emacs/CONTRIBUTE

* How developers contribute to GNU Emacs

Here is how software developers can contribute to Emacs.  (Non-developers: see
https://www.gnu.org/software/emacs/manual/html_node/emacs/Contributing.html
or run the shell command 'info "(emacs)Contributing"'.)

** The Emacs repository

Emacs development uses Git on Savannah for its main repository.
To configure Git for Emacs development, you can run the following:

	git config --global user.name 'Your Name'
	git config --global user.email 'your.name@example.com'
	git config --global transfer.fsckObjects true

The following shell commands then build and run Emacs from scratch:

	git clone git://git.sv.gnu.org/emacs.git
	cd emacs
	./autogen.sh
	./configure
	make
	src/emacs

For more details, see
https://www.emacswiki.org/emacs/GitQuickStartForEmacsDevs and
https://www.emacswiki.org/emacs/GitForEmacsDevs or see the file
admin/notes/git-workflow.

** Getting involved with development

Discussion about Emacs development takes place on emacs-devel@gnu.org.
You can subscribe to the emacs-devel@gnu.org mailing list.
If you want to get only the important mails (for things like
feature freezes), choose to receive only the 'emacs-announce' topic
(although so far this feature has not been well or consistently used).
See https://lists.gnu.org/mailman/listinfo/emacs-devel for mailing list
instructions and archives.  You can develop and commit changes in your
own copy of the repository, and discuss proposed changes on the
mailing list.  Frequent contributors to Emacs can request write access
there.

Bug reports and fixes, feature requests and patches/implementations
should be sent to bug-gnu-emacs@gnu.org, the bug/feature list.  This
is coupled to the https://debbugs.gnu.org tracker.  It is best to use
the command 'M-x report-emacs-bug RET' to report issues to the tracker
(described below).  Be prepared to receive comments and requests for
changes in your patches, following your submission.

The Savannah info page https://savannah.gnu.org/mail/?group=emacs
describes how to subscribe to the mailing lists, or see the list
archives.

To email a patch you can use a shell command like 'git format-patch -1'
to create a file, and then attach the file to your email.  This nicely
packages the patch's commit message and changes, and makes sure the
format and whitespace are not munged in transit by the various mail
agents.  To send just one such patch without additional remarks, it is
also possible to use a command like

  git send-email --to=bug-gnu-emacs@gnu.org 0001-DESCRIPTION.patch

However, we prefer the 'git format-patch' method with attachment, as
doing so delivers patches in the correct and easily-recognizable format
more reliably, and makes the job of applying the patches easier and less
error-prone.  It also allows sending patches whose author is someone
other than the email sender.

Once the cumulative amount of your submissions exceeds a dozen or so
lines of non-trivial changes, we will need you to assign to the FSF
the copyright for your contributions.  (To see how many lines were
non-trivially changed, count only added and modified lines in the
patched code.  Consider an added or changed line non-trivial if it
includes at least one identifier, string, or substantial comment.)
In most cases, to start the assignment process you should download
https://git.savannah.gnu.org/cgit/gnulib.git/plain/doc/Copyright/request-assign.future
and return the completed information to the address at the top.
(There are other assignment options, but they are much less commonly used.)
If you have questions about the assignment process, you can ask the
address listed on the form, and/or emacs-devel@gnu.org.

** Issue tracker (a.k.a. "bug tracker")

The Emacs issue tracker at https://debbugs.gnu.org lets you view bug
reports and search the database for bugs matching several criteria.
Messages posted to the bug-gnu-emacs@gnu.org mailing list, mentioned
above, are recorded by the tracker with the corresponding
bugs/issues.  If a message to the bug tracker contains a patch, please
include the string "[PATCH]" in the subject of the message in order to
let the bug tracker tag the bug properly.

GNU ELPA has a 'debbugs' package that allows accessing the tracker
database from Emacs.

Bugs needs regular attention.  A large backlog of bugs is
disheartening to the developers, and a culture of ignoring bugs is
harmful to users, who expect software that works.  Bugs have to be
regularly looked at and acted upon.  Not all bugs are critical, but at
the least, each bug needs to be regularly re-reviewed to make sure it
is still reproducible.

The process of going through old or new bugs and acting on them is
called bug triage.  This process is described in the file
admin/notes/bug-triage.

** Documenting your changes

Any change that matters to end-users should have an entry in etc/NEWS.
Try to start each NEWS entry with a sentence that summarizes the entry
and takes just one line -- this will allow to read NEWS in Outline
mode after hiding the body of each entry.

Doc-strings should be updated together with the code.

New defcustom's should always have a ':version' tag stating the first
Emacs version in which they will appear.  Likewise with defcustom's
whose value is changed -- update their ':version' tag.

Think about whether your change requires updating the manuals.  If you
know it does not, mark the NEWS entry with "---".  If you know
that *all* the necessary documentation updates have been made as part
of your changes or those by others, mark the entry with "+++".
Otherwise do not mark it.

If your change requires updating the manuals to document new
functions/commands/variables/faces, then use the proper Texinfo
command to index them; for instance, use @vindex for variables and
@findex for functions/commands.  For the full list of predefined indices, see
https://www.gnu.org/software/texinfo/manual/texinfo/html_node/Predefined-Indices.html
or run the shell command 'info "(texinfo)Predefined Indices"'.

We prefer American English both in doc strings and in the manuals.
That includes both spelling (e.g., "behavior", not "behaviour") and
the convention of leaving 2 spaces between sentences.

For more specific tips on Emacs's doc style, see
https://www.gnu.org/software/emacs/manual/html_node/elisp/Documentation-Tips.html
Use 'checkdoc' to check for documentation errors before submitting a patch.

** Testing your changes

Please test your changes before committing them or sending them to the
list.  If possible, add a new test along with any bug fix or new
functionality you commit (of course, some changes cannot be easily
tested).

Emacs uses ERT, Emacs Lisp Regression Testing, for testing.  See
https://www.gnu.org/software/emacs/manual/html_node/ert/
or run 'info "(ert)"' for more information on writing and running
tests.

If your test lasts longer than some few seconds, mark it in its
'ert-deftest' definition with ":tags '(:expensive-test)".

To run tests on the entire Emacs tree, run "make check" from the
top-level directory.  Most tests are in the directory "test/".  From
the "test/" directory, run "make <filename>" to run the tests for
<filename>.el(c).  See "test/README" for more information.

If you're making changes that involve the Emacs build system, please
test 'out-of-tree' builds as well, i.e.:

    mkdir emacs-build
    cd emacs-build
    ../path-to-emacs-sources/configure
    make

** Commit messages

Ordinarily, a change you commit should contain a log entry in its
commit message and should not touch the repository's ChangeLog files.
Here is an example commit message (indented):

	Deactivate shifted region

	Do not silently extend a region that is not highlighted;
	this can happen after a shift (Bug#19003).
	* doc/emacs/mark.texi (Shift Selection): Document the change.
	* lisp/window.el (handle-select-window):
	* src/frame.c (Fhandle_switch_frame, Fselected_frame):
	Deactivate the mark.

Occasionally, commit messages are collected and prepended to a
ChangeLog file, where they can be corrected.  It saves time to get
them right the first time, so here are guidelines for formatting them:

- Start with a single unindented summary line explaining the change;
  do not end this line with a period.  If possible, try to keep the
  summary line to 50 characters or fewer; this is for compatibility
  with certain Git commands that print that line in width-constrained
  contexts.

  If the summary line starts with a semicolon and a space "; ", the
  commit message will be ignored when generating the ChangeLog file.
  Use this for minor commits that do not need separate ChangeLog
  entries, such as changes in etc/NEWS.

- After the summary line, there should be an empty line.

- Unindented ChangeLog entries normally come next.  However, if the
  commit couldn't be properly summarized in the brief summary line,
  you can put a paragraph (after the empty line and before the
  individual ChangeLog entries) that further describes the commit.

- Limit lines in commit messages to 78 characters, unless they consist
  of a single word of at most 140 characters; this is enforced by a
  commit hook.

- If only a single file is changed, the summary line can be the normal
  file first line (starting with the asterisk).  Then there is no
  individual files section.

- If the commit has more than one author, the commit message should
  contain separate lines to mention the other authors, like the
  following:

	Co-authored-by: Joe Schmoe <j.schmoe@example.org>

- If the commit is a tiny change that is exempt from copyright paperwork,
  the commit message should contain a separate line like the following:

	Copyright-paperwork-exempt: yes

- The commit message should contain "Bug#NNNNN" if it is related to
  bug number NNNNN in the debbugs database.  This string is often
  parenthesized, as in "(Bug#19003)".

- When citing URLs, prefer https: to http: when either will do.  In
  particular, gnu.org and fsf.org URLs should start with "https:".

- Commit messages should contain only printable UTF-8 characters.

- Commit messages should not contain the "Signed-off-by:" lines that
  are used in some other projects.

- Any lines of the commit message that start with "; " are omitted
  from the generated ChangeLog.

- Explaining the rationale for a design choice is best done in comments
  in the source code.  However, sometimes it is useful to describe just
  the rationale for a change; that can be done in the commit message
  between the summary line and the file entries.

- Emacs generally follows the GNU coding standards for ChangeLogs: see
  https://www.gnu.org/prep/standards/html_node/Change-Logs.html
  or run 'info "(standards)Change Logs"'.  One exception is that
  commits still sometimes quote `like-this' (as the standards used to
  recommend) rather than 'like-this' or ‘like this’ (as they do now),
  as `...' is so widely used elsewhere in Emacs.

- Some commenting rules in the GNU coding standards also apply
  to ChangeLog entries: they must be in English, and be complete
  sentences starting with a capital and ending with a period (except
  the summary line should not end in a period).  See
  https://www.gnu.org/prep/standards/html_node/Comments.html
  or run 'info "(standards)Comments"'.  American English is preferred
  in Emacs; that includes spelling and leaving 2 blanks between
  sentences.

  They are preserved indefinitely, and have a reasonable chance of
  being read in the future, so it's better that they have good
  presentation.

- Use the present tense; describe "what the change does", not "what
  the change did".

- Preferred form for several entries with the same content:

	* lisp/menu-bar.el (clipboard-yank, clipboard-kill-ring-save)
	(clipboard-kill-region):
	* lisp/eshell/esh-io.el (eshell-virtual-targets)
	(eshell-clipboard-append):
	Replace option gui-select-enable-clipboard with
	select-enable-clipboard; renamed October 2014.  (Bug#25145)

  (Rather than anything involving "ditto" and suchlike.)

- There is no standard or recommended way to identify revisions in
  ChangeLog entries.  Using Git SHA1 values limits the usability of
  the references to Git, and will become much less useful if Emacs
  switches to a different VCS.  So we recommend against doing only that.

  One way to identify revisions is by quoting their summary line.
  Prefixing the summary with the commit date can give useful context
  (use 'git show -s "--pretty=format:%cd \"%s\"" --date=short HASH' to
  produce that).  Often, "my previous commit" will suffice.

- There is no need to mention files such as NEWS and MAINTAINERS, or
  to indicate regeneration of files such as 'lib/gnulib.mk', in the
  ChangeLog entry.  "There is no need" means you don't have to, but
  you can if you want to.

** Generating ChangeLog entries

- If you use Emacs VC, you can use 'C-c C-w' to generate formatted
  blank ChangeLog entries from the diff being committed, then use
  'M-q' to combine and fill them.  See 'info "(emacs) Log Buffer"'.

- Alternatively, you can use Emacs functions for ChangeLog files; see
  https://www.gnu.org/software/emacs/manual/html_node/emacs/Change-Log-Commands.html
  or run 'info "(emacs)Change Log Commands"'.

  To format ChangeLog entries with Emacs VC, create a top-level
  ChangeLog file manually, and update it with 'C-x 4 a' as usual.  Do
  not register the ChangeLog file under git; instead, use 'C-c C-a' to
  insert its contents into your *vc-log* buffer.  Or if
  'log-edit-hook' includes 'log-edit-insert-changelog' (which it does
  by default), they will be filled in for you automatically.

- Instead of Emacs VC, you can use the vc-dwim command to maintain commit
  messages.  When you create a source directory, run the shell command
  'git-changelog-symlink-init' to create a symbolic link from
  ChangeLog to .git/c/ChangeLog.  Edit this ChangeLog via its symlink
  with Emacs commands like 'C-x 4 a', and commit the change using the
  shell command 'vc-dwim --commit'.  Type 'vc-dwim --help' for more.

** Committing your changes.

When you commit changes, Git invokes several scripts that test the
commit for validity, and may abort the commit of some of the tests
fail.  These scripts live in the '.git/hooks/' subdirectory of the
top-level directory of the repository, and they perform the following
tests:

- commit log message must not be empty;
- the first line of the commit log message doesn't start with
  whitespace characters;
- the second line of the commit log message must be empty;
- commit log message should include only valid printable ASCII and
  UTF-8 characters;
- commit log message lines must be shorter than 79 characters, unless
  a line consists of a single long word, in which case that word can
  be up to 140 characters long;
- there shouldn't be any "Signed-off-by:" tags in the commit log
  message, and "git commit" should not be invoked with the '-s' option
  (which automatically adds "Signed-off-by:");
- if the commit adds new files, the file names must not begin with
  '-' and must consist of ASCII letters, digits, and characters of the
  set [-+./_];
- the changes don't include unresolved merge conflict markers;
- the changes don't introduce whitespace errors: trailing whitespace,
  lines that include nothing but whitespace characters, and indented
  lines where a SPC character is immediately followed by a TAB in the
  line's initial indentation

** Committing changes by others

If committing changes written by someone else, commit in their name,
not yours.  You can use 'git commit --author="AUTHOR"' to specify a
change's author.  Note that the validity checks described in the
previous section are still applied, so you will have to correct any
problems they uncover in the changes submitted by others.

** Branches

Future development normally takes place on the master branch.
Sometimes specialized features are developed on other branches before
possibly being merged to the master.  Release branches are named
"emacs-NN" where NN is the major version number, and are mainly
intended for more-conservative changes such as bug fixes.  Typically,
collective development is active on the master branch and possibly on
the current release branch.  Periodically, the current release branch
is merged into the master, using the gitmerge function described in
admin/notes/git-workflow.

If you are fixing a bug that exists in the current release, you should
generally commit it to the release branch; it will be merged to the
master branch later by the gitmerge function.  However, when the
release branch is for Emacs version NN.2 and later, or when it is for
Emacs version NN.1 that is in the very last stages of its pretest,
that branch is considered to be in a feature freeze: only bug fixes
that are "safe" or are fixing major problems should go to the release
branch, the rest should be committed to the master branch.  This is so
to avoid destabilizing the next Emacs release.  If you are unsure
whether your bug fix is "safe" enough for the release branch, ask on
the emacs-devel mailing list.

Documentation fixes (in doc strings, in manuals, in NEWS, and in
comments) should always go to the release branch, if the documentation
to be fixed exists and is relevant to the release-branch codebase.
Doc fixes are always considered "safe" -- even when a release branch
is in feature freeze, it can still receive doc fixes.  However, this
rule is limited to fixing real problems in the documentation; cleanups
and stylistic changes are excluded.

When you know that the change will be difficult to merge to the
master (e.g., because the code on master has changed a lot), you can
apply the change to both master and branch yourself.  It could also
happen that a change is cherry-picked from master to the release
branch, and so doesn't need to be merged back.  In these cases,
say in the release branch commit message that there is no need to merge
the commit to master, by starting the commit message with "Backport:".
The gitmerge function excludes these commits from the merge to the master.

Some changes should not be merged to master at all, for whatever
reasons.  These should be marked by including something like "Do not
merge to master" or anything that matches gitmerge-skip-regexp (see
admin/gitmerge.el) in the commit message.

** Some packages in Emacs are maintained externally

Sometimes a package that ships as part of GNU Emacs is maintained as a
separate project, with its own upstream repository, its own maintainer
group, its own development conventions, etc.  The upstream project's
code is periodically merged into Emacs (exactly when and how such
merges happen depends on the package).

So when you are making a contribution -- such as fixing a bug or
proposing an enhancement -- to one of these externally maintained
packages, you sometimes need to deal with that package at its upstream
source.

In the section "Externally maintained packages" in "admin/MAINTAINERS"
we maintain a list of such packages.

** GNU ELPA

This repository does not contain the Emacs Lisp package archive
(elpa.gnu.org).  See admin/notes/elpa for how to access the GNU ELPA
repository.

** Understanding Emacs internals

The best way to understand Emacs internals is to read the code.  Some
source files, such as xdisp.c, have extensive comments describing the
design and implementation.  The following resources may also help:

https://www.gnu.org/software/emacs/manual/html_node/elisp/Tips.html
https://www.gnu.org/software/emacs/manual/html_node/elisp/GNU-Emacs-Internals.html

or run 'info "(elisp)Tips"' or 'info "(elisp)GNU Emacs Internals"'.

The file etc/DEBUG describes how to debug Emacs bugs.

*** Non-ASCII characters in Emacs files

If you introduce non-ASCII characters into Emacs source files, use the
UTF-8 encoding unless it cannot do the job for some good reason.
Although it is generally a good idea to add 'coding:' cookies to
non-ASCII source files, cookies are not needed in UTF-8-encoded *.el
files intended for use only with Emacs version 24.5 and later.

*** Useful files in the admin/ directory

See all the files in admin/notes/* .  In particular, see
admin/notes/newfile, see admin/notes/repo.

The file admin/MAINTAINERS records the areas of interest of frequent
Emacs contributors.  If you are making changes in one of the files
mentioned there, it is a good idea to consult the person who expressed
an interest in that file, and/or get his/her feedback for the changes.
If you are a frequent contributor and have interest in maintaining
specific files, please record those interests in that file, so that
others could be aware of that.

*** git vs rename

Git does not explicitly represent a file renaming; it uses a percent
changed heuristic to deduce that a file was renamed.  So if you are
planning to make extensive changes to a file after renaming it (or
moving it to another directory), you should:

- Create a feature branch.

- Commit the rename without any changes.

- Make other changes.

- Merge the feature branch to the master branch, instead of squashing
  the commits into one.  The commit message on this merge should
  summarize the renames and all the changes.



This file is part of GNU Emacs.

GNU Emacs is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

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

You should have received a copy of the GNU General Public License
along with GNU Emacs.  If not, see <https://www.gnu.org/licenses/>.

Local variables:
mode: outline
paragraph-separate: "[ 	]*$"
coding: utf-8
end: