emacs/doc/lispref/files.texi

@c -*-texinfo-*-
@c This is part of the GNU Emacs Lisp Reference Manual.
@c Copyright (C) 1990--1995, 1998--1999, 2001--2023 Free Software
@c Foundation, Inc.
@c See the file elisp.texi for copying conditions.
@node Files
@chapter Files

  This chapter describes the Emacs Lisp functions and variables to
find, create, view, save, and otherwise work with files and
directories.  A few other file-related functions are described in
@ref{Buffers}, and those related to backups and auto-saving are
described in @ref{Backups and Auto-Saving}.

  Many of the file functions take one or more arguments that are file
names.  A file name is a string.  Most of these functions expand file
name arguments using the function @code{expand-file-name}, so that
@file{~} is handled correctly, as are relative file names (including
@file{../} and the empty string).  @xref{File Name Expansion}.

  In addition, certain @dfn{magic} file names are handled specially.
For example, when a remote file name is specified, Emacs accesses the
file over the network via an appropriate protocol.  @xref{Remote
Files,, Remote Files, emacs, The GNU Emacs Manual}.  This handling is
done at a very low level, so you may assume that all the functions
described in this chapter accept magic file names as file name
arguments, except where noted.  @xref{Magic File Names}, for details.

  When file I/O functions signal Lisp errors, they usually use the
condition @code{file-error} (@pxref{Handling Errors}).  The error
message is in most cases obtained from the operating system, according
to locale @code{system-messages-locale}, and decoded using coding system
@code{locale-coding-system} (@pxref{Locales}).

@menu
* Visiting Files::           Reading files into Emacs buffers for editing.
* Saving Buffers::           Writing changed buffers back into files.
* Reading from Files::       Reading files into buffers without visiting.
* Writing to Files::         Writing new files from parts of buffers.
* File Locks::               Locking and unlocking files, to prevent
                               simultaneous editing by two people.
* Information about Files::  Testing existence, accessibility, size of files.
* Changing Files::           Renaming files, changing permissions, etc.
* Files and Storage::        Surviving power and media failures
* File Names::               Decomposing and expanding file names.
* Contents of Directories::  Getting a list of the files in a directory.
* Create/Delete Dirs::       Creating and Deleting Directories.
* Magic File Names::         Special handling for certain file names.
* Format Conversion::        Conversion to and from various file formats.
@end menu

@node Visiting Files
@section Visiting Files
@cindex finding files
@cindex visiting files

  Visiting a file means reading a file into a buffer.  Once this is
done, we say that the buffer is @dfn{visiting} that file, and call the
file @dfn{the visited file} of the buffer.

  A file and a buffer are two different things.  A file is information
recorded permanently in the computer (unless you delete it).  A
buffer, on the other hand, is information inside of Emacs that will
vanish at the end of the editing session (or when you kill the
buffer).  When a buffer is visiting a file, it contains information
copied from the file.  The copy in the buffer is what you modify with
editing commands.  Changes to the buffer do not change the file; to
make the changes permanent, you must @dfn{save} the buffer, which
means copying the altered buffer contents back into the file.

  Despite the distinction between files and buffers, people often
refer to a file when they mean a buffer and vice-versa.  Indeed, we
say, ``I am editing a file'', rather than, ``I am editing a buffer
that I will soon save as a file of the same name''.  Humans do not
usually need to make the distinction explicit.  When dealing with a
computer program, however, it is good to keep the distinction in mind.

@menu
* Visiting Functions::         The usual interface functions for visiting.
* Subroutines of Visiting::    Lower-level subroutines that they use.
@end menu

@node Visiting Functions
@subsection Functions for Visiting Files
@cindex visiting files, functions for
@cindex how to visit files

  This section describes the functions normally used to visit files.
For historical reasons, these functions have names starting with
@samp{find-} rather than @samp{visit-}.  @xref{Buffer File Name}, for
functions and variables that access the visited file name of a buffer or
that find an existing buffer by its visited file name.

  In a Lisp program, if you want to look at the contents of a file but
not alter it, the fastest way is to use @code{insert-file-contents} in a
temporary buffer.  Visiting the file is not necessary and takes longer.
@xref{Reading from Files}.

@deffn Command find-file filename &optional wildcards
This command selects a buffer visiting the file @var{filename},
using an existing buffer if there is one, and otherwise creating a
new buffer and reading the file into it.  It also returns that buffer.

Aside from some technical details, the body of the @code{find-file}
function is basically equivalent to:

@smallexample
(switch-to-buffer (find-file-noselect filename nil nil wildcards))
@end smallexample

@noindent
(See @code{switch-to-buffer} in @ref{Switching Buffers}.)

If @var{wildcards} is non-@code{nil}, which is always true in an
interactive call, then @code{find-file} expands wildcard characters in
@var{filename} and visits all the matching files.

When @code{find-file} is called interactively, it prompts for
@var{filename} in the minibuffer.
@end deffn

@deffn Command find-file-literally filename
This command visits @var{filename}, like @code{find-file} does, but it
does not perform any format conversions (@pxref{Format Conversion}),
character code conversions (@pxref{Coding Systems}), or end-of-line
conversions (@pxref{Coding System Basics, End of line conversion}).
The buffer visiting the file is made unibyte, and its major mode is
Fundamental mode, regardless of the file name.  File local variable
specifications  in the file (@pxref{File Local Variables}) are
ignored, and automatic decompression and adding a newline at the end
of the file due to @code{require-final-newline} (@pxref{Saving
Buffers, require-final-newline}) are also disabled.

Note that if Emacs already has a buffer visiting the same file
non-literally, it will not visit the same file literally, but instead
just switch to the existing buffer.  If you want to be sure of
accessing a file's contents literally, you should create a temporary
buffer and then read the file contents into it using
@code{insert-file-contents-literally} (@pxref{Reading from Files}).
@end deffn

@defun find-file-noselect filename &optional nowarn rawfile wildcards
This function is the guts of all the file-visiting functions.  It
returns a buffer visiting the file @var{filename}.  You may make the
buffer current or display it in a window if you wish, but this
function does not do so.

The function returns an existing buffer if there is one; otherwise it
creates a new buffer and reads the file into it.  When
@code{find-file-noselect} uses an existing buffer, it first verifies
that the file has not changed since it was last visited or saved in
that buffer.  If the file has changed, this function asks the user
whether to reread the changed file.  If the user says @samp{yes}, any
edits previously made in the buffer are lost.

Reading the file involves decoding the file's contents (@pxref{Coding
Systems}), including end-of-line conversion, and format conversion
(@pxref{Format Conversion}).  If @var{wildcards} is non-@code{nil},
then @code{find-file-noselect} expands wildcard characters in
@var{filename} and visits all the matching files.

This function displays warning or advisory messages in various peculiar
cases, unless the optional argument @var{nowarn} is non-@code{nil}.  For
example, if it needs to create a buffer, and there is no file named
@var{filename}, it displays the message @samp{(New file)} in the echo
area, and leaves the buffer empty.

The @code{find-file-noselect} function normally calls
@code{after-find-file} after reading the file (@pxref{Subroutines of
Visiting}).  That function sets the buffer major mode, parses local
variables, warns the user if there exists an auto-save file more recent
than the file just visited, and finishes by running the functions in
@code{find-file-hook}.

If the optional argument @var{rawfile} is non-@code{nil}, then
@code{after-find-file} is not called, and the
@code{find-file-not-found-functions} are not run in case of failure.
What's more, a non-@code{nil} @var{rawfile} value suppresses coding
system conversion and format conversion.

The @code{find-file-noselect} function usually returns the buffer that
is visiting the file @var{filename}.  But, if wildcards are actually
used and expanded, it returns a list of buffers that are visiting the
various files.

@example
@group
(find-file-noselect "/etc/fstab")
     @result{} #<buffer fstab>
@end group
@end example
@end defun

@deffn Command find-file-other-window filename &optional wildcards
This command selects a buffer visiting the file @var{filename}, but
does so in a window other than the selected window.  It may use
another existing window or split a window; see @ref{Switching
Buffers}.

When this command is called interactively, it prompts for
@var{filename}.
@end deffn

@deffn Command find-file-read-only filename &optional wildcards
This command selects a buffer visiting the file @var{filename}, like
@code{find-file}, but it marks the buffer as read-only.  @xref{Read Only
Buffers}, for related functions and variables.

When this command is called interactively, it prompts for
@var{filename}.
@end deffn

@defopt find-file-wildcards
If this variable is non-@code{nil}, then the various @code{find-file}
commands check for wildcard characters and visit all the files that
match them (when invoked interactively or when their @var{wildcards}
argument is non-@code{nil}).  If this option is @code{nil}, then
the @code{find-file} commands ignore their @var{wildcards} argument
and never treat wildcard characters specially.
@end defopt

@defopt find-file-hook
The value of this variable is a list of functions to be called after a
file is visited.  The file's local-variables specification (if any) will
have been processed before the hooks are run.  The buffer visiting the
file is current when the hook functions are run.

This variable is a normal hook.  @xref{Hooks}.
@end defopt

@defvar find-file-not-found-functions
The value of this variable is a list of functions to be called when
@code{find-file} or @code{find-file-noselect} is passed a nonexistent
file name.  @code{find-file-noselect} calls these functions as soon as
it detects a nonexistent file.  It calls them in the order of the list,
until one of them returns non-@code{nil}.  @code{buffer-file-name} is
already set up.

This is not a normal hook because the values of the functions are
used, and in many cases only some of the functions are called.
@end defvar

@defvar find-file-literally
This buffer-local variable, if set to a non-@code{nil} value, makes
@code{save-buffer} behave as if the buffer were visiting its file
literally, i.e., without conversions of any kind.  The command
@code{find-file-literally} sets this variable's local value, but other
equivalent functions and commands can do that as well, e.g., to avoid
automatic addition of a newline at the end of the file.  This variable
is permanent local, so it is unaffected by changes of major modes.
@end defvar

@node Subroutines of Visiting
@subsection Subroutines of Visiting

  The @code{find-file-noselect} function uses two important subroutines
which are sometimes useful in user Lisp code: @code{create-file-buffer}
and @code{after-find-file}.  This section explains how to use them.

@c FIXME This does not describe the default behavior, because
@c uniquify is enabled by default and advises this function.
@c This is confusing.  uniquify should be folded into the function proper.
@defun create-file-buffer filename
This function creates a suitably named buffer for visiting
@var{filename}, and returns it.  It uses @var{filename} (sans directory)
as the name if that name is free; otherwise, it appends a string such as
@samp{<2>} to get an unused name.  See also @ref{Creating Buffers}.
Note that the @file{uniquify} library affects the result of this
function.  @xref{Uniquify,,, emacs, The GNU Emacs Manual}.

@strong{Please note:} @code{create-file-buffer} does @emph{not}
associate the new buffer with a file and does not select the buffer.
It also does not use the default major mode.

@example
@group
(create-file-buffer "foo")
     @result{} #<buffer foo>
@end group
@group
(create-file-buffer "foo")
     @result{} #<buffer foo<2>>
@end group
@group
(create-file-buffer "foo")
     @result{} #<buffer foo<3>>
@end group
@end example

This function is used by @code{find-file-noselect}.
It uses @code{generate-new-buffer} (@pxref{Creating Buffers}).
@end defun

@defun after-find-file &optional error warn noauto after-find-file-from-revert-buffer nomodes
This function sets the buffer major mode, and parses local variables
(@pxref{Auto Major Mode}).  It is called by @code{find-file-noselect}
and by the default revert function (@pxref{Reverting}).

@cindex new file message
@cindex file open error
If reading the file got an error because the file does not exist, but
its directory does exist, the caller should pass a non-@code{nil} value
for @var{error}.  In that case, @code{after-find-file} issues a warning:
@samp{(New file)}.  For more serious errors, the caller should usually not
call @code{after-find-file}.

If @var{warn} is non-@code{nil}, then this function issues a warning
if an auto-save file exists and is more recent than the visited file.

If @var{noauto} is non-@code{nil}, that says not to enable or disable
Auto-Save mode.  The mode remains enabled if it was enabled before.

If @var{after-find-file-from-revert-buffer} is non-@code{nil}, that
means this call was from @code{revert-buffer}.  This has no direct
effect, but some mode functions and hook functions check the value
of this variable.

If @var{nomodes} is non-@code{nil}, that means don't alter the buffer's
major mode, don't process local variables specifications in the file,
and don't run @code{find-file-hook}.  This feature is used by
@code{revert-buffer} in some cases.

The last thing @code{after-find-file} does is call all the functions
in the list @code{find-file-hook}.
@end defun

@node Saving Buffers
@section Saving Buffers
@cindex saving buffers

  When you edit a file in Emacs, you are actually working on a buffer
that is visiting that file---that is, the contents of the file are
copied into the buffer and the copy is what you edit.  Changes to the
buffer do not change the file until you @dfn{save} the buffer, which
means copying the contents of the buffer into the file.  Buffers which
are not visiting a file can still be ``saved'', in a sense, using
functions in the buffer-local @code{write-contents-functions} hook.

@deffn Command save-buffer &optional backup-option
This function saves the contents of the current buffer in its visited
file if the buffer has been modified since it was last visited or saved.
Otherwise it does nothing.

@code{save-buffer} is responsible for making backup files.  Normally,
@var{backup-option} is @code{nil}, and @code{save-buffer} makes a backup
file only if this is the first save since visiting the file.  Other
values for @var{backup-option} request the making of backup files in
other circumstances:

@itemize @bullet
@item
With an argument of 4 or 64, reflecting 1 or 3 @kbd{C-u}'s, the
@code{save-buffer} function marks this version of the file to be
backed up when the buffer is next saved.

@item
With an argument of 16 or 64, reflecting 2 or 3 @kbd{C-u}'s, the
@code{save-buffer} function unconditionally backs up the previous
version of the file before saving it.

@item
With an argument of 0, unconditionally do @emph{not} make any backup file.
@end itemize
@end deffn

@deffn Command save-some-buffers &optional save-silently-p pred
@anchor{Definition of save-some-buffers}
This command saves some modified file-visiting buffers.  Normally it
asks the user about each buffer.  But if @var{save-silently-p} is
non-@code{nil}, it saves all the file-visiting buffers without
querying the user.

@vindex save-some-buffers-default-predicate
The optional @var{pred} argument provides a predicate that controls
which buffers to ask about (or to save silently if
@var{save-silently-p} is non-@code{nil}).  If @var{pred} is
@code{nil}, that means to use the value of
@code{save-some-buffers-default-predicate} instead of @var{pred}.  If
the result is @code{nil}, it means ask only about file-visiting
buffers.  If it is @code{t}, that means also offer to save certain
other non-file buffers---those that have a non-@code{nil} buffer-local
value of @code{buffer-offer-save} (@pxref{Killing Buffers}).  A user
who says @samp{yes} to saving a non-file buffer is asked to specify
the file name to use.  The @code{save-buffers-kill-emacs} function
passes the value @code{t} for @var{pred}.

If the predicate is neither @code{t} nor @code{nil}, then it should be
a function of no arguments.  It will be called in each buffer to decide
whether to offer to save that buffer.  If it returns a non-@code{nil}
value in a certain buffer, that means do offer to save that buffer.
@end deffn

@deffn Command write-file filename &optional confirm
@anchor{Definition of write-file}
This function writes the current buffer into file @var{filename}, makes
the buffer visit that file, and marks it not modified.  Then it renames
the buffer based on @var{filename}, appending a string like @samp{<2>}
if necessary to make a unique buffer name.  It does most of this work by
calling @code{set-visited-file-name} (@pxref{Buffer File Name}) and
@code{save-buffer}.

If @var{confirm} is non-@code{nil}, that means to ask for confirmation
before overwriting an existing file.  Interactively, confirmation is
required, unless the user supplies a prefix argument.

If @var{filename} is a directory name (@pxref{Directory Names}),
@code{write-file} uses the name of the visited file, in directory
@var{filename}.  If the buffer is not visiting a file, it uses the
buffer name instead.
@end deffn

  Saving a buffer runs several hooks.  It also performs format
conversion (@pxref{Format Conversion}).  Note that these hooks,
described below, are only run by @code{save-buffer}, they are not run
by other primitives and functions that write buffer text to files, and
in particular auto-saving (@pxref{Auto-Saving}) doesn't run these
hooks.

@defvar write-file-functions
The value of this variable is a list of functions to be called before
writing out a buffer to its visited file.  If one of them returns
non-@code{nil}, the file is considered already written and the rest of
the functions are not called, nor is the usual code for writing the file
executed.

If a function in @code{write-file-functions} returns non-@code{nil}, it
is responsible for making a backup file (if that is appropriate).
To do so, execute the following code:

@example
(or buffer-backed-up (backup-buffer))
@end example

You might wish to save the file modes value returned by
@code{backup-buffer} and use that (if non-@code{nil}) to set the mode
bits of the file that you write.  This is what @code{save-buffer}
normally does.  @xref{Making Backups,, Making Backup Files}.

The hook functions in @code{write-file-functions} are also responsible
for encoding the data (if desired): they must choose a suitable coding
system and end-of-line conversion (@pxref{Lisp and Coding Systems}),
perform the encoding (@pxref{Explicit Encoding}), and set
@code{last-coding-system-used} to the coding system that was used
(@pxref{Encoding and I/O}).

If you set this hook locally in a buffer, it is assumed to be
associated with the file or the way the contents of the buffer were
obtained.  Thus the variable is marked as a permanent local, so that
changing the major mode does not alter a buffer-local value.  On the
other hand, calling @code{set-visited-file-name} will reset it.
If this is not what you want, you might like to use
@code{write-contents-functions} instead.

Even though this is not a normal hook, you can use @code{add-hook} and
@code{remove-hook} to manipulate the list.  @xref{Hooks}.
@end defvar

@defvar write-contents-functions
This works just like @code{write-file-functions}, but it is intended
for hooks that pertain to the buffer's contents, not to the particular
visited file or its location, and can be used to create arbitrary save
processes for buffers that aren't visiting files at all.  Such hooks
are usually set up by major modes, as buffer-local bindings for this
variable.  This variable automatically becomes buffer-local whenever
it is set; switching to a new major mode always resets this variable,
but calling @code{set-visited-file-name} does not.

If any of the functions in this hook returns non-@code{nil}, the file
is considered already written and the rest are not called and neither
are the functions in @code{write-file-functions}.

When using this hook to save buffers that are not visiting files (for
instance, special-mode buffers), keep in mind that, if the function
fails to save correctly and returns a @code{nil} value,
@code{save-buffer} will go on to prompt the user for a file to save
the buffer in.  If this is undesirable, consider having the function
fail by raising an error.
@end defvar

@defopt before-save-hook
This normal hook runs before a buffer is saved in its visited file,
regardless of whether that is done normally or by one of the hooks
described above.  For instance, the @file{copyright.el} program uses
this hook to make sure the file you are saving has the current year in
its copyright notice.
@end defopt

@defopt after-save-hook
This normal hook runs after a buffer has been saved in its visited file.
@end defopt

@defopt file-precious-flag
If this variable is non-@code{nil}, then @code{save-buffer} protects
against I/O errors while saving by writing the new file to a temporary
name instead of the name it is supposed to have, and then renaming it to
the intended name after it is clear there are no errors.  This procedure
prevents problems such as a lack of disk space from resulting in an
invalid file.

As a side effect, backups are necessarily made by copying.  @xref{Rename
or Copy}.  Yet, at the same time, saving a precious file always breaks
all hard links between the file you save and other file names.

Some modes give this variable a non-@code{nil} buffer-local value
in particular buffers.
@end defopt

@defopt require-final-newline
This variable determines whether files may be written out that do
@emph{not} end with a newline.  If the value of the variable is
@code{t}, then @code{save-buffer} silently adds a newline at the end
of the buffer whenever it does not already end in one.  If the value
is @code{visit}, Emacs adds a missing newline just after it visits the
file.  If the value is @code{visit-save}, Emacs adds a missing newline
both on visiting and on saving.  For any other non-@code{nil} value,
@code{save-buffer} asks the user whether to add a newline each time
the case arises.

If the value of the variable is @code{nil}, then @code{save-buffer}
doesn't add newlines at all.  @code{nil} is the default value, but a few
major modes set it to @code{t} in particular buffers.
@end defopt

  See also the function @code{set-visited-file-name} (@pxref{Buffer File
Name}).

@node Reading from Files
@section Reading from Files
@cindex reading from files

  To copy the contents of a file into a buffer, use the function
@code{insert-file-contents}.  (Don't use the command
@code{insert-file} in a Lisp program, as that sets the mark.)

@defun insert-file-contents filename &optional visit beg end replace
This function inserts the contents of file @var{filename} into the
current buffer after point.  It returns a list of the absolute file name
and the length of the data inserted.  An error is signaled if
@var{filename} is not the name of a file that can be read.

This function checks the file contents against the defined file
formats, and converts the file contents if appropriate and also calls
the functions in the list @code{after-insert-file-functions}.
@xref{Format Conversion}.  Normally, one of the functions in the
@code{after-insert-file-functions} list determines the coding system
(@pxref{Coding Systems}) used for decoding the file's contents,
including end-of-line conversion.  However, if the file contains null
bytes, it is by default visited without any code conversions.
@xref{Lisp and Coding Systems, inhibit-null-byte-detection}.

If @var{visit} is non-@code{nil}, this function additionally marks the
buffer as unmodified and sets up various fields in the buffer so that it
is visiting the file @var{filename}: these include the buffer's visited
file name and its last save file modtime.  This feature is used by
@code{find-file-noselect} and you probably should not use it yourself.

If @var{beg} and @var{end} are non-@code{nil}, they should be numbers
that are byte offsets specifying the portion of the file to insert.
In this case, @var{visit} must be @code{nil}.  For example,

@example
(insert-file-contents filename nil 0 500)
@end example

@noindent
inserts the characters coded by the first 500 bytes of a file.

If @var{beg} or @var{end} happens to be in the middle of a character's
multibyte sequence, Emacs's character code conversion will insert one
or more eight-bit characters (a.k.a.@: ``raw bytes'')
(@pxref{Character Sets}) into the buffer.  If you want to read part of
a file this way, we recommend to bind @code{coding-system-for-read} to
a suitable value around the call to this function (@pxref{Specifying
Coding Systems}), and to write Lisp code which will check for raw
bytes at the boundaries, read the entire sequence of these bytes, and
convert them back to valid characters.

If the argument @var{replace} is non-@code{nil}, it means to replace the
contents of the buffer (actually, just the accessible portion) with the
contents of the file.  This is better than simply deleting the buffer
contents and inserting the whole file, because (1) it preserves some
marker positions and (2) it puts less data in the undo list.

It is possible to read a special file (such as a FIFO or an I/O
device) with @code{insert-file-contents}, as long as @var{replace},
and @var{visit} and @var{beg} are @code{nil}.  However, you should
normally use an @var{end} argument for these files to avoid inserting
(potentially) unlimited data into the buffer (for instance, when
inserting data from @file{/dev/urandom}).
@end defun

@defun insert-file-contents-literally filename &optional visit beg end replace
This function works like @code{insert-file-contents} except that each
byte in the file is handled separately, being converted into an
eight-bit character if needed.  It does not run
@code{after-insert-file-functions}, and does not do format decoding,
character code conversion, automatic uncompression, and so on.
@end defun

If you want to pass a file name to another process so that another
program can read the file, use the function @code{file-local-copy}; see
@ref{Magic File Names}.

@node Writing to Files
@section Writing to Files
@cindex writing to files

  You can write the contents of a buffer, or part of a buffer, directly
to a file on disk using the @code{append-to-file} and
@code{write-region} functions.  Don't use these functions to write to
files that are being visited; that could cause confusion in the
mechanisms for visiting.

@deffn Command append-to-file start end filename
This function appends the contents of the region delimited by
@var{start} and @var{end} in the current buffer to the end of file
@var{filename}.  If that file does not exist, it is created.  This
function returns @code{nil}.

An error is signaled if you cannot write or create @var{filename}.

When called from Lisp, this function is completely equivalent to:

@example
(write-region start end filename t)
@end example
@end deffn

@deffn Command write-region start end filename &optional append visit lockname mustbenew
This function writes the region delimited by @var{start} and @var{end}
in the current buffer into the file specified by @var{filename}.

If @var{start} is @code{nil}, then the command writes the entire buffer
contents (@emph{not} just the accessible portion) to the file and
ignores @var{end}.

If @var{start} is a string, then @code{write-region} writes or appends
that string, rather than text from the buffer.  @var{end} is ignored in
this case.

If @var{append} is non-@code{nil}, then the specified text is appended
to the existing file contents (if any).  If @var{append} is a
number, @code{write-region} seeks to that byte offset from the start
of the file and writes the data from there.

If @var{mustbenew} is non-@code{nil}, then @code{write-region} asks
for confirmation if @var{filename} names an existing file.  If
@var{mustbenew} is the symbol @code{excl}, then @code{write-region}
does not ask for confirmation, but instead it signals an error
@code{file-already-exists} if the file already exists.  Although
@code{write-region} normally follows a symbolic link and creates the
pointed-to file if the symbolic link is dangling, it does not follow
symbolic links if @var{mustbenew} is @code{excl}.

The test for an existing file, when @var{mustbenew} is @code{excl}, uses
a special system feature.  At least for files on a local disk, there is
no chance that some other program could create a file of the same name
before Emacs does, without Emacs's noticing.

If @var{visit} is @code{t}, then Emacs establishes an association
between the buffer and the file: the buffer is then visiting that file.
It also sets the last file modification time for the current buffer to
@var{filename}'s modtime, and marks the buffer as not modified.  This
feature is used by @code{save-buffer}, but you probably should not use
it yourself.

If @var{visit} is a string, it specifies the file name to visit.  This
way, you can write the data to one file (@var{filename}) while recording
the buffer as visiting another file (@var{visit}).  The argument
@var{visit} is used in the echo area message and also for file locking;
@var{visit} is stored in @code{buffer-file-name}.  This feature is used
to implement @code{file-precious-flag}; don't use it yourself unless you
really know what you're doing.

The optional argument @var{lockname}, if non-@code{nil}, specifies the
file name to use for purposes of locking and unlocking, overriding
@var{filename} and @var{visit} for that purpose.

The function @code{write-region} converts the data which it writes to
the appropriate file formats specified by @code{buffer-file-format}
and also calls the functions in the list
@code{write-region-annotate-functions}.
@xref{Format Conversion}.

Normally, @code{write-region} displays the message @samp{Wrote
@var{filename}} in the echo area.  This message is inhibited if
@var{visit} is neither @code{t} nor @code{nil} nor a string, or if
Emacs is operating in batch mode (@pxref{Batch Mode}).  This
feature is useful for programs that use files for internal purposes,
files that the user does not need to know about.
@end deffn

@defvar write-region-inhibit-fsync
If this variable's value is @code{nil}, @code{write-region} uses the
@code{fsync} system call after writing a file.  If the value is
@code{t}, Emacs does not use @code{fsync}.  The default value is
@code{t}.  @xref{Files and Storage}.
@end defvar

@defmac with-temp-file file body@dots{}
@anchor{Definition of with-temp-file}
The @code{with-temp-file} macro evaluates the @var{body} forms with a
temporary buffer as the current buffer; then, at the end, it writes the
buffer contents into file @var{file}.  It kills the temporary buffer
when finished, restoring the buffer that was current before the
@code{with-temp-file} form.  Then it returns the value of the last form
in @var{body}.

The current buffer is restored even in case of an abnormal exit via
@code{throw} or error (@pxref{Nonlocal Exits}).

Like @code{with-temp-buffer} (@pxref{Definition of with-temp-buffer,,
Current Buffer}), the temporary buffer used by this macro does not run
the hooks @code{kill-buffer-hook}, @code{kill-buffer-query-functions}
(@pxref{Killing Buffers}), and @code{buffer-list-update-hook}
(@pxref{Buffer List}).
@end defmac

@node File Locks
@section File Locks
@cindex file locks
@cindex lock file
@cindex .#, lock file names

  When two users edit the same file at the same time, they are likely
to interfere with each other.  Emacs tries to prevent this situation
from arising by recording a @dfn{file lock} when a file is being
modified.
Emacs can then detect the first attempt to modify a buffer visiting a
file that is locked by another Emacs job, and ask the user what to do.
The file lock is really a file, a symbolic link with a special name,
stored in the same directory as the file you are editing.  The name is
constructed by prepending @file{.#} to the file name of the buffer.
The target of the symbolic link will be of the form
@code{@var{user}@@@var{host}.@var{pid}:@var{boot}}, where @var{user}
is replaced with the current username (from @code{user-login-name}),
@var{host} with the name of the host where Emacs is running (from
@code{system-name}), @var{pid} with Emacs's process id, and @var{boot}
with the time since the last reboot.  @code{:@var{boot}} is omitted if
the boot time is unavailable.  (On file systems that do not support
symbolic links, a regular file is used instead, with contents of the
form @code{@var{user}@@@var{host}.@var{pid}:@var{boot}}.)

  When you access files using NFS, there may be a small probability that
you and another user will both lock the same file simultaneously.
If this happens, it is possible for the two users to make changes
simultaneously, but Emacs will still warn the user who saves second.
Also, the detection of modification of a buffer visiting a file changed
on disk catches some cases of simultaneous editing; see
@ref{Modification Time}.

@defun file-locked-p filename
This function returns @code{nil} if the file @var{filename} is not
locked.  It returns @code{t} if it is locked by this Emacs process, and
it returns the name of the user who has locked it if it is locked by
some other job.

@example
@group
(file-locked-p "foo")
     @result{} nil
@end group
@end example
@end defun

@defun lock-buffer &optional filename
This function locks the file @var{filename}, if the current buffer is
modified.  The argument @var{filename} defaults to the current buffer's
visited file.  Nothing is done if the current buffer is not visiting a
file, or is not modified, or if the option @code{create-lockfiles} is
@code{nil}.
@end defun

@defun unlock-buffer
This function unlocks the file being visited in the current buffer,
if the buffer is modified.  If the buffer is not modified, then
the file should not be locked, so this function does nothing.  It also
does nothing if the current buffer is not visiting a file, or is not locked.
This function handles file system errors by calling @code{display-warning}
and otherwise ignores the error.
@end defun

@defopt create-lockfiles
If this variable is @code{nil}, Emacs does not lock files.
@end defopt

@defopt lock-file-name-transforms
By default, Emacs creates the lock files in the same directory as the
files that are being locked.  This can be changed by customizing this
variable.  Is has the same syntax as
@code{auto-save-file-name-transforms} (@pxref{Auto-Saving}).  For
instance, to make Emacs write all the lock files to @file{/var/tmp/},
you could say something like:

@lisp
(setq lock-file-name-transforms
      '(("\\`/.*/\\([^/]+\\)\\'" "/var/tmp/\\1" t)))
@end lisp
@end defopt

@defun ask-user-about-lock file other-user
This function is called when the user tries to modify @var{file}, but it
is locked by another user named @var{other-user}.  The default
definition of this function asks the user to say what to do.  The value
this function returns determines what Emacs does next:

@itemize @bullet
@item
A value of @code{t} says to grab the lock on the file.  Then
this user may edit the file and @var{other-user} loses the lock.

@item
A value of @code{nil} says to ignore the lock and let this
user edit the file anyway.

@item
@kindex file-locked
This function may instead signal a @code{file-locked} error, in which
case the change that the user was about to make does not take place.

The error message for this error looks like this:

@example
@error{} File is locked: @var{file} @var{other-user}
@end example

@noindent
where @code{file} is the name of the file and @var{other-user} is the
name of the user who has locked the file.
@end itemize

If you wish, you can replace the @code{ask-user-about-lock} function
with your own version that makes the decision in another way.
@end defun

@defopt remote-file-name-inhibit-locks
You can prevent the creation of remote lock files by setting the
variable @code{remote-file-name-inhibit-locks} to @code{t}.
@end defopt

@deffn Command lock-file-mode
This command, called interactively, toggles the local value of
@code{create-lockfiles} in the current buffer.
@end deffn

@node Information about Files
@section Information about Files
@cindex file, information about

  This section describes the functions for retrieving various types of
information about files (or directories or symbolic links), such as
whether a file is readable or writable, and its size.  These functions
all take arguments which are file names.  Except where noted, these
arguments need to specify existing files, or an error is signaled.

@cindex file names, trailing whitespace
@cindex trailing blanks in file names
  Be careful with file names that end in spaces.  On some filesystems
(notably, MS-Windows), trailing whitespace characters in file names
are silently and automatically ignored.

@menu
* Testing Accessibility::   Is a given file readable?  Writable?
* Kinds of Files::          Is it a directory?  A symbolic link?
* Truenames::               Eliminating symbolic links from a file name.
* File Attributes::         File sizes, modification times, etc.
* Extended Attributes::     Extended file attributes for access control.
* Locating Files::          How to find a file in standard places.
@end menu

@node Testing Accessibility
@subsection Testing Accessibility
@cindex accessibility of a file
@cindex file accessibility

  These functions test for permission to access a file for reading,
writing, or execution.  Unless explicitly stated otherwise, they
follow symbolic links.  @xref{Kinds of Files}.

  On some operating systems, more complex sets of access permissions
can be specified, via mechanisms such as Access Control Lists (ACLs).
@xref{Extended Attributes}, for how to query and set those
permissions.

@defun file-exists-p filename
This function returns @code{t} if a file named @var{filename} appears
to exist.  This does not mean you can necessarily read the file, only
that you can probably find out its attributes.  (On GNU and other POSIX-like
systems, this is true if the file exists and you have execute
permission on the containing directories, regardless of the
permissions of the file itself.)

If the file does not exist, or if there was trouble determining
whether the file exists, this function returns @code{nil}.

@cindex empty file name, and @code{file-exists-p}
Since a file name that is an empty string is interpreted relative to
the current buffer's default directory (@pxref{Relative File Names}),
calling @code{file-exists-p} with an argument that is an empty string
will report about the buffer's default directory.

@cindex dangling symlinks, testing for existence
Directories are files, so @code{file-exists-p} can return @code{t}
when given a directory.  However, because @code{file-exists-p} follows
symbolic links, it returns @code{t} for a symbolic link name only if
the target of the link exists; if your Lisp program needs to consider
@dfn{dangling symlinks} whose target doesn't exist as existing files,
use @code{file-attributes} (@pxref{File Attributes}) instead of
@code{file-exists-p}.
@end defun

@defun file-readable-p filename
This function returns @code{t} if a file named @var{filename} exists
and you can read it.  It returns @code{nil} otherwise.
@end defun

@defun file-executable-p filename
This function returns @code{t} if a file named @var{filename} exists
and you can execute it.  It returns @code{nil} otherwise.  On GNU and
other POSIX-like systems, if the file is a directory, execute
permission means you can check the existence and attributes of files
inside the directory, and open those files if their modes permit.
@end defun

@defun file-writable-p filename
This function returns @code{t} if the file @var{filename} can be written
or created by you, and @code{nil} otherwise.  A file is writable if the
file exists and you can write it.  It is creatable if it does not exist,
but its parent directory does exist and you can write in that
directory.

In the example below, @file{foo} is not writable because the parent
directory does not exist, even though the user could create such a
directory.

@example
@group
(file-writable-p "~/no-such-dir/foo")
     @result{} nil
@end group
@end example
@end defun

@defun file-accessible-directory-p dirname
This function returns @code{t} if you have permission to open existing
files in the directory whose name as a file is @var{dirname};
otherwise (e.g., if there is no such directory), it returns @code{nil}.
The value of @var{dirname} may be either a directory name (such as
@file{/foo/}) or the file name of a file which is a directory
(such as @file{/foo}, without the final slash).

For example, from the following we deduce that any attempt to read a
file in @file{/foo/} will give an error:

@example
(file-accessible-directory-p "/foo")
     @result{} nil
@end example
@end defun

@defmac with-existing-directory body@dots{}
This macro ensures that @code{default-directory} is bound to an
existing directory before executing @var{body}.  If
@code{default-directory} already exists, that's preferred, and
otherwise some other directory is used.  This macro can be useful, for
instance, when calling an external command that requires that it's
running in a directory that exists.  The chosen directory is not
guaranteed to be writable.
@end defmac

@defun access-file filename string
If you can read @var{filename} this function returns @code{nil};
otherwise it signals an error
using @var{string} as the error message text.
@end defun

@defun file-ownership-preserved-p filename &optional group
This function returns @code{t} if deleting the file @var{filename} and
then creating it anew would keep the file's owner unchanged.  It also
returns @code{t} for nonexistent files.

If the optional argument @var{group} is non-@code{nil}, this function
also checks that the file's group would be unchanged.

This function does not follow symbolic links.
@end defun

@defun file-modes filename &optional flag
@cindex mode bits
@cindex file permissions
@cindex permissions, file
@cindex file modes
This function returns the @dfn{mode bits} of @var{filename}---an
integer summarizing its read, write, and execution permissions.
This function follows symbolic links.  If the file does not exist, the
return value is @code{nil}.

@xref{File permissions,,, coreutils, The @sc{gnu} @code{Coreutils}
Manual}, for a description of mode bits.  For example, if the
low-order bit is 1, the file is executable by all users; if the
second-lowest-order bit is 1, the file is writable by all users; etc.
The highest possible value is 4095 (7777 octal), meaning that everyone
has read, write, and execute permission, the @acronym{SUID} bit is set
for both others and group, and the sticky bit is set.

By default this function follows symbolic links.  However, if the
optional argument @var{flag} is the symbol @code{nofollow}, this
function does not follow @var{filename} if it is a symbolic link;
this can help prevent inadvertently obtaining the mode bits of a file
somewhere else, and is more consistent with @code{file-attributes}
(@pxref{File Attributes}).

@xref{Changing Files}, for the @code{set-file-modes} function, which
can be used to set these permissions.

@example
@group
(file-modes "~/junk/diffs" 'nofollow)
     @result{} 492               ; @r{Decimal integer.}
@end group
@group
(format "%o" 492)
     @result{} "754"             ; @r{Convert to octal.}
@end group

@group
(set-file-modes "~/junk/diffs" #o666 'nofollow)
     @result{} nil
@end group

@group
$ ls -l diffs
-rw-rw-rw- 1 lewis lewis 3063 Oct 30 16:00 diffs
@end group
@end example

@cindex MS-DOS and file modes
@cindex file modes and MS-DOS
@strong{MS-DOS note:} On MS-DOS, there is no such thing as an
executable file mode bit.  So @code{file-modes} considers a file
executable if its name ends in one of the standard executable
extensions, such as @file{.com}, @file{.bat}, @file{.exe}, and some
others.  Files that begin with the POSIX-standard @samp{#!} signature,
such as shell and Perl scripts, are also considered executable.
Directories are also reported as executable, for compatibility with
POSIX@.  These conventions are also followed by @code{file-attributes}
(@pxref{File Attributes}).
@end defun

@node Kinds of Files
@subsection Distinguishing Kinds of Files
@cindex file classification
@cindex classification of file types
@cindex symbolic links

  This section describes how to distinguish various kinds of files, such
as directories, symbolic links, and ordinary files.

  Symbolic links are ordinarily followed wherever they appear.  For
example, to interpret the file name @file{a/b/c}, any of @file{a},
@file{a/b}, and @file{a/b/c} can be symbolic links that are followed,
possibly recursively if the link targets are themselves symbolic
links.  However, a few functions do not follow symbolic links at the
end of a file name (@file{a/b/c} in this example).  Such a function
is said to @dfn{not follow symbolic links}.

@defun file-symlink-p filename
@cindex symbolic links
If the file @var{filename} is a symbolic link, this function does not
follow it and instead returns its link target
as a string.  (The link target string is not necessarily the full
absolute file name of the target; determining the full file name that
the link points to is nontrivial, see below.)

If the file @var{filename} is not a symbolic link, or does not exist,
or if there is trouble determining whether it is a symbolic link,
@code{file-symlink-p} returns @code{nil}.

Here are a few examples of using this function:

@example
@group
(file-symlink-p "not-a-symlink")
     @result{} nil
@end group
@group
(file-symlink-p "sym-link")
     @result{} "not-a-symlink"
@end group
@group
(file-symlink-p "sym-link2")
     @result{} "sym-link"
@end group
@group
(file-symlink-p "/bin")
     @result{} "/pub/bin"
@end group
@end example

Note that in the third example, the function returned @file{sym-link},
but did not proceed to resolve it, although that file is itself a
symbolic link.  That is because this function does not follow symbolic
links---the process of following the symbolic links does not apply to
the last component of the file name.

The string that this function returns is what is recorded in the
symbolic link; it may or may not include any leading directories.
This function does @emph{not} expand the link target to produce a
fully-qualified file name, and in particular does not use the leading
directories, if any, of the @var{filename} argument if the link target
is not an absolute file name.  Here's an example:

@example
@group
(file-symlink-p "/foo/bar/baz")
     @result{} "some-file"
@end group
@end example

@noindent
Here, although @file{/foo/bar/baz} was given as a fully-qualified file
name, the result is not, and in fact does not have any leading
directories at all.  And since @file{some-file} might itself be a
symbolic link, you cannot simply prepend leading directories to it,
nor even naively use @code{expand-file-name} (@pxref{File Name
Expansion}) to produce its absolute file name.

For this reason, this function is seldom useful if you need to
determine more than just the fact that a file is or isn't a symbolic
link.  If you actually need the file name of the link target, use
@code{file-chase-links} or @code{file-truename}, described in
@ref{Truenames}.
@end defun

@defun file-directory-p filename
This function returns @code{t} if @var{filename} is the name of an
existing directory.  It returns @code{nil} if @var{filename} does
not name a directory, or if there is trouble determining whether
it is a directory.
This function follows symbolic links.

@example
@group
(file-directory-p "~rms")
     @result{} t
@end group
@group
(file-directory-p "~rms/lewis/files.texi")
     @result{} nil
@end group
@group
(file-directory-p "~rms/lewis/no-such-file")
     @result{} nil
@end group
@group
(file-directory-p "$HOME")
     @result{} nil
@end group
@group
(file-directory-p
 (substitute-in-file-name "$HOME"))
     @result{} t
@end group
@end example
@end defun

@defun file-regular-p filename
This function returns @code{t} if the file @var{filename} exists and is
a regular file (not a directory, named pipe, terminal, or
other I/O device).
It returns @code{nil} if @var{filename} does not exist or is not a regular
file, or if there is trouble determining whether it is a regular file.
This function follows symbolic links.
@end defun

@node Truenames
@subsection Truenames
@cindex truename (of file)

  The @dfn{truename} of a file is the name that you get by following
symbolic links at all levels until none remain, then simplifying away
@samp{.}@: and @samp{..}@: appearing as name components.  This results
in a sort of canonical name for the file.  A file does not always have a
unique truename; the number of distinct truenames a file has is equal to
the number of hard links to the file.  However, truenames are useful
because they eliminate symbolic links as a cause of name variation.

@defun file-truename filename
This function returns the truename of the file @var{filename}.  If the
argument is not an absolute file name, this function first expands it
against @code{default-directory}.

This function does not expand environment variables.  Only
@code{substitute-in-file-name} does that.  @xref{Definition of
substitute-in-file-name}.

If you may need to follow symbolic links preceding @samp{..}@:
appearing as a name component, call @code{file-truename} without prior
direct or indirect calls to @code{expand-file-name}.  Otherwise, the
file name component immediately preceding @samp{..} will be
simplified away before @code{file-truename} is called.  To
eliminate the need for a call to @code{expand-file-name},
@code{file-truename} handles @samp{~} in the same way that
@code{expand-file-name} does.

If the target of a symbolic links has remote file name syntax,
@code{file-truename} returns it quoted.  @xref{File Name Expansion,,
Functions that Expand Filenames}.
@end defun

@defun file-chase-links filename &optional limit
This function follows symbolic links, starting with @var{filename},
until it finds a file name which is not the name of a symbolic link.
Then it returns that file name.  This function does @emph{not} follow
symbolic links at the level of parent directories.

If you specify a number for @var{limit}, then after chasing through
that many links, the function just returns what it has even if that is
still a symbolic link.
@end defun

  To illustrate the difference between @code{file-chase-links} and
@code{file-truename}, suppose that @file{/usr/foo} is a symbolic link to
the directory @file{/home/foo}, and @file{/home/foo/hello} is an
ordinary file (or at least, not a symbolic link) or nonexistent.  Then
we would have:

@example
(file-chase-links "/usr/foo/hello")
     ;; @r{This does not follow the links in the parent directories.}
     @result{} "/usr/foo/hello"
(file-truename "/usr/foo/hello")
     ;; @r{Assuming that @file{/home} is not a symbolic link.}
     @result{} "/home/foo/hello"
@end example

@defun file-equal-p file1 file2
This function returns @code{t} if the files @var{file1} and
@var{file2} name the same file.  This is similar to comparing their
truenames, except that remote file names are also handled in an
appropriate manner.  If @var{file1} or @var{file2} does not exist, the
return value is unspecified.
@end defun

@defun file-name-case-insensitive-p filename
Sometimes file names or their parts need to be compared as strings, in
which case it's important to know whether the underlying filesystem is
case-insensitive.  This function returns @code{t} if file
@var{filename} is on a case-insensitive filesystem.  It always returns
@code{t} on MS-DOS and MS-Windows.  On Cygwin and macOS,
filesystems may or may not be case-insensitive, and the function tries
to determine case-sensitivity by a runtime test.  If the test is
inconclusive, the function returns @code{t} on Cygwin and @code{nil}
on macOS.

Currently this function always returns @code{nil} on platforms other
than MS-DOS, MS-Windows, Cygwin, and macOS@.  It does not detect
case-insensitivity of mounted filesystems, such as Samba shares or
NFS-mounted Windows volumes.  On remote hosts, it assumes @code{t} for
the @samp{smb} method.  For all other connection methods, runtime
tests are performed.
@end defun

@defun vc-responsible-backend file
This function determines the responsible VC backend of the given
@var{file}.  For example, if @file{emacs.c} is a file tracked by Git,
@w{@code{(vc-responsible-backend "emacs.c")}} returns @samp{Git}.
Note that if @var{file} is a symbolic link,
@code{vc-responsible-backend} will not resolve it---the backend of the
symbolic link file itself is reported.  To get the backend VC of the
file to which @var{file} refers, wrap @var{file} with a symbolic link
resolving function such as @code{file-chase-links}:

@smallexample
(vc-responsible-backend (file-chase-links "emacs.c"))
@end smallexample
@end defun

@node File Attributes
@subsection File Attributes
@cindex file attributes

  This section describes the functions for getting detailed
information about a file, including the owner and group numbers, the
number of names, the inode number, the size, and the times of access
and modification.

@defun file-newer-than-file-p filename1 filename2
@cindex file age
@cindex file modification time
This function returns @code{t} if the file @var{filename1} is
newer than file @var{filename2}.  If @var{filename1} does not
exist, it returns @code{nil}.  If @var{filename1} does exist, but
@var{filename2} does not, it returns @code{t}.

In the following example, assume that the file @file{aug-19} was written
on the 19th, @file{aug-20} was written on the 20th, and the file
@file{no-file} doesn't exist at all.

@example
@group
(file-newer-than-file-p "aug-19" "aug-20")
     @result{} nil
@end group
@group
(file-newer-than-file-p "aug-20" "aug-19")
     @result{} t
@end group
@group
(file-newer-than-file-p "aug-19" "no-file")
     @result{} t
@end group
@group
(file-newer-than-file-p "no-file" "aug-19")
     @result{} nil
@end group
@end example
@end defun

@defun file-has-changed-p filename tag
This function returns non-@code{nil} if the time stamp of
@var{filename} has changed since the last call.  When called for the
first time for some @var{filename}, it records the last modification
time and size of the file, and returns non-@code{nil} when
@var{filename} exists.  Thereafter, when called for the same
@var{filename}, it compares the current time stamp and size with the
recorded ones, and returns non-@code{nil} only if either the time
stamp or the size (or both) are different.  This is useful when a Lisp
program wants to re-read a file whenever it changes.  With an optional
argument @var{tag}, which must be a symbol, the size and modification
time comparisons are limited to calls with the same tag.
@end defun

@defun file-attributes filename &optional id-format
@anchor{Definition of file-attributes}
This function returns a list of attributes of file @var{filename}.  If
the specified file does not exist, it returns @code{nil}.
This function does not follow symbolic links.
The optional parameter @var{id-format} specifies the preferred format
of attributes @acronym{UID} and @acronym{GID} (see below)---the
valid values are @code{'string} and @code{'integer}.  The latter is
the default, but we plan to change that, so you should specify a
non-@code{nil} value for @var{id-format} if you use the returned
@acronym{UID} or @acronym{GID}.

On GNU platforms when operating on a local file, this function is
atomic: if the filesystem is simultaneously being changed by some
other process, this function returns the file's attributes either
before or after the change.  Otherwise this function is not atomic,
and might return @code{nil} if it detects the race condition, or might
return a hodgepodge of the previous and current file attributes.

Accessor functions are provided to access the elements in this list.
The accessors are mentioned along with the descriptions of the
elements below.

The elements of the list, in order, are:

@enumerate 0
@item
@code{t} for a directory, a string for a symbolic link (the name
linked to), or @code{nil} for a text file
(@code{file-attribute-type}).

@c Wordy so as to prevent an overfull hbox.  --rjc 15mar92
@item
The number of names the file has (@code{file-attribute-link-number}).
Alternate names, also known as hard links, can be created by using the
@code{add-name-to-file} function (@pxref{Changing Files}).

@item
The file's @acronym{UID}, normally as a string
(@code{file-attribute-user-id}).  However, if it does not correspond
to a named user, the value is an integer.

@item
The file's @acronym{GID}, likewise (@code{file-attribute-group-id}).

@item
The time of last access as a Lisp timestamp
(@code{file-attribute-access-time}).  The timestamp is in the
style of @code{current-time} (@pxref{Time of Day}) and is truncated
to that of the filesystem's timestamp resolution; for example, on some
FAT-based filesystems, only the date of last access is recorded, so
this time will always hold the midnight of the day of the last access.

@cindex modification time of file
@item
The time of last modification as a Lisp timestamp
(@code{file-attribute-modification-time}).  This is the last time when
the file's contents were modified.

@item
The time of last status change as a Lisp timestamp
(@code{file-attribute-status-change-time}).  This is the time of the
last change to the file's access mode bits, its owner and group, and
other information recorded in the filesystem for the file, beyond the
file's contents.

@item
The size of the file in bytes (@code{file-attribute-size}).

@item
The file's modes, as a string of ten letters or dashes, as in
@samp{ls -l} (@code{file-attribute-modes}).

@item
An unspecified value, present for backward compatibility.

@item
The file's inode number (@code{file-attribute-inode-number}),
a nonnegative integer.

@item
The filesystem's identifier of the device that the file is on
(@code{file-attribute-device-number}), an integer or a cons cell of
two integers.  The latter is sometimes used by remote files, in order
to distinguish remote filesystems from local ones.
@end enumerate

The file's inode and device together give enough information
to distinguish any two files on the system---no two files can have the
same values for both of these attributes.  This tuple that uniquely
identifies the file is returned by @code{file-attribute-file-identifier}.

For example, here are the file attributes for @file{files.texi}:

@example
@group
(file-attributes "files.texi" 'string)
     @result{}  (nil 1 "lh" "users"
          (20614 64019 50040 152000)
          (20000 23 0 0)
          (20614 64555 902289 872000)
          122295 "-rw-rw-rw-"
          t 6473924464520138
          1014478468)
@end group
@end example

@noindent
and here is how the result is interpreted:

@table @code
@item nil
is neither a directory nor a symbolic link.

@item 1
has only one name (the name @file{files.texi} in the current default
directory).

@item "lh"
is owned by the user with name @samp{lh}.

@item "users"
is in the group with name @samp{users}.

@item (20614 64019 50040 152000)
was last accessed on October 23, 2012, at 20:12:03.050040152 UTC@.
(This timestamp is @code{(1351023123050040152 . 1000000000)}
if @code{current-time-list} is @code{nil}.)

@item (20000 23 0 0)
was last modified on July 15, 2001, at 08:53:43.000000000 UTC@.
(This timestamp is @code{(1310720023000000000 . 1000000000)}
if @code{current-time-list} is @code{nil}.)

@item (20614 64555 902289 872000)
last had its status changed on October 23, 2012, at 20:20:59.902289872 UTC@.
(This timestamp is @code{(1351023659902289872 . 1000000000)}
if @code{current-time-list} is @code{nil}.)

@item 122295
is 122295 bytes long.  (It may not contain 122295 characters, though,
if some of the bytes belong to multibyte sequences, and also if the
end-of-line format is CR-LF.)

@item "-rw-rw-rw-"
has a mode of read and write access for the owner, group, and world.

@item t
is merely a placeholder; it carries no information.

@item 6473924464520138
has an inode number of 6473924464520138.

@item 1014478468
is on the file-system device whose number is 1014478468.
@end table
@end defun

@defun file-nlinks filename
This function returns the number of names (i.e., hard links) that
file @var{filename} has.  If the file does not exist, this function
returns @code{nil}.  Note that symbolic links have no effect on this
function, because they are not considered to be names of the files
they link to.  This function does not follow symbolic links.

@example
@group
$ ls -l foo*
-rw-rw-rw- 2 rms rms 4 Aug 19 01:27 foo
-rw-rw-rw- 2 rms rms 4 Aug 19 01:27 foo1
@end group

@group
(file-nlinks "foo")
     @result{} 2
@end group
@group
(file-nlinks "doesnt-exist")
     @result{} nil
@end group
@end example
@end defun

@node Extended Attributes
@subsection Extended File Attributes
@cindex extended file attributes

On some operating systems, each file can be associated with arbitrary
@dfn{extended file attributes}.  At present, Emacs supports querying
and setting two specific sets of extended file attributes: Access
Control Lists (ACLs) and SELinux contexts.  These extended file
attributes are used, on some systems, to impose more sophisticated
file access controls than the basic Unix-style permissions
discussed in the previous sections.

@cindex access control list
@cindex ACL entries
@cindex SELinux context
  A detailed explanation of ACLs and SELinux is beyond the scope of
this manual.  For our purposes, each file can be associated with an
@dfn{ACL}, which specifies its properties under an ACL-based file
control system, and/or an @dfn{SELinux context}, which specifies its
properties under the SELinux system.

@defun file-acl filename
This function returns the ACL for the file @var{filename}.  The exact
Lisp representation of the ACL is unspecified (and may change in
future Emacs versions), but it is the same as what @code{set-file-acl}
takes for its @var{acl} argument (@pxref{Changing Files}).

The underlying ACL implementation is platform-specific; on GNU/Linux
and BSD, Emacs uses the POSIX ACL interface, while on MS-Windows Emacs
emulates the POSIX ACL interface with native file security APIs.

If ACLs are not supported or the file does not exist,
then the return value is @code{nil}.
@end defun

@defun file-selinux-context filename
This function returns the SELinux context of the file @var{filename},
as a list of the form @code{(@var{user} @var{role} @var{type}
@var{range})}.  The list elements are the context's user, role, type,
and range respectively, as Lisp strings; see the SELinux documentation
for details about what these actually mean.  The return value has the
same form as what @code{set-file-selinux-context} takes for its
@var{context} argument (@pxref{Changing Files}).

If SELinux is not supported or the file does not exist,
then the return value is @code{(nil nil nil nil)}.
@end defun

@defun file-extended-attributes filename
This function returns an alist of the Emacs-recognized extended
attributes of file @var{filename}.  Currently, it serves as a
convenient way to retrieve both the ACL and SELinux context; you can
then call the function @code{set-file-extended-attributes}, with the
returned alist as its second argument, to apply the same file access
attributes to another file (@pxref{Changing Files}).

One of the elements is @code{(acl . @var{acl})}, where @var{acl} has
the same form returned by @code{file-acl}.

Another element is @code{(selinux-context . @var{context})}, where
@var{context} is the SELinux context, in the same form returned by
@code{file-selinux-context}.
@end defun

@node Locating Files
@subsection Locating Files in Standard Places
@cindex locate file in path
@cindex find file in path

  This section explains how to search for a file in a list of
directories (a @dfn{path}), or for an executable file in the standard
list of executable file directories.

  To search for a user-specific configuration file, @xref{Standard
File Names}, for the @code{locate-user-emacs-file} function.

@defun locate-file filename path &optional suffixes predicate
This function searches for a file whose name is @var{filename} in a
list of directories given by @var{path}, trying the suffixes in
@var{suffixes}.  If it finds such a file, it returns the file's
absolute file name (@pxref{Relative File Names}); otherwise it returns
@code{nil}.

The optional argument @var{suffixes} gives the list of file-name
suffixes to append to @var{filename} when searching.
@code{locate-file} tries each possible directory with each of these
suffixes.  If @var{suffixes} is @code{nil}, or @code{("")}, then there
are no suffixes, and @var{filename} is used only as-is.  Typical
values of @var{suffixes} are @code{exec-suffixes} (@pxref{Subprocess
Creation}), @code{load-suffixes}, @code{load-file-rep-suffixes} and
the return value of the function @code{get-load-suffixes} (@pxref{Load
Suffixes}).

Typical values for @var{path} are @code{exec-path} (@pxref{Subprocess
Creation}) when looking for executable programs, or @code{load-path}
(@pxref{Library Search}) when looking for Lisp files.  If
@var{filename} is absolute, @var{path} has no effect, but the suffixes
in @var{suffixes} are still tried.

The optional argument @var{predicate}, if non-@code{nil}, specifies a
predicate function for testing whether a candidate file is suitable.
The predicate is passed the candidate file name as its single
argument.  If @var{predicate} is @code{nil} or omitted,
@code{locate-file} uses @code{file-readable-p} as the predicate.
@xref{Kinds of Files}, for other useful predicates, e.g.,
@code{file-executable-p} and @code{file-directory-p}.

This function will normally skip directories, so if you want it to
find directories, make sure the @var{predicate} function returns
@code{dir-ok} for them.  For example:

@example
(locate-file "html" '("/var/www" "/srv") nil
             (lambda (f) (if (file-directory-p f) 'dir-ok)))
@end example


For compatibility, @var{predicate} can also be one of the symbols
@code{executable}, @code{readable}, @code{writable}, @code{exists}, or
a list of one or more of these symbols.
@end defun

@defun executable-find program &optional remote
This function searches for the executable file of the named
@var{program} and returns the absolute file name of the executable,
including its file-name extensions, if any.  It returns @code{nil} if
the file is not found.  The function searches in all the directories
in @code{exec-path}, and tries all the file-name extensions in
@code{exec-suffixes} (@pxref{Subprocess Creation}).

If @var{remote} is non-@code{nil}, and @code{default-directory} is a
remote directory, @var{program} is searched on the respective remote host.
@end defun

@node Changing Files
@section Changing File Names and Attributes
@c @cindex renaming files  Duplicates rename-file
@cindex copying files
@cindex deleting files
@cindex linking files
@cindex setting modes of files

  The functions in this section rename, copy, delete, link, and set
the modes (permissions) of files.  Typically, they signal a
@code{file-error} error if they fail to perform their function,
reporting the system-dependent error message that describes the reason
for the failure.  If they fail because a file is missing, they signal
a @code{file-missing} error instead.

  For performance, the operating system may cache or alias changes
made by these functions instead of writing them immediately to
secondary storage.  @xref{Files and Storage}.

  In the functions that have an argument @var{newname}, if this
argument is a directory name it is treated as if the nondirectory part
of the source name were appended.  Typically, a directory name is one
that ends in @samp{/} (@pxref{Directory Names}).  For example, if the
old name is @file{a/b/c}, the @var{newname} @file{d/e/f/} is treated
as if it were @file{d/e/f/c}.  This special treatment does not apply
if @var{newname} is not a directory name but names a file that is a
directory; for example, the @var{newname} @file{d/e/f} is left as-is
even if @file{d/e/f} happens to be a directory.

  In the functions that have an argument @var{newname}, if a file by the
name of @var{newname} already exists, the actions taken depend on the
value of the argument @var{ok-if-already-exists}:

@itemize @bullet
@item
Signal a @code{file-already-exists} error if
@var{ok-if-already-exists} is @code{nil}.

@item
Request confirmation if @var{ok-if-already-exists} is a number.

@item
Replace the old file without confirmation if @var{ok-if-already-exists}
is any other value.
@end itemize

@deffn Command add-name-to-file oldname newname &optional ok-if-already-exists
@cindex file with multiple names
@cindex file hard link
This function gives the file named @var{oldname} the additional name
@var{newname}.  This means that @var{newname} becomes a new hard
link to @var{oldname}.

If @var{newname} is a symbolic link, its directory entry is replaced,
not the directory entry it points to.  If @var{oldname} is a symbolic
link, this function might or might not follow the link; it does not
follow the link on GNU platforms.  If @var{oldname} is a directory,
this function typically fails, although for the superuser on a few
old-fashioned non-GNU platforms it can succeed and create a filesystem
that is not tree-structured.

In the first part of the following example, we list two files,
@file{foo} and @file{foo3}.

@example
@group
$ ls -li fo*
81908 -rw-rw-rw- 1 rms rms 29 Aug 18 20:32 foo
84302 -rw-rw-rw- 1 rms rms 24 Aug 18 20:31 foo3
@end group
@end example

Now we create a hard link, by calling @code{add-name-to-file}, then list
the files again.  This shows two names for one file, @file{foo} and
@file{foo2}.

@example
@group
(add-name-to-file "foo" "foo2")
     @result{} nil
@end group

@group
$ ls -li fo*
81908 -rw-rw-rw- 2 rms rms 29 Aug 18 20:32 foo
81908 -rw-rw-rw- 2 rms rms 29 Aug 18 20:32 foo2
84302 -rw-rw-rw- 1 rms rms 24 Aug 18 20:31 foo3
@end group
@end example

Finally, we evaluate the following:

@example
(add-name-to-file "foo" "foo3" t)
@end example

@noindent
and list the files again.  Now there are three names
for one file: @file{foo}, @file{foo2}, and @file{foo3}.  The old
contents of @file{foo3} are lost.

@example
@group
(add-name-to-file "foo1" "foo3")
     @result{} nil
@end group

@group
$ ls -li fo*
81908 -rw-rw-rw- 3 rms rms 29 Aug 18 20:32 foo
81908 -rw-rw-rw- 3 rms rms 29 Aug 18 20:32 foo2
81908 -rw-rw-rw- 3 rms rms 29 Aug 18 20:32 foo3
@end group
@end example

This function is meaningless on operating systems where multiple names
for one file are not allowed.  Some systems implement multiple names
by copying the file instead.

See also @code{file-nlinks} in @ref{File Attributes}.
@end deffn

@deffn Command rename-file filename newname &optional ok-if-already-exists
This command renames the file @var{filename} as @var{newname}.

If @var{filename} has additional names aside from @var{filename}, it
continues to have those names.  In fact, adding the name @var{newname}
with @code{add-name-to-file} and then deleting @var{filename} has the
same effect as renaming, aside from momentary intermediate states and
treatment of errors, directories and symbolic links.

This command does not follow symbolic links.  If @var{filename} is a
symbolic link, this command renames the symbolic link, not the file it
points to.  If @var{newname} is a symbolic link, its directory entry
is replaced, not the directory entry it points to.

This command does nothing if @var{filename} and @var{newname} are the
same directory entry, i.e., if they refer to the same parent directory
and give the same name within that directory.  Otherwise, if
@var{filename} and @var{newname} name the same file, this command does
nothing on POSIX-conforming systems, and removes @var{filename} on
some non-POSIX systems.

If @var{newname} exists, then it must be an empty directory if
@var{oldname} is a directory and a non-directory otherwise.
@end deffn

@deffn Command copy-file oldname newname &optional ok-if-already-exists time preserve-uid-gid preserve-extended-attributes
This command copies the file @var{oldname} to @var{newname}.  An
error is signaled if @var{oldname} is not a regular file.  If @var{newname}
names a directory, it copies @var{oldname} into that directory,
preserving its final name component.
@c FIXME: See Bug#27986 for how the previous sentence might change.

This function follows symbolic links, except that it does not follow a
dangling symbolic link to create @var{newname}.

If @var{time} is non-@code{nil}, then this function gives the new file
the same last-modified time that the old one has.  (This works on only
some operating systems.)  If setting the time gets an error,
@code{copy-file} signals a @code{file-date-error} error.  In an
interactive call, a prefix argument specifies a non-@code{nil} value
for @var{time}.

If argument @var{preserve-uid-gid} is @code{nil}, we let the operating
system decide the user and group ownership of the new file (this is
usually set to the user running Emacs).  If @var{preserve-uid-gid} is
non-@code{nil}, we attempt to copy the user and group ownership of the
file.  This works only on some operating systems, and only if you have
the correct permissions to do so.

If the optional argument @var{preserve-permissions} is non-@code{nil},
this function copies the file modes (or ``permissions'') of
@var{oldname} to @var{newname}, as well as the Access Control List and
SELinux context (if any).  @xref{Information about Files}.

Otherwise, the file modes of @var{newname} are left unchanged if it is
an existing file, and set to those of @var{oldname}, masked by the
default file permissions (see @code{set-default-file-modes} below), if
@var{newname} is to be newly created.  The Access Control List or
SELinux context are not copied over in either case.
@end deffn

@deffn Command make-symbolic-link target linkname &optional ok-if-already-exists
@pindex ln
@kindex file-already-exists
This command makes a symbolic link to @var{target}, named
@var{linkname}.  This is like the shell command @samp{ln -s
@var{target} @var{linkname}}.  The @var{target} argument
is treated only as a string; it need not name an existing file.
If @var{ok-if-already-exists} is an integer, indicating interactive
use, then leading @samp{~} is expanded and leading @samp{/:} is
stripped in the @var{target} string.

If @var{target} is a relative file name, the resulting symbolic link
is interpreted relative to the directory containing the symbolic link.
@xref{Relative File Names}.

If both @var{target} and @var{linkname} have remote file name syntax,
and if both remote identifications are equal, the symbolic link points
to the local file name part of @var{target}.

This function is not available on systems that don't support symbolic
links.
@end deffn

@cindex trash
@vindex delete-by-moving-to-trash
@deffn Command delete-file filename &optional trash
@pindex rm
This command deletes the file @var{filename}.  If the file has
multiple names, it continues to exist under the other names.  If
@var{filename} is a symbolic link, @code{delete-file} deletes only the
symbolic link and not its target.

A suitable kind of @code{file-error} error is signaled if the file
does not exist, or is not deletable.  (On GNU and other POSIX-like
systems, a file is deletable if its directory is writable.)

If the optional argument @var{trash} is non-@code{nil} and the
variable @code{delete-by-moving-to-trash} is non-@code{nil}, this
command moves the file into the system Trash instead of deleting it.
@xref{Misc File Ops,,Miscellaneous File Operations, emacs, The GNU
Emacs Manual}.  When called interactively, @var{trash} is @code{t} if
no prefix argument is given, and @code{nil} otherwise.

See also @code{delete-directory} in @ref{Create/Delete Dirs}.
@end deffn

@defopt remote-file-name-inhibit-delete-by-moving-to-trash
If this variable is non-@code{nil}, remote files are never moved to
the Trash.  They are deleted instead.
@end defopt

@cindex file permissions, setting
@cindex permissions, file
@cindex file modes, setting
@deffn Command set-file-modes filename mode &optional flag
This function sets the @dfn{file mode} (or @dfn{permissions}) of
@var{filename} to @var{mode}.

By default this function follows symbolic links.  However, if the
optional argument @var{flag} is the symbol @code{nofollow}, this
function does not follow @var{filename} if it is a symbolic link;
this can help prevent inadvertently changing the mode bits of a file
somewhere else.  On platforms that do not support changing mode bits
on a symbolic link, this function signals an error when @var{filename}
is a symbolic link and @var{flag} is @code{nofollow}.

If called non-interactively, @var{mode} must be an integer.  Only the
lowest 12 bits of the integer are used; on most systems, only the
lowest 9 bits are meaningful.  You can use the Lisp construct for
octal numbers to enter @var{mode}.  For example,

@example
(set-file-modes "myfile" #o644 'nofollow)
@end example

@noindent
specifies that the file should be readable and writable for its owner,
readable for group members, and readable for all other users.
@xref{File permissions,,, coreutils, The @sc{gnu} @code{Coreutils}
Manual}, for a description of mode bit specifications.

Interactively, @var{mode} is read from the minibuffer using
@code{read-file-modes} (see below), which lets the user type in either
an integer or a string representing the permissions symbolically.

@xref{Testing Accessibility}, for the function @code{file-modes}, which
returns the permissions of a file.
@end deffn

@defun set-default-file-modes mode
@cindex umask
This function sets the default permissions for new files created by
Emacs and its subprocesses.  Every file created with Emacs initially
has these permissions, or a subset of them (@code{write-region} will
not grant execute permissions even if the default file permissions
allow execution).  On GNU and other POSIX-like systems, the default
permissions are given by the bitwise complement of the @samp{umask}
value, i.e.@: each bit that is set in the argument @var{mode} will be
@emph{reset} in the default permissions with which Emacs creates
files.

The argument @var{mode} should be an integer which specifies the
permissions, similar to @code{set-file-modes} above.  Only the lowest
9 bits are meaningful.

The default file permissions have no effect when you save a modified
version of an existing file; saving a file preserves its existing
permissions.
@end defun

@defmac with-file-modes mode body@dots{}
This macro evaluates the @var{body} forms with the default
permissions for new files temporarily set to @var{modes} (whose value
is as for @code{set-file-modes} above).  When finished, it restores
the original default file permissions, and returns the value of the
last form in @var{body}.

This is useful for creating private files, for example.
@end defmac

@defun default-file-modes
This function returns the default file permissions, as an integer.
@end defun

@defun read-file-modes &optional prompt base-file
This function reads a set of file mode bits from the minibuffer.  The
first optional argument @var{prompt} specifies a non-default prompt.
Second second optional argument @var{base-file} is the name of a file
on whose permissions to base the mode bits that this function returns,
if what the user types specifies mode bits relative to permissions of
an existing file.

If user input represents an octal number, this function returns that
number.  If it is a complete symbolic specification of mode bits, as
in @code{"u=rwx"}, the function converts it to the equivalent numeric
value using @code{file-modes-symbolic-to-number} and returns the
result.  If the specification is relative, as in @code{"o+g"}, then
the permissions on which the specification is based are taken from the
mode bits of @var{base-file}.  If @var{base-file} is omitted or
@code{nil}, the function uses @code{0} as the base mode bits.  The
complete and relative specifications can be combined, as in
@code{"u+r,g+rx,o+r,g-w"}.  @xref{File permissions,,, coreutils, The
@sc{gnu} @code{Coreutils} Manual}, for a description of file mode
specifications.
@end defun

@defun file-modes-symbolic-to-number modes &optional base-modes
This function converts a symbolic file mode specification in
@var{modes} into the equivalent integer.  If the symbolic
specification is based on an existing file, that file's mode bits are
taken from the optional argument @var{base-modes}; if that argument is
omitted or @code{nil}, it defaults to 0, i.e., no access rights at
all.
@end defun

@defun file-modes-number-to-symbolic modes
This function converts a numeric file mode specification in
@var{modes} into the equivalent symbolic form.
@end defun

@defun set-file-times filename &optional time flag
This function sets the access and modification times of @var{filename}
to @var{time}.  The return value is @code{t} if the times are successfully
set, otherwise it is @code{nil}.  @var{time} defaults to the current
time and must be a time value (@pxref{Time of Day}).

By default this function follows symbolic links.  However, if the
optional argument @var{flag} is the symbol @code{nofollow}, this
function does not follow @var{filename} if it is a symbolic link;
this can help prevent inadvertently changing the times of a file
somewhere else.  On platforms that do not support changing times
on a symbolic link, this function signals an error when @var{filename}
is a symbolic link and @var{flag} is @code{nofollow}.
@end defun

@defun set-file-extended-attributes filename attribute-alist
This function sets the Emacs-recognized extended file attributes for
@var{filename}.  The second argument @var{attribute-alist} should be
an alist of the same form returned by @code{file-extended-attributes}.
The return value is @code{t} if the attributes are successfully set,
otherwise it is @code{nil}.
@xref{Extended Attributes}.
@end defun

@defun set-file-selinux-context filename context
This function sets the SELinux security context for @var{filename} to
@var{context}.  The @var{context} argument should be a list
@code{(@var{user} @var{role} @var{type} @var{range})}, where each
element is a string.  @xref{Extended Attributes}.

The function returns @code{t} if it succeeds in setting the SELinux
context of @var{filename}.  It returns @code{nil} if the context was
not set (e.g., if SELinux is disabled, or if Emacs was compiled
without SELinux support).
@end defun

@defun set-file-acl filename acl
This function sets the Access Control List for @var{filename} to
@var{acl}.  The @var{acl} argument should have the same form returned
by the function @code{file-acl}.  @xref{Extended Attributes}.

The function returns @code{t} if it successfully sets the ACL of
@var{filename}, @code{nil} otherwise.
@end defun

@node Files and Storage
@section Files and Secondary Storage
@cindex secondary storage

After Emacs changes a file, there are two reasons the changes might
not survive later failures of power or media, both having to do with
efficiency.  First, the operating system might alias written data with
data already stored elsewhere on secondary storage until one file or
the other is later modified; this will lose both files if the only
copy on secondary storage is lost due to media failure.  Second, the
operating system might not write data to secondary storage
immediately, which will lose the data if power is lost
or if there is a media failure.

@findex write-region
Although both sorts of failures can largely be avoided by a suitably
configured system, such systems are typically more expensive or
less efficient.  In lower-end systems, to survive media failure you
can copy the file to a different device, and to survive a power
failure (or be immediately notified of a media failure) you can use
the @code{write-region} function with the
@code{write-region-inhibit-fsync} variable set to @code{nil}.
Although this variable is ordinarily @code{t} because that can
significantly improve performance, it may make sense to temporarily
bind it to @code{nil} if using Emacs to implement database-like
transactions that survive power failure on lower-end systems.
@xref{Writing to Files}.

On some platforms when Emacs changes a file other processes might not
be notified of the change immediately.  Setting
@code{write-region-inhibit-fsync} to @code{nil} may improve
notification speed in this case, though there are no guarantees.

@node File Names
@section File Names
@cindex file names

  Files are generally referred to by their names, in Emacs as elsewhere.
File names in Emacs are represented as strings.  The functions that
operate on a file all expect a file name argument.

  In addition to operating on files themselves, Emacs Lisp programs
often need to operate on file names; i.e., to take them apart and to use
part of a name to construct related file names.  This section describes
how to manipulate file names.

  The functions in this section do not actually access files, so they
can operate on file names that do not refer to an existing file or
directory.

@findex cygwin-convert-file-name-from-windows
@findex cygwin-convert-file-name-to-windows
@cindex MS-Windows file-name syntax
@cindex converting file names from/to MS-Windows syntax
  On MS-DOS and MS-Windows, these functions (like the function that
actually operate on files) accept MS-DOS or MS-Windows file-name syntax,
where backslashes separate the components, as well as POSIX syntax; but
they always return POSIX syntax.  This enables Lisp programs to specify
file names in POSIX syntax and work properly on all systems without
change.@footnote{In MS-Windows versions of Emacs compiled for the Cygwin
environment, you can use the functions
@code{cygwin-convert-file-name-to-windows} and
@code{cygwin-convert-file-name-from-windows} to convert between the
two file-name syntaxes.}

@menu
* File Name Components::  The directory part of a file name, and the rest.
* Relative File Names::   Some file names are relative to a current directory.
* Directory Names::       A directory's name as a directory
                            is different from its name as a file.
* File Name Expansion::   Converting relative file names to absolute ones.
* Unique File Names::     Generating names for temporary files.
* File Name Completion::  Finding the completions for a given file name.
* Standard File Names::   If your package uses a fixed file name,
                            how to handle various operating systems simply.
@end menu

@node File Name Components
@subsection File Name Components
@cindex directory part (of file name)
@cindex nondirectory part (of file name)
@cindex version number (in file name)

  The operating system groups files into directories.  To specify a
file, you must specify the directory and the file's name within that
directory.  Therefore, Emacs considers a file name as having two main
parts: the @dfn{directory name} part, and the @dfn{nondirectory} part
(or @dfn{file name within the directory}).  Either part may be empty.
Concatenating these two parts reproduces the original file name.
@footnote{Emacs follows the GNU convention to use the term @emph{file name}
instead of the term @emph{pathname}.  We use the term @emph{path} only for
search paths, which are lists of directory names.}

  On most systems, the directory part is everything up to and including
the last slash (backslash is also allowed in input on MS-DOS or
MS-Windows); the nondirectory part is the rest.

  For some purposes, the nondirectory part is further subdivided into
the name proper and the @dfn{version number}.  On most systems, only
backup files have version numbers in their names.

@defun file-name-directory filename
This function returns the directory part of @var{filename}, as a
directory name (@pxref{Directory Names}), or @code{nil} if
@var{filename} does not include a directory part.

On GNU and other POSIX-like systems, a string returned by this function always
ends in a slash.  On MS-DOS it can also end in a colon.

@example
@group
(file-name-directory "lewis/foo")  ; @r{GNU example}
     @result{} "lewis/"
@end group
@group
(file-name-directory "foo")        ; @r{GNU example}
     @result{} nil
@end group
@end example
@end defun

@defun file-name-nondirectory filename
This function returns the nondirectory part of @var{filename}.

@example
@group
(file-name-nondirectory "lewis/foo")
     @result{} "foo"
@end group
@group
(file-name-nondirectory "foo")
     @result{} "foo"
@end group
@group
(file-name-nondirectory "lewis/")
     @result{} ""
@end group
@end example
@end defun

@defun file-name-sans-versions filename &optional keep-backup-version
This function returns @var{filename} with any file version numbers,
backup version numbers, or trailing tildes discarded.

If @var{keep-backup-version} is non-@code{nil}, then true file version
numbers understood as such by the file system are discarded from the
return value, but backup version numbers are kept.

@example
@group
(file-name-sans-versions "~rms/foo.~1~")
     @result{} "~rms/foo"
@end group
@group
(file-name-sans-versions "~rms/foo~")
     @result{} "~rms/foo"
@end group
@group
(file-name-sans-versions "~rms/foo")
     @result{} "~rms/foo"
@end group
@end example
@end defun

@defun file-name-extension filename &optional period
This function returns @var{filename}'s final extension, if any,
after applying @code{file-name-sans-versions} to remove any
version/backup part.  The extension, in a file name, is the part that
follows the last @samp{.} in the last name component (minus any
version/backup part).

This function returns @code{nil} for extensionless file names such as
@file{foo}.  It returns @code{""} for null extensions, as in
@file{foo.}.  If the last component of a file name begins with a
@samp{.}, that @samp{.}  doesn't count as the beginning of an
extension.  Thus, @file{.emacs}'s extension is @code{nil}, not
@samp{.emacs}.

If @var{period} is non-@code{nil}, then the returned value includes
the period that delimits the extension, and if @var{filename} has no
extension, the value is @code{""}.
@end defun

@defun file-name-with-extension filename extension
This function returns @var{filename} with its extension set to
@var{extension}.  A single leading dot in the @var{extension} will be
stripped if there is one.  For example:

@example
(file-name-with-extension "file" "el")
     @result{} "file.el"
(file-name-with-extension "file" ".el")
     @result{} "file.el"
(file-name-with-extension "file.c" "el")
     @result{} "file.el"
@end example

Note that this function will error if @var{filename} or
@var{extension} are empty, or if the @var{filename} is shaped like a
directory (i.e., if @code{directory-name-p} returns non-@code{nil}).
@end defun

@defun file-name-sans-extension filename
This function returns @var{filename} minus its extension, if any.  The
version/backup part, if present, is only removed if the file has an
extension.  For example,

@example
(file-name-sans-extension "foo.lose.c")
     @result{} "foo.lose"
(file-name-sans-extension "big.hack/foo")
     @result{} "big.hack/foo"
(file-name-sans-extension "/my/home/.emacs")
     @result{} "/my/home/.emacs"
(file-name-sans-extension "/my/home/.emacs.el")
     @result{} "/my/home/.emacs"
(file-name-sans-extension "~/foo.el.~3~")
     @result{} "~/foo"
(file-name-sans-extension "~/foo.~3~")
     @result{} "~/foo.~3~"
@end example

Note that the @samp{.~3~} in the two last examples is the backup part,
not an extension.
@end defun

@defun file-name-base filename
This function is the composition of @code{file-name-sans-extension}
and @code{file-name-nondirectory}.  For example,

@example
(file-name-base "/my/home/foo.c")
    @result{} "foo"
@end example
@end defun

@defun file-name-split filename
This function splits a file name into its components, and can be
thought of as the inverse of @code{string-join} with the appropriate
directory separator.  For example,

@example
(file-name-split "/tmp/foo.txt")
    @result{} ("" "tmp" "foo.txt")
(string-join (file-name-split "/tmp/foo.txt") "/")
    @result{} "/tmp/foo.txt"
@end example
@end defun

@node Relative File Names
@subsection Absolute and Relative File Names
@cindex absolute file name
@cindex relative file name

  All the directories in the file system form a tree starting at the
root directory.  A file name can specify all the directory names
starting from the root of the tree; then it is called an
@dfn{absolute} file name.  Or it can specify the position of the file
in the tree relative to a default directory; then it is called a
@dfn{relative} file name.  On GNU and other POSIX-like systems,
after any leading @samp{~} has been expanded, an absolute file name
starts with a @samp{/}
(@pxref{abbreviate-file-name}), and a relative one does not.  On
MS-DOS and MS-Windows, an absolute file name starts with a slash or a
backslash, or with a drive specification @samp{@var{x}:/}, where
@var{x} is the @dfn{drive letter}.

@defun file-name-absolute-p filename
This function returns @code{t} if file @var{filename} is an absolute
file name, @code{nil} otherwise.  A file name is considered to be
absolute if its first component is @samp{~}, or is @samp{~@var{user}}
where @var{user} is a valid login name.  In the following examples,
assume that there is a user named @samp{rms} but no user named
@samp{nosuchuser}.

@example
@group
(file-name-absolute-p "~rms/foo")
     @result{} t
@end group
@group
(file-name-absolute-p "~nosuchuser/foo")
     @result{} nil
@end group
@group
(file-name-absolute-p "rms/foo")
     @result{} nil
@end group
@group
(file-name-absolute-p "/user/rms/foo")
     @result{} t
@end group
@end example
@end defun

  Given a possibly relative file name, you can expand any
leading @samp{~} and convert the result to an
absolute name using @code{expand-file-name} (@pxref{File Name
Expansion}).  This function converts absolute file names to relative
names:

@defun file-relative-name filename &optional directory
This function tries to return a relative name that is equivalent to
@var{filename}, assuming the result will be interpreted relative to
@var{directory} (an absolute directory name or directory file name).
If @var{directory} is omitted or @code{nil}, it defaults to the
current buffer's default directory.

On some operating systems, an absolute file name begins with a device
name.  On such systems, @var{filename} has no relative equivalent based
on @var{directory} if they start with two different device names.  In
this case, @code{file-relative-name} returns @var{filename} in absolute
form.

@example
(file-relative-name "/foo/bar" "/foo/")
     @result{} "bar"
(file-relative-name "/foo/bar" "/hack/")
     @result{} "../foo/bar"
@end example
@end defun

@cindex empty file name
  A file name that is an empty string stands for the current buffer's
default directory.

@node Directory Names
@subsection Directory Names
@cindex directory name
@cindex directory file name
@cindex file name of directory

  A @dfn{directory name} is a string that must name a directory if it
names any file at all.  A directory is actually a kind of file, and it
has a file name (called the @dfn{directory file name}), which is
related to the directory name but is typically not identical.  (This
is not quite the same as the usual POSIX terminology.)  These two
names for the same entity are related by a syntactic transformation.
On GNU and other POSIX-like systems, this is simple: to obtain a
directory name, append a @samp{/} to a directory file name that does
not already end in @samp{/}.  On MS-DOS the relationship is more
complicated.

  The difference between a directory name and a directory file name is
subtle but crucial.  When an Emacs variable or function argument is
described as being a directory name, a directory file name is not
acceptable.  When @code{file-name-directory} returns a string, that is
always a directory name.

  The following two functions convert between directory names and
directory file names.  They do nothing special with environment
variable substitutions such as @samp{$HOME}, and the constructs
@samp{~}, @samp{.} and @samp{..}.

@defun file-name-as-directory filename
This function returns a string representing @var{filename} in a form
that the operating system will interpret as the name of a directory (a
directory name).  On most systems, this means appending a slash to the
string (if it does not already end in one).

@example
@group
(file-name-as-directory "~rms/lewis")
     @result{} "~rms/lewis/"
@end group
@end example
@end defun

@defun directory-name-p filename
This function returns non-@code{nil} if @var{filename} ends with a
directory separator character.  This is the forward slash @samp{/} on
GNU and other POSIX-like systems; MS-Windows and MS-DOS recognize both
the forward slash and the backslash @samp{\} as directory separators.
@end defun

@defun directory-file-name dirname
This function returns a string representing @var{dirname} in a form
that the operating system will interpret as the name of a file (a
directory file name).  On most systems, this means removing the final
directory separators from the string, unless the string consists
entirely of directory separators.

@example
@group
(directory-file-name "~lewis/")
     @result{} "~lewis"
@end group
@end example
@end defun

@defun file-name-concat directory &rest components
Concatenate @var{components} to @var{directory}, inserting a slash
before the components if @var{directory} or the preceding component
didn't end with a slash.

@example
@group
(file-name-concat "/tmp" "foo")
     @result{} "/tmp/foo"
@end group
@end example

A @var{directory} or components that are @code{nil} or the empty
string are ignored---they are filtered out first and do not affect the
results in any way.

This is almost the same as using @code{concat}, but @var{dirname} (and
the non-final components) may or may not end with slash characters,
and this function will not double those characters.
@end defun

  To convert a directory name to its abbreviation, use this
function:

@cindex file name abbreviations
@cindex abbreviated file names
@vindex directory-abbrev-alist
@defun abbreviate-file-name filename
@anchor{abbreviate-file-name}
This function returns an abbreviated form of @var{filename}.  It
applies the abbreviations specified in @code{directory-abbrev-alist}
(@pxref{File Aliases,,File Aliases, emacs, The GNU Emacs Manual}),
then substitutes @samp{~} for the user's home directory if the
argument names a file in the home directory or one of its
subdirectories.  If the home directory is a root directory, it is not
replaced with @samp{~}, because this does not make the result shorter
on many systems.

You can use this function for directory names and for file names,
because it recognizes abbreviations even as part of the name.
@end defun

@defun file-name-parent-directory filename
This function returns the directory name of the parent directory of
@var{filename}.  If @var{filename} is at the root directory of the
filesystem, it returns @code{nil}.  A relative @var{filename} is
assumed to be relative to @code{default-directory}, and the return
value will also be relative in that case.  If the return value is
non-@code{nil}, it ends in a slash.
@end defun

@node File Name Expansion
@subsection Functions that Expand Filenames
@cindex expansion of file names

  @dfn{Expanding} a file name means converting a relative file name to
an absolute one.  Since this is done relative to a default directory,
you must specify the default directory as well as the file name
to be expanded.  It also involves expanding abbreviations like
@file{~/}
@ifnottex
(@pxref{abbreviate-file-name}),
@end ifnottex
and eliminating redundancies like @file{./} and @file{@var{name}/../}.

@defun expand-file-name filename &optional directory
This function converts @var{filename} to an absolute file name.  If
@var{directory} is supplied, it is the default directory to start with
if @var{filename} is relative and does not start with @samp{~}.
(The value of @var{directory} should
itself be an absolute directory name or directory file name; it may
start with @samp{~}.)  Otherwise, the current buffer's value of
@code{default-directory} is used.  For example:

@example
@group
(expand-file-name "foo")
     @result{} "/xcssun/users/rms/lewis/foo"
@end group
@group
(expand-file-name "../foo")
     @result{} "/xcssun/users/rms/foo"
@end group
@group
(expand-file-name "foo" "/usr/spool/")
     @result{} "/usr/spool/foo"
@end group
@end example

If the part of @var{filename} before the first slash is
@samp{~}, it expands to your home directory, which is typically
specified by the value of the @env{HOME} environment variable
(@pxref{General Variables,,, emacs, The GNU Emacs Manual}).
If the part before the first
slash is @samp{~@var{user}} and if @var{user} is a valid login name,
it expands to @var{user}'s home directory.
If you do not want this expansion for a relative @var{filename} that
might begin with a literal @samp{~}, you can use @code{(concat
(file-name-as-directory directory) filename)} instead of
@code{(expand-file-name filename directory)}.

File names containing @samp{.} or @samp{..} are simplified to their
canonical form:

@example
@group
(expand-file-name "bar/../foo")
     @result{} "/xcssun/users/rms/lewis/foo"
@end group
@end example

In some cases, a leading @samp{..} component can remain in the output:

@example
@group
(expand-file-name "../home" "/")
     @result{} "/../home"
@end group
@end example

@noindent
This is for the sake of filesystems that have the concept of a
superroot above the root directory @file{/}.  On other filesystems,
@file{/../} is interpreted exactly the same as @file{/}.

@cindex empty file names, and @code{expand-file-name}
Expanding @file{.} or the empty string returns the default directory:

@example
@group
(expand-file-name "." "/usr/spool/")
     @result{} "/usr/spool"
(expand-file-name "" "/usr/spool/")
     @result{} "/usr/spool"
@end group
@end example

Note that @code{expand-file-name} does @emph{not} expand environment
variables; only @code{substitute-in-file-name} does that:

@example
@group
(expand-file-name "$HOME/foo")
     @result{} "/xcssun/users/rms/lewis/$HOME/foo"
@end group
@end example

Note also that @code{expand-file-name} does not follow symbolic links
at any level.  This results in a difference between the way
@code{file-truename} and @code{expand-file-name} treat @samp{..}.
Assuming that @samp{/tmp/bar} is a symbolic link to the directory
@samp{/tmp/foo/bar} we get:

@example
@group
(file-truename "/tmp/bar/../myfile")
     @result{} "/tmp/foo/myfile"
@end group
@group
(expand-file-name "/tmp/bar/../myfile")
     @result{} "/tmp/myfile"
@end group
@end example

If you may need to follow symbolic links preceding @samp{..}, you
should make sure to call @code{file-truename} without prior direct or
indirect calls to @code{expand-file-name}.  @xref{Truenames}.
@end defun

@defvar default-directory
The value of this buffer-local variable is the default directory for the
current buffer.  It should be an absolute directory name; it may start
with @samp{~}.  This variable is buffer-local in every buffer.

@code{expand-file-name} uses the default directory when its second
argument is @code{nil}.

The value is always a string ending with a slash.

@example
@group
default-directory
     @result{} "/user/lewis/manual/"
@end group
@end example
@end defvar

@defun substitute-in-file-name filename
@anchor{Definition of substitute-in-file-name}
This function replaces environment variable references in
@var{filename} with the environment variable values.  Following
standard Unix shell syntax, @samp{$} is the prefix to substitute an
environment variable value.  If the input contains @samp{$$}, that is
converted to @samp{$}; this gives the user a way to quote a
@samp{$}.

The environment variable name is the series of alphanumeric characters
(including underscores) that follow the @samp{$}.  If the character following
the @samp{$} is a @samp{@{}, then the variable name is everything up to the
matching @samp{@}}.

Calling @code{substitute-in-file-name} on output produced by
@code{substitute-in-file-name} tends to give incorrect results.  For
instance, use of @samp{$$} to quote a single @samp{$} won't work
properly, and @samp{$} in an environment variable's value could lead
to repeated substitution.  Therefore, programs that call this function
and put the output where it will be passed to this function need to
double all @samp{$} characters to prevent subsequent incorrect
results.

@c Wordy to avoid overfull hbox.  --rjc 15mar92
Here we assume that the environment variable @env{HOME}, which holds
the user's home directory, has value @samp{/xcssun/users/rms}.

@example
@group
(substitute-in-file-name "$HOME/foo")
     @result{} "/xcssun/users/rms/foo"
@end group
@end example

After substitution, if a @samp{~} or a @samp{/} appears immediately
after another @samp{/}, the function discards everything before it (up
through the immediately preceding @samp{/}).

@example
@group
(substitute-in-file-name "bar/~/foo")
     @result{} "~/foo"
@end group
@group
(substitute-in-file-name "/usr/local/$HOME/foo")
     @result{} "/xcssun/users/rms/foo"
     ;; @r{@file{/usr/local/} has been discarded.}
@end group
@end example

@end defun

  Sometimes, it is not desired to expand file names.  In such cases,
the file name can be quoted to suppress the expansion, and to handle
the file name literally.  Quoting happens by prefixing the file name
with @samp{/:}.

@defmac file-name-quote name
This macro adds the quotation prefix @samp{/:} to the file @var{name}.
For a local file @var{name}, it prefixes @var{name} with @samp{/:}.
If @var{name} is a remote file name, the local part of @var{name}
(@pxref{Magic File Names}) is quoted.  If @var{name} is already a
quoted file name, @var{name} is returned unchanged.

@example
@group
(substitute-in-file-name (file-name-quote "bar/~/foo"))
     @result{} "/:bar/~/foo"
@end group

@group
(substitute-in-file-name (file-name-quote "/ssh:host:bar/~/foo"))
     @result{} "/ssh:host:/:bar/~/foo"
@end group
@end example

The macro cannot be used to suppress file name handlers from magic
file names (@pxref{Magic File Names}).
@end defmac

@defmac file-name-unquote name
This macro removes the quotation prefix @samp{/:} from the file
@var{name}, if any. If @var{name} is a remote file name, the local
part of @var{name} is unquoted.
@end defmac

@defmac file-name-quoted-p name
This macro returns non-@code{nil}, when @var{name} is quoted with the
prefix @samp{/:}.  If @var{name} is a remote file name, the local part
of @var{name} is checked.
@end defmac


@node Unique File Names
@subsection Generating Unique File Names
@cindex unique file names
@cindex temporary files

  Some programs need to write temporary files.  Here is the usual way to
construct a name for such a file:

@example
(make-temp-file @var{name-of-application})
@end example

@noindent
The job of @code{make-temp-file} is to prevent two different users or
two different jobs from trying to use the exact same file name.

@defun make-temp-file prefix &optional dir-flag suffix text
This function creates a temporary file and returns its name.  Emacs
creates the temporary file's name by adding to @var{prefix} some
random characters that are different in each Emacs job.  The result is
guaranteed to be a newly created file, containing @var{text} if that's
given as a string and empty otherwise. On MS-DOS, this function can
truncate @var{prefix} to fit into the 8+3 file-name limits.  If
@var{prefix} is a relative file name, it is expanded against
@code{temporary-file-directory}.

@example
@group
(make-temp-file "foo")
     @result{} "/tmp/foo232J6v"
@end group
@end example

When @code{make-temp-file} returns, the file has been created and is
empty.  At that point, you should write the intended contents into the
file.

If @var{dir-flag} is non-@code{nil}, @code{make-temp-file} creates an
empty directory instead of an empty file.  It returns the file name,
not the directory name, of that directory.  @xref{Directory Names}.

If @var{suffix} is non-@code{nil}, @code{make-temp-file} adds it at
the end of the file name.

If @var{text} is a string, @code{make-temp-file} inserts it in the file.

To prevent conflicts among different libraries running in the same
Emacs, each Lisp program that uses @code{make-temp-file} should have its
own @var{prefix}.  The number added to the end of @var{prefix}
distinguishes between the same application running in different Emacs
jobs.  Additional added characters permit a large number of distinct
names even in one Emacs job.
@end defun

  The default directory for temporary files is controlled by the
variable @code{temporary-file-directory}.  This variable gives the user
a uniform way to specify the directory for all temporary files.  Some
programs use @code{small-temporary-file-directory} instead, if that is
non-@code{nil}.  To use it, you should expand the prefix against
the proper directory before calling @code{make-temp-file}.

@defopt temporary-file-directory
@cindex @env{TMPDIR} environment variable
@cindex @env{TMP} environment variable
@cindex @env{TEMP} environment variable
This variable specifies the directory name for creating temporary files.
Its value should be a directory name (@pxref{Directory Names}), but it
is good for Lisp programs to cope if the value is a directory's file
name instead.  Using the value as the second argument to
@code{expand-file-name} is a good way to achieve that.

The default value is determined in a reasonable way for your operating
system; it is based on the @env{TMPDIR}, @env{TMP} and @env{TEMP}
environment variables, with a fall-back to a system-dependent name if
none of these variables is defined.

Even if you do not use @code{make-temp-file} to create the temporary
file, you should still use this variable to decide which directory to
put the file in.  However, if you expect the file to be small, you
should use @code{small-temporary-file-directory} first if that is
non-@code{nil}.
@end defopt

@defopt small-temporary-file-directory
This variable specifies the directory name for
creating certain temporary files, which are likely to be small.

If you want to write a temporary file which is likely to be small, you
should compute the directory like this:

@example
(make-temp-file
  (expand-file-name @var{prefix}
                    (or small-temporary-file-directory
                        temporary-file-directory)))
@end example
@end defopt

@defun make-temp-name base-name
This function generates a string that might be a unique file
name.  The name starts with @var{base-name}, and has several random
characters appended to it, which are different in each Emacs job.  It
is like @code{make-temp-file} except that (i) it just constructs a
name and does not create a file, (ii) @var{base-name} should be an
absolute file name that is not magic, and (iii) if the returned file
name is magic, it might name an existing file.  @xref{Magic File
Names}.

@strong{Warning:} In most cases, you should not use this function; use
@code{make-temp-file} instead!  This function is susceptible to a race
condition, between the @code{make-temp-name} call and the creation of
the file, which in some cases may cause a security hole.
@end defun

Sometimes, it is necessary to create a temporary file on a remote host
or a mounted directory.  The following two functions support this.

@cindex temporary file on a remote host
@defun make-nearby-temp-file prefix &optional dir-flag suffix
This function is similar to @code{make-temp-file}, but it creates a
temporary file as close as possible to @code{default-directory}.  If
@var{prefix} is a relative file name, and @code{default-directory} is
a remote file name or located on a mounted file systems, the temporary
file is created in the directory returned by the function
@code{temporary-file-directory}.  Otherwise, the function
@code{make-temp-file} is used.  @var{prefix}, @var{dir-flag} and
@var{suffix} have the same meaning as in @code{make-temp-file}.

@example
@group
(let ((default-directory "/ssh:remotehost:"))
  (make-nearby-temp-file "foo"))
     @result{} "/ssh:remotehost:/tmp/foo232J6v"
@end group
@end example
@end defun

@defun temporary-file-directory
The directory for writing temporary files via
@code{make-nearby-temp-file}.  In case of a remote
@code{default-directory}, this is a directory for temporary files on
that remote host.  If such a directory does not exist, or
@code{default-directory} ought to be located on a mounted file system
(see @code{mounted-file-systems}), the function returns
@code{default-directory}.  For a non-remote and non-mounted
@code{default-directory}, the value of the variable
@code{temporary-file-directory} is returned.
@end defun

In order to extract the local part of the file's name of a temporary
file, use @code{file-local-name} (@pxref{Magic File Names}).

@node File Name Completion
@subsection File Name Completion
@cindex file name completion subroutines
@cindex completion, file name

  This section describes low-level subroutines for completing a file
name.  For higher level functions, see @ref{Reading File Names}.

@defun file-name-all-completions partial-filename directory
This function returns a list of all possible completions for a file
whose name starts with @var{partial-filename} in directory
@var{directory}.  The order of the completions is the order of the files
in the directory, which is unpredictable and conveys no useful
information.

The argument @var{partial-filename} must be a file name containing no
directory part and no slash (or backslash on some systems).  The current
buffer's default directory is prepended to @var{directory}, if
@var{directory} is not absolute.

In the following example, suppose that @file{~rms/lewis} is the current
default directory, and has five files whose names begin with @samp{f}:
@file{foo}, @file{file~}, @file{file.c}, @file{file.c.~1~}, and
@file{file.c.~2~}.

@example
@group
(file-name-all-completions "f" "")
     @result{} ("foo" "file~" "file.c.~2~"
                "file.c.~1~" "file.c")
@end group

@group
(file-name-all-completions "fo" "")
     @result{} ("foo")
@end group
@end example
@end defun

@defun file-name-completion filename directory &optional predicate
This function completes the file name @var{filename} in directory
@var{directory}.  It returns the longest prefix common to all file names
in directory @var{directory} that start with @var{filename}.  If
@var{predicate} is non-@code{nil} then it ignores possible completions
that don't satisfy @var{predicate}, after calling that function
with one argument, the expanded absolute file name.

If only one match exists and @var{filename} matches it exactly, the
function returns @code{t}.  The function returns @code{nil} if directory
@var{directory} contains no name starting with @var{filename}.

In the following example, suppose that the current default directory
has five files whose names begin with @samp{f}: @file{foo},
@file{file~}, @file{file.c}, @file{file.c.~1~}, and
@file{file.c.~2~}.

@example
@group
(file-name-completion "fi" "")
     @result{} "file"
@end group

@group
(file-name-completion "file.c.~1" "")
     @result{} "file.c.~1~"
@end group

@group
(file-name-completion "file.c.~1~" "")
     @result{} t
@end group

@group
(file-name-completion "file.c.~3" "")
     @result{} nil
@end group
@end example
@end defun

@defopt completion-ignored-extensions
@code{file-name-completion} usually ignores file names that end in any
string in this list.  It does not ignore them when all the possible
completions end in one of these suffixes.  This variable has no effect
on @code{file-name-all-completions}.

A typical value might look like this:

@example
@group
completion-ignored-extensions
     @result{} (".o" ".elc" "~" ".dvi")
@end group
@end example

If an element of @code{completion-ignored-extensions} ends in a slash
@samp{/}, it signals a directory.  The elements which do @emph{not} end
in a slash will never match a directory; thus, the above value will not
filter out a directory named @file{foo.elc}.
@end defopt

@node Standard File Names
@subsection Standard File Names

  Sometimes, an Emacs Lisp program needs to specify a standard file
name for a particular use---typically, to hold configuration data
specified by the current user.  Usually, such files should be located
in the directory specified by @code{user-emacs-directory}, which is
typically @file{~/.config/emacs/} or @file{~/.emacs.d/} by default (@pxref{Find
Init,,How Emacs Finds Your Init File, emacs, The GNU Emacs Manual}).
For example, abbrev definitions are stored by default in
@file{~/.config/emacs/abbrev_defs} or @file{~/.emacs.d/abbrev_defs}.
The easiest way to specify such a file name is to use the function
@code{locate-user-emacs-file}.

@defun locate-user-emacs-file base-name &optional old-name
This function returns an absolute file name for an Emacs-specific
configuration or data file.  The argument @file{base-name} should be a
relative file name.  The return value is the absolute name of a file
in the directory specified by @code{user-emacs-directory}; if that
directory does not exist, this function creates it.

If the optional argument @var{old-name} is non-@code{nil}, it
specifies a file in the user's home directory,
@file{~/@var{old-name}}.  If such a file exists, the return value is
the absolute name of that file, instead of the file specified by
@var{base-name}.  This argument is intended to be used by Emacs
packages to provide backward compatibility.  For instance, prior to
the introduction of @code{user-emacs-directory}, the abbrev file was
located in @file{~/.abbrev_defs}.  Here is the definition of
@code{abbrev-file-name}:

@example
(defcustom abbrev-file-name
  (locate-user-emacs-file "abbrev_defs" ".abbrev_defs")
  "Default name of file from which to read abbrevs."
  @dots{}
  :type 'file)
@end example
@end defun

  A lower-level function for standardizing file names, which
@code{locate-user-emacs-file} uses as a subroutine, is
@code{convert-standard-filename}.

@defun convert-standard-filename filename
This function returns a file name based on @var{filename}, which fits
the conventions of the current operating system.

On GNU and other POSIX-like systems, this simply returns @var{filename}.
On other operating systems, it may enforce system-specific file name
conventions; for example, on MS-DOS this function performs a variety
of changes to enforce MS-DOS file name limitations, including
converting any leading @samp{.} to @samp{_} and truncating to three
characters after the @samp{.}.

The recommended way to use this function is to specify a name which
fits the conventions of GNU and Unix systems, and pass it to
@code{convert-standard-filename}.
@end defun

@node Contents of Directories
@section Contents of Directories
@cindex directory-oriented functions
@cindex file names in directory

  A directory is a kind of file that contains other files entered under
various names.  Directories are a feature of the file system.

  Emacs can list the names of the files in a directory as a Lisp list,
or display the names in a buffer using the @code{ls} shell command.  In
the latter case, it can optionally display information about each file,
depending on the options passed to the @code{ls} command.

@defun directory-files directory &optional full-name match-regexp nosort count
This function returns a list of the names of the files in the directory
@var{directory}.  By default, the list is in alphabetical order.

If @var{full-name} is non-@code{nil}, the function returns the files'
absolute file names.  Otherwise, it returns the names relative to
the specified directory.

If @var{match-regexp} is non-@code{nil}, this function returns only
those file names whose non-directory part contain a match for that
regular expression---the other file names are excluded from the list.
On case-insensitive filesystems, the regular expression matching is
case-insensitive.

If @var{nosort} is non-@code{nil}, @code{directory-files} does not sort
the list, so you get the file names in no particular order.  Use this if
you want the utmost possible speed and don't care what order the files
are processed in.  If the order of processing is visible to the user,
then the user will probably be happier if you do sort the names.

If @var{count} is non-@code{nil}, the function will return names of
first @var{count} number of files, or names of all files, whichever
occurs first.  @var{count} has to be an integer greater than zero.

@example
@group
(directory-files "~lewis")
     @result{} ("#foo#" "#foo.el#" "." ".."
         "dired-mods.el" "files.texi"
         "files.texi.~1~")
@end group
@end example

An error is signaled if @var{directory} is not the name of a directory
that can be read.
@end defun

@defun directory-empty-p directory
This utility function returns @code{t} if given @var{directory} is an
accessible directory and it does not contain any files, i.e., is an
empty directory.  It will ignore @samp{.} and @samp{..} on systems
that return them as files in a directory.

Symbolic links to directories count as directories.
See @var{file-symlink-p} to distinguish symlinks.
@end defun

@cindex recursive traverse of directory tree
@defun directory-files-recursively directory regexp &optional include-directories predicate follow-symlinks
Return all files under @var{directory} whose names match @var{regexp}.
This function searches the specified @var{directory} and its
sub-directories, recursively, for files whose basenames (i.e., without
the leading directories) match the specified @var{regexp}, and returns
a list of the absolute file names of the matching files
(@pxref{Relative File Names, absolute file names}).  The file names
are returned in depth-first order, meaning that files in some
sub-directory are returned before the files in its parent directory.
In addition, matching files found in each subdirectory are sorted
alphabetically by their basenames.  By default, directories whose
names match @var{regexp} are omitted from the list, but if the
optional argument @var{include-directories} is non-@code{nil}, they
are included.

By default, all subdirectories are descended into.  If @var{predicate}
is @code{t}, errors when trying to descend into a subdirectory (for
instance, if it's not readable by this user) are ignored.  If it's
neither @code{nil} nor @code{t}, it should be a function that takes
one parameter (the subdirectory name) and should return non-@code{nil}
if the directory is to be descended into.

Symbolic links to subdirectories are not followed by default, but if
@var{follow-symlinks} is non-@code{nil}, they are followed.
@end defun


@defun locate-dominating-file file name
Starting at @var{file}, go up the directory tree hierarchy looking for
the first directory where @var{name}, a string, exists, and return that
directory.  If @var{file} is a file, its directory will serve as the
starting point for the search; otherwise @var{file} should be a
directory from which to start.  The function looks in the starting
directory, then in its parent, then in its parent's parent, etc.,
until it either finds a directory with @var{name} or reaches the root
directory of the filesystem without finding @var{name} -- in the
latter case the function returns @code{nil}.

The argument @code{name} can also be a predicate function.  The
predicate is called for every directory examined by the function,
starting from @var{file} (even if @var{file} is not a directory).  It
is called with one argument (the file or directory) and should return
non-@code{nil} if that directory is the one it is looking for.
@end defun

@cindex parent directory of file
@cindex ancestor directory of file
@cindex file, ancestor directory of
@defun file-in-directory-p file dir
This function returns @code{t} if @var{file} is a file in directory
@var{dir}, or in a subdirectory of @var{dir}.  It also returns
@code{t} if @var{file} and @var{dir} are the same directory.  It
compares the truenames of the two directories.  If @var{dir} does not
name an existing directory, the return value is @code{nil}.
@end defun

@defun directory-files-and-attributes directory &optional full-name match-regexp nosort id-format count
This is similar to @code{directory-files} in deciding which files
to report on and how to report their names.  However, instead
of returning a list of file names, it returns for each file a
list @code{(@var{filename} . @var{attributes})}, where @var{attributes}
is what @code{file-attributes} returns for that file.
The optional argument @var{id-format} has the same meaning as the
corresponding argument to @code{file-attributes} (@pxref{Definition
of file-attributes}).
@end defun

@defvr Constant directory-files-no-dot-files-regexp
This regular expression matches any file name except @samp{.} and
@samp{..}.  More precisely, it matches parts of any nonempty string
except those two.  It is useful as the @var{match-regexp} argument to
@code{directory-files} and @code{directory-files-and-attributes}:

@example
(directory-files "/foo" nil directory-files-no-dot-files-regexp)
@end example

returns @code{nil}, if directory @samp{/foo} is empty.
@end defvr

@defun file-expand-wildcards pattern &optional full regexp
This function expands the wildcard pattern @var{pattern}, returning
a list of file names that match it.

@var{pattern} is, by default, a ``glob''/wildcard string, e.g.,
@samp{"/tmp/*.png"} or @samp{"/*/*/foo.png"}, but can also be a
regular expression if the optional @var{regexp} parameter is non-@code{nil}.
In any case, the matches are applied per sub-directory, so a match
can't span a parent/sub directory.

If @var{pattern} is written as an absolute file name,
the values are absolute also.

If @var{pattern} is written as a relative file name, it is interpreted
relative to the current default directory.  The file names returned are
normally also relative to the current default directory.  However, if
@var{full} is non-@code{nil}, they are absolute.
@end defun

@defun insert-directory file switches &optional wildcard full-directory-p
This function inserts (in the current buffer) a directory listing for
directory @var{file}, formatted with @code{ls} according to
@var{switches}.  It leaves point after the inserted text.
@var{switches} may be a string of options, or a list of strings
representing individual options.

The argument @var{file} may be either a directory or a file
specification including wildcard characters.  If @var{wildcard} is
non-@code{nil}, that means treat @var{file} as a file specification with
wildcards.

If @var{full-directory-p} is non-@code{nil}, that means the directory
listing is expected to show the full contents of a directory.  You
should specify @code{t} when @var{file} is a directory and switches do
not contain @samp{-d}.  (The @samp{-d} option to @code{ls} says to
describe a directory itself as a file, rather than showing its
contents.)

On most systems, this function works by running a directory listing
program whose name is in the variable @code{insert-directory-program}.
If @var{wildcard} is non-@code{nil}, it also runs the shell specified by
@code{shell-file-name}, to expand the wildcards.

MS-DOS and MS-Windows systems usually lack the standard Unix program
@code{ls}, so this function emulates the standard Unix program @code{ls}
with Lisp code.

As a technical detail, when @var{switches} contains the long
@samp{--dired} option, @code{insert-directory} treats it specially,
for the sake of dired.  However, the normally equivalent short
@samp{-D} option is just passed on to @code{insert-directory-program},
as any other option.
@end defun

@defvar insert-directory-program
This variable's value is the program to run to generate a directory listing
for the function @code{insert-directory}.  It is ignored on systems
which generate the listing with Lisp code.
@end defvar

@node Create/Delete Dirs
@section Creating, Copying and Deleting Directories
@cindex creating, copying and deleting directories

  Most Emacs Lisp file-manipulation functions get errors when used on
files that are directories.  For example, you cannot delete a directory
with @code{delete-file}.  These special functions exist to create and
delete directories.

@findex mkdir
@deffn Command make-directory dirname &optional parents
This command creates a directory named @var{dirname}.  If
@var{parents} is non-@code{nil}, as is always the case in an
interactive call, that means to create the parent directories first,
if they don't already exist.
As a function, @code{make-directory} returns non-@code{nil} if @var{dirname}
already exists as a directory and @var{parents} is non-@code{nil},
and returns @code{nil} if it successfully created @var{dirname}.
@code{mkdir} is an alias for this.
@end deffn

@deffn Command make-empty-file filename &optional parents
This command creates an empty file named @var{filename}.
As @code{make-directory}, this command creates parent directories
if @var{parents} is non-@code{nil}.
If @var{filename} already exists, this command signals an error.
@end deffn

@deffn Command copy-directory dirname newname &optional keep-time parents copy-contents
This command copies the directory named @var{dirname} to
@var{newname}.  If @var{newname} is a directory name,
@var{dirname} will be copied to a subdirectory there.
@xref{Directory Names}.

It always sets the file modes of the copied files to match the
corresponding original file.

The third argument @var{keep-time} non-@code{nil} means to preserve the
modification time of the copied files.  A prefix arg makes
@var{keep-time} non-@code{nil}.

The fourth argument @var{parents} says whether to
create parent directories if they don't exist.  Interactively,
this happens by default.

The fifth argument @var{copy-contents}, if non-@code{nil}, means to
copy the contents of @var{dirname} directly into @var{newname} if the
latter is a directory name, instead of copying @var{dirname} into
it as a subdirectory.
@end deffn

@cindex trash
@vindex delete-by-moving-to-trash
@deffn Command delete-directory dirname &optional recursive trash
This command deletes the directory named @var{dirname}.  The function
@code{delete-file} does not work for files that are directories; you
must use @code{delete-directory} for them.  If @var{recursive} is
@code{nil}, and the directory contains any files,
@code{delete-directory} signals an error.
If recursive is non-@code{nil}, there is no error merely because the
directory or its files are deleted by some other process before
@code{delete-directory} gets to them.

@code{delete-directory} only follows symbolic links at the level of
parent directories.

If the optional argument @var{trash} is non-@code{nil} and the
variable @code{delete-by-moving-to-trash} is non-@code{nil}, this
command moves the file into the system Trash instead of deleting it.
@xref{Misc File Ops,,Miscellaneous File Operations, emacs, The GNU
Emacs Manual}.  When called interactively, @var{trash} is @code{t} if
no prefix argument is given, and @code{nil} otherwise.
@end deffn

@node Magic File Names
@section Making Certain File Names ``Magic''
@cindex magic file names

  You can implement special handling for certain file names.  This is
called making those names @dfn{magic}.  The principal use for this
feature is in implementing access to remote files (@pxref{Remote Files,,
Remote Files, emacs, The GNU Emacs Manual}).

  To define a kind of magic file name, you must supply a regular
expression to define the class of names (all those that match the
regular expression), plus a handler that implements all the primitive
Emacs file operations for file names that match.

@cindex file name handler
@vindex file-name-handler-alist
  The variable @code{file-name-handler-alist} holds a list of handlers,
together with regular expressions that determine when to apply each
handler.  Each element has this form:

@example
(@var{regexp} . @var{handler})
@end example

@noindent
All the Emacs primitives for file access and file name transformation
check the given file name against @code{file-name-handler-alist}.  If
the file name matches @var{regexp}, the primitives handle that file by
calling @var{handler}.

  The first argument given to @var{handler} is the name of the
primitive, as a symbol; the remaining arguments are the arguments that
were passed to that primitive.  (The first of these arguments is most
often the file name itself.)  For example, if you do this:

@example
(file-exists-p @var{filename})
@end example

@noindent
and @var{filename} has handler @var{handler}, then @var{handler} is
called like this:

@example
(funcall @var{handler} 'file-exists-p @var{filename})
@end example

  When a function takes two or more arguments that must be file names,
it checks each of those names for a handler.  For example, if you do
this:

@example
(expand-file-name @var{filename} @var{dirname})
@end example

@noindent
then it checks for a handler for @var{filename} and then for a handler
for @var{dirname}.  In either case, the @var{handler} is called like
this:

@example
(funcall @var{handler} 'expand-file-name @var{filename} @var{dirname})
@end example

@noindent
The @var{handler} then needs to figure out whether to handle
@var{filename} or @var{dirname}.

  If the specified file name matches more than one handler, the one
whose match starts last in the file name gets precedence.  This rule
is chosen so that handlers for jobs such as uncompression are handled
first, before handlers for jobs such as remote file access.

  Here are the operations that a magic file name handler gets to handle:

@ifnottex
@noindent
@code{abbreviate-file-name}, @code{access-file},
@code{add-name-to-file}, @code{byte-compiler-base-file-name},@*
@code{copy-directory}, @code{copy-file},
@code{delete-directory}, @code{delete-file},
@code{diff-latest-backup-file},
@code{directory-file-name},
@code{directory-files},
@code{directory-files-and-attributes},
@code{dired-compress-file}, @code{dired-uncache},
@code{exec-path}, @code{expand-file-name},@*
@code{file-accessible-directory-p},
@code{file-acl},
@code{file-attributes},
@code{file-directory-p},
@code{file-equal-p},
@code{file-executable-p}, @code{file-exists-p},
@code{file-in-directory-p},
@code{file-local-copy}, @code{file-locked-p},
@code{file-modes}, @code{file-name-all-completions},
@code{file-name-as-directory},
@code{file-name-case-insensitive-p},
@code{file-name-completion},
@code{file-name-directory},
@code{file-name-nondirectory},
@code{file-name-sans-versions}, @code{file-newer-than-file-p},
@code{file-notify-add-watch}, @code{file-notify-rm-watch},
@code{file-notify-valid-p},
@code{file-ownership-preserved-p},
@code{file-readable-p}, @code{file-regular-p},
@code{file-remote-p}, @code{file-selinux-context},
@code{file-symlink-p}, @code{file-system-info},
@code{file-truename}, @code{file-user-uid},
@code{file-writable-p},
@code{find-backup-file-name},@*
@code{get-file-buffer},
@code{insert-directory},
@code{insert-file-contents},@*
@code{list-system-processes},
@code{load}, @code{lock-file},
@code{make-auto-save-file-name},
@code{make-directory},
@code{make-lock-file-name},
@code{make-nearby-temp-file},
@code{make-process},
@code{make-symbolic-link},@*
@code{memory-info}, @code{process-attributes}, @code{process-file},
@code{rename-file}, @code{set-file-acl}, @code{set-file-modes},
@code{set-file-selinux-context}, @code{set-file-times},
@code{set-visited-file-modtime}, @code{shell-command},
@code{start-file-process},
@code{substitute-in-file-name},@*
@code{temporary-file-directory},
@code{unhandled-file-name-directory},
@code{unlock-file},
@code{vc-registered},
@code{verify-visited-file-modtime},@*
@code{write-region}.
@end ifnottex
@iftex
@noindent
@flushleft
@code{abbreviate-file-name}, @code{access-file},
@code{add-name-to-file},
@code{byte-com@discretionary{}{}{}piler-base-file-name},
@code{copy-directory}, @code{copy-file},
@code{delete-directory}, @code{delete-file},
@code{diff-latest-backup-file},
@code{directory-file-name},
@code{directory-files},
@code{directory-files-and-at@discretionary{}{}{}tributes},
@code{dired-compress-file}, @code{dired-uncache},
@code{exec-path}, @code{expand-file-name},
@code{file-accessible-direc@discretionary{}{}{}tory-p},
@code{file-acl},
@code{file-attributes},
@code{file-direc@discretionary{}{}{}tory-p},
@code{file-equal-p},
@code{file-executable-p}, @code{file-exists-p},
@code{file-in-directory-p},
@code{file-local-copy}, @code{file-locked-p},
@code{file-modes}, @code{file-name-all-completions},
@code{file-name-as-directory},
@code{file-name-case-insensitive-p},
@code{file-name-completion},
@code{file-name-directory},
@code{file-name-nondirec@discretionary{}{}{}tory},
@code{file-name-sans-versions}, @code{file-newer-than-file-p},
@code{file-notify-add-watch}, @code{file-notify-rm-watch},
@code{file-notify-valid-p},
@code{file-ownership-pre@discretionary{}{}{}served-p},
@code{file-readable-p}, @code{file-regular-p},
@code{file-remote-p}, @code{file-selinux-context},
@code{file-symlink-p}, @code{file-system-info},
@code{file-truename}, @code{file-user-uid},
@code{file-writable-p},
@code{find-backup-file-name},
@code{get-file-buffer},
@code{insert-directory},
@code{insert-file-contents},
@code{list-system-processes},
@code{load}, @code{lock-file},
@code{make-auto-save-file-name},
@code{make-direc@discretionary{}{}{}tory},
@code{make-lock-file-name},
@code{make-nearby-temp-file},
@code{make-process},
@code{make-symbolic-link},
@code{memory-info}, @code{process-attributes}, @code{process-file},
@code{rename-file}, @code{set-file-acl}, @code{set-file-modes},
@code{set-file-selinux-context}, @code{set-file-times},
@code{set-visited-file-modtime}, @code{shell-command},
@code{start-file-process},
@code{substitute-in-file-name},
@code{temporary-file-directory},
@code{unhandled-file-name-directory},
@code{unlock-file},
@code{vc-regis@discretionary{}{}{}tered},
@code{verify-visited-file-modtime},
@code{write-region}.
@end flushleft
@end iftex

  Handlers for @code{insert-file-contents} typically need to clear the
buffer's modified flag, with @code{(set-buffer-modified-p nil)}, if the
@var{visit} argument is non-@code{nil}.  This also has the effect of
unlocking the buffer if it is locked.

  The handler function must handle all of the above operations, and
possibly others to be added in the future.  It need not implement all
these operations itself---when it has nothing special to do for a
certain operation, it can reinvoke the primitive, to handle the
operation in the usual way.  It should always reinvoke the primitive
for an operation it does not recognize.  Here's one way to do this:

@smallexample
(defun my-file-handler (operation &rest args)
  ;; @r{First check for the specific operations}
  ;; @r{that we have special handling for.}
  (cond ((eq operation 'insert-file-contents) @dots{})
        ((eq operation 'write-region) @dots{})
        @dots{}
        ;; @r{Handle any operation we don't know about.}
        (t (let ((inhibit-file-name-handlers
                  (cons 'my-file-handler
                        (and (eq inhibit-file-name-operation operation)
                             inhibit-file-name-handlers)))
                 (inhibit-file-name-operation operation))
             (apply operation args)))))
@end smallexample

  When a handler function decides to call the ordinary Emacs primitive for
the operation at hand, it needs to prevent the primitive from calling
the same handler once again, thus leading to an infinite recursion.  The
example above shows how to do this, with the variables
@code{inhibit-file-name-handlers} and
@code{inhibit-file-name-operation}.  Be careful to use them exactly as
shown above; the details are crucial for proper behavior in the case of
multiple handlers, and for operations that have two file names that may
each have handlers.

@kindex safe-magic @r{(property)}
  Handlers that don't really do anything special for actual access to the
file---such as the ones that implement completion of host names for
remote file names---should have a non-@code{nil} @code{safe-magic}
property.  For instance, Emacs normally protects directory names
@c FIXME I don't think this means the PATH environment variable?
it finds in @code{PATH} from becoming magic, if they look like magic
file names, by prefixing them with @samp{/:}.  But if the handler that
would be used for them has a non-@code{nil} @code{safe-magic}
property, the @samp{/:} is not added.

@kindex operations @r{(property)}
  A file name handler can have an @code{operations} property to
declare which operations it handles in a nontrivial way.  If this
property has a non-@code{nil} value, it should be a list of
operations; then only those operations will call the handler.  This
avoids inefficiency, but its main purpose is for autoloaded handler
functions, so that they won't be loaded except when they have real
work to do.

  Simply deferring all operations to the usual primitives does not
work.  For instance, if the file name handler applies to
@code{file-exists-p}, then it must handle @code{load} itself, because
the usual @code{load} code won't work properly in that case.  However,
if the handler uses the @code{operations} property to say it doesn't
handle @code{file-exists-p}, then it need not handle @code{load}
nontrivially.

@defvar inhibit-file-name-handlers
This variable holds a list of handlers whose use is presently inhibited
for a certain operation.
@end defvar

@defvar inhibit-file-name-operation
The operation for which certain handlers are presently inhibited.
@end defvar

@defun find-file-name-handler file operation
This function returns the handler function for file name @var{file},
or @code{nil} if there is none.  The argument @var{operation} should
be the operation to be performed on the file---the value you will pass
to the handler as its first argument when you call it.  If
@var{operation} equals @code{inhibit-file-name-operation}, or if it is
not found in the @code{operations} property of the handler, this
function returns @code{nil}.
@end defun

@defun file-local-copy filename
This function copies file @var{filename} to an ordinary non-magic file
on the local machine, if it isn't on the local machine already.  Magic
file names should handle the @code{file-local-copy} operation if they
refer to files on other machines.  A magic file name that is used for
other purposes than remote file access should not handle
@code{file-local-copy}; then this function will treat the file as
local.

If @var{filename} is local, whether magic or not, this function does
nothing and returns @code{nil}.  Otherwise it returns the file name
of the local copy file.
@end defun

@defun file-remote-p filename &optional identification connected
This function tests whether @var{filename} is a remote file.  If
@var{filename} is local (not remote), the return value is @code{nil}.
If @var{filename} is indeed remote, the return value is a string that
identifies the remote system.

This identifier string can include a host name and a user name, as
well as characters designating the method used to access the remote
system.  For example, the remote identifier string for the file name
@code{/sudo::/some/file} is @code{/sudo:root@@localhost:}.

If @code{file-remote-p} returns the same identifier for two different
file names, that means they are stored on the same file system and can
be accessed locally with respect to each other.  This means, for
example, that it is possible to start a remote process accessing both
files at the same time.  Implementers of file name handlers need to
ensure this principle is valid.

@var{identification} specifies which part of the identifier shall be
returned as string.  @var{identification} can be the symbol
@code{method}, @code{user} or @code{host}; any other value is handled
like @code{nil} and means to return the complete identifier string.
In the example above, the remote @code{user} identifier string would
be @code{root}.

If @var{connected} is non-@code{nil}, this function returns @code{nil}
even if @var{filename} is remote, if Emacs has no network connection
to its host.  This is useful when you want to avoid the delay of
making connections when they don't exist.
@end defun

@defun unhandled-file-name-directory filename
This function returns the name of a directory that is not magic.  For
a non-magic @var{filename} it returns the corresponding directory name
(@pxref{Directory Names}).  For a magic @var{filename}, it invokes the
file name handler, which therefore decides what value to return.  If
@var{filename} is not accessible from a local process, then the file
name handler should indicate that by returning @code{nil}.

This is useful for running a subprocess; every subprocess must have a
non-magic directory to serve as its current directory, and this function
is a good way to come up with one.
@end defun

@cindex local part of remote file name
@defun file-local-name filename
This function returns the @dfn{local part} of @var{filename}.  This is
the part of the file's name that identifies it on the remote host, and
is typically obtained by removing from the remote file name the parts
that specify the remote host and the method of accessing it.  For
example:

@smallexample
(file-local-name "/ssh:@var{user}@@@var{host}:/foo/bar")
     @result{} "/foo/bar"
@end smallexample

For a remote @var{filename}, this function returns a file name which
could be used directly as an argument of a remote process
(@pxref{Asynchronous Processes}, and @pxref{Synchronous Processes}),
and as the program to run on the remote host.  If @var{filename} is
local, this function returns it unchanged.
@end defun

@defopt remote-file-name-inhibit-cache
The attributes of remote files can be cached for better performance.  If
they are changed outside of Emacs's control, the cached values become
invalid, and must be reread.

When this variable is set to @code{nil}, cached values are never
expired.  Use this setting with caution, only if you are sure nothing
other than Emacs ever changes the remote files.  If it is set to
@code{t}, cached values are never used.  This is the safest value, but
could result in performance degradation.

A compromise is to set it to a positive number.  This means that
cached values are used for that amount of seconds since they were
cached.  If a remote file is checked regularly, it might be a good
idea to let-bind this variable to a value less than the time period
between consecutive checks.  For example:

@example
(defun display-time-file-nonempty-p (file)
  (let ((remote-file-name-inhibit-cache
         (- display-time-interval 5)))
    (and (file-exists-p file)
         (< 0 (file-attribute-size
               (file-attributes
                (file-chase-links file)))))))
@end example
@end defopt

@node Format Conversion
@section File Format Conversion

@cindex file format conversion
@cindex encoding file formats
@cindex decoding file formats
@cindex text properties in files
@cindex saving text properties
  Emacs performs several steps to convert the data in a buffer (text,
text properties, and possibly other information) to and from a
representation suitable for storing into a file.  This section describes
the fundamental functions that perform this @dfn{format conversion},
namely @code{insert-file-contents} for reading a file into a buffer,
and @code{write-region} for writing a buffer into a file.

@menu
* Overview: Format Conversion Overview.     @code{insert-file-contents} and @code{write-region}.
* Round-Trip: Format Conversion Round-Trip. Using @code{format-alist}.
* Piecemeal: Format Conversion Piecemeal.   Specifying non-paired conversion.
@end menu

@node Format Conversion Overview
@subsection Overview
@noindent
The function @code{insert-file-contents}:

@itemize
@item initially, inserts bytes from the file into the buffer;
@item decodes bytes to characters as appropriate;
@item processes formats as defined by entries in @code{format-alist}; and
@item calls functions in @code{after-insert-file-functions}.
@end itemize

@noindent
The function @code{write-region}:

@itemize
@item initially, calls functions in @code{write-region-annotate-functions};
@item processes formats as defined by entries in @code{format-alist};
@item encodes characters to bytes as appropriate; and
@item modifies the file with the bytes.
@end itemize

  This shows the symmetry of the lowest-level operations; reading and
writing handle things in opposite order.  The rest of this section
describes the two facilities surrounding the three variables named
above, as well as some related functions.  @ref{Coding Systems}, for
details on character encoding and decoding.

@node Format Conversion Round-Trip
@subsection Round-Trip Specification

  The most general of the two facilities is controlled by the variable
@code{format-alist}, a list of @dfn{file format} specifications, which
describe textual representations used in files for the data in an Emacs
buffer.  The descriptions for reading and writing are paired, which is
why we call this ``round-trip'' specification
(@pxref{Format Conversion Piecemeal}, for non-paired specification).

@defvar format-alist
This list contains one format definition for each defined file format.
Each format definition is a list of this form:

@example
(@var{name} @var{doc-string} @var{regexp} @var{from-fn} @var{to-fn} @var{modify} @var{mode-fn} @var{preserve})
@end example
@end defvar

@cindex format definition
@noindent
Here is what the elements in a format definition mean:

@table @var
@item name
The name of this format.

@item doc-string
A documentation string for the format.

@item regexp
A regular expression which is used to recognize files represented in
this format.  If @code{nil}, the format is never applied automatically.

@item from-fn
A shell command or function to decode data in this format (to convert
file data into the usual Emacs data representation).

A shell command is represented as a string; Emacs runs the command as a
filter to perform the conversion.

If @var{from-fn} is a function, it is called with two arguments, @var{begin}
and @var{end}, which specify the part of the buffer it should convert.
It should convert the text by editing it in place.  Since this can
change the length of the text, @var{from-fn} should return the modified
end position.

One responsibility of @var{from-fn} is to make sure that the beginning
of the file no longer matches @var{regexp}.  Otherwise it is likely to
get called again.  Also, @var{from-fn} must not involve buffers or
files other than the one being decoded, otherwise the internal buffer
used for formatting might be overwritten.

@item to-fn
A shell command or function to encode data in this format---that is, to
convert the usual Emacs data representation into this format.

If @var{to-fn} is a string, it is a shell command; Emacs runs the
command as a filter to perform the conversion.

If @var{to-fn} is a function, it is called with three arguments:
@var{begin} and @var{end}, which specify the part of the buffer it
should convert, and @var{buffer}, which specifies which buffer.  There
are two ways it can do the conversion:

@itemize @bullet
@item
By editing the buffer in place.  In this case, @var{to-fn} should
return the end-position of the range of text, as modified.

@item
By returning a list of annotations.  This is a list of elements of the
form @code{(@var{position} . @var{string})}, where @var{position} is an
integer specifying the relative position in the text to be written, and
@var{string} is the annotation to add there.  The list must be sorted in
order of position when @var{to-fn} returns it.

When @code{write-region} actually writes the text from the buffer to the
file, it intermixes the specified annotations at the corresponding
positions.  All this takes place without modifying the buffer.
@end itemize

@var{to-fn} must not involve buffers or files other than the one being
encoded, otherwise the internal buffer used for formatting might be
overwritten.

@item modify
A flag, @code{t} if the encoding function modifies the buffer, and
@code{nil} if it works by returning a list of annotations.

@item mode-fn
A minor-mode function to call after visiting a file converted from this
format.  The function is called with one argument, the integer 1;
that tells a minor-mode function to enable the mode.

@item preserve
A flag, @code{t} if @code{format-write-file} should not remove this format
from @code{buffer-file-format}.
@end table

The function @code{insert-file-contents} automatically recognizes file
formats when it reads the specified file.  It checks the text of the
beginning of the file against the regular expressions of the format
definitions, and if it finds a match, it calls the decoding function for
that format.  Then it checks all the known formats over again.
It keeps checking them until none of them is applicable.

Visiting a file, with @code{find-file-noselect} or the commands that use
it, performs conversion likewise (because it calls
@code{insert-file-contents}); it also calls the mode function for each
format that it decodes.  It stores a list of the format names in the
buffer-local variable @code{buffer-file-format}.

@defvar buffer-file-format
This variable states the format of the visited file.  More precisely,
this is a list of the file format names that were decoded in the course
of visiting the current buffer's file.  It is always buffer-local in all
buffers.
@end defvar

When @code{write-region} writes data into a file, it first calls the
encoding functions for the formats listed in @code{buffer-file-format},
in the order of appearance in the list.

@deffn Command format-write-file file format &optional confirm
This command writes the current buffer contents into the file @var{file}
in a format based on @var{format}, which is a list of format names.  It
constructs the actual format starting from @var{format}, then appending
any elements from the value of @code{buffer-file-format} with a
non-@code{nil} @var{preserve} flag (see above), if they are not already
present in @var{format}.  It then updates @code{buffer-file-format} with
this format, making it the default for future saves.  Except for the
@var{format} argument, this command is similar to @code{write-file}.  In
particular, @var{confirm} has the same meaning and interactive treatment
as the corresponding argument to @code{write-file}.  @xref{Definition of
write-file}.
@end deffn

@deffn Command format-find-file file format
This command finds the file @var{file}, converting it according to
format @var{format}.  It also makes @var{format} the default if the
buffer is saved later.

The argument @var{format} is a list of format names.  If @var{format} is
@code{nil}, no conversion takes place.  Interactively, typing just
@key{RET} for @var{format} specifies @code{nil}.
@end deffn

@deffn Command format-insert-file file format &optional beg end
This command inserts the contents of file @var{file}, converting it
according to format @var{format}.  If @var{beg} and @var{end} are
non-@code{nil}, they specify which part of the file to read, as in
@code{insert-file-contents} (@pxref{Reading from Files}).

The return value is like what @code{insert-file-contents} returns: a
list of the absolute file name and the length of the data inserted
(after conversion).

The argument @var{format} is a list of format names.  If @var{format} is
@code{nil}, no conversion takes place.  Interactively, typing just
@key{RET} for @var{format} specifies @code{nil}.
@end deffn

@defvar buffer-auto-save-file-format
This variable specifies the format to use for auto-saving.  Its value is
a list of format names, just like the value of
@code{buffer-file-format}; however, it is used instead of
@code{buffer-file-format} for writing auto-save files.  If the value
is @code{t}, the default, auto-saving uses the same format as a
regular save in the same buffer.  This variable is always buffer-local
in all buffers.
@end defvar

@node Format Conversion Piecemeal
@subsection Piecemeal Specification

  In contrast to the round-trip specification described in the previous
subsection (@pxref{Format Conversion Round-Trip}), you can use the variables
@code{after-insert-file-functions} and @code{write-region-annotate-functions}
to separately control the respective reading and writing conversions.

  Conversion starts with one representation and produces another
representation.  When there is only one conversion to do, there is no
conflict about what to start with.  However, when there are multiple
conversions involved, conflict may arise when two conversions need to
start with the same data.

  This situation is best understood in the context of converting text
properties during @code{write-region}.  For example, the character at
position 42 in a buffer is @samp{X} with a text property @code{foo}.  If
the conversion for @code{foo} is done by inserting into the buffer, say,
@samp{FOO:}, then that changes the character at position 42 from
@samp{X} to @samp{F}.  The next conversion will start with the wrong
data straight away.

  To avoid conflict, cooperative conversions do not modify the buffer,
but instead specify @dfn{annotations}, a list of elements of the form
@code{(@var{position} . @var{string})}, sorted in order of increasing
@var{position}.

  If there is more than one conversion, @code{write-region} merges their
annotations destructively into one sorted list.  Later, when the text
from the buffer is actually written to the file, it intermixes the
specified annotations at the corresponding positions.  All this takes
place without modifying the buffer.

@c ??? What about "overriding" conversions like those allowed
@c ??? for 'write-region-annotate-functions', below?  --ttn

  In contrast, when reading, the annotations intermixed with the text
are handled immediately.  @code{insert-file-contents} sets point to
the beginning of some text to be converted, then calls the conversion
functions with the length of that text.  These functions should always
return with point at the beginning of the inserted text.  This
approach makes sense for reading because annotations removed by the
first converter can't be mistakenly processed by a later converter.
Each conversion function should scan for the annotations it
recognizes, remove the annotation, modify the buffer text (to set a
text property, for example), and return the updated length of the
text, as it stands after those changes.  The value returned by one
function becomes the argument to the next function.

@defvar write-region-annotate-functions
A list of functions for @code{write-region} to call.  Each function in
the list is called with two arguments: the start and end of the region
to be written.  These functions should not alter the contents of the
buffer.  Instead, they should return annotations.

As a special case, a function may return with a different buffer
current.  Emacs takes this to mean that the current buffer contains
altered text to be output.  It therefore changes the @var{start} and
@var{end} arguments of the @code{write-region} call, giving them the
values of @code{point-min} and @code{point-max} in the new buffer,
respectively.  It also discards all previous annotations, because they
should have been dealt with by this function.
@end defvar

@defvar write-region-post-annotation-function
The value of this variable, if non-@code{nil}, should be a function.
This function is called, with no arguments, after @code{write-region}
has completed.

If any function in @code{write-region-annotate-functions} returns with
a different buffer current, Emacs calls
@code{write-region-post-annotation-function} more than once.  Emacs
calls it with the last buffer that was current, and again with the
buffer before that, and so on back to the original buffer.

Thus, a function in @code{write-region-annotate-functions} can create
a buffer, give this variable the local value of @code{kill-buffer} in
that buffer, set up the buffer with altered text, and make the buffer
current.  The buffer will be killed after @code{write-region} is done.
@end defvar

@defvar after-insert-file-functions
Each function in this list is called by @code{insert-file-contents}
with one argument, the number of characters inserted, and with point
at the beginning of the inserted text.  Each function should leave
point unchanged, and return the new character count describing the
inserted text as modified by the function.
@c ??? The docstring mentions a handler from 'file-name-handler-alist'
@c     "intercepting" 'insert-file-contents'.  Hmmm.  --ttn
@end defvar

  We invite users to write Lisp programs to store and retrieve text
properties in files, using these hooks, and thus to experiment with
various data formats and find good ones.  Eventually we hope users
will produce good, general extensions we can install in Emacs.

  We suggest not trying to handle arbitrary Lisp objects as text property
names or values---because a program that general is probably difficult
to write, and slow.  Instead, choose a set of possible data types that
are reasonably flexible, and not too hard to encode.