This chapter describes all the commands that you can use in PCL-CVS.
@end iftex
@ifnottex
The nodes in this menu contains explanations about all the commands that
you can use in PCL-CVS@.  They are grouped together by type.
@end ifnottex

@menu
* Entering PCL-CVS::            Commands to invoke PCL-CVS
* Setting flags::               Setting flags for CVS commands
* Updating the buffer::
* Movement commands::           How to move up and down in the buffer
* Marking files::               How to mark files that other commands
                                will later operate on.
* Committing changes::          Checking in your modifications to the
                                CVS repository.
* Editing files::               Loading files into Emacs.
* Getting info about files::    Display the log and status of files.
* Adding and removing files::   Adding and removing files
* Undoing changes::             Undoing changes
* Removing handled entries::    Uninteresting lines can easily be removed.
* Ignoring files::              Telling CVS to ignore generated files.
* Viewing differences::         Commands to @samp{diff} different versions.
* Invoking Ediff::              Running @samp{ediff} from @file{*cvs*} buffer.
* Updating files::              Updating files that Need-update.
* Tagging files::               Tagging files.
* Miscellaneous commands::      Miscellaneous commands.
@end menu


@node Entering PCL-CVS
@section Entering PCL-CVS
@findex cvs-update
@findex cvs-examine
@findex cvs-status
@findex cvs-checkout
@findex cvs-quickdir
@cindex Creating the *cvs* buffer

Most commands in PCL-CVS require that you have a @file{*cvs*}
buffer.  The commands that you use to get one are listed below.
For each, a @samp{cvs} process will be run, the output will be parsed by
PCL-CVS, and the result will be printed in the @file{*cvs*} buffer (see
@ref{Buffer contents}, for a description of the buffer's contents).

@table @kbd
@item M-x cvs-update
Run a @samp{cvs update} command.  You will be asked for the directory
in which the @samp{cvs update} will be run.

@item M-x cvs-examine
Run a @samp{cvs -n update} command.  This is identical to the previous
command, except that it will only check what needs to be done but will
not change anything.  You will be asked for the directory in
which the @samp{cvs -n update} will be run.

@item M-x cvs-status
Run a @samp{cvs status} command.  You will be asked for the directory
in which the @samp{cvs status} will be run.

@item M-x cvs-checkout
Run a @samp{cvs checkout} command.  You will be asked for the directory
in which the @samp{cvs update} will be run and the module to be checked
out.

@item M-x cvs-quickdir
Populate the @file{*cvs*} buffer by just looking at the @file{CVS/Entries}
files.  This is very much like @code{cvs-examine} except that it does
not access the CVS repository, which is a major advantage when the
repository is far away.  But of course, it will not be able to detect
when a file needs to be updated or merged.
@end table

@findex cvs-dired-action
@findex cvs-dired-use-hook
The first four of
those commands are also reachable from the menu bar
under @samp{Tools->PCL-CVS}.  Finally, an alternative way is to visit
the CVS administrative subdirectory in your work area with a simple
prefix argument.  For example @kbd{C-u C-x C-f ~/my/work/CVS @key{RET}}.  This
by default runs @code{cvs-quickdir} but the specific behavior can be
changed with @code{cvs-dired-action} and @code{cvs-dired-use-hook}.

By default, the commands above will descend recursively into
subdirectories.  You can avoid that behavior by including @samp{-l} in
the flags for the command.  These flags can be set by giving a prefix
argument to the command (e.g., by typing
@kbd{C-u M-x cvs-update @key{RET} -l @key{RET}}).


@node Setting flags
@section Setting flags for CVS commands
@cindex Optional switches to CVS
@cindex Command-line options to CVS

This section describes the convention used by nearly all PCL-CVS
commands for setting optional flags sent to CVS@.  A single @kbd{C-u}
prefix argument is used to cause the command to prompt for flags to be
used for the current invocation of the command only.  Two @kbd{C-u} prefix
arguments are used to prompt for flags which will be set permanently, for the
current invocation and all that follow, until the flags are changed, or
unless temporary flags are set which override them.

Perhaps an example or two is in order.  Say you are about to add a
binary file to the repository, and want to specify the flags @samp{-kb}
to @samp{cvs add}.  You can type @kbd{C-u a -kb @key{RET}},
and the file will be added.  Subsequent @samp{cvs add}
commands will use the previously prevailing flags.

As a second example, say you are about to perform a diff and want to see
the result in unified diff format, i.e., you'd like to pass the flag
@samp{-u} to both @samp{cvs diff} and @samp{diff}.  You'd also like all
subsequent diffs to use this flag.  You can type @kbd{C-u C-u = -u @key{RET}}
and the diff will be performed, and the default flags will be set to
@code{("-u")}.  You can of course override this flag for a single diff
by using a single @kbd{C-u} prefix argument.

@cindex Special prefix
In addition to this, some commands can take @dfn{special prefix} arguments.
These work as follows: When called with a @kbd{C-u} prefix, the user is
prompted for a new value of the special prefix and the special prefix is
activated for the next command.  When called without the @kbd{C-u}
prefix, the special prefix is re-activated (with the same value as last
time) for the next command.  Calling the prefix command again when it's
already activated deactivates it.  Calling it with the @kbd{C-u C-u}
prefix activates it for all subsequent commands until you deactivate it
explicitly.  The special prefixes are:

@table @kbd
@item T
Toggles whether or not marks will be active in the next command.

@item b
Provide the next command with a branch (can be any version
specifier) to work on.

@item B
Secondary branch argument.  Only meaningful if @kbd{b} is also used.
It can be used to provide a second branch argument to
@code{cvs-mode-diff} or to @code{cvs-mode-update}.

@item M-f
Forces the next command to apply to every selected file rather than only
to the ones PCL-CVS thinks are relevant.
@end table

@node Updating the buffer
@section Updating the @file{*cvs*} buffer
@findex cvs-update
@findex cvs-examine
@findex cvs-status
@findex cvs-mode-update
@findex cvs-mode-examine
@findex cvs-mode-status

The following commands can be used from within the @file{*cvs*} buffer
to update the display:

@table @kbd
@item M-u
Runs the command @samp{cvs-update}.

@item M-e
Runs the command @samp{cvs-examine}.

@item M-s
Runs the command @samp{cvs-status}.
@end table

In addition to the above commands which operate on the whole module,
you can run the equivalent CVS command on just a subset of the
files/directories with these keys:

@table @kbd
@item O
Runs @code{cvs-mode-update} on the selected files.  When run on the
top-level directory, this is equivalent to @kbd{M-u}.

@item e
Runs @code{cvs-mode-examine} on the selected files.  When run on the
top-level directory, this is equivalent to @kbd{M-e}.

@findex cvs-status-mode
@item s
Runs @code{cvs-mode-status} on the selected files.  When run on the
top-level directory, this is equivalent to @kbd{M-s}, except that
CVS output will be shown in a @file{*cvs-info*} buffer that will be
put in @samp{cvs-status-mode}.
@end table


@node Movement commands
@section Movement Commands
@cindex Movement Commands
@findex cvs-mode-next-line
@findex cvs-mode-previous-line
@kindex SPC@r{--Move down one file}
@kindex n@r{--Move down one file}
@kindex p@r{--Move up one file}

You can use most normal Emacs commands to move forward and backward in
the buffer.  Some keys are rebound to functions that take advantage of
the fact that the buffer is a PCL-CVS buffer:


@table @kbd
@item @key{SPC}
@itemx n
These keys move the cursor one file forward, towards the end of the
buffer (@code{cvs-mode-next-line}).

@item p
This key moves one file backward, towards the beginning of the buffer
(@code{cvs-mode-previous-line}).
@end table


@node Marking files
@section Marking files
@cindex Selecting files (commands to mark files)
@cindex Marking files
@kindex m@r{--marking a file}
@kindex M@r{--marking all files}
@kindex u@r{--unmark a file}
@kindex ESC DEL@r{--unmark all files}
@kindex DEL@r{--unmark previous file}
@kindex %@r{--mark files matching regexp}
@kindex S@r{--mark files in a particular state}
@kindex T@r{--toggle marks}
@findex cvs-mode-mark
@findex cvs-mode-unmark
@findex cvs-mode-mark-all-files
@findex cvs-mode-unmark-all-files
@findex cvs-mode-unmark-up
@findex cvs-mode-mark-matching-files
@findex cvs-mode-mark-on-state
@findex cvs-mode-toggle-marks

PCL-CVS works on a set of @dfn{selected files} (@pxref{Selected files}).
You can mark and unmark files with these commands:

@table @kbd
@item m
This marks the file that the cursor is positioned on.  If the cursor is
positioned on a directory all files in that directory are marked
(@code{cvs-mode-mark}).

@item u
Unmark the file that the cursor is positioned on. If the cursor is on a
directory, all files in that directory are unmarked
(@code{cvs-mode-unmark}).

@item M
Mark @emph{all} files in the buffer (@code{cvs-mode-mark-all-files}).

@item M-@key{DEL}
Unmark @emph{all} files (@code{cvs-mode-unmark-all-files}).

@item @key{DEL}
Unmark the file on the previous line, and move point to that line
(@code{cvs-mode-unmark-up}).

@item %
Mark all files matching a regular expression
(@code{cvs-mode-mark-matching-files}).

@item S
Mark all files in a particular state, such as ``Modified'' or
``Removed'' (@code{cvs-mode-mark-on-state}).

@item T
Toggle use of marks for the next command (@code{cvs-mode-toggle-marks}).
@end table


@node Committing changes
@section Committing changes
@cindex Committing changes
@findex cvs-mode-commit
@findex cvs-mode-commit-setup
@kindex c@r{--commit files}
@kindex C@r{--commit files with @file{ChangeLog} message}
@vindex cvs-auto-revert@r{ (variable)}
@cindex Commit buffer
@cindex Edit buffer
@cindex Erasing commit message
@cindex Reverting buffers after commit

Committing changes basically works as follows:

@enumerate
@item
After having selected the files you want to commit, you type either
@kbd{c} or @kbd{C} which brings up a special buffer
@file{*cvs-commit*}.

@item
You type in the log message describing the changes you're about to
commit (@pxref{Log Edit Mode}).

@item
When you're happy with it, you type @kbd{C-c C-c} to do the actual
commit.
@end enumerate

There's no hidden state, so you can abort the process or pick it up
again at any time.

@vindex log-edit-confirm@r{ (variable)}
The set of files actually committed is really decided only during the
very last step, which is a mixed blessing.  It allows you to go back and
change your mind about which files to commit, but it also means that you
might inadvertently change the set of selected files.  To reduce the
risk of error, @kbd{C-c C-c} will ask for confirmation if the set of
selected files has changed between the first step and the last.  You can
change this last detail with @code{log-edit-confirm}.

As for the difference between @kbd{c} (i.e., @code{cvs-mode-commit}) and
@kbd{C} (i.e., @code{cvs-mode-commit-setup}) is that the first gets you
straight to @file{*cvs-commit*} without erasing it or changing anything
to its content, while the second first erases @file{*cvs-commit*}
and tries to initialize it with a sane default (it does that by either
using a template provided by the CVS administrator or by extracting a
relevant log message from a @file{ChangeLog} file).

If you are editing the files in your Emacs, an automatic
@samp{revert-buffer} will be performed.  (If the file contains
@samp{$@asis{Id}$} keywords, @samp{cvs commit} will write a new file with
the new values substituted.  The auto-revert makes sure that you get
them into your buffer.)  The revert will not occur if you have modified
your buffer, or if @samp{cvs-auto-revert} is set to
@samp{nil}.


@node Editing files
@section Editing files
@cindex Editing files
@cindex Finding files
@cindex Loading files
@cindex Dired
@cindex Invoking dired
@findex cvs-mode-find-file
@findex cvs-mode-find-file-other-window
@findex cvs-mode-add-change-log-entry-other-window
@kindex f@r{--find file or directory}
@kindex o@r{--find file in other window}
@kindex A@r{--add @file{ChangeLog} entry}

There are currently three commands that can be used to find a file (that
is, load it into a buffer and start editing it there).  These commands
work on the line that the cursor is situated at.  They always ignore any marked
files.

@table @kbd
@item f
Find the file that the cursor points to (@code{cvs-mode-find-file}).  If
the cursor points to a directory, run @code{dired} on that directory;
@inforef{Dired, , emacs}.

@item o
Like @kbd{f}, but use another window
(@code{cvs-mode-find-file-other-window}).

@item A
Invoke @samp{add-change-log-entry-other-window} to edit a
@file{ChangeLog} file.  The @file{ChangeLog} file will be found in the
directory of the file the cursor points to, or in a parent of that
directory (@code{cvs-mode-add-change-log-entry-other-window}).
@end table


@node Getting info about files
@section Getting info about files
@cindex Status (cvs command)
@cindex Log (RCS/cvs command)
@cindex Getting status
@kindex l@r{--run @samp{cvs log}}
@kindex s@r{--run @samp{cvs status}}
@findex cvs-mode-log
@findex cvs-mode-status

@table @kbd
@item l
Call the command @code{cvs-mode-log} which runs @samp{cvs log} on all
selected files, and show the result in a temporary buffer
@file{*cvs-info*} (@pxref{Log View Mode}).

@item s
Call the command @code{cvs-mode-status} which runs @samp{cvs status} on
all selected files, and show the result in a temporary buffer
@file{*cvs-info*}.
@c Fixme: reinstate when node is written:
@c (@pxref{CVS Status Mode}).
@end table


@node Adding and removing files
@section Adding and removing files
@cindex Adding files
@cindex Removing files
@cindex Resurrecting files
@cindex Deleting files
@cindex Putting files under CVS control
@kindex a@r{--add a file}
@kindex r@r{--remove a file}
@findex cvs-mode-add
@findex cvs-mode-remove-file

The following commands are available to make it easy to add files to
and remove them from the CVS repository.

@table @kbd
@item a
Add all selected files.  This command can be used on @samp{Unknown}
files (@pxref{Buffer contents}).  The status of the file will change to
@samp{Added}, and you will have to use @kbd{c} (@samp{cvs-mode-commit}
@pxref{Committing changes}), to really add the file to the
repository.

This command can also be used on @samp{Removed} files (before you commit
them) to resurrect them.

The command that is run is @code{cvs-mode-add}.

@item r
This command removes the selected files (after prompting for
confirmation).  The files are deleted from your directory and
(unless the status was @samp{Unknown}; @pxref{Buffer contents}) they will
also be @samp{cvs remove}d.  If the files' status was @samp{Unknown}
they will disappear from the buffer.  Otherwise their status will change to
@samp{Removed}, and you must use @kbd{c} (@samp{cvs-mode-commit},
@pxref{Committing changes}) to commit the removal.

The command that is run is @code{cvs-mode-remove-file}.
@end table


@node Undoing changes
@section Undoing changes
@cindex Undo changes
@cindex Flush changes
@kindex U@r{--undo changes}
@findex cvs-mode-undo-local-changes

@table @kbd
@item U
If you have modified a file, and for some reason decide that you don't
want to keep the changes, you can undo them with this command.  It works
by removing your working copy of the file and then getting the latest
version from the repository (@code{cvs-mode-undo-local-changes}).
@end table


@node Removing handled entries
@section Removing handled entries
@cindex Expunging uninteresting entries
@cindex Uninteresting entries, getting rid of them
@cindex Getting rid of uninteresting lines
@cindex Removing uninteresting (processed) lines
@cindex Handled lines, removing them
@kindex x@r{--remove processed entries}
@kindex C-k@r{--remove selected entries}
@findex cvs-mode-remove-handled
@findex cvs-mode-acknowledge
@findex cvs-mode-ignore

@table @kbd
@item x
This command allows you to remove all entries that you have processed.
More specifically, the lines for @samp{Up-to-date} files (@pxref{Buffer
contents}) are removed from the buffer.  If a directory becomes empty
the heading for that directory is also removed.  This makes it easier to
get an overview of what needs to be done.

@vindex cvs-mode-remove-handled@r{ (variable)}
@kbd{x} invokes @code{cvs-mode-remove-handled}.  If
@samp{cvs-auto-remove-handled} is set to non-@code{nil}, this will
automatically be performed after every commit.

@item C-k
This command can be used for lines that @samp{cvs-mode-remove-handled} would
not delete, but that you want to delete (@code{cvs-mode-acknowledge}).
@end table


@node Ignoring files
@section Ignoring files
@cindex Ignoring files
@kindex i@r{--ignoring files}
@findex cvs-mode-ignore

@table @kbd
@item i
Arrange so that CVS will ignore the selected files.  The file names are
added to the @file{.cvsignore} file in the corresponding directory.  If
the @file{.cvsignore} file doesn't exist, it will be created.

The @file{.cvsignore} file should normally be added to the repository,
but you could ignore it as well, if you like it better that way.

This runs @code{cvs-mode-ignore}.
@end table

@node Viewing differences
@section Viewing differences
@cindex Diff
@cindex Invoking @code{diff}
@cindex Conflicts, how to resolve them
@cindex Viewing differences
@kindex d=@r{--run @samp{cvs diff}}
@kindex =@r{--run @samp{cvs diff}}
@kindex db@r{--diff against base version}
@kindex dh@r{--diff against head of repository}
@kindex dr@r{--diff between base and head of repository}
@kindex dv@r{--diff against vendor branch}
@kindex dy@r{--diff against yesterday's head}
@findex cvs-mode-diff
@findex cvs-mode-diff-backup
@findex cvs-mode-diff-head
@findex cvs-mode-diff-repository
@findex cvs-mode-diff-vendor
@findex cvs-mode-diff-yesterday
@vindex cvs-invert-ignore-marks@r{ (variable)}

@table @kbd
@item =
@itemx d =
Display a @samp{cvs diff} between the selected files and the version
that they are based on (@code{cvs-mode-diff}).

@item d b
If CVS finds a conflict while merging two versions of a file (during a
@samp{cvs update}, @pxref{Updating the buffer}) it will save the
original file in a file called @file{.#@var{file}.@var{version}} where
@var{file} is the name of the file, and @var{version} is the revision
number that @var{file} was based on.

With the @kbd{d b} command you can run a @samp{diff} on the files
@file{.#@var{file}.@var{version}} and @file{@var{file}}.

@item d h
Display a @samp{cvs diff} between the selected files and the head
revision (the most recent version on the current
branch) in the repository (@code{cvs-mode-diff-head}).

@item d r
Display a @samp{cvs diff} between the base revision of the selected
files and the head revision in the repository.  This displays the
changes anyone has committed to the repository since you last executed
a checkout, update or commit operation
(@code{cvs-mode-diff-repository}).

@item d v
Display a @samp{cvs diff} between the selected files and the head
revision of the vendor branch in the repository
(@code{cvs-mode-diff-vendor}).

@item d y
Display a @samp{cvs diff} between the selected files and yesterday's
head revision in the repository
(@code{cvs-mode-diff-yesterday}).
@end table

By default, @samp{diff} commands ignore the marks.  This can be changed
with @code{cvs-invert-ignore-marks}.

@node Invoking Ediff
@section Running ediff
@cindex Ediff
@cindex Invoking ediff
@cindex Viewing differences
@cindex Conflicts, how to resolve them
@cindex Resolving conflicts
@kindex e@r{--invoke @samp{ediff}}
@findex cvs-mode-idiff
@findex cvs-mode-imerge

@table @kbd
@vindex cvs-idiff-imerge-handlers@r{ (variable)}
@item d e
This uses @code{ediff} (or @code{emerge}, depending on
@samp{cvs-idiff-imerge-handlers}) to allow you to view diffs.
If a prefix argument is given, PCL-CVS will prompt for a revision against
which the diff should be made, else the default will be to use the BASE
revision.

@cindex Merging with @code{ediff} and @code{emerge}
@item d E
This command use @code{ediff} (or @code{emerge}, see above) to allow you
to do an interactive 3-way merge.

@strong{Please note:}  when the file status is @samp{Conflict},
CVS has already performed a merge.  The resulting file is not used in
any way if you use this command.  If you use the @kbd{q} command inside
@samp{ediff} (to successfully terminate a merge) the file that CVS
created will be overwritten.
@end table

@node Updating files
@section Updating files
@findex cvs-mode-update
@cindex Updating files
@kindex O@r{--update files}

@table @kbd
@item O
Update all selected files with status @samp{Need-update} by running
@samp{cvs update} on them (@code{cvs-mode-update}).
@end table


@node Tagging files
@section Tagging files
@findex cvs-mode-tag
@findex cvs-mode-untag
@findex cvs-rtag
@cindex Tagging files
@kindex M-t@r{--repository tag files}
@kindex t@r{--tag files}
@vindex cvs-invert-ignore-marks@r{ (variable)}
@vindex cvs-force-dir-tag@r{ (variable)}

@table @kbd
@item t
Tag all selected files by running @samp{cvs tag} on
them (@code{cvs-mode-tag}).  It's usually preferable to tag a directory
at a time.  Rather than selecting all files (which too often doesn't
select all files but only the few that are displayed), clear the
selection with @kbd{M-DEL} (@code{cvs-mode-unmark-all-files}), position
the cursor on the directory you want to tag and hit @kbd{t}.
@end table

By default, @samp{tag} commands ignore the marks.  This can be changed
with @code{cvs-invert-ignore-marks}.  Also, by default @samp{tag} can
only be applied to directories, see @code{cvs-force-dir-tag} if you want
to change this behavior.


@node Miscellaneous commands
@section Miscellaneous commands
@findex cvs-mode-byte-compile-files
@cindex Recompiling elisp files
@cindex Byte compilation
@findex cvs-mode-delete-lock
@cindex Getting rid of lock files
@cindex Lock files
@kindex q@r{--bury the PCL-CVS buffer}
@findex cvs-bury-buffer
@findex cvs-mode-quit
@cindex Quitting
@kindex h@r{--help}
@kindex ?@r{--help}
@findex cvs-help
@cindex Help

@table @kbd
@item M-x cvs-mode-byte-compile-files
Byte compile all selected files that end in @file{.el}.

@item M-x cvs-mode-delete-lock
This command deletes the lock files that
the @file{*cvs*} buffer informs you about.  You should normally never have to
use this command, since CVS tries very carefully to always remove the
lock files itself.

You can only use this command when a message in the @file{*cvs*} buffer tells
you so.  You should wait a while before using this command in case
someone else is running a @code{cvs} command.

Also note that this only works if the repository is local.

@item ?
@itemx h
Show a summary of common command key bindings in the echo
area (@code{cvs-help}).

@item q
Bury the PCL-CVS buffer (@code{cvs-bury-buffer}).

@item M-x cvs-mode-quit
Quit PCL-CVS, killing the @file{*cvs*} buffer.
@end table

@node Log Edit Mode
@chapter Editing a Log Message

@cindex Log Edit mode
@cindex mode, Log Edit
Buffers for entering/editing log messages for changes which are about
to be committed are put into Log Edit mode.

Sometimes the log buffer contains default text when you enter it,
typically the last log message entered.  If it does, mark and point
are set around the entire contents of the buffer so that it is easy to
kill the contents of the buffer with @kbd{C-w}.

@findex log-edit-insert-changelog
If you work by writing entries in the @file{ChangeLog}
(@pxref{Change Log,,, emacs, The GNU Emacs Manual}) and then commit the change under revision
control, you can generate the Log Edit text from the ChangeLog using
@kbd{C-c C-a} (@kbd{log-edit-insert-changelog}).  This looks for
entries for the file(s) concerned in the top entry in the ChangeLog
and uses those paragraphs as the log text.  This text is only inserted
if the top entry was made under your user name on the current date.
@xref{Change Logs and VC,,, emacs, The GNU Emacs Manual}, for the opposite way of
working---generating ChangeLog entries from the revision control log.

In the Log Edit buffer, @kbd{C-c C-f} (@kbd{M-x log-edit-show-files})
shows the list of files to be committed in case you need to check
that.

When you have finished editing the log message, type @kbd{C-c C-c} to
exit the buffer and commit the change.

@c Fixme: customization variables

@node Log View Mode
@chapter Browsing a Log of Changes

@cindex Log View mode
@cindex mode, Log View
@cindex output, logs

@findex cvs-mode-log
@findex vc-print-log
Log View mode provides a few useful commands for navigating revision
control log output.  It is used for the output buffers of both
@code{cvs-mode-log} and @code{vc-print-log}.

In this mode, @kbd{n} goes to the next message and @kbd{p} goes to the
previous message and @kbd{N} and @kbd{P} go to the next and previous
files, respectively, in multi-file output.  With a numeric prefix
argument, these commands move that many messages of files.

@c @node CVS Status Mode
@c @chapter Viewing CVS' Status output

@node Customization
@chapter Customization
@vindex log-edit-changelog-full-paragraphs@r{ (variable)}
@vindex cvs-auto-remove-handled@r{ (variable)}
@vindex cvs-auto-remove-directories@r{ (variable)}
@vindex cvs-update-prog-output-skip-regexp@r{ (variable)}
@vindex cvs-cvsroot@r{ (variable)}
@vindex cvs-auto-revert@r{ (variable)}
@vindex log-edit-require-final-newline@r{ (variable)}
@vindex cvs-sort-ignore-file@r{ (variable)}
@cindex Customization
@cindex Variables, list of all
@cindex Erasing input buffer
@cindex Context diff, how to get
@cindex Unidiff, how to get
@cindex Automatically remove handled files
@cindex @samp{-u} option in modules file
@cindex Modules file (@samp{-u} option)
@cindex Update program (@samp{-u} option in modules file)
@cindex Reverting buffers after commit
@cindex Require final newline
@cindex Automatically inserting newline
@cindex Commit message, inserting newline
@cindex Sorting @file{.cvsignore} file
@cindex @file{.cvsignore} file, sorting
@cindex Automatically sorting @file{.cvsignore}
@cindex @samp{CVSROOT}, overriding

If you have an idea about any customization that would be handy but
isn't present in this list, please tell us!
For info on how to reach us, see @ref{Bugs}.

@table @samp
@item cvs-auto-remove-handled
If this variable is set to any non-@code{nil} value,
@samp{cvs-mode-remove-handled} will be called every time you check in
files, after the check-in is ready.  @xref{Removing handled
entries}.

@item cvs-auto-remove-directories
If this variable is set to any non-@code{nil} value, directories that do
not contain any files to be checked in will not be listed in the
@file{*cvs*} buffer.

@item cvs-auto-revert
If this variable is set to any non-@samp{nil} value any buffers you have
that visit a file that is committed will be automatically reverted.
This variable defaults to @samp{t}. @xref{Committing changes}.

@item cvs-update-prog-output-skip-regexp
The @samp{-u} flag in the @file{modules} file can be used to run a command
whenever a @samp{cvs update} is performed (see @code{cvs(5)}).  This regexp
is used to search for the last line in that output.  It is normally set
to @samp{$}.  That setting is only correct if the command outputs
nothing.  Note that PCL-CVS will get very confused if the command
outputs @emph{anything} to @code{stderr}.

@item cvs-cvsroot
This variable can be set to override @samp{CVSROOT}.  It should be a
string.  If it is set, then every time a @code{cvs} command is run, it
will be called as @samp{cvs -d @var{cvs-cvsroot}@dots{}}.  This can be
useful if your site has several repositories.

@item log-edit-require-final-newline
@c wordy to avoid underfull hbox
When you enter a log message by typing into the
@file{*cvs-commit-message*} buffer, PCL-CVS normally automatically
inserts a trailing newline, unless there already is one.  This behavior
can be controlled via @samp{cvs-commit-buffer-require-final-newline}.
If it is @samp{t} (the default behavior), a newline will always be
appended.  If it is @samp{nil}, newlines will never be appended.  Any
other value causes PCL-CVS to ask the user whenever there is no trailing
newline in the commit message buffer.

@findex cvs-mode-changelog-commit
@item log-edit-changelog-full-paragraphs
If this variable is non-@code{nil}, include full @file{ChangeLog}
paragraphs in the CVS log created by @samp{cvs-mode-changelog-commit}.
This may be set in the local variables section of a @file{ChangeLog}
file, to indicate the policy for that @file{ChangeLog}.

@cindex @file{ChangeLog} paragraphs
A @dfn{@file{ChangeLog} paragraph} is a bunch of log text containing no
blank lines; a paragraph usually describes a set of changes with a
single purpose, but perhaps spanning several functions in several files.
Changes in different paragraphs are unrelated.

You could argue that the CVS log entry for a file should contain the
full @file{ChangeLog} paragraph mentioning the change to the file, even though
it may mention other files, because that gives you the full context you
need to understand the change.  This is the behavior you get when this
variable is set to @code{t}, the default.

On the other hand, you could argue that the CVS log entry for a change
should contain only the text for the changes which occurred in that
file, because the CVS log is per-file.  This is the behavior you get
when this variable is set to @code{nil}.

@findex cvs-mode-ignore@r{, and @file{.cvsignore} sorting}
@item cvs-sort-ignore-file
If this variable is set to any non-@samp{nil} value, the
@file{.cvsignore} file will always be sorted whenever you use
@samp{cvs-mode-ignore} to add a file to it.  This option is on by
default.
@end table


@menu
* Customizing Faces::
@end menu

@node Customizing Faces
@section Customizing Faces
@vindex cvs-header (face)
@vindex cvs-filename (face)
@vindex cvs-unknown (face)
@vindex cvs-handled (face)
@vindex cvs-need-action (face)
@vindex cvs-marked (face)
@vindex cvs-msg (face)

PCL-CVS adds a few extra features, including menus, mouse bindings, and
fontification of the @file{*cvs*} buffer.  The faces defined for
fontification are listed below:

@table @samp
@item cvs-header
used to highlight directory changes.

@item cvs-filename
Used to highlight file names.

@item cvs-unknown
Used to highlight the status of files which are @samp{Unknown}.

@item cvs-handled
Used to highlight the status of files which are handled and
need no further action.

@item cvs-need-action
Used to highlight the status of files which still need action.

@item cvs-marked
Used to highlight the marked file indicator (@samp{*}).

@item cvs-msg
Used to highlight CVS messages.
@end table


@node   Bugs
@chapter Bugs (known and unknown)
@cindex Reporting bugs and ideas
@cindex Bugs, how to report them
@cindex Author, how to reach
@cindex Email to the author
@cindex Known bugs
@cindex Bugs, known
@cindex FAQ
@cindex Problems, list of common

If you find a bug or misfeature, don't hesitate to tell us!
Use @kbd{M-x report-emacs-bug} to send us a report.
You can follow the same process for feature requests.
We prefer discussing one thing at a time.  If you find several unrelated
bugs, please report them separately.  If you are running PCL-CVS under
XEmacs, you should also send a copy of bug reports to
the @url{http://lists.xemacs.org/mailman/listinfo/xemacs-beta,
XEmacs mailing list}.

If you have problems using PCL-CVS or other questions, send them to
the @url{http://lists.gnu.org/mailman/listinfo/help-gnu-emacs,
help-gnu-emacs mailing list}.  This is a good place to get help, as is
the @url{http://lists.nongnu.org/mailman/listinfo/info-cvs, info-cvs list}.

If you have ideas for improvements, or if you have written some
extensions to this package, we would like to hear from you.  We hope that
you find this package useful!

Below is a partial list of currently known problems with PCL-CVS.

@table @asis
@item Unexpected output from CVS
Unexpected output from CVS may confuse PCL-CVS@.  It will create
warning messages in the @file{*cvs*} buffer alerting you to any parse errors.
If you get these messages, please send a bug report to the email
addresses listed above.  Include the contents of the @file{*cvs*} buffer, the
output of the CVS process (which should be found in the @file{ *cvs-tmp*}
buffer), and the versions of Emacs, PCL-CVS and CVS you are using.
@end table

@node GNU Free Documentation License
@appendix GNU Free Documentation License
@include doclicense.texi



@node Function and Variable Index
@unnumbered Function and Variable Index

This is an index of all the functions and variables documented in this
manual.

@printindex fn

@node     Concept Index
@unnumbered Concept Index

This is an index of concepts discussed in this manual.

@printindex cp

@node     Key Index
@unnumbered Key Index

This index includes an entry for each PCL-CVS key sequence documented in
this manual.

@printindex ky

@bye