Erlang logo
Reference Manual
Release Notes
PDF
Top

Kernel
Reference Manual
Version 5.2


Expand All
Contract All

Table of Contents

file

MODULE

file

MODULE SUMMARY

File interface module.

DESCRIPTION

This module provides an interface to the file system.

On operating systems with thread support, file operations can be performed in threads of their own, allowing other Erlang processes to continue executing in parallel with the file operations. See command-line flag +A in erl(1).

Regarding filename encoding, the Erlang VM can operate in two modes. The current mode can be queried using function native_name_encoding/0. It returns latin1 or utf8.

In latin1 mode, the Erlang VM does not change the encoding of filenames. In utf8 mode, filenames can contain Unicode characters greater than 255 and the VM converts filenames back and forth to the native filename encoding (usually UTF-8, but UTF-16 on Windows).

The default mode depends on the operating system. Windows and MacOS X enforce consistent filename encoding and therefore the VM uses utf8 mode.

On operating systems with transparent naming (for example, all Unix systems except MacOS X), default is utf8 if the terminal supports UTF-8, otherwise latin1. The default can be overridden using +fnl (to force latin1 mode) or +fnu (to force utf8 mode) when starting erts:erl.

On operating systems with transparent naming, files can be inconsistently named, for example, some files are encoded in UTF-8 while others are encoded in ISO Latin-1. The concept of raw filenames is introduced to handle file systems with inconsistent naming when running in utf8 mode.

A raw filename is a filename specified as a binary. The Erlang VM does not translate a filename specified as a binary on systems with transparent naming.

When running in utf8 mode, functions list_dir/1 and read_link/1 never return raw filenames. To return all filenames including raw filenames, use functions list_dir_all/1 and read_link_all/1.

See also section Notes About Raw Filenames in the STDLIB User's Guide.

DATA TYPES

deep_list() = [char() | atom() | deep_list()]

fd()

A file descriptor representing a file opened in raw mode.

filename() = string()

filename_all() = string() | binary()

io_device() = pid() | fd()

As returned by open/2; pid() is a process handling I/O-protocols.

name() = string() | atom() | deep_list()

If VM is in Unicode filename mode, string() and char() are allowed to be > 255.

name_all() =
    string() | atom() | deep_list() | (RawFilename :: binary())

If VM is in Unicode filename mode, string() and char() are allowed to be > 255. RawFilename is a filename not subject to Unicode translation, meaning that it can contain characters not conforming to the Unicode encoding expected from the file system (that is, non-UTF-8 characters although the VM is started in Unicode filename mode).

posix() =
    eacces |
    eagain |
    ebadf |
    ebusy |
    edquot |
    eexist |
    efault |
    efbig |
    eintr |
    einval |
    eio |
    eisdir |
    eloop |
    emfile |
    emlink |
    enametoolong |
    enfile |
    enodev |
    enoent |
    enomem |
    enospc |
    enotblk |
    enotdir |
    enotsup |
    enxio |
    eperm |
    epipe |
    erofs |
    espipe |
    esrch |
    estale |
    exdev

An atom that is named from the POSIX error codes used in Unix, and in the runtime libraries of most C compilers.

date_time() = calendar:datetime()

Must denote a valid date and time.

file_info() =
    #file_info{size = integer() >= 0 | undefined,
               type =
                   device |
                   directory |
                   other |
                   regular |
                   symlink |
                   undefined,
               access =
                   read | write | read_write | none | undefined,
               atime =
                   file:date_time() |
                   integer() >= 0 |
                   undefined,
               mtime =
                   file:date_time() |
                   integer() >= 0 |
                   undefined,
               ctime =
                   file:date_time() |
                   integer() >= 0 |
                   undefined,
               mode = integer() >= 0 | undefined,
               links = integer() >= 0 | undefined,
               major_device = integer() >= 0 | undefined,
               minor_device = integer() >= 0 | undefined,
               inode = integer() >= 0 | undefined,
               uid = integer() >= 0 | undefined,
               gid = integer() >= 0 | undefined}

location() =
    integer() |
    {bof, Offset :: integer()} |
    {cur, Offset :: integer()} |
    {eof, Offset :: integer()} |
    bof |
    cur |
    eof

mode() =
    read |
    write |
    append |
    exclusive |
    raw |
    binary |
    {delayed_write,
     Size :: integer() >= 0,
     Delay :: integer() >= 0} |
    delayed_write |
    {read_ahead, Size :: integer() >= 1} |
    read_ahead |
    compressed |
    {encoding, unicode:encoding()} |
    sync

file_info_option() =
    {time, local} | {time, universal} | {time, posix} | raw

EXPORTS

advise(IoDevice, Offset, Length, Advise) -> ok | {error, Reason}

Types:

IoDevice = io_device()
Offset = Length = integer()
Advise = posix_file_advise()
Reason = posix() | badarg
posix_file_advise() =
    normal |
    sequential |
    random |
    no_reuse |
    will_need |
    dont_need

advise/4 can be used to announce an intention to access file data in a specific pattern in the future, thus allowing the operating system to perform appropriate optimizations.

On some platforms, this function might have no effect.

allocate(File, Offset, Length) -> ok | {error, posix()}

Types:

Offset = Length = integer() >= 0

allocate/3 can be used to preallocate space for a file.

This function only succeeds in platforms that provide this feature. When it succeeds, space is preallocated for the file but the file size might not be updated. This behaviour depends on the preallocation implementation. To guarantee that the file size is updated, truncate the file to the new size.

change_group(Filename, Gid) -> ok | {error, Reason}

Types:

Filename = name_all()
Gid = integer()
Reason = posix() | badarg

Changes group of a file. See write_file_info/2.

change_mode(Filename, Mode) -> ok | {error, Reason}

Types:

Filename = name_all()
Mode = integer()
Reason = posix() | badarg

Changes permissions of a file. See write_file_info/2.

change_owner(Filename, Uid) -> ok | {error, Reason}

Types:

Filename = name_all()
Uid = integer()
Reason = posix() | badarg

Changes owner of a file. See write_file_info/2.

change_owner(Filename, Uid, Gid) -> ok | {error, Reason}

Types:

Filename = name_all()
Uid = Gid = integer()
Reason = posix() | badarg

Changes owner and group of a file. See write_file_info/2.

change_time(Filename, Mtime) -> ok | {error, Reason}

Types:

Filename = name_all()
Mtime = date_time()
Reason = posix() | badarg

Changes the modification and access times of a file. See write_file_info/2.

change_time(Filename, Atime, Mtime) -> ok | {error, Reason}

Types:

Filename = name_all()
Atime = Mtime = date_time()
Reason = posix() | badarg

Changes the modification and last access times of a file. See write_file_info/2.

close(IoDevice) -> ok | {error, Reason}

Types:

IoDevice = io_device()
Reason = posix() | badarg | terminated

Closes the file referenced by IoDevice. It mostly returns ok, except for some severe errors such as out of memory.

Notice that if option delayed_write was used when opening the file, close/1 can return an old write error and not even try to close the file. See open/2.

consult(Filename) -> {ok, Terms} | {error, Reason}

Types:

Filename = name_all()
Terms = [term()]
Reason =
    posix() |
    badarg |
    terminated |
    system_limit |
    {Line :: integer(), Mod :: module(), Term :: term()}

Reads Erlang terms, separated by '.', from Filename. Returns one of the following:

{ok, Terms}

The file was successfully read.

{error, atom()}

An error occurred when opening the file or reading it. For a list of typical error codes, see open/2.

{error, {Line, Mod, Term}}

An error occurred when interpreting the Erlang terms in the file. To convert the three-element tuple to an English description of the error, use format_error/1.

Example:

f.txt:  {person, "kalle", 25}.
        {person, "pelle", 30}.
1> file:consult("f.txt").
{ok,[{person,"kalle",25},{person,"pelle",30}]}

The encoding of Filename can be set by a comment, as described in epp(3).

copy(Source, Destination) -> {ok, BytesCopied} | {error, Reason}
copy(Source, Destination, ByteCount) ->
        {ok, BytesCopied} | {error, Reason}

Types:

Source = Destination = io_device() | Filename | {Filename, Modes}
Filename = name_all()
Modes = [mode()]
ByteCount = integer() >= 0 | infinity
BytesCopied = integer() >= 0
Reason = posix() | badarg | terminated

Copies ByteCount bytes from Source to Destination. Source and Destination refer to either filenames or IO devices from, for example, open/2. ByteCount defaults to infinity, denoting an infinite number of bytes.

Argument Modes is a list of possible modes, see open/2, and defaults to [].

If both Source and Destination refer to filenames, the files are opened with [read, binary] and [write, binary] prepended to their mode lists, respectively, to optimize the copy.

If Source refers to a filename, it is opened with read mode prepended to the mode list before the copy, and closed when done.

If Destination refers to a filename, it is opened with write mode prepended to the mode list before the copy, and closed when done.

Returns {ok, BytesCopied}, where BytesCopied is the number of bytes that was copied, which can be less than ByteCount if end of file was encountered on the source. If the operation fails, {error, Reason} is returned.

Typical error reasons: as for open/2 if a file had to be opened, and as for read/2 and write/2.

datasync(IoDevice) -> ok | {error, Reason}

Types:

IoDevice = io_device()
Reason = posix() | badarg | terminated

Ensures that any buffers kept by the operating system (not by the Erlang runtime system) are written to disk. In many ways it resembles fsync but it does not update some of the metadata of the file, such as the access time. On some platforms this function has no effect.

Applications that access databases or log files often write a tiny data fragment (for example, one line in a log file) and then call fsync() immediately to ensure that the written data is physically stored on the hard disk. Unfortunately, fsync() always initiates two write operations: one for the newly written data and another one to update the modification time stored in the inode. If the modification time is not a part of the transaction concept, fdatasync() can be used to avoid unnecessary inode disk write operations.

Available only in some POSIX systems, this call results in a call to fsync(), or has no effect in systems not providing the fdatasync() syscall.

del_dir(Dir) -> ok | {error, Reason}

Types:

Reason = posix() | badarg

Tries to delete directory Dir. The directory must be empty before it can be deleted. Returns ok if successful.

Typical error reasons:

eacces

Missing search or write permissions for the parent directories of Dir.

eexist

The directory is not empty.

enoent

The directory does not exist.

enotdir

A component of Dir is not a directory. On some platforms, enoent is returned instead.

einval

Attempt to delete the current directory. On some platforms, eacces is returned instead.

delete(Filename) -> ok | {error, Reason}

Types:

Filename = name_all()
Reason = posix() | badarg

Tries to delete file Filename. Returns ok if successful.

Typical error reasons:

enoent

The file does not exist.

eacces

Missing permission for the file or one of its parents.

eperm

The file is a directory and the user is not superuser.

enotdir

A component of the filename is not a directory. On some platforms, enoent is returned instead.

einval

Filename has an improper type, such as tuple.

Warning

In a future release, a bad type for argument Filename will probably generate an exception.

eval(Filename) -> ok | {error, Reason}

Types:

Filename = name_all()
Reason =
    posix() |
    badarg |
    terminated |
    system_limit |
    {Line :: integer(), Mod :: module(), Term :: term()}

Reads and evaluates Erlang expressions, separated by '.' (or ',', a sequence of expressions is also an expression) from Filename. The result of the evaluation is not returned; any expression sequence in the file must be there for its side effect. Returns one of the following:

ok

The file was read and evaluated.

{error, atom()}

An error occurred when opening the file or reading it. For a list of typical error codes, see open/2.

{error, {Line, Mod, Term}}

An error occurred when interpreting the Erlang expressions in the file. To convert the three-element tuple to an English description of the error, use format_error/1.

The encoding of Filename can be set by a comment, as described in epp(3).

eval(Filename, Bindings) -> ok | {error, Reason}

Types:

Filename = name_all()
Reason =
    posix() |
    badarg |
    terminated |
    system_limit |
    {Line :: integer(), Mod :: module(), Term :: term()}

The same as eval/1, but the variable bindings Bindings are used in the evaluation. For information about the variable bindings, see erl_eval(3).

format_error(Reason) -> Chars

Types:

Reason =
    posix() |
    badarg |
    terminated |
    system_limit |
    {Line :: integer(), Mod :: module(), Term :: term()}
Chars = string()

Given the error reason returned by any function in this module, returns a descriptive string of the error in English.

get_cwd() -> {ok, Dir} | {error, Reason}

Types:

Reason = posix()

Returns {ok, Dir}, where Dir is the current working directory of the file server.

Note

In rare circumstances, this function can fail on Unix. It can occur if read permission does not exist for the parent directories of the current directory.

A typical error reason:

eacces

Missing read permission for one of the parents of the current directory.

get_cwd(Drive) -> {ok, Dir} | {error, Reason}

Types:

Drive = string()
Reason = posix() | badarg

Returns {ok, Dir} or {error, Reason}, where Dir is the current working directory of the specified drive.

Drive is to be of the form "Letter:", for example, "c:".

Returns {error, enotsup} on platforms that have no concept of current drive (Unix, for example).

Typical error reasons:

enotsup

The operating system has no concept of drives.

eacces

The drive does not exist.

einval

The format of Drive is invalid.

list_dir(Dir) -> {ok, Filenames} | {error, Reason}

Types:

Filenames = [filename()]
Reason =
    posix() |
    badarg |
    {no_translation, Filename :: unicode:latin1_binary()}

Lists all files in a directory, except files with raw filenames. Returns {ok, Filenames} if successful, otherwise {error, Reason}. Filenames is a list of the names of all the files in the directory. The names are not sorted.

Typical error reasons:

eacces

Missing search or write permissions for Dir or one of its parent directories.

enoent

The directory does not exist.

{no_translation, Filename}

Filename is a binary() with characters coded in ISO Latin-1 and the VM was started with parameter +fnue.

list_dir_all(Dir) -> {ok, Filenames} | {error, Reason}

Types:

Filenames = [filename_all()]
Reason = posix() | badarg

Lists all the files in a directory, including files with raw filenames. Returns {ok, Filenames} if successful, otherwise {error, Reason}. Filenames is a list of the names of all the files in the directory. The names are not sorted.

Typical error reasons:

eacces

Missing search or write permissions for Dir or one of its parent directories.

enoent

The directory does not exist.

make_dir(Dir) -> ok | {error, Reason}

Types:

Reason = posix() | badarg

Tries to create directory Dir. Missing parent directories are not created. Returns ok if successful.

Typical error reasons:

eacces

Missing search or write permissions for the parent directories of Dir.

eexist

A file or directory named Dir exists already.

enoent

A component of Dir does not exist.

enospc

No space is left on the device.

enotdir

A component of Dir is not a directory. On some platforms, enoent is returned instead.

make_link(Existing, New) -> ok | {error, Reason}

Types:

Existing = New = name_all()
Reason = posix() | badarg

Makes a hard link from Existing to New on platforms supporting links (Unix and Windows). This function returns ok if the link was successfully created, otherwise {error, Reason}. On platforms not supporting links, {error,enotsup} is returned.

Typical error reasons:

eacces

Missing read or write permissions for the parent directories of Existing or New.

eexist

New already exists.

enotsup

Hard links are not supported on this platform.

make_symlink(Existing, New) -> ok | {error, Reason}

Types:

Existing = New = name_all()
Reason = posix() | badarg

Creates a symbolic link New to the file or directory Existing on platforms supporting symbolic links (most Unix systems and Windows, beginning with Vista). Existing does not need to exist. Returns ok if the link is successfully created, otherwise {error, Reason}. On platforms not supporting symbolic links, {error, enotsup} is returned.

Typical error reasons:

eacces

Missing read or write permissions for the parent directories of Existing or New.

eexist

New already exists.

enotsup

Symbolic links are not supported on this platform.

eperm

User does not have privileges to create symbolic links (SeCreateSymbolicLinkPrivilege on Windows).

native_name_encoding() -> latin1 | utf8

Returns the filename encoding mode. If it is latin1, the system translates no filenames. If it is utf8, filenames are converted back and forth to the native filename encoding (usually UTF-8, but UTF-16 on Windows).

open(File, Modes) -> {ok, IoDevice} | {error, Reason}

Types:

File = Filename | iodata()
Filename = name_all()
Modes = [mode() | ram]
IoDevice = io_device()
Reason = posix() | badarg | system_limit

Opens file File in the mode determined by Modes, which can contain one or more of the following options:

read

The file, which must exist, is opened for reading.

write

The file is opened for writing. It is created if it does not exist. If the file exists and write is not combined with read, the file is truncated.

append

The file is opened for writing. It is created if it does not exist. Every write operation to a file opened with append takes place at the end of the file.

exclusive

The file is opened for writing. It is created if it does not exist. If the file exists, {error, eexist} is returned.

Warning

This option does not guarantee exclusiveness on file systems not supporting O_EXCL properly, such as NFS. Do not depend on this option unless you know that the file system supports it (in general, local file systems are safe).

raw

Allows faster access to a file, as no Erlang process is needed to handle the file. However, a file opened in this way has the following limitations:

  • The functions in the io module cannot be used, as they can only talk to an Erlang process. Instead, use functions read/2, read_line/1, and write/2.

  • Especially if read_line/1 is to be used on a raw file, it is recommended to combine this option with option {read_ahead, Size} as line-oriented I/O is inefficient without buffering.

  • Only the Erlang process that opened the file can use it.

  • A remote Erlang file server cannot be used. The computer on which the Erlang node is running must have access to the file system (directly or through NFS).

binary

Read operations on the file return binaries rather than lists.

{delayed_write, Size, Delay}

Data in subsequent write/2 calls is buffered until at least Size bytes are buffered, or until the oldest buffered data is Delay milliseconds old. Then all buffered data is written in one operating system call. The buffered data is also flushed before some other file operation than write/2 is executed.

The purpose of this option is to increase performance by reducing the number of operating system calls. Thus, the write/2 calls must be for sizes significantly less than Size, and not interspersed by too many other file operations.

When this option is used, the result of write/2 calls can prematurely be reported as successful, and if a write error occurs, the error is reported as the result of the next file operation, which is not executed.

For example, when delayed_write is used, after a number of write/2 calls, close/1 can return {error, enospc}, as there is not enough space on the disc for previously written data. close/1 must probably be called again, as the file is still open.

delayed_write

The same as {delayed_write, Size, Delay} with reasonable default values for Size and Delay (roughly some 64 KB, 2 seconds).

{read_ahead, Size}

Activates read data buffering. If read/2 calls are for significantly less than Size bytes, read operations to the operating system are still performed for blocks of Size bytes. The extra data is buffered and returned in subsequent read/2 calls, giving a performance gain as the number of operating system calls is reduced.

The read_ahead buffer is also highly used by function read_line/1 in raw mode, therefore this option is recommended (for performance reasons) when accessing raw files using that function.

If read/2 calls are for sizes not significantly less than, or even greater than Size bytes, no performance gain can be expected.

read_ahead

The same as {read_ahead, Size} with a reasonable default value for Size (roughly some 64 KB).

compressed

Makes it possible to read or write gzip compressed files. Option compressed must be combined with read or write, but not both. Notice that the file size obtained with read_file_info/1 does probably not match the number of bytes that can be read from a compressed file.

{encoding, Encoding}

Makes the file perform automatic translation of characters to and from a specific (Unicode) encoding. Notice that the data supplied to write/2 or returned by read/2 still is byte-oriented; this option denotes only how data is stored in the disk file.

Depending on the encoding, different methods of reading and writing data is preferred. The default encoding of latin1 implies using this module (file) for reading and writing data as the interfaces provided here work with byte-oriented data. Using other (Unicode) encodings makes the io(3) functions get_chars, get_line, and put_chars more suitable, as they can work with the full Unicode range.

If data is sent to an io_device() in a format that cannot be converted to the specified encoding, or if data is read by a function that returns data in a format that cannot cope with the character range of the data, an error occurs and the file is closed.

Allowed values for Encoding:

latin1

The default encoding. Bytes supplied to the file, that is, write/2 are written "as is" on the file. Likewise, bytes read from the file, that is, read/2 are returned "as is". If module io(3) is used for writing, the file can only cope with Unicode characters up to code point 255 (the ISO Latin-1 range).

unicode or utf8

Characters are translated to and from UTF-8 encoding before they are written to or read from the file. A file opened in this way can be readable using function read/2, as long as no data stored on the file lies beyond the ISO Latin-1 range (0..255), but failure occurs if the data contains Unicode code points beyond that range. The file is best read with the functions in the Unicode aware module io(3).

Bytes written to the file by any means are translated to UTF-8 encoding before being stored on the disk file.

utf16 or {utf16,big}

Works like unicode, but translation is done to and from big endian UTF-16 instead of UTF-8.

{utf16,little}

Works like unicode, but translation is done to and from little endian UTF-16 instead of UTF-8.

utf32 or {utf32,big}

Works like unicode, but translation is done to and from big endian UTF-32 instead of UTF-8.

{utf32,little}

Works like unicode, but translation is done to and from little endian UTF-32 instead of UTF-8.

The Encoding can be changed for a file "on the fly" by using function io:setopts/2. So a file can be analyzed in latin1 encoding for, for example, a BOM, positioned beyond the BOM and then be set for the right encoding before further reading. For functions identifying BOMs, see module unicode(3).

This option is not allowed on raw files.

ram

File must be iodata(). Returns an fd(), which lets module file operate on the data in-memory as if it is a file.

sync

On platforms supporting it, enables the POSIX O_SYNC synchronous I/O flag or its platform-dependent equivalent (for example, FILE_FLAG_WRITE_THROUGH on Windows) so that writes to the file block until the data is physically written to disk. However, be aware that the exact semantics of this flag differ from platform to platform. For example, none of Linux or Windows guarantees that all file metadata are also written before the call returns. For precise semantics, check the details of your platform documentation. On platforms with no support for POSIX O_SYNC or equivalent, use of the sync flag causes open to return {error, enotsup}.

Returns:

{ok, IoDevice}

The file is opened in the requested mode. IoDevice is a reference to the file.

{error, Reason}

The file cannot be opened.

IoDevice is really the pid of the process that handles the file. This process is linked to the process that originally opened the file. If any process to which the IoDevice is linked terminates, the file is closed and the process itself is terminated. An IoDevice returned from this call can be used as an argument to the I/O functions (see io(3)).

Note

In previous versions of file, modes were specified as one of the atoms read, write, or read_write instead of a list. This is still allowed for reasons of backwards compatibility, but is not to be used for new code. Also note that read_write is not allowed in a mode list.

Typical error reasons:

enoent

The file does not exist.

eacces

Missing permission for reading the file or searching one of the parent directories.

eisdir

The named file is not a regular file. It can be a directory, a FIFO, or a device.

enotdir

A component of the filename is not a directory. On some platforms, enoent is returned instead.

enospc

There is no space left on the device (if write access was specified).

path_consult(Path, Filename) ->
                {ok, Terms, FullName} | {error, Reason}

Types:

Path = [Dir]
Dir = Filename = name_all()
Terms = [term()]
FullName = filename_all()
Reason =
    posix() |
    badarg |
    terminated |
    system_limit |
    {Line :: integer(), Mod :: module(), Term :: term()}

Searches the path Path (a list of directory names) until the file Filename is found. If Filename is an absolute filename, Path is ignored. Then reads Erlang terms, separated by '.', from the file.

Returns one of the following:

{ok, Terms, FullName}

The file is successfully read. FullName is the full name of the file.

{error, enoent}

The file cannot be found in any of the directories in Path.

{error, atom()}

An error occurred when opening the file or reading it. For a list of typical error codes, see open/2.

{error, {Line, Mod, Term}}

An error occurred when interpreting the Erlang terms in the file. Use format_error/1 to convert the three-element tuple to an English description of the error.

The encoding of Filename can be set by a comment as described in epp(3).

path_eval(Path, Filename) -> {ok, FullName} | {error, Reason}

Types:

Path = [Dir :: name_all()]
Filename = name_all()
FullName = filename_all()
Reason =
    posix() |
    badarg |
    terminated |
    system_limit |
    {Line :: integer(), Mod :: module(), Term :: term()}

Searches the path Path (a list of directory names) until the file Filename is found. If Filename is an absolute filename, Path is ignored. Then reads and evaluates Erlang expressions, separated by '.' (or ',', a sequence of expressions is also an expression), from the file. The result of evaluation is not returned; any expression sequence in the file must be there for its side effect.

Returns one of the following:

{ok, FullName}

The file is read and evaluated. FullName is the full name of the file.

{error, enoent}

The file cannot be found in any of the directories in Path.

{error, atom()}

An error occurred when opening the file or reading it. For a list of typical error codes, see open/2.

{error, {Line, Mod, Term}}

An error occurred when interpreting the Erlang expressions in the file. Use format_error/1 to convert the three-element tuple to an English description of the error.

The encoding of Filename can be set by a comment as described in epp(3).

path_open(Path, Filename, Modes) ->
             {ok, IoDevice, FullName} | {error, Reason}

Types:

Path = [Dir :: name_all()]
Filename = name_all()
Modes = [mode()]
IoDevice = io_device()
FullName = filename_all()
Reason = posix() | badarg | system_limit

Searches the path Path (a list of directory names) until the file Filename is found. If Filename is an absolute filename, Path is ignored. Then opens the file in the mode determined by Modes.

Returns one of the following:

{ok, IoDevice, FullName}

The file is opened in the requested mode. IoDevice is a reference to the file and FullName is the full name of the file.

{error, enoent}

The file cannot be found in any of the directories in Path.

{error, atom()}

The file cannot be opened.

path_script(Path, Filename) ->
               {ok, Value, FullName} | {error, Reason}

Types:

Path = [Dir :: name_all()]
Filename = name_all()
Value = term()
FullName = filename_all()
Reason =
    posix() |
    badarg |
    terminated |
    system_limit |
    {Line :: integer(), Mod :: module(), Term :: term()}

Searches the path Path (a list of directory names) until the file Filename is found. If Filename is an absolute filename, Path is ignored. Then reads and evaluates Erlang expressions, separated by '.' (or ',', a sequence of expressions is also an expression), from the file.

Returns one of the following:

{ok, Value, FullName}

The file is read and evaluated. FullName is the full name of the file and Value the value of the last expression.

{error, enoent}

The file cannot be found in any of the directories in Path.

{error, atom()}

An error occurred when opening the file or reading it. For a list of typical error codes, see open/2.

{error, {Line, Mod, Term}}

An error occurred when interpreting the Erlang expressions in the file. Use format_error/1 to convert the three-element tuple to an English description of the error.

The encoding of Filename can be set by a comment as described in epp(3).

path_script(Path, Filename, Bindings) ->
               {ok, Value, FullName} | {error, Reason}

Types:

Path = [Dir :: name_all()]
Filename = name_all()
Value = term()
FullName = filename_all()
Reason =
    posix() |
    badarg |
    terminated |
    system_limit |
    {Line :: integer(), Mod :: module(), Term :: term()}

The same as path_script/2 but the variable bindings Bindings are used in the evaluation. See erl_eval(3) about variable bindings.

pid2name(Pid) -> {ok, Filename} | undefined

Types:

Filename = filename_all()
Pid = pid()

If Pid is an I/O device, that is, a pid returned from open/2, this function returns the filename, or rather:

{ok, Filename}

If the file server of this node is not a slave, the file was opened by the file server of this node (this implies that Pid must be a local pid) and the file is not closed. Filename is the filename in flat string format.

undefined

In all other cases.

Warning

This function is intended for debugging only.

position(IoDevice, Location) ->
            {ok, NewPosition} | {error, Reason}

Types:

IoDevice = io_device()
Location = location()
NewPosition = integer()
Reason = posix() | badarg | terminated

Sets the position of the file referenced by IoDevice to Location. Returns {ok, NewPosition} (as absolute offset) if successful, otherwise {error, Reason}. Location is one of the following:

Offset

The same as {bof, Offset}.

{bof, Offset}

Absolute offset.

{cur, Offset}

Offset from the current position.

{eof, Offset}

Offset from the end of file.

bof | cur | eof

The same as above with Offset 0.

Notice that offsets are counted in bytes, not in characters. If the file is opened using some other encoding than latin1, one byte does not correspond to one character. Positioning in such a file can only be done to known character boundaries. That is, to a position earlier retrieved by getting a current position, to the beginning/end of the file or to some other position known to be on a correct character boundary by some other means (typically beyond a byte order mark in the file, which has a known byte-size).

A typical error reason is:

einval

Either Location is illegal, or it is evaluated to a negative offset in the file. Notice that if the resulting position is a negative value, the result is an error, and after the call the file position is undefined.

pread(IoDevice, LocNums) -> {ok, DataL} | eof | {error, Reason}

Types:

IoDevice = io_device()
LocNums =
    [{Location :: location(), Number :: integer() >= 0}]
DataL = [Data]
Data = string() | binary() | eof
Reason = posix() | badarg | terminated

Performs a sequence of pread/3 in one operation, which is more efficient than calling them one at a time. Returns {ok, [Data, ...]} or {error, Reason}, where each Data, the result of the corresponding pread, is either a list or a binary depending on the mode of the file, or eof if the requested position is beyond end of file.

As the position is specified as a byte-offset, take special caution when working with files where encoding is set to something else than latin1, as not every byte position is a valid character boundary on such a file.

pread(IoDevice, Location, Number) ->
         {ok, Data} | eof | {error, Reason}

Types:

IoDevice = io_device()
Location = location()
Number = integer() >= 0
Data = string() | binary()
Reason = posix() | badarg | terminated

Combines position/2 and read/2 in one operation, which is more efficient than calling them one at a time. If IoDevice is opened in raw mode, some restrictions apply:

  • Location is only allowed to be an integer.
  • The current position of the file is undefined after the operation.

As the position is specified as a byte-offset, take special caution when working with files where encoding is set to something else than latin1, as not every byte position is a valid character boundary on such a file.

pwrite(IoDevice, LocBytes) -> ok | {error, {N, Reason}}

Types:

IoDevice = io_device()
LocBytes = [{Location :: location(), Bytes :: iodata()}]
N = integer() >= 0
Reason = posix() | badarg | terminated

Performs a sequence of pwrite/3 in one operation, which is more efficient than calling them one at a time. Returns ok or {error, {N, Reason}}, where N is the number of successful writes done before the failure.

When positioning in a file with other encoding than latin1, caution must be taken to set the position on a correct character boundary. For details, see position/2.

pwrite(IoDevice, Location, Bytes) -> ok | {error, Reason}

Types:

IoDevice = io_device()
Location = location()
Bytes = iodata()
Reason = posix() | badarg | terminated

Combines position/2 and write/2 in one operation, which is more efficient than calling them one at a time. If IoDevice has been opened in raw mode, some restrictions apply:

  • Location is only allowed to be an integer.
  • The current position of the file is undefined after the operation.

When positioning in a file with other encoding than latin1, caution must be taken to set the position on a correct character boundary. For details, see position/2.

read(IoDevice, Number) -> {ok, Data} | eof | {error, Reason}

Types:

IoDevice = io_device() | atom()
Number = integer() >= 0
Data = string() | binary()
Reason =
    posix() |
    badarg |
    terminated |
    {no_translation, unicode, latin1}

Reads Number bytes/characters from the file referenced by IoDevice. The functions read/2, pread/3, and read_line/1 are the only ways to read from a file opened in raw mode (although they work for normally opened files, too).

For files where encoding is set to something else than latin1, one character can be represented by more than one byte on the file. The parameter Number always denotes the number of characters read from the file, while the position in the file can be moved much more than this number when reading a Unicode file.

Also, if encoding is set to something else than latin1, the read/3 call fails if the data contains characters larger than 255, which is why module io(3) is to be preferred when reading such a file.

The function returns:

{ok, Data}

If the file was opened in binary mode, the read bytes are returned in a binary, otherwise in a list. The list or binary is shorter than the number of bytes requested if end of file was reached.

eof

Returned if Number>0 and end of file was reached before anything at all could be read.

{error, Reason}

An error occurred.

Typical error reasons:

ebadf

The file is not opened for reading.

{no_translation, unicode, latin1}

The file is opened with another encoding than latin1 and the data in the file cannot be translated to the byte-oriented data that this function returns.

read_file(Filename) -> {ok, Binary} | {error, Reason}

Types:

Filename = name_all()
Binary = binary()
Reason = posix() | badarg | terminated | system_limit

Returns {ok, Binary}, where Binary is a binary data object that contains the contents of Filename, or {error, Reason} if an error occurs.

Typical error reasons:

enoent

The file does not exist.

eacces

Missing permission for reading the file, or for searching one of the parent directories.

eisdir

The named file is a directory.

enotdir

A component of the filename is not a directory. On some platforms, enoent is returned instead.

enomem

There is not enough memory for the contents of the file.

read_file_info(Filename) -> {ok, FileInfo} | {error, Reason}
read_file_info(Filename, Opts) -> {ok, FileInfo} | {error, Reason}

Types:

Filename = name_all()
FileInfo = file_info()
Reason = posix() | badarg

Retrieves information about a file. Returns {ok, FileInfo} if successful, otherwise {error, Reason}. FileInfo is a record file_info, defined in the Kernel include file file.hrl. Include the following directive in the module from which the function is called:

 -include_lib("kernel/include/file.hrl").

The time type returned in atime, mtime, and ctime is dependent on the time type set in Opts :: {time, Type} as follows:

local

Returns local time.

universal

Returns universal time.

posix

Returns seconds since or before Unix time epoch, which is 1970-01-01 00:00 UTC.

Default is {time, local}.

If the option raw is set, the file server is not called and only information about local files is returned.

Note

As file times are stored in POSIX time on most OS, it is faster to query file information with option posix.

The record file_info contains the following fields:

size = integer() >= 0

Size of file in bytes.

type = device | directory | other | regular | symlink

The type of the file.

access = read | write | read_write | none

The current system access to the file.

atime = date_time() | integer() >= 0

The last time the file was read.

mtime = date_time() | integer() >= 0

The last time the file was written.

ctime = date_time() | integer() >=0

The interpretation of this time field depends on the operating system. On Unix, it is the last time the file or the inode was changed. In Windows, it is the create time.

mode = integer() >= 0

The file permissions as the sum of the following bit values:

8#00400

read permission: owner

8#00200

write permission: owner

8#00100

execute permission: owner

8#00040

read permission: group

8#00020

write permission: group

8#00010

execute permission: group

8#00004

read permission: other

8#00002

write permission: other

8#00001

execute permission: other

16#800

set user id on execution

16#400

set group id on execution

On Unix platforms, other bits than those listed above may be set.

links = integer() >= 0

Number of links to the file (this is always 1 for file systems that have no concept of links).

major_device = integer() >= 0

Identifies the file system where the file is located. In Windows, the number indicates a drive as follows: 0 means A:, 1 means B:, and so on.

minor_device = integer() >= 0

Only valid for character devices on Unix. In all other cases, this field is zero.

inode = integer() >= 0

Gives the inode number. On non-Unix file systems, this field is zero.

uid = integer() >= 0

Indicates the owner of the file. On non-Unix file systems, this field is zero.

gid = integer() >= 0

Gives the group that the owner of the file belongs to. On non-Unix file systems, this field is zero.

Typical error reasons:

eacces

Missing search permission for one of the parent directories of the file.

enoent

The file does not exist.

enotdir

A component of the filename is not a directory. On some platforms, enoent is returned instead.

read_line(IoDevice) -> {ok, Data} | eof | {error, Reason}

Types:

IoDevice = io_device() | atom()
Data = string() | binary()
Reason =
    posix() |
    badarg |
    terminated |
    {no_translation, unicode, latin1}

Reads a line of bytes/characters from the file referenced by IoDevice. Lines are defined to be delimited by the linefeed (LF, \n) character, but any carriage return (CR, \r) followed by a newline is also treated as a single LF character (the carriage return is silently ignored). The line is returned including the LF, but excluding any CR immediately followed by an LF. This behaviour is consistent with the behaviour of io:get_line/2. If end of file is reached without any LF ending the last line, a line with no trailing LF is returned.

The function can be used on files opened in raw mode. However, it is inefficient to use it on raw files if the file is not opened with option {read_ahead, Size} specified. Thus, combining raw and {read_ahead, Size} is highly recommended when opening a text file for raw line-oriented reading.

If encoding is set to something else than latin1, the read_line/1 call fails if the data contains characters larger than 255, why module io(3) is to be preferred when reading such a file.

The function returns:

{ok, Data}

One line from the file is returned, including the trailing LF, but with CRLF sequences replaced by a single LF (see above).

If the file is opened in binary mode, the read bytes are returned in a binary, otherwise in a list.

eof

Returned if end of file was reached before anything at all could be read.

{error, Reason}

An error occurred.

Typical error reasons:

ebadf

The file is not opened for reading.

{no_translation, unicode, latin1}

The file is opened with another encoding than latin1 and the data on the file cannot be translated to the byte-oriented data that this function returns.

read_link(Name) -> {ok, Filename} | {error, Reason}

Types:

Name = name_all()
Filename = filename()
Reason = posix() | badarg

Returns {ok, Filename} if Name refers to a symbolic link that is not a raw filename, or {error, Reason} otherwise. On platforms that do not support symbolic links, the return value is {error,enotsup}.

Typical error reasons:

einval

Name does not refer to a symbolic link or the name of the file that it refers to does not conform to the expected encoding.

enoent

The file does not exist.

enotsup

Symbolic links are not supported on this platform.

read_link_all(Name) -> {ok, Filename} | {error, Reason}

Types:

Name = name_all()
Filename = filename_all()
Reason = posix() | badarg

Returns {ok, Filename} if Name refers to a symbolic link or {error, Reason} otherwise. On platforms that do not support symbolic links, the return value is {error,enotsup}.

Notice that Filename can be either a list or a binary.

Typical error reasons:

einval

Name does not refer to a symbolic link.

enoent

The file does not exist.

enotsup

Symbolic links are not supported on this platform.

read_link_info(Name) -> {ok, FileInfo} | {error, Reason}
read_link_info(Name, Opts) -> {ok, FileInfo} | {error, Reason}

Types:

Name = name_all()
FileInfo = file_info()
Reason = posix() | badarg

Works like read_file_info/1,2 except that if Name is a symbolic link, information about the link is returned in the file_info record and the type field of the record is set to symlink.

If the option raw is set, the file server is not called and only information about local files is returned.

If Name is not a symbolic link, this function returns the same result as read_file_info/1. On platforms that do not support symbolic links, this function is always equivalent to read_file_info/1.

rename(Source, Destination) -> ok | {error, Reason}

Types:

Source = Destination = name_all()
Reason = posix() | badarg

Tries to rename the file Source to Destination. It can be used to move files (and directories) between directories, but it is not sufficient to specify the destination only. The destination filename must also be specified. For example, if bar is a normal file and foo and baz are directories, rename("foo/bar", "baz") returns an error, but rename("foo/bar", "baz/bar") succeeds. Returns ok if it is successful.

Note

Renaming of open files is not allowed on most platforms (see eacces below).

Typical error reasons:

eacces

Missing read or write permissions for the parent directories of Source or Destination. On some platforms, this error is given if either Source or Destination is open.

eexist

Destination is not an empty directory. On some platforms, also given when Source and Destination are not of the same type.

einval

Source is a root directory, or Destination is a subdirectory of Source.

eisdir

Destination is a directory, but Source is not.

enoent

Source does not exist.

enotdir

Source is a directory, but Destination is not.

exdev

Source and Destination are on different file systems.

script(Filename) -> {ok, Value} | {error, Reason}

Types:

Filename = name_all()
Value = term()
Reason =
    posix() |
    badarg |
    terminated |
    system_limit |
    {Line :: integer(), Mod :: module(), Term :: term()}

Reads and evaluates Erlang expressions, separated by '.' (or ',', a sequence of expressions is also an expression), from the file.

Returns one of the following:

{ok, Value}

The file is read and evaluated. Value is the value of the last expression.

{error, atom()}

An error occurred when opening the file or reading it. For a list of typical error codes, see open/2.

{error, {Line, Mod, Term}}

An error occurred when interpreting the Erlang expressions in the file. Use format_error/1 to convert the three-element tuple to an English description of the error.

The encoding of Filename can be set by a comment as described in epp(3).

script(Filename, Bindings) -> {ok, Value} | {error, Reason}

Types:

Filename = name_all()
Value = term()
Reason =
    posix() |
    badarg |
    terminated |
    system_limit |
    {Line :: integer(), Mod :: module(), Term :: term()}

The same as script/1 but the variable bindings Bindings are used in the evaluation. See erl_eval(3) about variable bindings.

sendfile(Filename, Socket) ->
            {ok, integer() >= 0} |
            {error, inet:posix() | closed | badarg | not_owner}

Types:

Filename = name_all()
Socket = inet:socket()

Sends the file Filename to Socket. Returns {ok, BytesSent} if successful, otherwise {error, Reason}.

sendfile(RawFile, Socket, Offset, Bytes, Opts) ->
            {ok, integer() >= 0} |
            {error, inet:posix() | closed | badarg | not_owner}

Types:

RawFile = fd()
Socket = inet:socket()
Offset = Bytes = integer() >= 0
Opts = [sendfile_option()]
sendfile_option() =
    {chunk_size, integer() >= 0} | {use_threads, boolean()}

Sends Bytes from the file referenced by RawFile beginning at Offset to Socket. Returns {ok, BytesSent} if successful, otherwise {error, Reason}. If Bytes is set to 0 all data after the specified Offset is sent.

The file used must be opened using the raw flag, and the process calling sendfile must be the controlling process of the socket. See gen_tcp:controlling_process/2.

If the OS used does not support sendfile, an Erlang fallback using read/2 and gen_tcp:send/2 is used.

The option list can contain the following options:

chunk_size

The chunk size used by the Erlang fallback to send data. If using the fallback, set this to a value that comfortably fits in the systems memory. Default is 20 MB.

use_threads

Instructs the emulator to use the async thread pool for the sendfile system call. This can be useful if the OS you are running on does not properly support non-blocking sendfile calls. Notice that using async threads potentially makes your system vulnerable to slow client attacks. If set to true and no async threads are available, the sendfile call returns {error,einval}. Introduced in Erlang/OTP 17.0. Default is false.

set_cwd(Dir) -> ok | {error, Reason}

Types:

Dir = name() | EncodedBinary
EncodedBinary = binary()
Reason = posix() | badarg | no_translation

Sets the current working directory of the file server to Dir. Returns ok if successful.

The functions in the module file usually treat binaries as raw filenames, that is, they are passed "as is" even when the encoding of the binary does not agree with native_name_encoding(). However, this function expects binaries to be encoded according to the value returned by native_name_encoding().

Typical error reasons are:

enoent

The directory does not exist.

enotdir

A component of Dir is not a directory. On some platforms, enoent is returned.

eacces

Missing permission for the directory or one of its parents.

badarg

Dir has an improper type, such as tuple.

no_translation

Dir is a binary() with characters coded in ISO-latin-1 and the VM is operating with unicode filename encoding.

Warning

In a future release, a bad type for argument Dir will probably generate an exception.

sync(IoDevice) -> ok | {error, Reason}

Types:

IoDevice = io_device()
Reason = posix() | badarg | terminated

Ensures that any buffers kept by the operating system (not by the Erlang runtime system) are written to disk. On some platforms, this function might have no effect.

A typical error reason is:

enospc

Not enough space left to write the file.

truncate(IoDevice) -> ok | {error, Reason}

Types:

IoDevice = io_device()
Reason = posix() | badarg | terminated

Truncates the file referenced by IoDevice at the current position. Returns ok if successful, otherwise {error, Reason}.

write(IoDevice, Bytes) -> ok | {error, Reason}

Types:

IoDevice = io_device() | atom()
Bytes = iodata()
Reason = posix() | badarg | terminated

Writes Bytes to the file referenced by IoDevice. This function is the only way to write to a file opened in raw mode (although it works for normally opened files too). Returns ok if successful, and {error, Reason} otherwise.

If the file is opened with encoding set to something else than latin1, each byte written can result in many bytes being written to the file, as the byte range 0..255 can represent anything between one and four bytes depending on value and UTF encoding type.

Typical error reasons:

ebadf

The file is not opened for writing.

enospc

No space is left on the device.

write_file(Filename, Bytes) -> ok | {error, Reason}

Types:

Filename = name_all()
Bytes = iodata()
Reason = posix() | badarg | terminated | system_limit

Writes the contents of the iodata term Bytes to file Filename. The file is created if it does not exist. If it exists, the previous contents are overwritten. Returns ok if successful, otherwise {error, Reason}.

Typical error reasons:

enoent

A component of the filename does not exist.

enotdir

A component of the filename is not a directory. On some platforms, enoent is returned instead.

enospc

No space is left on the device.

eacces

Missing permission for writing the file or searching one of the parent directories.

eisdir

The named file is a directory.

write_file(Filename, Bytes, Modes) -> ok | {error, Reason}

Types:

Filename = name_all()
Bytes = iodata()
Modes = [mode()]
Reason = posix() | badarg | terminated | system_limit

Same as write_file/2, but takes a third argument Modes, a list of possible modes, see open/2. The mode flags binary and write are implicit, so they are not to be used.

write_file_info(Filename, FileInfo) -> ok | {error, Reason}
write_file_info(Filename, FileInfo, Opts) -> ok | {error, Reason}

Types:

Filename = name_all()
FileInfo = file_info()
Reason = posix() | badarg

Changes file information. Returns ok if successful, otherwise {error, Reason}. FileInfo is a record file_info, defined in the Kernel include file file.hrl. Include the following directive in the module from which the function is called:

 -include_lib("kernel/include/file.hrl").

The time type set in atime, mtime, and ctime depends on the time type set in Opts :: {time, Type} as follows:

local

Interprets the time set as local.

universal

Interprets it as universal time.

posix

Must be seconds since or before Unix time epoch, which is 1970-01-01 00:00 UTC.

Default is {time, local}.

If the option raw is set, the file server is not called and only information about local files is returned.

The following fields are used from the record, if they are specified:

atime = date_time() | integer() >= 0

The last time the file was read.

mtime = date_time() | integer() >= 0

The last time the file was written.

ctime = date_time() | integer() >= 0

On Unix, any value specified for this field is ignored (the "ctime" for the file is set to the current time). On Windows, this field is the new creation time to set for the file.

mode = integer() >= 0

The file permissions as the sum of the following bit values:

8#00400

Read permission: owner

8#00200

Write permission: owner

8#00100

Execute permission: owner

8#00040

Read permission: group

8#00020

Write permission: group

8#00010

Execute permission: group

8#00004

Read permission: other

8#00002

Write permission: other

8#00001

Execute permission: other

16#800

Set user id on execution

16#400

Set group id on execution

On Unix platforms, other bits than those listed above may be set.

uid = integer() >= 0

Indicates the file owner. Ignored for non-Unix file systems.

gid = integer() >= 0

Gives the group that the file owner belongs to. Ignored for non-Unix file systems.

Typical error reasons:

eacces

Missing search permission for one of the parent directories of the file.

enoent

The file does not exist.

enotdir

A component of the filename is not a directory. On some platforms, enoent is returned instead.

POSIX Error Codes

  • eacces - Permission denied
  • eagain - Resource temporarily unavailable
  • ebadf - Bad file number
  • ebusy - File busy
  • edquot - Disk quota exceeded
  • eexist - File already exists
  • efault - Bad address in system call argument
  • efbig - File too large
  • eintr - Interrupted system call
  • einval - Invalid argument
  • eio - I/O error
  • eisdir - Illegal operation on a directory
  • eloop - Too many levels of symbolic links
  • emfile - Too many open files
  • emlink - Too many links
  • enametoolong - Filename too long
  • enfile - File table overflow
  • enodev - No such device
  • enoent - No such file or directory
  • enomem - Not enough memory
  • enospc - No space left on device
  • enotblk - Block device required
  • enotdir - Not a directory
  • enotsup - Operation not supported
  • enxio - No such device or address
  • eperm - Not owner
  • epipe - Broken pipe
  • erofs - Read-only file system
  • espipe - Invalid seek
  • esrch - No such process
  • estale - Stale remote file handle
  • exdev - Cross-domain link

Performance

Some operating system file operations, for example, a sync/1 or close/1 on a huge file, can block their calling thread for seconds. If this affects the emulator main thread, the response time is no longer in the order of milliseconds, depending on the definition of "soft" in soft real-time system.

If the device driver thread pool is active, file operations are done through those threads instead, so the emulator can go on executing Erlang processes. Unfortunately, the time for serving a file operation increases because of the extra scheduling required from the operating system.

If the device driver thread pool is disabled or of size 0, large file reads and writes are segmented into many smaller, which enable the emulator to serve other processes during the file operation. This has the same effect as when using the thread pool, but with larger overhead. Other file operations, for example, sync/1 or close/1 on a huge file, still are a problem.

For increased performance, raw files are recommended. Raw files use the file system of the host machine of the node.

Note

For normal files (non-raw), the file server is used to find the files, and if the node is running its file server as slave to the file server of another node, and the other node runs on some other host machine, they can have different file systems. However, this is seldom a problem.

A normal file is really a process so it can be used as an I/O device (see io). Therefore, when data is written to a normal file, the sending of the data to the file process, copies all data that are not binaries. Opening the file in binary mode and writing binaries is therefore recommended. If the file is opened on another node, or if the file server runs as slave to the file server of another node, also binaries are copied.

Caching data to reduce the number of file operations, or rather the number of calls to the file driver, generally increases performance. The following function writes 4 MBytes in 23 seconds when tested:

create_file_slow(Name, N) when integer(N), N >= 0 ->
    {ok, FD} = file:open(Name, [raw, write, delayed_write, binary]),
    ok = create_file_slow(FD, 0, N),
    ok = ?FILE_MODULE:close(FD),
    ok.
      
create_file_slow(FD, M, M) ->
    ok;
create_file_slow(FD, M, N) ->
    ok = file:write(FD, <<M:32/unsigned>>),
    create_file_slow(FD, M+1, N).

The following, functionally equivalent, function collects 1024 entries into a list of 128 32-byte binaries before each call to write/2 and so does the same work in 0.52 seconds, which is 44 times faster:

create_file(Name, N) when integer(N), N >= 0 ->
    {ok, FD} = file:open(Name, [raw, write, delayed_write, binary]),
    ok = create_file(FD, 0, N),
    ok = ?FILE_MODULE:close(FD),
    ok.
      
create_file(FD, M, M) ->
    ok;
create_file(FD, M, N) when M + 1024 =&lt; N ->
    create_file(FD, M, M + 1024, []),
    create_file(FD, M + 1024, N);
create_file(FD, M, N) ->
    create_file(FD, M, N, []).
      
create_file(FD, M, M, R) ->
    ok = file:write(FD, R);
create_file(FD, M, N0, R) when M + 8 =&lt; N0 ->
    N1  = N0-1,  N2  = N0-2,  N3  = N0-3,  N4  = N0-4, 
    N5  = N0-5,  N6  = N0-6,  N7  = N0-7,  N8  = N0-8, 
    create_file(FD, M, N8, 
                [<<N8:32/unsigned,  N7:32/unsigned, 
                   N6:32/unsigned,  N5:32/unsigned, 
                   N4:32/unsigned,  N3:32/unsigned, 
                   N2:32/unsigned,  N1:32/unsigned>> | R]);
create_file(FD, M, N0, R) ->
    N1 = N0-1,
    create_file(FD, M, N1, [<<N1:32/unsigned>> | R]).
Note

Trust only your own benchmarks. If the list length in create_file/2 above is increased, it runs slightly faster, but consumes more memory and causes more memory fragmentation. How much this affects your application is something that this simple benchmark cannot predict.

If the size of each binary is increased to 64 bytes, it also runs slightly faster, but the code is then twice as clumsy. In the current implementation, binaries larger than 64 bytes are stored in memory common to all processes and not copied when sent between processes, while these smaller binaries are stored on the process heap and copied when sent like any other term.

So, with a binary size of 68 bytes, create_file/2 runs 30 percent slower than with 64 bytes, and causes much more memory fragmentation. Notice that if the binaries were to be sent between processes (for example, a non-raw file), the results would probably be completely different.

A raw file is really a port. When writing data to a port, it is efficient to write a list of binaries. It is not needed to flatten a deep list before writing. On Unix hosts, scatter output, which writes a set of buffers in one operation, is used when possible. In this way write(FD, [Bin1, Bin2 | Bin3]) writes the contents of the binaries without copying the data at all, except for perhaps deep down in the operating system kernel.

For raw files, pwrite/2 and pread/2 are efficiently implemented. The file driver is called only once for the whole operation, and the list iteration is done in the file driver.

The options delayed_write and read_ahead to open/2 make the file driver cache data to reduce the number of operating system calls. The function create_file/2 in the recent example takes 60 seconds without option delayed_write, which is 2.6 times slower.

As a bad example, create_file_slow/2 without options raw, binary, and delayed_write, meaning it calls open(Name, [write]), needs 1 min 20 seconds for the job, which is 3.5 times slower than the first example, and 150 times slower than the optimized create_file/2.

Warning

If an error occurs when accessing an open file with module io, the process handling the file exits. The dead file process can hang if a process tries to access it later. This will be fixed in a future release.

See Also