@c This is part of the Emacs manual., Abbrevs, This is part of the Emacs manual., Top @c Copyright (C) 1985--1987, 1993--1995, 1997, 1999--2024 Free Software @c Foundation, Inc. @c See file emacs.texi for copying conditions. @node Maintaining @chapter Maintaining Large Programs @cindex maintaining large programs @cindex large programming projects, maintaining This chapter describes Emacs features for maintaining medium- to large-size programs and packages. These features include: @itemize @minus @item Unified interface to Support for Version Control Systems (@acronym{VCS}) that record the history of changes to source files. @item Commands for handling programming projects. @item A specialized mode for maintaining @file{ChangeLog} files that provide a chronological log of program changes. @item @acronym{Xref}, a set of commands for displaying definitions of symbols (a.k.a.@: ``identifiers'') and their references. @item @acronym{EDE}, the Emacs's own IDE. @ifnottex @item A mode for merging changes to program sources made on separate branches of development. @end ifnottex @item A minor-mode for highlighting bug references and visiting the referenced bug reports in their issue tracker. @end itemize If you are maintaining a large Lisp program, then in addition to the features described here, you may find the Emacs Lisp Regression Testing (@acronym{ERT}) library useful (@pxref{Top,,ERT,ert, Emacs Lisp Regression Testing}). @menu * Version Control:: Using version control systems. * Projects:: Commands for handling source files in a project. * Change Log:: Maintaining a change history for your program. * Xref:: Find definitions and references of any function, method, struct, macro, @dots{} in your program. * EDE:: An integrated development environment for Emacs. @ifnottex * Emerge:: A convenient way of merging two versions of a program. @end ifnottex * Bug Reference:: Highlighting references to bug reports and browsing them in their issue trackers. @end menu @node Version Control @section Version Control @cindex version control A @dfn{version control system} is a program that can record multiple versions of a source file, storing information such as the creation time of each version, who made it, and a description of what was changed. @cindex VC The Emacs version control interface is called @dfn{VC}@. VC commands work with several different version control systems; currently, it supports Bazaar, CVS, Git, Mercurial, Monotone, RCS, SRC, SCCS/CSSC, and Subversion. Of these, the GNU project distributes CVS, RCS, and Bazaar. VC is enabled automatically whenever you visit a file governed by a version control system. To disable VC entirely, set the customizable variable @code{vc-handled-backends} to @code{nil} @iftex (@pxref{Customizing VC,,,emacs-xtra, Specialized Emacs Features}). @end iftex @ifnottex (@pxref{Customizing VC}). @end ifnottex @findex vc-refresh-state @findex vc-state-refresh To update the VC state information for the file visited in the current buffer, use the command @code{vc-refresh-state}. This command is useful when you perform version control commands outside Emacs (e.g., from the shell prompt), or if you put the buffer's file under a different version control system, or remove it from version control entirely. @cindex VC commands, in Dired buffers @cindex filesets, VC, in Dired buffers VC is also enabled automatically in Dired buffers (@pxref{Dired}) showing directories whose files are controlled by a VCS@. All VC commands described in this section can be invoked from any Dired buffer showing a directory with VC-controlled files; any files that are marked in a Dired buffer (@pxref{Marks vs Flags}) are considered to belong to the current fileset, and VC commands operate on the files in this fileset. This allows you to construct VC filesets including any files you want, regardless of their VC state. (If no files are marked when a VC command is invoked from a Dired buffer, the file shown on the current line in the buffer is considered the only file in the fileset.) @menu * Introduction to VC:: How version control works in general. * VC Mode Line:: How the mode line shows version control status. * Basic VC Editing:: How to edit a file under version control. * Log Buffer:: Features available in log entry buffers. * Registering:: Putting a file under version control. * Old Revisions:: Examining and comparing old versions. * VC Change Log:: Viewing the VC Change Log. * VC Undo:: Canceling changes before or after committing. * VC Ignore:: Ignore files under version control system. * VC Directory Mode:: Listing files managed by version control. * Branches:: Multiple lines of development. @ifnottex * Miscellaneous VC:: Various other commands and features of VC. * Customizing VC:: Variables that change VC's behavior. @end ifnottex @end menu @node Introduction to VC @subsection Introduction to Version Control VC allows you to use a version control system from within Emacs, integrating the version control operations smoothly with editing. It provides a uniform interface for common operations in many version control operations. Some uncommon or intricate version control operations, such as altering repository settings, are not supported in VC@. You should perform such tasks outside VC, e.g., via the command line. This section provides a general overview of version control, and describes the version control systems that VC supports. You can skip this section if you are already familiar with the version control system you want to use. @menu * Why Version Control?:: Understanding the problems it addresses. * Version Control Systems:: Supported version control back-end systems. * VCS Concepts:: Words and concepts related to version control. * VCS Merging:: How file conflicts are handled. * VCS Changesets:: How changes are grouped. * VCS Repositories:: Where version control repositories are stored. * Types of Log File:: The VCS log in contrast to the ChangeLog. @end menu @node Why Version Control? @subsubsection Understanding the Problems it Addresses Version control systems provide you with three important capabilities: @itemize @bullet @item @dfn{Reversibility}: the ability to back up to a previous state if you discover that some modification you did was a mistake or a bad idea. @item @dfn{Concurrency}: the ability to have many people modifying the same collection of files knowing that conflicting modifications can be detected and resolved. @item @dfn{History}: the ability to attach historical data to your data, such as explanatory comments about the intention behind each change. Even for a programmer working solo, change histories are an important aid to memory; for a multi-person project, they are a vitally important form of communication among developers. @end itemize @node Version Control Systems @subsubsection Supported Version Control Systems @cindex back end (version control) VC currently works with many different version control systems, which it refers to as @dfn{back ends}: @itemize @bullet @cindex git @item Git is a decentralized version control system originally invented by Linus Torvalds to support development of Linux (his kernel). VC supports many common Git operations, but others, such as repository syncing, must be done from the command line. @cindex CVS @item CVS is the free version control system that was, until circa 2008, used by the majority of free software projects. Since then, it has been superseded by newer systems. CVS allows concurrent multi-user development either locally or over the network. Unlike newer systems, it lacks support for atomic commits and file moving/renaming. VC supports all basic editing operations under CVS. @cindex SVN @cindex Subversion @item Subversion (svn) is a free version control system designed to be similar to CVS but without its problems (e.g., it supports atomic commits of filesets, and versioning of directories, symbolic links, meta-data, renames, copies, and deletes). @cindex SCCS @item SCCS was the first version control system ever built, and was long ago superseded by more advanced ones. VC compensates for certain features missing in SCCS (e.g., tag names for releases) by implementing them itself. Other VC features, such as multiple branches, are simply unavailable. Since SCCS is non-free, we recommend avoiding it. @cindex CSSC @item CSSC is a free replacement for SCCS@. You should use CSSC only if, for some reason, you cannot use a more recent and better-designed version control system. @cindex RCS @item RCS is the free version control system around which VC was initially built. It is relatively primitive: it cannot be used over the network, and works at the level of individual files. Almost everything you can do with RCS can be done through VC. @cindex hg @cindex Mercurial @item Mercurial (hg) is a decentralized version control system broadly resembling Git. VC supports most Mercurial commands, with the exception of repository sync operations. @cindex bzr @cindex Bazaar @item Bazaar (bzr) is a decentralized version control system that supports both repository-based and decentralized versioning. VC supports most basic editing operations under Bazaar. @cindex SRC @cindex src @item SRC (src) is RCS, reloaded---a specialized version-control system designed for single-file projects worked on by only one person. It allows multiple files with independent version-control histories to exist in one directory, and is thus particularly well suited for maintaining small documents, scripts, and dotfiles. While it uses RCS for revision storage, it presents a modern user interface featuring lockless operation and integer sequential version numbers. VC supports almost all SRC operations. @end itemize @node VCS Concepts @subsubsection Concepts of Version Control @cindex repository @cindex registered file When a file is under version control, we say that it is @dfn{registered} in the version control system. The system has a @dfn{repository} which stores both the file's present state and its change history---enough to reconstruct the current version or any earlier version. The repository also contains other information, such as @dfn{log entries} that describe the changes made to each file. @cindex work file @cindex checking out files The copy of a version-controlled file that you actually edit is called the @dfn{work file}. You can change each work file as you would an ordinary file. After you are done with a set of changes, you may @dfn{commit} (or @dfn{check in}) the changes; this records the changes in the repository, along with a descriptive log entry. @cindex working tree A directory tree of work files is called a @dfn{working tree}. @cindex revision @cindex revision ID Each commit creates a new @dfn{revision} in the repository. The version control system keeps track of all past revisions and the changes that were made in each revision. Each revision is named by a @dfn{revision ID}, whose format depends on the version control system; in the simplest case, it is just an integer. To go beyond these basic concepts, you will need to understand three aspects in which version control systems differ. As explained in the next three sections, they can be lock-based or merge-based; file-based or changeset-based; and centralized or decentralized. VC handles all these modes of operation, but it cannot hide the differences. @node VCS Merging @subsubsection Merge-based vs Lock-based Version Control A version control system typically has some mechanism to coordinate between users who want to change the same file. There are two ways to do this: merging and locking. @cindex merging-based version In a version control system that uses merging, each user may modify a work file at any time. The system lets you @dfn{merge} your work file, which may contain changes that have not been committed, with the latest changes that others have committed. @cindex locking-based version Older version control systems use a @dfn{locking} scheme instead. Here, work files are normally read-only. To edit a file, you ask the version control system to make it writable for you by @dfn{locking} it; only one user can lock a given file at any given time. This procedure is analogous to, but different from, the locking that Emacs uses to detect simultaneous editing of ordinary files (@pxref{Interlocking}). When you commit your changes, that unlocks the file, and the work file becomes read-only again. Other users may then lock the file to make their own changes. Both locking and merging systems can have problems when multiple users try to modify the same file at the same time. Locking systems have @dfn{lock conflicts}; a user may try to check a file out and be unable to because it is locked. In merging systems, @dfn{merge conflicts} happen when you commit a change to a file that conflicts with a change committed by someone else after your checkout. Both kinds of conflict have to be resolved by human judgment and communication. Experience has shown that merging is superior to locking, both in convenience to developers and in minimizing the number and severity of conflicts that actually occur. SCCS always uses locking. RCS is lock-based by default but can be told to operate in a merging style. CVS and Subversion are merge-based by default but can be told to operate in a locking mode. Decentralized version control systems, such as Git and Mercurial, are exclusively merging-based. VC mode supports both locking and merging version control. The terms ``commit'' and ``update'' are used in newer version control systems; older lock-based systems use the terms ``check in'' and ``check out''. VC hides the differences between them as much as possible. @node VCS Changesets @subsubsection Changeset-based vs File-based Version Control @cindex file-based version control On SCCS, RCS, CVS, and other early version control systems (and also in SRC), version control operations are @dfn{file-based}: each file has its own comment and revision history separate from that of all other files. Newer systems, beginning with Subversion, are @dfn{changeset-based}: a commit may include changes to several files, and the entire set of changes is handled as a unit. Any comment associated with the change does not belong to a single file, but to the changeset itself. @cindex changeset-based version control Changeset-based version control is more flexible and powerful than file-based version control; usually, when a change to multiple files has to be reversed, it's good to be able to easily identify and remove all of it. @node VCS Repositories @subsubsection Decentralized vs Centralized Repositories @cindex centralized version control @cindex decentralized version control @cindex distributed version control Early version control systems were designed around a @dfn{centralized} model in which each project has only one repository used by all developers. SCCS, RCS, CVS, Subversion, and SRC share this kind of model. One of its drawbacks is that the repository is a choke point for reliability and efficiency. GNU Arch pioneered the concept of @dfn{distributed} or @dfn{decentralized} version control, later implemented in Git, Mercurial, and Bazaar. A project may have several different repositories, and these systems support a sort of super-merge between repositories that tries to reconcile their change histories. In effect, there is one repository for each developer, and repository merges take the place of commit operations. VC helps you manage the traffic between your personal workfiles and a repository. Whether the repository is a single master, or one of a network of peer repositories, is not something VC has to care about. @node Types of Log File @subsubsection Types of Log File @cindex types of log file @cindex log File, types of @cindex version control log Projects that use a version control system can have two types of log for changes. One is the log maintained by the version control system: each time you commit a change, you fill out a @dfn{log entry} for the change (@pxref{Log Buffer}). This is called the @dfn{version control log}. The other kind of log is the file @file{ChangeLog} (@pxref{Change Log}). It provides a chronological record of all changes to a large portion of a program---typically one directory and its subdirectories. A small program would use one @file{ChangeLog} file; a large program may have a @file{ChangeLog} file in each major directory. @xref{Change Log}. Programmers have used change logs since long before version control systems. Changeset-based version systems typically maintain a changeset-based modification log for the entire system, which makes change log files somewhat redundant. One advantage that they retain is that it is sometimes useful to be able to view the transaction history of a single directory separately from those of other directories. Another advantage is that commit logs can't be fixed in many version control systems. A project maintained with version control can use just the version control log, or it can use both kinds of logs. It can handle some files one way and some files the other way. Each project has its policy, which you should follow. When the policy is to use both, you typically want to write an entry for each change just once, then put it into both logs. You can write the entry in @file{ChangeLog}, then copy it to the log buffer with @kbd{C-c C-a} when committing the change (@pxref{Log Buffer}). Or you can write the entry in the log buffer while committing the change (with the help of @kbd{C-c C-w}), and later use the @kbd{C-x v a} command to copy it to @file{ChangeLog} @iftex (@pxref{Change Logs and VC,,,emacs-xtra, Specialized Emacs Features}). @end iftex @ifnottex (@pxref{Change Logs and VC}). @end ifnottex @node VC Mode Line @subsection Version Control and the Mode Line @cindex VC mode line indicator When you visit a file that is under version control, Emacs indicates this on the mode line. For example, @samp{Bzr-1223} says that Bazaar is used for that file, and the current revision ID is 1223. @cindex version control status The character between the back-end name and the revision ID indicates the @dfn{version control status} of the work file. In a merge-based version control system, a @samp{-} character indicates that the work file is unmodified, and @samp{:} indicates that it has been modified. @samp{!} indicates that the file contains conflicts as result of a recent merge operation (@pxref{Merging}), or that the file was removed from the version control, or that it is versioned but also @dfn{ignored}, something that usually should not happen (@pxref{VC Ignore}). Finally, @samp{?} means that the file is under version control, but is missing from the working tree. In a lock-based system, @samp{-} indicates an unlocked file, and @samp{:} a locked file; if the file is locked by another user (for instance, @samp{jim}), that is displayed as @samp{RCS:jim:1.3}. @samp{@@} means that the file was locally added, but not yet committed to the master repository. On a graphical display, you can move the mouse over this mode line indicator to pop up a tool-tip, which displays a more verbose description of the version control status. Pressing @kbd{mouse-1} over the indicator pops up a menu of VC commands, identical to @samp{Tools / Version Control} on the menu bar. @vindex auto-revert-check-vc-info When Auto Revert mode (@pxref{Reverting}) reverts a buffer that is under version control, it updates the version control information in the mode line. However, Auto Revert mode may not properly update this information if the version control status changes without changes to the work file, from outside the current Emacs session. If you set @code{auto-revert-check-vc-info} to @code{t}, Auto Revert mode updates the version control status information every @code{auto-revert-interval} seconds, even if the work file itself is unchanged. The resulting CPU usage depends on the version control system, but is usually not excessive. @node Basic VC Editing @subsection Basic Editing under Version Control @cindex filesets, VC @cindex VC filesets Most VC commands operate on @dfn{VC filesets}. A VC fileset is a collection of one or more files that a VC operation acts upon. When you type VC commands in a buffer visiting a version-controlled file, the VC fileset is simply that one file. When you type them in a VC Directory buffer, and some files in it are marked, the VC fileset consists of the marked files (@pxref{VC Directory Mode}). Likewise, when you invoke a VC command from a Dired buffer, the VC fileset consists of the marked files (@pxref{Marks vs Flags}), defaulting to the file shown on the current line if no files are marked. With modern changeset-based version control systems (@pxref{VCS Changesets}), such as Git, Mercurial, and Bazaar, VC commands handle multi-file VC filesets as a group. For example, committing a multi-file VC fileset generates a single revision, containing the changes to all those files. On older file-based version control systems like CVS, each file in a multi-file VC fileset is handled individually; thus, committing a fileset generates one revision for each changed file in the fileset. @table @kbd @item C-x v v Perform the next appropriate version control operation on the current VC fileset. @end table @findex vc-next-action @kindex C-x v v The principal VC command is a multi-purpose command, @kbd{C-x v v} (@code{vc-next-action}), which performs the most appropriate action on the current VC fileset: either registering it with a version control system, or committing it, or unlocking it, or merging changes into it. The precise actions for each situation are described in detail in the following subsections. You can use @kbd{C-x v v} either in a file-visiting buffer, in a Dired buffer, or in a VC Directory buffer; in the latter two cases the command operates on the fileset consisting of the marked files. You can also use @kbd{C-x v v}, in a buffer with patches under Diff Mode (@pxref{Diff Mode}), in which case the command operates on the files whose diffs are shown in the buffer. Note that VC filesets are distinct from the named filesets used for viewing and visiting files in functional groups (@pxref{Filesets}). Unlike named filesets, VC filesets are not named and don't persist across sessions. @menu * VC With A Merging VCS:: Without locking: default mode for Git, Hg, SVN, CVS. * VC With A Locking VCS:: RCS in its default mode, SCCS, and optionally CVS. * Advanced C-x v v:: Advanced features available with a prefix argument. @end menu @node VC With A Merging VCS @subsubsection Basic Version Control with Merging With a modern merging-based version control system (such as Git and Hg; @pxref{VCS Merging}), @kbd{C-x v v} does the following when invoked from a buffer that visits a version-controlled file or a VC Directory or Dired buffer: @itemize @bullet @item If there is more than one file in the VC fileset and the files have inconsistent version control statuses, signal an error. (Note, however, that a fileset is allowed to include both newly-added files and modified files; @pxref{Registering}.) Also signal an error if the files in the fileset are missing (removed from the filesystem, but still tracked by version control), or are ignored by version control. @item If every file in the VC fileset is registered and unchanged with respect to the last revision, do nothing. @item If none of the files in the VC fileset are registered with a version control system, register the newly-added files in the VC fileset, i.e., place them under version control. @xref{Registering}. If Emacs cannot find a system to register under, it prompts for a repository type, creates a new repository, and registers the VC fileset with it. You can also specify the system explicitly, see @ref{Advanced C-x v v}. Note that registering the files doesn't commit them; you must invoke @w{@kbd{C-x v v}} again to commit, see below. @item If every file in the VC fileset has been either newly-added or modified, commit the changed files. To do this, Emacs pops up a @file{*vc-log*} buffer; type the desired log entry for the changes, followed by @kbd{C-c C-c} to commit. @xref{Log Buffer}. With modern decentralized version control systems (Git, Mercurial, etc.), the changes are committed locally and not automatically propagated to the upstream repository (which is usually on a remote host). In these cases, if the repository has been changed since your last update, the commit may fail. In that case, you must update from upstream and then try again. Use @kbd{C-x v +} (@pxref{Pulling / Pushing}) or @kbd{C-x v m} (@pxref{Merging}) for that. With a centralized version control system, if the commit fails due to upstream changes, type @kbd{C-x v v} again to merge in the upstream repository changes. @item Finally, if you are using a centralized version control system, if any file in the VC fileset is outdated with respect to the upstream repository, offer to update the fileset from the repository. @end itemize These rules also apply when you use RCS in its non-locking mode, except that changes are not automatically merged from the repository. Nothing informs you if another user has committed changes in the same file since you began editing it; when you commit your revision, that other user's changes are removed (however, they remain in the repository and are thus not irrevocably lost). Therefore, you must verify that the current revision is unchanged before committing your changes. In addition, locking is possible with RCS even in this mode: @kbd{C-x v v} with an unmodified file locks the file, just as it does with RCS in its normal locking mode (@pxref{VC With A Locking VCS}). If @kbd{C-x v v} is invoked from a buffer under Diff Mode, the command assumes the buffer holds a set of patches for one or more files. It then applies the changes to the respective files and commits the changes after popping up the @file{*vc-log*} buffer to allow you to type a suitable commit log message. @node VC With A Locking VCS @subsubsection Basic Version Control with Locking With a locking-based version control system (such as SCCS, and RCS in its default mode), @kbd{C-x v v} does the following: @itemize @bullet @item If there is more than one file in the VC fileset and the files have inconsistent version control statuses, signal an error. Also signal an error if the files in the fileset are missing (removed from the filesystem, but still tracked by version control). @item If each file in the VC fileset is not registered with a version control system, register the newly-added files in the fileset. @xref{Registering}. If Emacs cannot find a system to register under, it prompts for a repository type, creates a new repository, and registers the VC fileset with it. You can also specify the system explicitly, see @ref{Advanced C-x v v}. @item If each file is registered and unlocked, check the files out: lock each one and make it writable, so that you can begin to edit it. @item If each file is locked by you and contains changes, commit (a.k.a.@: ``check-in'') the changes. To do this, Emacs pops up a @file{*vc-log*} buffer; type the desired log entry for the new revision, followed by @kbd{C-c C-c} to commit (@pxref{Log Buffer}). @item If each file is locked by you, but you have not changed it, release the lock and make the file read-only again. This undoes previous check-out operation for files that were not changed since the checkout. @item If each file is locked by another user, ask whether you want to steal the lock. If you say yes, the file becomes locked by you, and a warning message is sent to the user who had formerly locked the file. @item If files in the fileset are unlocked, but have changes with respect to their last revision, offer to claim the lock for each such file or to revert the file to the last checked-in revision. (This situation is exceptional and should not normally happen.) @end itemize These rules also apply when you use CVS in locking mode, except that CVS does not support stealing locks. @node Advanced C-x v v @subsubsection Advanced Control in @kbd{C-x v v} @cindex revision ID in version control When you give a prefix argument to @code{vc-next-action} (@kbd{C-u C-x v v}), it still performs the next logical version control operation, but accepts additional arguments to specify precisely how to do the operation. @itemize @bullet @item @cindex specific version control system You can specify the name of a version control system. This is useful if the fileset can be managed by more than one version control system, and Emacs fails to detect the correct one. @item Otherwise, if using CVS, RCS or SRC, you can specify a revision ID. If the fileset is modified (or locked), this makes Emacs commit the files with that revision ID@. You can create a new branch by supplying an appropriate revision ID (@pxref{Branches}). If the fileset is unmodified (and unlocked), this checks out the specified revision into the working tree. You can also specify a revision on another branch by giving its revision or branch ID (@pxref{Switching Branches}). An empty argument (i.e., @kbd{C-u C-x v v @key{RET}}) checks out the latest (a.k.a.@: ``head'') revision on the current branch. Specifying revision ID in this manner is silently ignored by a decentralized version control system. Those systems do not let you specify your own revision IDs, nor do they use the concept of checking out individual files. @end itemize @node Log Buffer @subsection Features of the Log Entry Buffer @kindex C-c C-c @r{(Log Edit mode)} @findex log-edit-done When you tell VC to commit a change, it pops up a buffer named @file{*vc-log*}. In this buffer, you should write a @dfn{log entry} describing the changes you have made (@pxref{Why Version Control?}). After you are done, type @kbd{C-c C-c} (@code{log-edit-done}) to exit the buffer and commit the change, together with your log entry. @cindex Log Edit mode @cindex mode, Log Edit @vindex vc-log-mode-hook @c FIXME: Mention log-edit-mode-hook here? --xfq The major mode for the @file{*vc-log*} buffer is Log Edit mode, a variant of Text mode (@pxref{Text Mode}). On entering Log Edit mode, Emacs runs the hooks @code{text-mode-hook} and @code{vc-log-mode-hook} (@pxref{Hooks}). In the @file{*vc-log*} buffer, you can write one or more @dfn{header lines}, specifying additional information to be supplied to the version control system. Each header line must occupy a single line at the top of the buffer; the first line that is not a header line is treated as the start of the log entry. For example, the following header line states that the present change was not written by you, but by another developer: @smallexample Author: J. R. Hacker @end smallexample @noindent Apart from the @samp{Author} header, Emacs recognizes the headers @samp{Summary} (a one-line summary of the changeset), @samp{Date} (a manually-specified commit time), and @samp{Fixes} (a reference to a bug fixed by the change). Not all version control systems recognize all headers. If you specify a header for a system that does not support it, the header is treated as part of the log entry. @kindex C-c C-f @r{(Log Edit mode)} @findex log-edit-show-files @kindex C-c C-d @r{(Log Edit mode)} @findex log-edit-show-diff While in the @file{*vc-log*} buffer, the current VC fileset is considered to be the fileset that will be committed if you type @w{@kbd{C-c C-c}}. To view a list of the files in the VC fileset, type @w{@kbd{C-c C-f}} (@code{log-edit-show-files}). To view a diff of changes between the VC fileset and the version from which you started editing (@pxref{Old Revisions}), type @kbd{C-c C-d} (@code{log-edit-show-diff}). @kindex C-c C-w @r{(Log Edit mode)} @findex log-edit-generate-changelog-from-diff @vindex diff-add-log-use-relative-names To help generate ChangeLog entries, type @kbd{C-c C-w} (@code{log-edit-generate-changelog-from-diff}), to generate skeleton ChangeLog entries, listing all changed file and function names based on the diff of the VC fileset. Consecutive entries left empty will be combined by @kbd{M-q} (@code{fill-paragraph}). By default the skeleton will just include the file name, without any leading directories. If you wish to prepend the leading directories up to the VC root, customize @code{diff-add-log-use-relative-names}. @kindex C-c C-a @r{(Log Edit mode)} @findex log-edit-insert-changelog If the VC fileset includes one or more @file{ChangeLog} files (@pxref{Change Log}), type @kbd{C-c C-a} (@code{log-edit-insert-changelog}) to pull the relevant entries into the @file{*vc-log*} buffer. If the topmost item in each @file{ChangeLog} was made under your user name on the current date, this command searches that item for entries matching the file(s) to be committed, and inserts them. @ifnottex If you are using CVS or RCS, see @ref{Change Logs and VC}, for the opposite way of working---generating ChangeLog entries from the Log Edit buffer. @end ifnottex To abort a commit, just @emph{don't} type @kbd{C-c C-c} in that buffer. You can switch buffers and do other editing. As long as you don't try to make another commit, the entry you were editing remains in the @file{*vc-log*} buffer, and you can go back to that buffer at any time to complete the commit. @kindex M-n @r{(Log Edit mode)} @kindex M-p @r{(Log Edit mode)} @kindex M-s @r{(Log Edit mode)} @kindex M-r @r{(Log Edit mode)} You can also browse the history of previous log entries to duplicate a commit comment. This can be useful when you want to make several commits with similar comments. The commands @kbd{M-n}, @kbd{M-p}, @kbd{M-s} and @kbd{M-r} for doing this work just like the minibuffer history commands (@pxref{Minibuffer History}), except that they are used outside the minibuffer. @node Registering @subsection Registering a File for Version Control @table @kbd @item C-x v i Register the visited file for version control. @end table @kindex C-x v i @findex vc-register The command @kbd{C-x v i} (@code{vc-register}) @dfn{registers} each file in the current VC fileset, placing it under version control. This is essentially equivalent to the action of @kbd{C-x v v} on an unregistered VC fileset (@pxref{Basic VC Editing}), except that if the VC fileset is already registered, @kbd{C-x v i} signals an error whereas @kbd{C-x v v} performs some other action. To register a file, Emacs must choose a version control system. For a multi-file VC fileset, the VC Directory buffer specifies the system to use (@pxref{VC Directory Mode}). For a single-file VC fileset, if the file's directory already contains files registered in a version control system, or if the directory is part of a directory tree controlled by a version control system, Emacs chooses that system. In the event that more than one version control system is applicable, Emacs uses the one that appears first in the variable @iftex @code{vc-handled-backends}. @end iftex @ifnottex @code{vc-handled-backends} (@pxref{Customizing VC}). @end ifnottex If Emacs cannot find a version control system to register the file under, it prompts for a repository type, creates a new repository, and registers the file into that repository. @cindex added files, VC @cindex files added to VCS With most version control systems, registering a file with @w{@kbd{C-x v i}} or @w{@kbd{C-x v v}} adds it to the working tree, but does not commit it, i.e., doesn't add it to the repository. Such files are labeled as @dfn{added} in the VC Directory buffer, and the mode line of the buffers visiting such files shows a revision ID of @samp{@@@@}. To make the registration take effect in the repository, you must commit the newly-added files (@pxref{Basic VC Editing}). Note that a single commit can include both file additions and edits to files already known to the VCS. With a locking-based version control system (@pxref{VCS Merging}), registering a file leaves it unlocked and read-only. Type @kbd{C-x v v} to check-out the file and start editing it. @node Old Revisions @subsection Examining And Comparing Old Revisions @table @kbd @item C-x v = Compare the work files in the current VC fileset with the versions you started from (@code{vc-diff}). With a prefix argument, prompt for two revisions of the current VC fileset and compare them. You can also call this command from a Dired buffer (@pxref{Dired}). @ifnottex @item M-x vc-ediff Like @kbd{C-x v =}, but using Ediff. @xref{Top,, Ediff, ediff, The Ediff Manual}. @end ifnottex @item C-x v D Compare the entire working tree to the revision you started from (@code{vc-root-diff}). With a prefix argument, prompt for two revisions and compare their trees. @item C-x v ~ Prompt for a revision of the current file, and visit it in a separate buffer (@code{vc-revision-other-window}). @item C-x v g Display an annotated version of the current file: for each line, show the latest revision in which it was modified (@code{vc-annotate}). @end table @findex vc-diff @kindex C-x v = @kbd{C-x v =} (@code{vc-diff}) displays a @dfn{diff} which compares each work file in the current VC fileset to the version(s) from which you started editing. The diff is displayed in another window, in a Diff mode buffer (@pxref{Diff Mode}) named @file{*vc-diff*}. The usual Diff mode commands are available in this buffer. In particular, the @kbd{g} (@code{revert-buffer}) command performs the file comparison again, generating a new diff. @kindex C-u C-x v = To compare two arbitrary revisions of the current VC fileset, call @code{vc-diff} with a prefix argument: @kbd{C-u C-x v =}. This prompts for two revision IDs (@pxref{VCS Concepts}), and displays a diff between those versions of the fileset. This will not work reliably for multi-file VC filesets, if the version control system is file-based rather than changeset-based (e.g., CVS), since then revision IDs for different files would not be related in any meaningful way. Instead of the revision ID, some version control systems let you specify revisions in other formats. For instance, under Bazaar you can enter @samp{date:yesterday} for the argument to @kbd{C-u C-x v =} (and related commands) to specify the first revision committed after yesterday. See the documentation of the version control system for details. If you invoke @kbd{C-x v =} or @kbd{C-u C-x v =} from a Dired buffer (@pxref{Dired}), the file listed on the current line is treated as the current VC fileset. The VC fileset can also include several marked files. @ifnottex @findex vc-ediff @kbd{M-x vc-ediff} works like @kbd{C-x v =}, except that it uses an Ediff session. @xref{Top,, Ediff, ediff, The Ediff Manual}. @end ifnottex @findex vc-root-diff @kindex C-x v D @kbd{C-x v D} (@code{vc-root-diff}) is similar to @kbd{C-x v =}, but it displays the changes in the entire current working tree (i.e., the working tree containing the current VC fileset). If you invoke this command from a Dired buffer, it applies to the working tree containing the directory. @findex vc-root-version-diff @kindex C-u C-x v D To compare two arbitrary revisions of the whole trees, call @code{vc-root-diff} with a prefix argument: @kbd{C-u C-x v D}. This prompts for two revision IDs (@pxref{VCS Concepts}), and displays a diff between those versions of the entire version-controlled directory trees (RCS, SCCS, CVS, and SRC do not support this feature). @vindex vc-diff-switches You can customize the @command{diff} options that @kbd{C-x v =} and @kbd{C-x v D} use for generating diffs. The options used are taken from the first non-@code{nil} value amongst the variables @code{vc-@var{backend}-diff-switches}, @code{vc-diff-switches}, and @code{diff-switches} (@pxref{Comparing Files}), in that order. Here, @var{backend} stands for the relevant version control system, e.g., @code{bzr} for Bazaar. Since @code{nil} means to check the next variable in the sequence, either of the first two may use the value @code{t} to mean no switches at all. Most of the @code{vc-@var{backend}-diff-switches} variables default to @code{nil}, but some default to @code{t}; these are for version control systems whose @code{diff} implementations do not accept common diff options, such as Subversion. @findex vc-revision-other-window @kindex C-x v ~ To directly examine an older version of a file, visit the work file and type @kbd{C-x v ~ @var{revision} @key{RET}} (@code{vc-revision-other-window}). This retrieves the file version corresponding to @var{revision}, saves it to @file{@var{filename}.~@var{revision}~}, and visits it in a separate window. @findex vc-annotate @vindex vc-annotate-background-mode @kindex C-x v g Many version control systems allow you to view files @dfn{annotated} with per-line revision information, by typing @kbd{C-x v g} (@code{vc-annotate}). This creates a new ``annotate'' buffer displaying the file's text, with each line colored to show how old it is. Red text is new, blue is old, and intermediate colors indicate intermediate ages. By default, the color is scaled over the full range of ages, such that the oldest changes are blue, and the newest changes are red. If the variable @code{vc-annotate-background-mode} is non-@code{nil}, the colors expressing the age of each line are applied to the background color, leaving the foreground at its default color. @vindex vc-annotate-switches You can customize the @code{annotate} options that @kbd{C-x v g} uses by customizing @code{vc-@var{backend}-annotate-switches} and @code{vc-annotate-switches}. They function similarly to @code{vc-@var{backend}-diff-switches} and @code{vc-diff-switches}, described above. When you give a prefix argument to @kbd{C-x v g}, Emacs reads two arguments using the minibuffer: the revision to display and annotate (instead of the current file contents), and the time span in days the color range should cover. From the ``annotate'' buffer, these and other color scaling options are available from the @samp{VC-Annotate} menu. In this buffer, you can also use the following keys to browse the annotations of past revisions, view diffs, or view log entries: @table @kbd @item p Annotate the previous revision, i.e., the revision before the one currently annotated. A numeric prefix argument is a repeat count, so @kbd{C-u 10 p} would take you back 10 revisions. @item n Annotate the next revision, i.e., the revision after the one currently annotated. A numeric prefix argument is a repeat count. @item j Annotate the revision indicated by the current line. @item a Annotate the revision before the one indicated by the current line. This is useful to see the state the file was in before the change on the current line was made. @item f Show in a buffer the file revision indicated by the current line. @item d Display the diff between the current line's revision and the previous revision. This is useful to see what the current line's revision actually changed in the file. @item D Display the diff between the current line's revision and the previous revision for all files in the changeset (for VC systems that support changesets). This is useful to see what the current line's revision actually changed in the tree. @item l Show the log of the current line's revision. This is useful to see the author's description of the changes in the revision on the current line. @item w Annotate the working revision---the one you are editing. If you used @kbd{p} and @kbd{n} to browse to other revisions, use this key to return to your working revision. @item v Toggle the annotation visibility. This is useful for looking just at the file contents without distraction from the annotations. @end table @node VC Change Log @subsection VC Change Log @cindex VC change log @table @kbd @item C-x v l Display the change history for the current fileset (@code{vc-print-log}). @item C-x v L Display the change history for the current repository (@code{vc-print-root-log}). @item C-u 1 C-x v L @var{revision} @key{RET} Display the log entry and changes (diffs) of a single @var{revision} (@code{vc-print-root-log}). @item C-x v b l Display the change history for another branch (@code{vc-print-branch-log}). @item C-x v I Display the changes that a ``pull'' operation will retrieve (@code{vc-log-incoming}). @item C-x v O Display the changes that will be sent by the next ``push'' operation (@code{vc-log-outgoing}). @item C-x v h Display the history of changes made in the region of file visited by the current buffer (@code{vc-region-history}). @item M-x vc-log-search @key{RET} Search the change history for a specified pattern. @end table @kindex C-x v l @findex vc-print-log @kbd{C-x v l} (@code{vc-print-log}) displays a buffer named @file{*vc-change-log*}, showing the history of changes made to the current fileset in the long form, including who made the changes, the dates, and the log entry for each change (these are the same log entries you would enter via the @file{*vc-log*} buffer; @pxref{Log Buffer}). If invoked from a buffer visiting a file, the current fileset consists of that single file, and point in the displayed @file{*vc-change-log*} buffer is centered at the revision of that file. If invoked from a VC Directory buffer (@pxref{VC Directory Mode}) or from a Dired buffer (@pxref{Dired}), the fileset consists of all the marked files, defaulting to the file shown on the current line in the directory buffer if no file is marked. If the fileset includes one or more directories, the resulting @file{*vc-change-log*} buffer shows a short log of changes (one line for each change), if the VC backend supports that; otherwise it shows the log in the long form. With a prefix argument, the command prompts for the revision to center on in the @file{*vc-change-log*} buffer and for the maximum number of revisions to display. @kindex C-x v L @findex vc-print-root-log @findex log-view-toggle-entry-display @kbd{C-x v L} (@code{vc-print-root-log}) displays a @file{*vc-change-log*} buffer showing the history of the entire version-controlled directory tree (RCS, SCCS, CVS, and SRC do not support this feature). With a prefix argument, the command prompts for the maximum number of revisions to display. A numeric prefix argument specifies the maximum number of revisions without prompting. If the numeric prefix argument is 1, as in @w{@kbd{C-1 C-x v L}} or @w{@kbd{C-u 1 C-x v L}}, the command prompts for the revision ID, and displays the log entry of that revision together with the changes (diffs) it introduced. (Some less capable version control systems, such as RCS and CVS, don't have commands to show a revision log with its diffs; for them the command displays only the log entry, and you can request to show the diffs by typing @kbd{d} or @kbd{D}, see below.) The @kbd{C-x v L} history is shown in a compact form, usually showing only the first line of each log entry. However, you can type @key{RET} (@code{log-view-toggle-entry-display}) in the @file{*vc-change-log*} buffer to reveal the entire log entry for the revision at point. A second @key{RET} hides it again. @kindex C-x v b l @findex vc-print-branch-log @kbd{C-x v b l @var{branch-name} @key{RET}} (@code{vc-print-branch-log}) displays a @file{*vc-change-log*} buffer showing the history of the version-controlled directory tree, like @code{vc-print-root-log} does, but it shows the history of a branch other than the current one; it prompts for the branch whose history to display. @kindex C-x v I @kindex C-x v O @findex vc-log-incoming @findex vc-log-outgoing On a decentralized version control system, the @kbd{C-x v I} (@code{vc-log-incoming}) command displays a log buffer showing the changes that will be applied, the next time you run the version control system's pull command to get new revisions from another remote location (@pxref{Pulling / Pushing}). This other remote location is the default one from which changes are pulled, as defined by the version control system; with a prefix argument, @code{vc-log-incoming} prompts for a specific remote location. Similarly, @kbd{C-x v O} (@code{vc-log-outgoing}) shows the changes that will be sent to another remote location, the next time you run the push command; with a prefix argument, it prompts for a specific destination that in case of some version control system can be a branch name. @cindex VC log buffer, commands in @cindex vc-log buffer In the @file{*vc-change-log*} buffer, you can use the following keys to move between the logs of revisions and of files, and to examine and compare past revisions (@pxref{Old Revisions}): @table @kbd @item p Move to the previous revision entry. (Revision entries in the log buffer are usually in reverse-chronological order, so the previous revision-item usually corresponds to a newer revision.) A numeric prefix argument is a repeat count. @item n Move to the next revision entry. A numeric prefix argument is a repeat count. @item a Annotate the revision on the current line (@pxref{Old Revisions}). @item e Modify the change comment displayed at point. Note that not all VC systems support modifying change comments. @item f Visit the revision indicated at the current line. @item d Display a diff between the revision at point and the next earlier revision, for the specific file. @item D Display the changeset diff between the revision at point and the next earlier revision. This shows the changes to all files made in that revision. @item @key{RET} In a compact-style log buffer (e.g., the one created by @kbd{C-x v L}), toggle between showing and hiding the full log entry for the revision at point. @end table @vindex vc-log-show-limit Because fetching many log entries can be slow, the @file{*vc-change-log*} buffer displays no more than 2000 revisions by default. The variable @code{vc-log-show-limit} specifies this limit; if you set the value to zero, that removes the limit. You can also increase the number of revisions shown in an existing @file{*vc-change-log*} buffer by clicking on the @samp{Show 2X entries} or @samp{Show unlimited entries} buttons at the end of the buffer. However, RCS, SCCS, CVS, and SRC do not support this feature. @kindex C-x v h @findex vc-region-history A useful variant of examining history of changes is provided by the command @kbd{vc-region-history} (by default bound to @kbd{C-x v h}), which shows a @file{*VC-history*} buffer with the history of changes made in the region of the current buffer's file between point and the mark (@pxref{Mark}). The history of changes includes the commit log messages and also the changes themselves in the Diff format. Invoke this command after marking in the current buffer the region in whose changes you are interested. In the @file{*VC-history*} buffer it pops up, you can use all of the commands available in the @file{*vc-change-log*} buffer described above, and also the commands defined by Diff mode (@pxref{Diff Mode}). This command is currently available only with Git and Mercurial (hg). @findex vc-log-search The command @code{vc-log-search} allows searching for a pattern in the log of changes. It prompts for a pattern (a regular expression), and displays all entries in the change history whose log messages match the pattern. When invoked with a prefix argument, the command will also prompt for a specific VCS shell command to run for this purpose. @node VC Undo @subsection Undoing Version Control Actions @table @kbd @item C-x v u Revert the work file(s) in the current VC fileset to the last revision (@code{vc-revert}). @end table @kindex C-x v u @findex vc-revert @vindex vc-revert-show-diff If you want to discard all the changes you have made to the current VC fileset, type @kbd{C-x v u} (@code{vc-revert}). This will ask you for confirmation before discarding the changes. If you agree, the fileset is reverted. If @code{vc-revert-show-diff} is non-@code{nil}, this command will show you a diff between the work file(s) and the revision from which you started editing. Afterwards, the diff buffer will either be killed (if this variable is @code{kill}), or the buffer will be buried (any other non-@code{nil} value). If you don't want @kbd{C-x v u} to show a diff, set this variable to @code{nil} (you can still view the diff directly with @kbd{C-x v =}; @pxref{Old Revisions}). On locking-based version control systems, @kbd{C-x v u} leaves files unlocked; you must lock again to resume editing. You can also use @kbd{C-x v u} to unlock a file if you lock it and then decide not to change it. @node VC Ignore @subsection Ignore Version Control Files @table @kbd @item C-x v G Ignore a file under current version control system. (@code{vc-ignore}). @end table @kindex C-x v G @findex vc-ignore Many source trees contain some files that do not need to be versioned, such as editor backups, object or bytecode files, and built programs. You can simply not add them, but then they'll always crop up as unknown files. You can also tell the version control system to ignore these files by adding them to the ignore file at the top of the tree. @kbd{C-x v G} (@code{vc-ignore}) can help you do this. When called with a prefix argument, you can remove a file from the ignored file list. @node VC Directory Mode @subsection VC Directory Mode @cindex VC Directory buffer The @dfn{VC Directory buffer} is a specialized buffer for viewing the version control statuses of the files in a directory tree, and performing version control operations on those files. In particular, it is used to specify multi-file VC filesets for commands like @w{@kbd{C-x v v}} to act on (@pxref{VC Directory Commands}). @kindex C-x v d @findex vc-dir To use the VC Directory buffer, type @kbd{C-x v d} (@code{vc-dir}). This reads a directory's name using the minibuffer, and switches to a VC Directory buffer for that directory. By default, the buffer is named @file{*vc-dir*}. Its contents are described @iftex below. @end iftex @ifnottex in @ref{VC Directory Buffer}. @end ifnottex The @code{vc-dir} command automatically detects the version control system to be used in the specified directory. In the event that more than one system is being used in the directory, you should invoke the command with a prefix argument, @kbd{C-u C-x v d}; this prompts for the version control system which the VC Directory buffer should use. @ifnottex @cindex PCL-CVS @pindex cvs @cindex CVS directory mode In addition to the VC Directory buffer, Emacs has a similar facility called PCL-CVS which is specialized for CVS@. @xref{Top, , About PCL-CVS, pcl-cvs, PCL-CVS---The Emacs Front-End to CVS}. @end ifnottex You can also invoke VC commands from Dired buffers (@pxref{Dired}). In that case, any VC command you invoke considers the marked files as the current fileset (@pxref{Basic VC Editing}), defaulting to the file on the current line if no files are marked. @menu * Buffer: VC Directory Buffer. What the buffer looks like and means. * Commands: VC Directory Commands. Commands to use in a VC directory buffer. @end menu @node VC Directory Buffer @subsubsection The VC Directory Buffer The VC Directory buffer contains a list of version-controlled files and their version control statuses. It lists files in the current directory (the one specified when you called @kbd{C-x v d}) and its subdirectories, but only those with a noteworthy status. Files that are up-to-date (i.e., the same as in the repository) are omitted. If all the files in a subdirectory are up-to-date, the subdirectory is not listed either. As an exception, if a file has become up-to-date as a direct result of a VC command, it is listed. Here is an example of a VC Directory buffer listing: @smallexample @group ./ edited configure.ac * added README unregistered temp.txt src/ * edited src/main.c @end group @end smallexample @noindent Two work files have been modified but not committed: @file{configure.ac} in the current directory, and @file{main.c} in the @file{src/} subdirectory. The file named @file{README} has been added but is not yet committed, while @file{temp.txt} is not under version control (@pxref{Registering}). The @samp{*} characters next to the entries for @file{README} and @file{src/main.c} indicate that the user has marked these files as the current VC fileset @iftex (see below). @end iftex @ifnottex (@pxref{VC Directory Commands}). @end ifnottex The above example is typical for a decentralized version control system like Bazaar, Git, or Mercurial. Other systems can show other statuses. For instance, CVS shows the @samp{needs-update} status if the repository has changes that have not been applied to the work file. RCS and SCCS show the name of the user locking a file as its status. @ifnottex On CVS, the @code{vc-dir} command normally contacts the repository, which may be on a remote machine, to check for updates. If you change the variable @code{vc-cvs-stay-local} to @code{nil} (@pxref{CVS Options}), then Emacs avoids contacting a remote repository when generating the VC Directory buffer (it will still contact it when necessary, e.g., when doing a commit). This may be desirable if you are working offline or the network is slow. @end ifnottex @vindex vc-directory-exclusion-list The VC Directory buffer omits subdirectories listed in the variable @code{vc-directory-exclusion-list}. Its default value contains directories that are used internally by version control systems. @node VC Directory Commands @subsubsection VC Directory Commands Emacs provides several commands for navigating the VC Directory buffer, and for marking files as belonging to the current VC fileset. @table @kbd @item n @itemx @key{SPC} Move point to the next entry (@code{vc-dir-next-line}). @item p Move point to the previous entry (@code{vc-dir-previous-line}). @item @key{TAB} Move to the next directory entry (@code{vc-dir-next-directory}). @item S-@key{TAB} Move to the previous directory entry (@code{vc-dir-previous-directory}). @item @key{RET} @itemx f Visit the file or directory listed on the current line (@code{vc-dir-find-file}). @item o Visit the file or directory on the current line, in a separate window (@code{vc-dir-find-file-other-window}). @item m Mark the file or directory on the current line (@code{vc-dir-mark}), putting it in the current VC fileset. If the region is active, mark all files in the region. A file cannot be marked with this command if it is already in a marked directory, or one of its subdirectories. Similarly, a directory cannot be marked with this command if any file in its tree is marked. @item M If point is on a file entry, mark all files with the same status; if point is on a directory entry, mark all files in that directory tree (@code{vc-dir-mark-all-files}). With a prefix argument, mark all listed files and directories. @findex vc-dir-mark-by-regexp @item % m @itemx * % You can use this command to mark files by regexp (@code{vc-dir-mark-by-regexp}). If given a prefix, unmark files instead. @findex vc-dir-mark-registered-files @item * r You can use this command to mark files that are in one of registered states, including edited, added or removed. (@code{vc-dir-mark-registered-files}). @item G Add the file under point to the list of files that the VC should ignore (@code{vc-dir-ignore}). For instance, if the VC is Git, it will append this file to the @file{.gitignore} file. If given a prefix, do this with all the marked files. @item q Quit the VC Directory buffer, and bury it (@code{quit-window}). @item u Unmark the file or directory on the current line. If the region is active, unmark all the files in the region (@code{vc-dir-unmark}). @item U If point is on a file entry, unmark all files with the same status; if point is on a directory entry, unmark all files in that directory tree (@code{vc-dir-unmark-all-files}). With a prefix argument, unmark all files and directories. @item x Hide files with @samp{up-to-date} or @samp{ignored} status (@code{vc-dir-hide-up-to-date}). With a prefix argument, hide items whose state is that of the item at point. @end table @findex vc-dir-mark @findex vc-dir-mark-all-files While in the VC Directory buffer, all the files that you mark with @kbd{m} (@code{vc-dir-mark}) or @kbd{M} (@code{vc-dir-mark-all-files}) are in the current VC fileset. If you mark a directory entry with @kbd{m}, all the listed files in that directory tree are in the current VC fileset. The files and directories that belong to the current VC fileset are indicated with a @samp{*} character in the VC Directory buffer, next to their VC status. In this way, you can set up a multi-file VC fileset to be acted on by VC commands like @w{@kbd{C-x v v}} (@pxref{Basic VC Editing}), @w{@kbd{C-x v =}} (@pxref{Old Revisions}), and @w{@kbd{C-x v u}} (@pxref{VC Undo}). The VC Directory buffer also defines some single-key shortcuts for VC commands with the @kbd{C-x v} prefix: @kbd{=}, @kbd{+}, @kbd{l}, @kbd{i}, @kbd{D}, @kbd{L}, @kbd{G}, @kbd{I}, @kbd{O}, and @kbd{v}. For example, you can commit a set of edited files by opening a VC Directory buffer, where the files are listed with the @samp{edited} status; marking the files; and typing @kbd{v} or @kbd{C-x v v} (@code{vc-next-action}). If the version control system is changeset-based, Emacs will commit the files in a single revision. While in the VC Directory buffer, you can also perform search and replace on the current VC fileset, with the following commands: @table @kbd @item S Search the fileset (@code{vc-dir-search}). @item Q Do a regular expression query replace on the fileset (@code{vc-dir-query-replace-regexp}). @item M-s a C-s Do an incremental search on the fileset (@code{vc-dir-isearch}). @item M-s a C-M-s Do an incremental regular expression search on the fileset (@code{vc-dir-isearch-regexp}). @end table @noindent Apart from acting on multiple files, these commands behave much like their single-buffer counterparts (@pxref{Search}). The VC Directory buffer additionally defines some branch-related commands starting with the prefix @kbd{b}: @table @kbd @item b c Create a new branch (@code{vc-create-branch}). @xref{Creating Branches}. @item b l Prompt for the name of a branch and display the change history of that branch (@code{vc-print-branch-log}). @item b s Switch to a branch (@code{vc-switch-branch}). @xref{Switching Branches}. @item d Delete the marked files, or the current file if no marks (@code{vc-dir-clean-delete)}. The files will not be marked as deleted in the version control system, so this function is mostly useful for unregistered files. @end table @cindex stashes in version control @cindex shelves in version control The above commands are also available via the menu bar, and via a context menu invoked by @kbd{mouse-2}. Furthermore, some VC backends use the menu to provide extra backend-specific commands. For example, Git and Bazaar allow you to manipulate @dfn{stashes} and @dfn{shelves} (which are a way to temporarily put aside uncommitted changes, and bring them back at a later time). @node Branches @subsection Version Control Branches @cindex branch (version control) One use of version control is to support multiple independent lines of development, which are called @dfn{branches}. Amongst other things, branches can be used for maintaining separate stable and development versions of a program, and for developing unrelated features in isolation from one another. VC's support for branch operations is currently fairly limited. For decentralized version control systems, it provides commands for @dfn{updating} one branch with the contents of another, and for @dfn{merging} the changes made to two different branches (@pxref{Merging}). For centralized version control systems, it supports checking out different branches and committing into new or different branches. @menu * Switching Branches:: How to get to another existing branch. * Pulling / Pushing:: Receiving/sending changes from/to elsewhere. * Merging:: Transferring changes between branches. * Creating Branches:: How to start a new branch. @end menu @node Switching Branches @subsubsection Switching between Branches The various version control systems differ in how branches are implemented, and these differences cannot be entirely concealed by VC. On some decentralized version control systems, including Bazaar and Mercurial in its normal mode of operation, each branch has its own working directory tree, so switching between branches just involves switching directories. On Git, branches are normally @dfn{co-located} in the same directory, and switching between branches is done using the @kbd{git checkout} command, which changes the contents of the working tree to match the branch you switch to. Bazaar also supports co-located branches, in which case the @kbd{bzr switch} command will switch branches in the current directory. With Subversion, you switch to another branch using the @kbd{svn switch} command. With Mercurial, command @kbd{hg update} is used to switch to another branch. @kindex C-x v b s @findex vc-switch-branch The VC command to switch to another branch in the current directory is @kbd{C-x v b s @var{branch-name} @key{RET}} (@code{vc-switch-branch}). On centralized version control systems, you can also switch between branches by typing @kbd{C-u C-x v v} in an up-to-date work file (@pxref{Advanced C-x v v}), and entering the revision ID for a revision on another branch. On CVS, for instance, revisions on the @dfn{trunk} (the main line of development) normally have IDs of the form 1.1, 1.2, 1.3, @dots{}, while the first branch created from (say) revision 1.2 has revision IDs 1.2.1.1, 1.2.1.2, @dots{}, the second branch created from revision 1.2 has revision IDs 1.2.2.1, 1.2.2.2, @dots{}, and so forth. You can also specify the @dfn{branch ID}, which is a branch revision ID omitting its final component (e.g., 1.2.1), to switch to the latest revision on that branch. On a locking-based system, switching to a different branch also unlocks (write-protects) the working tree. Once you have switched to a branch, VC commands will apply to that branch until you switch away; for instance, any VC filesets that you commit will be committed to that specific branch. @node Pulling / Pushing @subsubsection Pulling/Pushing Changes into/from a Branch @table @kbd @cindex push changes to upstream (VC) @item C-x v P With a decentralized version control system, update another repository with locally-committed changes from the current branch (a.k.a.@: @dfn{push} changes). This concept does not exist for centralized version control systems @cindex pull changes from upstream (VC) @item C-x v + With a decentralized version control system, update the current branch of the local repository by @dfn{pulling in} changes from another repository. With a centralized version control system, update the current VC fileset from the repository. @end table @kindex C-x v P @findex vc-push @cindex upstream repository On a decentralized version control system, the command @kbd{C-x v P} (@code{vc-push}) updates another location, commonly known as the @dfn{upstream repository}, with locally-committed changes from the current branch. With a prefix argument, it prompts for the exact version control command to run, which lets you specify where to push changes; the default is @kbd{bzr push} with Bazaar, @kbd{git push} with Git, and @kbd{hg push} with Mercurial. The default commands always push to the repository in the default location determined by the version control system from your branch configuration. Prior to pushing, you can use @kbd{C-x v O} (@code{vc-log-outgoing}) to view a log buffer of the changes to be sent upstream. @xref{VC Change Log}. @cindex bound branch (Bazaar VCS) This command is currently supported only by Bazaar, Git, and Mercurial. The concept of ``pushing'' does not exist for centralized version control systems, where this operation is a part of committing a changeset, so invoking this command on a centralized VCS signals an error. This command also signals an error when attempted in a Bazaar @dfn{bound branch}, where committing a changeset automatically pushes the changes to the remote repository to which the local branch is bound. @kindex C-x v + @findex vc-pull With a decentralized version control system, the command @kbd{C-x v +} (@code{vc-pull}) updates the current branch of the local repository and it working tree with changes made in the upstream repository. It is typically used to update a copy (a.k.a.@: @dfn{clone}) of a remote branch. If you supply a prefix argument, the command prompts for the exact version control command to use, which lets you specify where to pull changes from. Otherwise, it pulls from the repository in the default location determined by the version control system from your branch configuration. Amongst decentralized version control systems, @kbd{C-x v +} is currently supported only by Bazaar, Git, and Mercurial. With Bazaar, it calls @kbd{bzr pull} for ordinary branches (to pull from a master branch into a mirroring branch), and @kbd{bzr update} for a bound branch (to pull from a central repository). With Git, it calls @kbd{git pull} to fetch changes from a remote repository and merge it into the current branch. With Mercurial, it calls @kbd{hg pull -u} to fetch changesets from the default remote repository and update the working directory. Prior to pulling, you can use @kbd{C-x v I} (@code{vc-log-incoming}) to view a log buffer of the changes to be applied. @xref{VC Change Log}. With a centralized version control system like CVS, @kbd{C-x v +} updates the current VC fileset from the repository. @node Merging @subsubsection Merging Branches @cindex merging changes @table @kbd @item C-x v m With a decentralized version control system, merge changes from another branch into the current one. With a centralized version control system, merge changes from another branch into the current VC fileset. @end table While developing a branch, you may sometimes need to @dfn{merge} in changes that have already been made in another branch. This is not a trivial operation, as overlapping and conflicting changes may have been made to the two branches. With a decentralized version control system, you merge changes with the command @kbd{C-x v m} (@code{vc-merge}). With Bazaar, this prompts for the exact arguments to pass to the @command{bzr merge} command, offering a sensible default if possible. With Git, this prompts for the name of a branch to merge from, with completion (based on the branch names known to the current repository). With Mercurial, this prompts for argument to pass to @command{hg merge}. The output from running the merge command is shown in a separate buffer. With a centralized version control system like CVS, @kbd{C-x v m} prompts for a branch ID, or a pair of revision IDs (@pxref{Switching Branches}); then it finds the changes from that branch, or the changes between the two revisions you specified, and merges those changes into the current VC fileset. If you just type @kbd{@key{RET}} at the prompt, Emacs simply merges any changes that were made on the same branch since you checked the file out. @cindex conflicts, VC @cindex resolving conflicts Immediately after performing a merge, only the working tree is modified, and you can review the changes produced by the merge with @kbd{C-x v D} and related commands (@pxref{Old Revisions}). If the two branches contained overlapping changes, merging produces a @dfn{conflict}; a warning appears in the output of the merge command, and @dfn{conflict markers} are inserted into each affected work file, surrounding the two sets of conflicting changes. You must then resolve the conflict by editing the conflicted files; by default, Emacs will place buffers with VC conflicts in the special Smerge mode, which provides special commands for resolving the merge conflicts. Once you are done with resolving the conflicts and have saved the files with resolved conflicts, those files must be committed in the usual way for the merge to take effect (@pxref{Basic VC Editing}). @node Creating Branches @subsubsection Creating New Branches On centralized version control systems like CVS, Emacs supports creating new branches as part of a commit operation. When committing a modified VC fileset, type @kbd{C-u C-x v v} (@code{vc-next-action} with a prefix argument; @pxref{Advanced C-x v v}). Then Emacs prompts for a revision ID for the new revision. You should specify a suitable branch ID for a branch starting at the current revision. For example, if the current revision is 2.5, the branch ID should be 2.5.1, 2.5.2, and so on, depending on the number of existing branches at that point. @kindex C-x v b c @findex vc-create-branch This procedure will not work for distributed version control systems like git or Mercurial. For those systems you should use the command @code{vc-create-branch} (@w{@kbd{C-x v b c @var{branch-name} @key{RET}}}) instead. To create a new branch at an older revision (one that is no longer the head of a branch), first select that revision (@pxref{Switching Branches}). Your procedure will then differ depending on whether you are using a locking or merging-based VCS. On a locking VCS, you will need to lock the old revision branch with @kbd{C-x v v}. You'll be asked to confirm, when you lock the old revision, that you really mean to create a new branch---if you say no, you'll be offered a chance to lock the latest revision instead. On a merging-based VCS you will skip this step. Then make your changes and type @kbd{C-x v v} again to commit a new revision. This creates a new branch starting from the selected revision. After the branch is created, subsequent commits create new revisions on that branch. To leave the branch, you must explicitly select a different revision with @kbd{C-u C-x v v}. @ifnottex @include vc1-xtra.texi @end ifnottex @node Projects @section Working with Projects @cindex projects @cindex project root A @dfn{project} is a collection of files used for producing one or more programs. Files that belong to a project are typically stored in a hierarchy of directories; the top-level directory of the hierarchy is known as the @dfn{project root}. @cindex project back-end Whether a given directory is a root of some project is determined by the project-specific infrastructure, known as @dfn{project back-end}. Emacs currently supports two such back-ends: VC-aware (@pxref{Version Control}), whereby a VCS repository is considered a project; and EDE (@pxref{EDE}). This is expected to be extended in the future to support additional types of projects. Which files do or don't belong to a project is also determined by the project back-end. For example, the VC-aware back-end doesn't consider ``ignored'' files (@pxref{VC Ignore}) to be part of the project. Also, the VC-aware Project back-end considers ``untracked'' files by default. That behavior is controllable with the variable @code{project-vc-include-untracked}. @cindex current project name on mode line @defopt project-mode-line If this user option is non-@code{nil}, Emacs displays the name of the current project (if any) on the mode line; clicking @kbd{mouse-1} on the project name pops up the menu with the project-related commands. The default value is @code{nil}. @end defopt @menu * Project File Commands:: Commands for handling project files. * Project Buffer Commands:: Commands for handling project buffers. * Switching Projects:: Switching between projects. * Managing Projects:: Managing the project list file. @end menu @node Project File Commands @subsection Project Commands That Operate on Files @table @kbd @item C-x p f Visit a file that belongs to the current project (@code{project-find-file}). @item C-x p g Find matches for a regexp in all files that belong to the current project (@code{project-find-regexp}). @item M-x project-search Interactively search for regexp matches in all files that belong to the current project. @item C-x p r Perform query-replace for a regexp in all files that belong to the current project (@code{project-query-replace-regexp}). @item C-x p d Run Dired in the current project's root directory (@code{project-dired}). @item C-x p v Run @code{vc-dir} in the current project's root directory (@code{project-vc-dir}). @item C-x p s Start an inferior shell in the current project's root directory (@code{project-shell}). @item C-x p e Start Eshell in the current project's root directory (@code{project-eshell}). @item C-x p c Run compilation in the current project's root directory (@code{project-compile}). @item C-x p ! Run shell command in the current project's root directory (@code{project-shell-command}). @item C-x p & Run shell command asynchronously in the current project's root directory (@code{project-async-shell-command}). @end table Emacs provides commands for handling project files conveniently. This subsection describes these commands. @cindex current project All of the commands described here share the notion of the @dfn{current project}. The current project is determined by the @code{default-directory} (@pxref{File Names}) of the buffer that is the current buffer when the command is invoked. If that directory doesn't seem to belong to a recognizable project, these commands prompt you for the project directory. @findex project-find-file @vindex vc-directory-exclusion-list The command @kbd{C-x p f} (@code{project-find-file}) is a convenient way of visiting files (@pxref{Visiting}) that belong to the current project. Unlike @kbd{C-x C-f}, this command doesn't require typing the full file name of the file to visit, you can type only the file's base name (i.e., omit the leading directories). In addition, the completion candidates considered by the command include only the files belonging to the current project, and nothing else. If there's a file name at point, this command offers that file as the first element of the ``future history''. If given a prefix, include all files under the project root, except for @acronym{VCS} directories listed in @code{vc-directory-exclusion-list}. @findex project-find-regexp The command @kbd{C-x p g} (@code{project-find-regexp}) is similar to @code{rgrep} (@pxref{Grep Searching}), but it searches only the files that belong to the current project. The command prompts for the regular expression to search, and pops up an Xref mode buffer with the search results, where you can select a match using the Xref mode commands (@pxref{Xref Commands}). When invoked with a prefix argument, this command additionally prompts for the base directory from which to start the search; this allows, for example, to limit the search only to project files under a certain subdirectory of the project root. The way this command displays the matches is affected by the value of @code{xref-auto-jump-to-first-xref} (@pxref{Identifier Search}). @findex project-search @kbd{M-x project-search} is a sequential variant of @code{project-find-regexp}. It prompts for a regular expression to search in the current project's files, but instead of finding all the matches and displaying them, it stops when it finds a match and visits the matched file at the locus of the match, allowing you to edit the matched file. To find the rest of the matches, type @w{@kbd{M-x fileloop-continue @key{RET}}}. @findex project-query-replace-regexp @kbd{C-x p r} (@code{project-query-replace-regexp}) is similar to @code{project-search}, but it prompts you for whether to replace each match it finds, like @code{query-replace} does (@pxref{Query Replace}), and continues to the next match after you respond. If your response causes Emacs to exit the query-replace loop, you can later continue with @w{@kbd{M-x fileloop-continue @key{RET}}}. @findex project-find-dir The command @kbd{C-x p d} (@code{project-find-dir}) prompts you to choose a directory inside the current project, with completion. And opens a Dired buffer (@pxref{Dired}) listing the files in it. @findex project-dired The command @kbd{C-x p D} (@code{project-dired}) opens a Dired buffer (@pxref{Dired}) listing the files in the current project's root directory. @findex project-vc-dir The command @kbd{C-x p v} (@code{project-vc-dir}) opens a VC Directory buffer (@pxref{VC Directory Mode}) listing the version control statuses of the files in a directory tree under the current project's root directory. @findex project-shell The command @kbd{C-x p s} (@code{project-shell}) starts a shell session (@pxref{Shell}) in a new buffer with the current project's root as the working directory. @findex project-eshell The command @kbd{C-x p e} (@code{project-eshell}) starts an Eshell session in a new buffer with the current project's root as the working directory. @xref{Top,Eshell,Eshell, eshell, Eshell: The Emacs Shell}. @findex project-compile The command @kbd{C-x p c} (@code{project-compile}) runs compilation (@pxref{Compilation}) in the current project's root directory. @findex project-shell-command The command @kbd{C-x p !} (@code{project-shell-command}) runs @code{shell-command} in the current project's root directory. @findex project-async-shell-command The command @kbd{C-x p &} (@code{project-async-shell-command}) runs @code{async-shell-command} in the current project's root directory. @node Project Buffer Commands @subsection Project Commands That Operate on Buffers @table @kbd @item C-x p b Switch to another buffer belonging to the current project (@code{project-switch-to-buffer}). @item C-x p C-b List the project buffers (@code{project-list-buffers}). @item C-x p k Kill all live buffers that belong to the current project (@code{project-kill-buffers}). @end table @findex project-switch-to-buffer Working on a project could potentially involve having many buffers visiting files that belong to the project, and also buffers that belong to the project, but don't visit any files (like the @file{*compilation*} buffer created by @code{project-compile}). The command @kbd{C-x p b} (@code{project-switch-to-buffer}) helps you switch between buffers that belong to the current project by prompting for a buffer to switch and considering only the current project's buffers as candidates for completion. @findex project-list-buffers Like the command @code{list-buffers} (@pxref{List Buffers}), the command @kbd{C-x p C-b} (@code{project-list-buffers}) displays a list of existing buffers, but only belonging to the current project. @findex project-kill-buffers @vindex project-kill-buffer-conditions @vindex project-kill-buffers-display-buffer-list When you finish working on the project, you may wish to kill all the buffers that belong to the project, to keep your Emacs session smaller. The command @kbd{C-x p k} (@code{project-kill-buffers}) accomplishes that: it kills all the buffers that belong to the current project that satisfy any of @code{project-kill-buffer-conditions}. If @code{project-kill-buffers-display-buffer-list} is non-@code{nil}, the buffers to be killed will be displayed first. @node Switching Projects @subsection Switching Projects @table @kbd @item C-x p p Run an Emacs command for another project (@code{project-switch-project}). @end table @findex project-switch-project @vindex project-switch-commands Commands that operate on project files (@pxref{Project File Commands}) will conveniently prompt you for a project directory when no project is current. When you are inside some project, but you want to operate on a different project, use the @kbd{C-x p p} command (@code{project-switch-project}). This command prompts you to choose a directory among known project roots, and then displays the menu of available commands to operate on the project you choose. The variable @code{project-switch-commands} controls which commands are available in the menu, and which key invokes each command. @vindex project-list-file The variable @code{project-list-file} names the file in which Emacs records the list of known projects. It defaults to the file @file{projects} in @code{user-emacs-directory} (@pxref{Find Init}). @node Managing Projects @subsection Managing the Project List File @table @kbd @item M-x project-forget-project Remove a known project from the @code{project-list-file}. @end table @findex project-forget-project Normally Emacs automatically adds and removes projects to and from the @code{project-list-file}, but sometimes you may want to manually edit the available projects. @kbd{M-x project-forget-project} prompts you to choose one of the available projects, and then removes it from the file. @node Change Log @section Change Logs @cindex change log Many software projects keep a @dfn{change log}. This is a file, normally named @file{ChangeLog}, containing a chronological record of when and how the program was changed. Sometimes, these files are automatically generated from the change log entries stored in version control systems, or are used to generate these change log entries. Sometimes, there are several change log files, each recording the changes in one directory or directory tree. @menu * Change Log Commands:: Commands for editing change log files. * Format of ChangeLog:: What the change log file looks like. @end menu @node Change Log Commands @subsection Change Log Commands @kindex C-x 4 a @findex add-change-log-entry-other-window The Emacs command @kbd{C-x 4 a} adds a new entry to the change log file for the file you are editing (@code{add-change-log-entry-other-window}). If that file is actually a backup file, it makes an entry appropriate for the file's parent---that is useful for making log entries for functions that have been deleted in the current version. @kbd{C-x 4 a} visits the change log file and creates a new entry unless the most recent entry is for today's date and your name. It also creates a new item for the current file. For many languages, it can even guess the name of the function or other object that was changed. @c Not worth it. @c @vindex change-log-directory-files To find the change log file, Emacs searches up the directory tree from the file you are editing. By default, it stops if it finds a directory that seems to be the root of a version-control repository. To change this, customize @code{change-log-directory-files}. @vindex add-log-keep-changes-together When the variable @code{add-log-keep-changes-together} is non-@code{nil}, @kbd{C-x 4 a} adds to any existing item for the file, rather than starting a new item. You can combine multiple changes of the same nature. If you don't enter any text after the initial @kbd{C-x 4 a}, any subsequent @kbd{C-x 4 a} adds another symbol to the change log entry. @vindex add-log-always-start-new-record If @code{add-log-always-start-new-record} is non-@code{nil}, @kbd{C-x 4 a} always makes a new entry, even if the last entry was made by you and on the same date. @vindex change-log-version-info-enabled @vindex change-log-version-number-regexp-list @cindex file version in change log entries If the value of the variable @code{change-log-version-info-enabled} is non-@code{nil}, @kbd{C-x 4 a} adds the file's version number to the change log entry. It finds the version number by searching the first ten percent of the file, using regular expressions from the variable @code{change-log-version-number-regexp-list}. @cindex Change Log mode @findex change-log-mode The change log file is visited in Change Log mode. In this major mode, each bunch of grouped items counts as one paragraph, and each entry is considered a page. This facilitates editing the entries. @kbd{C-j} and auto-fill indent each new line like the previous line; this is convenient for entering the contents of an entry. @findex change-log-goto-source You can use the command @code{change-log-goto-source} (by default bound to @kbd{C-c C-c}) to go to the source location of the change log entry near point, when Change Log mode is on. Then subsequent invocations of the @code{next-error} command (by default bound to @kbd{M-g M-n} and @kbd{C-x `}) will move between entries in the change log. You will jump to the actual site in the file that was changed, not just to the next change log entry. You can also use @code{previous-error} to move back through the change log entries. @findex change-log-merge You can use the command @kbd{M-x change-log-merge} to merge other log files into a buffer in Change Log Mode, preserving the date ordering of entries. @vindex add-log-dont-create-changelog-file Version control systems are another way to keep track of changes in your program and keep a change log. Many projects that use a VCS don't keep a separate versioned change log file nowadays, so you may wish to avoid having such a file in the repository. If the value of @code{add-log-dont-create-changelog-file} is non-@code{nil}, commands like @kbd{C-x 4 a} (@code{add-change-log-entry-other-window}) will record changes in a suitably named temporary buffer instead of a file, if such a file does not already exist. Whether you have a change log file or use a temporary buffer for change logs, you can type @kbd{C-c C-a} (@code{log-edit-insert-changelog}) in the VC Log buffer to insert the relevant change log entries, if they exist. @xref{Log Buffer}. @node Format of ChangeLog @subsection Format of ChangeLog A change log entry starts with a header line that contains the current date, your name (taken from the variable @code{add-log-full-name}), and your email address (taken from the variable @code{add-log-mailing-address}). Aside from these header lines, every line in the change log starts with a space or a tab. The bulk of the entry consists of @dfn{items}, each of which starts with a line starting with whitespace and a star. Here are two entries, both dated in May 1993, with two items and one item respectively. @iftex @medbreak @end iftex @smallexample 1993-05-25 Richard Stallman * man.el: Rename symbols 'man-*' to 'Man-*'. (manual-entry): Make prompt string clearer. * simple.el (blink-matching-paren-distance): Change default to 12,000. 1993-05-24 Richard Stallman * vc.el (minor-mode-map-alist): Don't use it if it's void. (vc-cancel-version): Doc fix. @end smallexample One entry can describe several changes; each change should have its own item, or its own line in an item. Normally there should be a blank line between items. When items are related (parts of the same change, in different places), group them by leaving no blank line between them. You should put a copyright notice and permission notice at the end of the change log file. Here is an example: @smallexample Copyright 1997, 1998 Free Software Foundation, Inc. Copying and distribution of this file, with or without modification, are permitted provided the copyright notice and this notice are preserved. @end smallexample @noindent Of course, you should substitute the proper years and copyright holder. @node Xref @section Find Identifier References @cindex xref @cindex tag An @dfn{identifier} is a name of a syntactical subunit of the program: a function, a subroutine, a method, a class, a data type, a macro, etc. In a programming language, each identifier is a symbol in the language's syntax. Identifiers are also known as @dfn{tags}. Program development and maintenance requires capabilities to quickly find where each identifier was defined and referenced, to rename identifiers across the entire project, etc. These capabilities are also useful for finding references in major modes other than those defined to support programming languages. For example, chapters, sections, appendices, etc.@: of a text or a @TeX{} document can be treated as subunits as well, and their names can be used as identifiers. In this chapter, we use the term ``identifiers'' to collectively refer to the names of any kind of subunits, in program source and in other kinds of text alike. Emacs provides a unified interface to these capabilities, called @samp{xref}. @cindex xref backend To do its job, @code{xref} needs to make use of information and to employ methods specific to the major mode. What files to search for identifiers, how to find references to identifiers, how to complete on identifiers---all this and more is mode-specific knowledge. @code{xref} delegates the mode-specific parts of its job to a @dfn{backend} provided by the mode; it also includes defaults for some of its commands, for those modes that don't provide their own. A backend can implement its capabilities in a variety of ways. Here are a few examples: @enumerate a @item Some major modes provide built-in means for looking up the language symbols. For example, Emacs Lisp symbols can be identified by searching the package load history, maintained by the Emacs Lisp interpreter, and by consulting the built-in documentation strings; the Emacs Lisp mode uses these facilities in its backend to allow finding definitions of symbols. (One disadvantage of this kind of backend is that it only knows about subunits that were loaded into the interpreter.) @item If Eglot is activated for the current buffer's project (@pxref{Projects}) and the current buffer's major mode, Eglot consults an external language server program and provides the data supplied by the server regarding the definitions of the identifiers in the project. @xref{Eglot Features,,, eglot, Eglot: The Emacs LSP Client}. @item An external program can extract references by scanning the relevant files, and build a database of these references. A backend can then access this database whenever it needs to list or look up references. The Emacs distribution includes @command{etags}, a command for tagging identifier definitions in programs, which supports many programming languages and other major modes, such as HTML, by extracting references into @dfn{tags tables}. @xref{Create Tags Table}. Major modes for languages supported by @command{etags} can use tags tables as basis for their backend. (One disadvantage of this kind of backend is that tags tables need to be kept reasonably up to date, by rebuilding them from time to time.) @end enumerate @menu * Find Identifiers:: Commands to find where an identifier is defined or referenced, to list identifiers, etc. * Tags Tables:: Tags table records which file defines a symbol. * Select Tags Table:: How to visit a specific tags table. @end menu @node Find Identifiers @subsection Find Identifiers This subsection describes the commands that find references to identifiers and perform various queries about identifiers. Each such reference could @emph{define} an identifier, e.g., provide the implementation of a program subunit or the text of a document section; or it could @emph{use} the identifier, e.g., call a function or a method, assign a value to a variable, mention a chapter in a cross-reference, etc. @menu * Looking Up Identifiers:: Commands to find the definition of an identifier. * Xref Commands:: Commands in the @file{*xref*} buffer. * Identifier Search:: Searching and replacing identifiers. * List Identifiers:: Listing identifiers and completing on them. @end menu @node Looking Up Identifiers @subsubsection Looking Up Identifiers @cindex find definition of symbols @cindex identifier, finding definition of @cindex find references to symbols The most important thing that @code{xref} enables you to do is to find the definition of a specific identifier. @table @kbd @item M-.@: Find definitions of an identifier (@code{xref-find-definitions}). @item C-M-. @var{pattern} @key{RET} Find all identifiers whose name matches @var{pattern} (@code{xref-find-apropos}). @item C-x 4 .@: @key{RET} Find definitions of identifier, but display it in another window (@code{xref-find-definitions-other-window}). @item C-x 5 .@: @key{RET} Find definition of identifier, and display it in a new frame (@code{xref-find-definitions-other-frame}). @item M-x xref-find-definitions-at-mouse Find definition of identifier at mouse click. @item M-, Go back to where you previously invoked @kbd{M-.} and friends (@code{xref-go-back}). @item C-M-, Go forward to where you previously invoked @kbd{M-,} (@code{xref-go-forward}). @item M-x xref-etags-mode Switch @code{xref} to use the @code{etags} backend. @end table @kindex M-. @findex xref-find-definitions @vindex xref-prompt-for-identifier @kbd{M-.}@: (@code{xref-find-definitions}) shows the definition of the identifier at point. With a prefix argument, or if there's no identifier at point, it prompts for the identifier. (If you want it to always prompt, customize @code{xref-prompt-for-identifier} to @code{t}.) When entering the identifier argument to @kbd{M-.}, you can use the usual minibuffer completion commands (@pxref{Completion}), with the known identifier names being the completion candidates. @kindex C-x 4 . @findex xref-find-definitions-other-window @kindex C-x 5 . @findex xref-find-definitions-other-frame Like most commands that can switch buffers, @code{xref-find-definitions} has a variant that displays the new buffer in another window, and one that makes a new frame for it. The former is @w{@kbd{C-x 4 .}} (@code{xref-find-definitions-other-window}), and the latter is @w{@kbd{C-x 5 .}} (@code{xref-find-definitions-other-frame}). The command @code{xref-find-definitions-at-mouse} works like @code{xref-find-definitions}, but it looks for the identifier name at or around the place of a mouse event. This command is intended to be bound to a mouse event, such as @kbd{C-M-mouse-1}, for example. @kindex C-M-. @findex xref-find-apropos @vindex tags-apropos-additional-actions The command @kbd{C-M-.}@: (@code{xref-find-apropos}) is like @code{apropos} for tags (@pxref{Apropos}). It displays a list of identifiers in the selected tags table whose names match the specified @var{regexp}. This is just like @kbd{M-.}, except that it does regexp matching of identifiers instead of matching symbol names as fixed strings. By default, the command pops up the @file{*xref*} buffer, like @kbd{M-.}, but you can display additional output by customizing the variable @code{tags-apropos-additional-actions}; see its documentation for details. @vindex xref-auto-jump-to-first-definition If any of the above commands finds more than one matching definition, it by default pops up the @file{*xref*} buffer showing the matching candidates and selects that buffer's window. (@kbd{C-M-.}@: @emph{always} pops up the @file{*xref*} buffer if it finds at least one match.) Each candidate is normally shown in that buffer as the name of a file and the matching identifier(s) in that file. In that buffer, you can select any of the candidates for display, and you have several additional commands, described in @ref{Xref Commands}. However, if the value of the variable @code{xref-auto-jump-to-first-definition} is @code{move}, Emacs automatically moves point to the first of these candidates in the @file{*xref*} buffer, so just typing @key{RET} will display the definition of that candidate. If the value of the variable is @code{t} or @code{show}, the first candidate is automatically shown in its own window; @code{t} also selects the window showing the first candidate's definition, while @code{show} leaves the window of the @file{*xfer*} buffer selected. The default value is @code{nil}, which just shows the candidates in the @file{*xref*} buffer, but neither selects any of them nor shows their definition, until you select a candidate in the @file{*xref*} buffer. @findex next-error, in @file{*xref*} buffer @findex previous-error, in @file{*xref*} buffer @kindex M-g M-n, for navigation in @file{*xref*} buffer @kindex M-g M-p, for navigation in @file{*xref*} buffer If you switch away of the window showing the @file{*xref*} buffer which displays several candidates, you can move from one candidate to another using the commands @w{@kbd{M-g M-n}} (@code{next-error}) and @w{@kbd{M-g M-p}} (@code{previous-error}). @xref{Compilation Mode}. @kindex M-, @findex xref-go-back To go back to places @emph{from where} you've displayed the definition, use @kbd{M-,} (@code{xref-go-back}). It jumps back to the point of the last invocation of @kbd{M-.}. Thus you can find and examine the definition of something with @kbd{M-.} and then return to where you were with @kbd{M-,}. @kbd{M-,} allows you to retrace the steps you made forward in the history of places, all the way to the first place in history, where you first invoked @kbd{M-.}, or to any place in-between. @kindex C-M-, @findex xref-go-forward If you previously went back too far with @kbd{M-,}, or want to re-examine a place from which you went back, you can use @kbd{C-M-,} (@code{xref-go-forward}) to go forward again. This is similar to using @kbd{M-.}, except that you don't need on each step to move point to the identifier whose definition you want to look up. @kbd{C-M-,} allows you to retrace all the steps you made back in the history of places, all the way to the last place in history, where you invoked @kbd{M-,}, or to any place in-between. @findex xref-etags-mode Some major modes install @code{xref} support facilities that might sometimes fail to find certain identifiers. For example, in Emacs Lisp mode (@pxref{Lisp Eval}) @kbd{M-.} will by default find only functions and variables from Lisp packages which are loaded into the current Emacs session or are auto-loaded (@pxref{Autoload,,, elisp, The Emacs Lisp Reference Manual}). If @kbd{M-.} fails to find some identifiers, you can try forcing @code{xref} to use the @code{etags} backend (@pxref{Xref}). To this end, turn on the Xref Etags minor mode with @w{@kbd{M-x xref-etags-mode}}, then invoke @kbd{M-.} again. (For this to work, be sure to run @command{etags} to create the tags table in the directory tree of the source files, see @ref{Create Tags Table}.) @node Xref Commands @subsubsection Commands Available in the @file{*xref*} Buffer @cindex commands in @file{*xref*} buffers @cindex XREF mode The following commands are provided in the @file{*xref*} buffer by the special XREF mode: @table @kbd @item @key{RET} @itemx mouse-1 Display the reference on the current line (@code{xref-goto-xref}). With prefix argument, also bury the @file{*xref*} buffer. @item mouse-2 @findex xref-select-and-show-xref The same as @code{mouse-1}, but make the window displaying the @file{*xref*} buffer the selected window (@code{xref-select-and-show-xref}). @item n @itemx . @findex xref-next-line Move to the next reference and display it in the other window (@code{xref-next-line}). @item N @findex xref-next-group Move to the first reference of the next reference group and display it in the other window (@code{xref-next-group}). @item p @itemx , @findex xref-prev-line Move to the previous reference and display it in the other window (@code{xref-prev-line}). @item P @findex xref-prev-group Move to the first reference of the previous reference group and display it in the other window (@code{xref-prev-group}). @item C-o @findex xref-show-location-at-point Display the reference on the current line in the other window (@code{xref-show-location-at-point}). @item r @var{pattern} @key{RET} @var{replacement} @key{RET} Perform interactive query-replace on references that match @var{pattern} (@code{xref-query-replace-in-results}), replacing the match with @var{replacement}. This command can only be used in @file{*xref*} buffers that show all the matches for an identifier in all the relevant files. @xref{Identifier Search}. @item g @findex xref-revert-buffer Refresh the contents of the @file{*xref*} buffer (@code{xref-revert-buffer}). @item M-, @findex xref-quit-and-pop-marker-stack Quit the window showing the @file{*xref*} buffer, and then jump to the previous Xref stack location (@code{xref-quit-and-pop-marker-stack}). @item q @findex xref-quit Quit the window showing the @file{*xref*} buffer (@code{xref-quit}). @end table In addition, the usual navigation commands, such as the arrow keys, @kbd{C-n}, and @kbd{C-p} are available for moving around the buffer without displaying the references. @node Identifier Search @subsubsection Searching and Replacing with Identifiers @cindex search and replace in multiple source files @cindex multiple source file search and replace The commands in this section perform various search and replace operations either on identifiers themselves or on files that reference them. @table @kbd @item M-? Find all the references for the identifier at point. @item r @itemx M-x xref-query-replace-in-results @key{RET} @var{replacement} @key{RET} @itemx C-u M-x xref-query-replace-in-results @key{RET} @var{regexp} @key{RET} @var{replacement} @key{RET} Interactively replace @var{regexp} with @var{replacement} in the names of all the identifiers shown in the @file{*xref*} buffer. @item M-x xref-find-references-and-replace @key{RET} @var{from} @key{RET} @var{to} @key{RET} Interactively rename all instances of the identifier @var{from} to the new name @var{to}. @item M-x tags-search @key{RET} @var{regexp} @key{RET} Search for @var{regexp} through the files in the selected tags table. @item M-x tags-query-replace @key{RET} @var{regexp} @key{RET} @var{replacement} @key{RET} Perform a @code{query-replace-regexp} on each file in the selected tags table. @item M-x fileloop-continue Restart one of the last 2 commands above, from the current location of point. @end table @kindex M-? @findex xref-find-references @kbd{M-?} finds all the references for the identifier at point, prompting for the identifier as needed, with completion. Depending on the current backend (@pxref{Xref}), the command may prompt even if it finds a valid identifier at point. When invoked with a prefix argument, it always prompts for the identifier. (If you want it to prompt always, customize the value of the variable @code{xref-prompt-for-identifier} to @code{t}; or set it to @code{nil} to prompt only if there's no usable identifier at point.) The command then presents the @file{*xref*} buffer with all the references to the identifier, showing the file name and the line where the identifier is referenced. The XREF mode commands are available in this buffer, see @ref{Xref Commands}. @vindex xref-auto-jump-to-first-xref If the value of the variable @code{xref-auto-jump-to-first-xref} is @code{t}, @code{xref-find-references} automatically jumps to the first result in the @file{*xref*} buffer and selects the window where that reference is displayed; you can select the other results with @w{@kbd{M-g M-n}} (@code{next-error}) and @w{@kbd{M-g M-p}} (@code{previous-error}) (@pxref{Compilation Mode}). If the value is @code{show}, the first result is displayed, but the window showing the @file{*xref*} buffer is left selected. If the value is @code{move}, the first result is selected in the @file{*xref*} buffer, but is not displayed; you can then use @key{RET} to actually display the reference. The default value is @code{nil}, which just shows the results in the @file{*xref*} buffer, but doesn't select any of them, and doesn't display the reference itself. @findex xref-query-replace-in-results @kbd{r} (@code{xref-query-replace-in-results}) reads a @var{replacement} string, just like ordinary @kbd{M-x query-replace-regexp}. It then renames the identifiers shown in the @file{*xref*} buffer in all the places in all the files where these identifiers are referenced, such that their new name is @var{replacement}. This is useful when you rename your identifiers as part of refactoring. This command should be invoked in the @file{*xref*} buffer generated by @kbd{M-?}. By default, the command replaces the entire name of each identifier with @var{replacement}, but if invoked with a prefix argument, the command prompts for a regexp to match identifier names, and replaces only the matches of that regexp in the names of the identifiers with @var{replacement}. @findex xref-find-references-and-replace @kbd{M-x xref-find-references-and-replace} works similarly to @code{xref-query-replace-in-results}, but is more convenient when you want to rename a single identifier specified by its name @var{from}. @findex tags-search @kbd{M-x tags-search} reads a regexp using the minibuffer, then searches for matches in all the files in the selected tags table, one file at a time. It displays the name of the file being searched so you can follow its progress. As soon as it finds an occurrence, @code{tags-search} returns. This command requires tags tables to be available (@pxref{Tags Tables}). @findex fileloop-continue Having found one match with @code{tags-search}, you probably want to find all the rest. @kbd{M-x fileloop-continue} resumes the @code{tags-search}, finding one more match. This searches the rest of the current buffer, followed by the remaining files of the tags table. @findex tags-query-replace @kbd{M-x tags-query-replace} performs a single @code{query-replace-regexp} through all the files in the tags table. It reads a regexp to search for and a string to replace with, just like ordinary @kbd{M-x query-replace-regexp}. It searches much like @kbd{M-x tags-search}, but repeatedly, processing matches according to your input. @xref{Query Replace}, for more information on query replace. @vindex tags-case-fold-search @cindex case-sensitivity and tags search You can control the case-sensitivity of tags search commands by customizing the value of the variable @code{tags-case-fold-search}. The default is to use the same setting as the value of @code{case-fold-search} (@pxref{Lax Search}). It is possible to get through all the files in the tags table with a single invocation of @kbd{M-x tags-query-replace}. But often it is useful to exit temporarily, which you can do with any input event that has no special query replace meaning. You can resume the query replace subsequently by typing @kbd{M-x fileloop-continue}; this command resumes the last tags search or replace command that you did. For instance, to skip the rest of the current file, you can type @w{@kbd{M-> M-x fileloop-continue}}. Note that the commands described above carry out much broader searches than the @code{xref-find-definitions} family. The @code{xref-find-definitions} commands search only for definitions of identifiers that match your string or regexp. The commands @code{xref-find-references}, @code{tags-search}, and @code{tags-query-replace} find every occurrence of the identifier or regexp, as ordinary search commands and replace commands do in the current buffer. As an alternative to @code{xref-find-references} and @code{tags-search}, you can run @command{grep} as a subprocess and have Emacs show you the matching lines one by one. @xref{Grep Searching}. @node List Identifiers @subsubsection Identifier Inquiries @table @kbd @item C-M-i @itemx M-@key{TAB} Perform completion on the text around point, possibly using the selected tags table if one is loaded (@code{completion-at-point}). @item M-x list-tags @key{RET} @var{file} @key{RET} Display a list of the identifiers defined in the program file @var{file}. @item C-M-. @var{regexp} @key{RET} Display a list of all identifiers matching @var{regexp} (@code{xref-find-apropos}). @xref{Looking Up Identifiers}. @item M-x tags-next-file Visit files recorded in the selected tags table. @end table In most programming language modes, you can type @kbd{C-M-i} or @kbd{M-@key{TAB}} (@code{completion-at-point}) to complete the symbol at point. Some modes provide specialized completion for this command tailored to the mode; for those that don't, if there is a tags table loaded, this command can use it to generate completion candidates. @xref{Symbol Completion}. @findex list-tags @kbd{M-x list-tags} reads the name of one of the files covered by the selected tags table, and displays a list of tags defined in that file. Do not include a directory as part of the file name unless the file name recorded in the tags table includes a directory. This command works only with the etags backend, and requires a tags table for the project to be available. @xref{Tags Tables}. If used interactively, the default tag is file name of the current buffer if used interactively. @findex tags-next-file @kbd{M-x tags-next-file} visits files covered by the selected tags table. The first time it is called, it visits the first file covered by the table. Each subsequent call visits the next covered file, unless a prefix argument is supplied, in which case it returns to the first file. This command requires a tags table to be selected. @node Tags Tables @subsection Tags Tables @cindex tags and tag tables A @dfn{tags table} records the tags@footnote{ A @dfn{tag} is a synonym for identifier reference. Commands and features based on the @code{etags} package traditionally use ``tag'' with this meaning, and this subsection follows that tradition. } extracted by scanning the source code of a certain program or a certain document. Tags extracted from generated files reference the original files, rather than the generated files that were scanned during tag extraction. Examples of generated files include C files generated from Cweb source files, from a Yacc parser, or from Lex scanner definitions; @file{.i} preprocessed C files; and Fortran files produced by preprocessing @file{.fpp} source files. @cindex etags To produce a tags table, you run the @command{etags} shell command on a document or the source code file. The @samp{etags} program writes the tags to a @dfn{tags table file}, or @dfn{tags file} in short. The conventional name for a tags file is @file{TAGS}@. @xref{Create Tags Table}. (It is also possible to create a tags table by using one of the commands from other packages that can produce such tables in the same format.) Emacs uses the tags tables via the @code{etags} package as one of the supported backends for @code{xref}. Because tags tables are produced by the @command{etags} command that is part of an Emacs distribution, we describe tags tables in more detail here. @cindex C++ class browser, tags @cindex tags, C++ @cindex class browser, C++ @cindex Ebrowse The Ebrowse facility is similar to @command{etags} but specifically tailored for C++. @xref{Top,, Ebrowse, ebrowse, Ebrowse User's Manual}. The Semantic package provides another way to generate and use tags, separate from the @command{etags} facility. @xref{Semantic}. @menu * Tag Syntax:: Tag syntax for various types of code and text files. * Create Tags Table:: Creating and updating tags tables with @command{etags}. * Etags Regexps:: Create arbitrary tags using regular expressions. @end menu @node Tag Syntax @subsubsection Source File Tag Syntax Here is how tag syntax is defined for the most popular languages: @itemize @bullet @item In C code, any C function or typedef is a tag, and so are definitions of @code{struct}, @code{union} and @code{enum}. @code{#define} macro definitions, @code{#undef} and @code{enum} constants are also tags, unless you specify @samp{--no-defines} when making the tags table. Similarly, global variables are tags, unless you specify @samp{--no-globals}, and so are struct members, unless you specify @samp{--no-members}. Use of @samp{--no-globals}, @samp{--no-defines} and @samp{--no-members} can make the tags table file much smaller. You can tag function declarations and external variables in addition to function definitions by giving the @samp{--declarations} option to @command{etags}. @item In C++ code, in addition to all the tag constructs of C code, member functions are also recognized; member variables are also recognized, unless you use the @samp{--no-members} option. @code{operator} definitions have tag names like @samp{operator+}. If you specify the @samp{--class-qualify} option, tags for variables and functions in classes are named @samp{@var{class}::@var{variable}} and @samp{@var{class}::@var{function}}. By default, class methods and members are not class-qualified, which facilitates identifying their names in the sources more accurately. @item In Java code, tags include all the constructs recognized in C++, plus the @code{interface}, @code{extends} and @code{implements} constructs. Tags for variables and functions in classes are named @samp{@var{class}.@var{variable}} and @samp{@var{class}.@var{function}}. @item In @LaTeX{} documents, the arguments for @code{\chapter}, @code{\section}, @code{\subsection}, @code{\subsubsection}, @code{\eqno}, @code{\label}, @code{\ref}, @code{\cite}, @code{\bibitem}, @code{\part}, @code{\appendix}, @code{\entry}, @code{\index}, @code{\def}, @code{\newcommand}, @code{\renewcommand}, @code{\newenvironment} and @code{\renewenvironment} are tags. Other commands can make tags as well, if you specify them in the environment variable @env{TEXTAGS} before invoking @command{etags}. The value of this environment variable should be a colon-separated list of command names. For example, @example TEXTAGS="mycommand:myothercommand" export TEXTAGS @end example @noindent specifies (using Bourne shell syntax) that the commands @samp{\mycommand} and @samp{\myothercommand} also define tags. @item In Lisp code, any function defined with @code{defun}, any variable defined with @code{defvar} or @code{defconst}, and in general the first argument of any expression that starts with @samp{(def} in column zero is a tag. As an exception, expressions of the form @code{(defvar @var{foo})} are treated as declarations, and are only tagged if the @samp{--declarations} option is given. @item In Scheme code, tags include anything defined with @code{def} or with a construct whose name starts with @samp{def}. They also include variables set with @code{set!} at top level in the file. @end itemize Several other languages are also supported: @itemize @bullet @item In Ada code, functions, procedures, packages, tasks and types are tags. Use the @samp{--packages-only} option to create tags for packages only. In Ada, the same name can be used for different kinds of entity (e.g., for a procedure and for a function). Also, for things like packages, procedures and functions, there is the spec (i.e., the interface) and the body (i.e., the implementation). To make it easier to pick the definition you want, Ada tag names have suffixes indicating the type of entity: @table @samp @item /b package body. @item /f function. @item /k task. @item /p procedure. @item /s package spec. @item /t type. @end table Thus, @kbd{M-x find-tag @key{RET} bidule/b @key{RET}} will go directly to the body of the package @code{bidule}, while @kbd{M-x find-tag @key{RET} bidule @key{RET}} will just search for any tag @code{bidule}. @item In assembler code, labels appearing at the start of a line, followed by a colon, are tags. @item In Bison or Yacc input files, each rule defines as a tag the nonterminal it constructs. The portions of the file that contain C code are parsed as C code. @item In Cobol code, tags are paragraph names; that is, any word starting in column 8 and followed by a period. @item In Erlang code, the tags are the functions, records and macros defined in the file. @item In Fortran code, functions, subroutines and block data are tags. @item In Go code, packages, functions, and types are tags. @item In HTML input files, the tags are the @code{title} and the @code{h1}, @code{h2}, @code{h3} headers. Also, tags are @code{name=} in anchors and all occurrences of @code{id=}. @item In Lua input files, all functions are tags. @item In makefiles, targets are tags; additionally, variables are tags unless you specify @samp{--no-globals}. @item In Objective C code, tags include Objective C definitions for classes, class categories, methods and protocols. Tags for variables and functions in classes are named @samp{@var{class}::@var{variable}} and @samp{@var{class}::@var{function}}. @item In Pascal code, the tags are the functions and procedures defined in the file. @item In Perl code, the tags are the packages, subroutines and variables defined by the @code{package}, @code{sub}, @code{use constant}, @code{my}, and @code{local} keywords. Use @samp{--globals} if you want to tag global variables. Tags for subroutines are named @samp{@var{package}::@var{sub}}. The name for subroutines defined in the default package is @samp{main::@var{sub}}. @item In PHP code, tags are functions, classes and defines. Vars are tags too, unless you use the @samp{--no-members} option. @item In PostScript code, the tags are the functions. @item In Prolog code, tags are predicates and rules at the beginning of line. @item In Python code, @code{def} or @code{class} at the beginning of a line generate a tag. @item In Ruby code, @code{def} or @code{class} or @code{module} at the beginning of a line generate a tag. Constants also generate tags. @item In Rust code, tags anything defined with @code{fn}, @code{enum}, @code{struct} or @code{macro_rules!}. @end itemize You can also generate tags based on regexp matching (@pxref{Etags Regexps}) to handle other formats and languages. @node Create Tags Table @subsubsection Creating Tags Tables @cindex @command{etags} program The @command{etags} program is used to create a tags table file. It knows the syntax of several languages, as described in @iftex the previous section. @end iftex @ifnottex @ref{Tag Syntax}. @end ifnottex Here is how to run @command{etags}: @example etags @var{inputfiles}@dots{} @end example @noindent The @command{etags} program reads the specified files, and writes a tags table named @file{TAGS} in the current working directory. You can optionally specify a different file name for the tags table by using the @samp{--output=@var{file}} option; specifying @file{-} as a file name prints the tags table to standard output. You can also append the newly created tags table to an existing file by using the @samp{--append} option. If the specified files don't exist, @command{etags} looks for compressed versions of them and uncompresses them to read them. Under MS-DOS, @command{etags} also looks for file names like @file{mycode.cgz} if it is given @samp{mycode.c} on the command line and @file{mycode.c} does not exist. If the tags table becomes outdated due to changes in the files described in it, you can update it by running the @command{etags} program again. If the tags table does not record a tag, or records it for the wrong file, then Emacs will not be able to find that definition until you update the tags table. But if the position recorded in the tags table becomes a little bit wrong (due to other editing), Emacs will still be able to find the right position, with a slight delay. Thus, there is no need to update the tags table after each edit. You should update a tags table when you define new tags that you want to have listed, or when you move tag definitions from one file to another, or when changes become substantial. You can make a tags table @dfn{include} another tags table, by passing the @samp{--include=@var{file}} option to @command{etags}. It then covers all the files covered by the included tags file, as well as its own. If you specify the source files with relative file names when you run @command{etags}, the tags file will contain file names relative to the directory where the tags file was initially written. This way, you can move an entire directory tree containing both the tags file and the source files, and the tags file will still refer correctly to the source files. If the tags file is @file{-} or is in the @file{/dev} directory, however, the file names are made relative to the current working directory. This is useful, for example, when writing the tags to the standard output. When using a relative file name, it should not be a symbolic link pointing to a tags file in a different directory, because this would generally render the file names invalid. If you specify absolute file names as arguments to @command{etags}, then the tags file will contain absolute file names. This way, the tags file will still refer to the same files even if you move it, as long as the source files remain in the same place. Absolute file names start with @samp{/}, or with @samp{@var{device}:/} on MS-DOS and MS-Windows. When you want to make a tags table from a great number of files, you may have problems listing them on the command line, because some systems have a limit on its length. You can circumvent this limit by telling @command{etags} to read the file names from its standard input, by typing a dash in place of the file names, like this: @smallexample find . -name "*.[chCH]" -print | etags - @end smallexample @command{etags} recognizes the language used in an input file based on its file name and contents. It first tries to match the file's name and extension to the ones commonly used with certain languages. Some languages have interpreters with known names (e.g., @command{perl} for Perl or @command{pl} for Prolog), so @command{etags} next looks for an interpreter specification of the form @samp{#!@var{interp}} on the first line of an input file, and matches that against known interpreters. If none of that works, or if you want to override the automatic detection of the language, you can specify the language explicitly with the @samp{--language=@var{name}} option. You can intermix these options with file names; each one applies to the file names that follow it. Specify @samp{--language=auto} to tell @command{etags} to resume guessing the language from the file names and file contents. Specify @samp{--language=none} to turn off language-specific processing entirely; then @command{etags} recognizes tags by regexp matching alone (@pxref{Etags Regexps}). This comes in handy when an input file uses a language not yet supported by @command{etags}, and you want to avoid having @command{etags} fall back on Fortran and C as the default languages. The option @samp{--parse-stdin=@var{file}} is mostly useful when calling @command{etags} from programs. It can be used (only once) in place of a file name on the command line. @command{etags} will read from standard input and mark the produced tags as belonging to the file @var{file}. For C and C++, if the source files don't observe the GNU Coding Standards' convention if having braces (@samp{@{} and @samp{@}}) in column zero only for top-level definitions, like functions and @code{struct} definitions, we advise that you use the @samp{--ignore-indentation} option, to prevent @command{etags} from incorrectly interpreting closing braces in column zero. @samp{etags --help} outputs the list of the languages @command{etags} knows, and the file name rules for guessing the language. It also prints a list of all the available @command{etags} options, together with a short explanation. If followed by one or more @samp{--language=@var{lang}} options, it outputs detailed information about how tags are generated for @var{lang}. @findex etags-regen-mode Instead of creating and updating the tags table by manually invoking @command{etags}, you can ask Emacs to do it for you automatically. The global minor mode @code{etags-regen-mode}, if enabled, generates tags tables automatically as needed, and takes care of updating them when you edit any of the source files that contribute tags. This mode uses the current project configuration (@pxref{Projects}) to determine which files to submit to @command{etags} for regenerating the tags table for the project. You can customize how this minor mode works using the following user options: @vtable @code @item etags-regen-program The program to regenerate tags table; defaults to @command{etags}. @item etags-regen-program-options Command-line options to pass to the program which regenerates tags tables. @item etags-regen-ignores List of glob patterns which specify files to ignore when regenerating tags tables. @end vtable @cindex tags-reset-tags-tables If you select a tags table manually, with @kbd{M-x visit-tags-table} (@pxref{Select Tags Table}), @code{etags-regen-mode} effectively disables itself: it will no longer automatically create and update tags tables, assuming that you prefer managing your tags tables manually. You can cancel this effect of using @code{visit-tags-table} by invoking the command @code{tags-reset-tags-tables}. @node Etags Regexps @subsubsection Etags Regexps The @samp{--regex} option to @command{etags} allows tags to be recognized by regular expression matching. You can intermix this option with file names; each one applies to the source files that follow it. If you specify multiple @samp{--regex} options, all of them are used in parallel. The syntax is: @smallexample --regex=[@var{@{language@}}]/@var{tagregexp}/[@var{nameregexp}/]@var{modifiers} @end smallexample @noindent The essential part of the option value is @var{tagregexp}, the regexp for matching tags. It is always used anchored, that is, it only matches at the beginning of a line. If you want to allow indented tags, use a regexp that matches initial whitespace; start it with @samp{[ \t]*}. In these regular expressions, @samp{\} quotes the next character, and all the C character escape sequences are supported: @samp{\a} for bell, @samp{\b} for back space, @samp{\e} for escape, @samp{\f} for formfeed, @samp{\n} for newline, @samp{\r} for carriage return, @samp{\t} for tab, and @samp{\v} for vertical tab. In addition, @samp{\d} stands for the @code{DEL} character. Ideally, @var{tagregexp} should not match more characters than are needed to recognize what you want to tag. If the syntax requires you to write @var{tagregexp} so it matches more characters beyond the tag itself, you should add a @var{nameregexp}, to pick out just the tag. This will enable Emacs to find tags more accurately and to do completion on tag names more reliably. In @var{nameregexp}, it is frequently convenient to use ``back references'' (@pxref{Regexp Backslash}) to parenthesized groupings @w{@samp{\( @dots{} \)}} in @var{tagregexp}. For example, @samp{\1} refers to the first such parenthesized grouping. You can find some examples of this below. The @var{modifiers} are a sequence of zero or more characters that modify the way @command{etags} does the matching. A regexp with no modifiers is applied sequentially to each line of the input file, in a case-sensitive way. The modifiers and their meanings are: @table @samp @item i Ignore case when matching this regexp. @item m Match this regular expression against the whole file, so that multi-line matches are possible. @item s Match this regular expression against the whole file, and allow @samp{.} in @var{tagregexp} to match newlines. @end table The @samp{-R} option cancels all the regexps defined by preceding @samp{--regex} options. It too applies to the file names following it. Here's an example: @smallexample etags --regex=/@var{reg1}/i voo.doo --regex=/@var{reg2}/m \ bar.ber -R --lang=lisp los.er @end smallexample @noindent Here @command{etags} chooses the parsing language for @file{voo.doo} and @file{bar.ber} according to their contents. @command{etags} also uses @var{reg1} to recognize additional tags in @file{voo.doo}, and both @var{reg1} and @var{reg2} to recognize additional tags in @file{bar.ber}. @var{reg1} is checked against each line of @file{voo.doo} and @file{bar.ber}, in a case-insensitive way, while @var{reg2} is checked against the whole @file{bar.ber} file, permitting multi-line matches, in a case-sensitive way. @command{etags} uses only the Lisp tags rules, with no user-specified regexp matching, to recognize tags in @file{los.er}. You can restrict a @samp{--regex} option to match only files of a given language by using the optional prefix @var{@{language@}}. (@samp{etags --help} prints the list of languages recognized by @command{etags}.) This is particularly useful when storing many predefined regular expressions for @command{etags} in a file. The following example tags the @code{DEFVAR} macros in the Emacs source files, for the C language only: @smallexample --regex='@{c@}/[ \t]*DEFVAR_[A-Z_ \t(]+"\([^"]+\)"/\1/' @end smallexample @noindent When you have complex regular expressions, you can store the list of them in a file. The following option syntax instructs @command{etags} to read two files of regular expressions. The regular expressions contained in the second file are matched without regard to case. @smallexample --regex=@@@var{case-sensitive-file} --ignore-case-regex=@@@var{ignore-case-file} @end smallexample @noindent A regex file for @command{etags} contains one regular expression per line. Empty lines, and lines beginning with space or tab are ignored. When the first character in a line is @samp{@@}, @command{etags} assumes that the rest of the line is the name of another file of regular expressions; thus, one such file can include another file. All the other lines are taken to be regular expressions. If the first non-whitespace text on the line is @samp{--}, that line is a comment. For example, we can create a file called @samp{emacs.tags} with the following contents: @smallexample -- This is for GNU Emacs C source files @{c@}/[ \t]*DEFVAR_[A-Z_ \t(]+"\([^"]+\)"/\1/ @end smallexample @noindent and then use it like this: @smallexample etags --regex=@@emacs.tags *.[ch] */*.[ch] @end smallexample Here are some more examples. The regexps are quoted to protect them from shell interpretation. @itemize @bullet @item Tag Octave files: @smallexample etags --language=none \ --regex='/[ \t]*function.*=[ \t]*\([^ \t]*\)[ \t]*(/\1/' \ --regex='/###key \(.*\)/\1/' \ --regex='/[ \t]*global[ \t].*/' \ *.m @end smallexample @noindent Note that tags are not generated for scripts, so that you have to add a line by yourself of the form @samp{###key @var{scriptname}} if you want to jump to it. @item Tag Tcl files: @smallexample etags --language=none --regex='/proc[ \t]+\([^ \t]+\)/\1/' *.tcl @end smallexample @item Tag VHDL files: @smallexample etags --language=none \ --regex='/[ \t]*\(ARCHITECTURE\|CONFIGURATION\) +[^ ]* +OF/' \ --regex='/[ \t]*\(ATTRIBUTE\|ENTITY\|FUNCTION\|PACKAGE\ \( BODY\)?\|PROCEDURE\|PROCESS\|TYPE\)[ \t]+\([^ \t(]+\)/\3/' @end smallexample @end itemize @node Select Tags Table @subsection Selecting a Tags Table @findex visit-tags-table Emacs has at any time at most one @dfn{selected} tags table. All the commands for working with tags tables use the selected one first. To select a tags table, type @kbd{M-x visit-tags-table}, which reads the tags table file name as an argument, with @file{TAGS} defaulting to the first directory that contains a file named @file{TAGS} encountered when recursively searching upward from the default directory. @vindex tags-file-name Emacs does not actually read in the tags table contents until you try to use them; all @code{visit-tags-table} does is store the file name in the variable @code{tags-file-name}, and not much more. The variable's initial value is @code{nil}; that value tells all the commands for working with tags tables that they must ask for a tags table file name to use. In addition to the selected tags table, Emacs maintains the list of several tags tables that you use together. For example, if you are working on a program that uses a library, you may wish to have the tags tables of both the program and the library available, so that Emacs could easily find identifiers from both. If the selected tags table doesn't have the identifier or doesn't mention the source file a tags command needs, the command will try using all the other tags tables in the current list of tags tables. Using @code{visit-tags-table} to load a new tags table when another tags table is already loaded gives you a choice: you can add the new tags table to the current list of tags tables, or discard the current list and start a new list. If you start a new list, the new tags table is used @emph{instead} of others. If you add the new table to the current list, it is used @emph{as well as} the others. @vindex tags-table-list You can specify a precise list of tags tables by setting the variable @code{tags-table-list} to a list of directory names, like this: @c keep this on two lines for formatting in smallbook @example @group (setq tags-table-list '("~/.emacs.d" "/usr/local/lib/emacs/src")) @end group @end example @noindent This tells the tags commands to look at the @file{TAGS} files in your @file{~/.emacs.d} directory and in the @file{/usr/local/lib/emacs/src} directory. The order depends on which file you are in and which tags table mentions that file. Do not set both @code{tags-file-name} and @code{tags-table-list}. @node EDE @section Emacs Development Environment @cindex EDE (Emacs Development Environment) @cindex Emacs Development Environment @cindex Integrated development environment EDE (@dfn{Emacs Development Environment}) is a package that simplifies the task of creating, building, and debugging large programs with Emacs. It provides some of the features of an IDE, or @dfn{Integrated Development Environment}, in Emacs. This section provides a brief description of EDE usage. @ifnottex For full details, see @ref{Top, EDE,, ede, Emacs Development Environment}. @end ifnottex @iftex For full details on Ede, type @kbd{C-h i} and then select the EDE manual. @end iftex EDE is implemented as a global minor mode (@pxref{Minor Modes}). To enable it, type @kbd{M-x global-ede-mode} or click on the @samp{Project Support (EDE)} item in the @samp{Tools} menu. You can also enable EDE each time you start Emacs, by adding the following line to your initialization file: @smallexample (global-ede-mode t) @end smallexample @noindent Activating EDE adds a menu named @samp{Development} to the menu bar. Many EDE commands, including the ones described below, can be invoked from this menu. EDE organizes files into @dfn{projects}, which correspond to directory trees. The @dfn{project root} is the topmost directory of a project. To define a new project, visit a file in the desired project root and type @kbd{M-x ede-new}. This command prompts for a @dfn{project type}, which refers to the underlying method that EDE will use to manage the project (@pxref{Creating a project, EDE,, ede, Emacs Development Environment}). The most common project types are @samp{Make}, which uses Makefiles, and @samp{Automake}, which uses GNU Automake (@pxref{Top, Automake,, automake, Automake}). In both cases, EDE also creates a file named @file{Project.ede}, which stores information about the project. A project may contain one or more @dfn{targets}. A target can be an object file, executable program, or some other type of file, which is built from one or more of the files in the project. To add a new @dfn{target} to a project, type @kbd{C-c . t} (@code{M-x ede-new-target}). This command also asks if you wish to add the current file to that target, which means that the target is to be built from that file. After you have defined a target, you can add more files to it by typing @kbd{C-c . a} (@code{ede-add-file}). To build a target, type @kbd{C-c . c} (@code{ede-compile-target}). To build all the targets in the project, type @kbd{C-c . C} (@code{ede-compile-project}). EDE uses the file types to guess how the target should be built. @ifnottex @include emerge-xtra.texi @end ifnottex @node Bug Reference @section Bug Reference @cindex bug reference Most projects with a certain amount of users track bug reports in some issue tracking software which assigns each report a unique and short number or identifier. Those are used to reference a given bug, e.g., in a source code comment above the code fixing some bug, in documentation files, or in discussions on some mailing list or IRC channel. @findex bug-reference-mode @findex bug-reference-prog-mode The minor modes @code{bug-reference-mode} and @code{bug-reference-prog-mode} highlight such bug references and make it possible to follow them to the corresponding bug report on the project's issue tracker. @code{bug-reference-prog-mode} is a variant of @code{bug-reference-mode} which highlights bug references only inside source code comments and strings. @vindex bug-reference-bug-regexp @vindex bug-reference-url-format For its working, bug reference mode needs to know the syntax of bug references (@code{bug-reference-bug-regexp}), and the URL of the tracker where bug reports can be looked up (@code{bug-reference-url-format}). Since those are typically different from project to project, it makes sense to specify them in @pxref{Directory Variables} or @pxref{File Variables}. For example, let's assume in our project, we usually write references to bug reports as bug#1234, or Bug-1234 and that this bug's page on the issue tracker is @url{https://project.org/issues/1234}, then these local variables section would do. @smallexample ;; Local Variables: ;; bug-reference-bug-regexp: "\\([Bb]ug[#-]\\([0-9]+\\)\\)" ;; bug-reference-url-format: "https://project.org/issues/%s" ;; End: @end smallexample The string captured by the first regexp group defines the bounds of the overlay bug-reference creates, i.e., the part which is highlighted and made clickable. The string captured by the second regexp group in @code{bug-reference-bug-regexp} is used to replace the @code{%s} template in the @code{bug-reference-url-format}. Note that @code{bug-reference-url-format} may also be a function in order to cater for more complex scenarios, e.g., when different parts of the bug reference have to be used to distinguish between issues and merge requests resulting in different URLs. @heading Automatic Setup @vindex bug-reference-auto-setup-functions If @code{bug-reference-mode} is activated, @code{bug-reference-mode-hook} has been run, and either @code{bug-reference-bug-regexp} or @code{bug-reference-url-format} is still @code{nil}, the mode will try to automatically find a suitable value for these two variables by calling the functions in @code{bug-reference-auto-setup-functions} one by one until one succeeds. @vindex bug-reference-setup-from-vc-alist @vindex bug-reference-forge-alist @vindex bug-reference-setup-from-mail-alist @vindex bug-reference-setup-from-irc-alist Right now, there are three types of setup functions. @enumerate @item Setup for version-controlled files configurable by the variables @code{bug-reference-forge-alist}, and @code{bug-reference-setup-from-vc-alist}. The defaults are able to set up GNU projects where @url{https://debbugs.gnu.org} is used as issue tracker and issues are usually referenced as @code{bug#13} (but many different notations are considered, too), as well as several other kinds of software forges such as GitLab, Gitea, SourceHut, and GitHub. If you deploy a self-hosted instance of such a forge, the easiest way to tell bug-reference about it is through @code{bug-reference-forge-alist}. @item Setup for email guessing from mail folder/mbox names, and mail header values configurable by the variable @code{bug-reference-setup-from-mail-alist}. The built-in news- and mailreader @ref{Gnus} and @ref{Rmail} are supported. @item Setup for IRC channels configurable by the variable @code{bug-reference-setup-from-irc-alist}. The built-in IRC clients Rcirc, @xref{Top, Rcirc,, rcirc, The Rcirc Manual}, and ERC, @xref{Top, ERC,, erc, The ERC Manual}, are supported. @end enumerate For almost all of those modes, it's enough to simply enable @code{bug-reference-mode}; only Rmail requires a slightly different setup. @smallexample ;; Use VC-based setup if file is under version control. (add-hook 'prog-mode-hook #'bug-reference-prog-mode) ;; Gnus (summary & article buffers) (add-hook 'gnus-mode-hook #'bug-reference-mode) ;; Rmail (add-hook 'rmail-show-message-hook #'bug-reference-mode-force-auto-setup) ;; Rcirc (add-hook 'rcirc-mode-hook #'bug-reference-mode) ;; ERC (add-hook 'erc-mode-hook #'bug-reference-mode) @end smallexample In the Rmail case, instead of the mode hook, the @code{rmail-show-message-hook} has to be used in combination with the function @code{bug-reference-mode-force-auto-setup} which activates @code{bug-reference-mode} and forces auto-setup. The reason is that with Rmail all messages reside in the same buffer but the setup needs to be performed whenever another messages is displayed. @heading Adding support for third-party packages @vindex bug-reference-auto-setup-functions Adding support for bug-reference auto-setup is usually quite straightforward: write a setup function of zero arguments which gathers the required information (e.g., List-Id/To/From/Cc mail header values in the case of a MUA), and then calls one of the following helper functions: @itemize @bullet @item @code{bug-reference-maybe-setup-from-vc}, which does the setup according to @code{bug-reference-setup-from-vc-alist}; @item @code{bug-reference-maybe-setup-from-mail}, which does the setup according to @code{bug-reference-setup-from-mail-alist}; and @item @code{bug-reference-maybe-setup-from-irc}, which does the setup according to @code{bug-reference-setup-from-irc-alist}. @end itemize A setup function should return non-@code{nil} if it could set up bug-reference mode, which is the case if the last thing the function does is call one of the helper functions above. Finally, the setup function has to be added to @code{bug-reference-auto-setup-functions}. Note that these auto-setup functions should check as a first step if they are applicable, e.g., by checking the value of @code{major-mode}. @heading Integration with the debbugs package @findex debbugs-browse-mode If your project's issues are tracked on the server @url{https://debbugs.gnu.org}, you can browse and reply to reports directly in Emacs using the @code{debbugs} package, which can be downloaded via the Package Menu (@pxref{Packages}). This package adds the minor mode @code{debbugs-browse-mode}, which can be activated on top of @code{bug-reference-mode} and @code{bug-reference-prog-mode} as follows: @smallexample (add-hook 'bug-reference-mode-hook 'debbugs-browse-mode) (add-hook 'bug-reference-prog-mode-hook 'debbugs-browse-mode) @end smallexample