IO

Class

An instance of class IO (commonly called a stream) represents an input/output stream in the underlying operating system. Class IO is the basis for input and output in Ruby.

Class File is the only class in the Ruby core that is a subclass of IO. Some classes in the Ruby standard library are also subclasses of IO; these include TCPSocket and UDPSocket.

The global constant ARGF (also accessible as $<) provides an IO-like stream that allows access to all file paths found in ARGV (or found in STDIN if ARGV is empty). ARGF is not itself a subclass of IO.

Class StringIO provides an IO-like stream that handles a String. StringIO is not itself a subclass of IO.

Important objects based on IO include:

  • $stdin.

  • $stdout.

  • $stderr.

  • Instances of class File.

An instance of IO may be created using:

  • IO.new: returns a new IO object for the given integer file descriptor.

  • IO.open: passes a new IO object to the given block.

  • IO.popen: returns a new IO object that is connected to the $stdin and $stdout of a newly-launched subprocess.

  • Kernel#open: Returns a new IO object connected to a given source: stream, file, or subprocess.

Like a File stream, an IO stream has:

  • A read/write mode, which may be read-only, write-only, or read/write; see Read/Write Mode.

  • A data mode, which may be text-only or binary; see Data Mode.

  • Internal and external encodings; see Encodings.

And like other IO streams, it has:

  • A position, which determines where in the stream the next read or write is to occur; see Position.

  • A line number, which is a special, line-oriented, “position” (different from the position mentioned above); see Line Number.

Extension io/console

Extension io/console provides numerous methods for interacting with the console; requiring it adds numerous methods to class IO.

Example Files

Many examples here use these variables:

# English text with newlines.
text = <<~EOT
  First line
  Second line

  Fourth line
  Fifth line
EOT

# Russian text.
russian = "\u{442 435 441 442}" # => "тест"

# Binary data.
data = "\u9990\u9991\u9992\u9993\u9994"

# Text file.
File.write('t.txt', text)

# File with Russian text.
File.write('t.rus', russian)

# File with binary data.
f = File.new('t.dat', 'wb:UTF-16')
f.write(data)
f.close

Open Options

A number of IO methods accept optional keyword arguments that determine how a new stream is to be opened:

  • :mode: Stream mode.

  • :flags: Integer file open flags; If mode is also given, the two are bitwise-ORed.

  • :external_encoding: External encoding for the stream.

  • :internal_encoding: Internal encoding for the stream. '-' is a synonym for the default internal encoding. If the value is nil no conversion occurs.

  • :encoding: Specifies external and internal encodings as 'extern:intern'.

  • :textmode: If a truthy value, specifies the mode as text-only, binary otherwise.

  • :binmode: If a truthy value, specifies the mode as binary, text-only otherwise.

  • :autoclose: If a truthy value, specifies that the fd will close when the stream closes; otherwise it remains open.

  • :path: If a string value is provided, it is used in inspect and is available as path method.

Also available are the options offered in String#encode, which may control conversion between external internal encoding.

Basic IO

You can perform basic stream IO with these methods, which typically operate on multi-byte strings:

  • IO#read: Reads and returns some or all of the remaining bytes from the stream.

  • IO#write: Writes zero or more strings to the stream; each given object that is not already a string is converted via to_s.

Position

An IO stream has a nonnegative integer position, which is the byte offset at which the next read or write is to occur. A new stream has position zero (and line number zero); method rewind resets the position (and line number) to zero.

The relevant methods:

  • IO#tell (aliased as #pos): Returns the current position (in bytes) in the stream.

  • IO#pos=: Sets the position of the stream to a given integer new_position (in bytes).

  • IO#seek: Sets the position of the stream to a given integer offset (in bytes), relative to a given position whence (indicating the beginning, end, or current position).

  • IO#rewind: Positions the stream at the beginning (also resetting the line number).

Open and Closed Streams

A new IO stream may be open for reading, open for writing, or both.

A stream is automatically closed when claimed by the garbage collector.

Attempted reading or writing on a closed stream raises an exception.

The relevant methods:

End-of-Stream

You can query whether a stream is positioned at its end:

  • IO#eof? (also aliased as #eof): Returns whether the stream is at end-of-stream.

You can reposition to end-of-stream by using method IO#seek:

f = File.new('t.txt')
f.eof? # => false
f.seek(0, :END)
f.eof? # => true
f.close

Or by reading all stream content (which is slower than using IO#seek):

f.rewind
f.eof? # => false
f.read # => "First line\nSecond line\n\nFourth line\nFifth line\n"
f.eof? # => true

Line IO

You can read an IO stream line-by-line using these methods:

Each of these reader methods accepts:

For each of these reader methods, reading may begin mid-line, depending on the stream’s position; see Position:

f = File.new('t.txt')
f.pos = 27
f.each_line {|line| p line }
f.close

Output:

"rth line\n"
"Fifth line\n"

You can write to an IO stream line-by-line using this method:

  • IO#puts: Writes objects to the stream.

Line Separator

Each of these methods uses a line separator, which is the string that delimits lines:

The default line separator is the given by the global variable $/, whose value is by default "\n". The line to be read next is all data from the current position to the next line separator:

f = File.new('t.txt')
f.gets # => "First line\n"
f.gets # => "Second line\n"
f.gets # => "\n"
f.gets # => "Fourth line\n"
f.gets # => "Fifth line\n"
f.close

You can specify a different line separator:

f = File.new('t.txt')
f.gets('l')   # => "First l"
f.gets('li')  # => "ine\nSecond li"
f.gets('lin') # => "ne\n\nFourth lin"
f.gets        # => "e\n"
f.close

There are two special line separators:

  • nil: The entire stream is read into a single string:

    f = File.new('t.txt')
    f.gets(nil) # => "First line\nSecond line\n\nFourth line\nFifth line\n"
    f.close
    
  • '' (the empty string): The next “paragraph” is read (paragraphs being separated by two consecutive line separators):

    f = File.new('t.txt')
    f.gets('') # => "First line\nSecond line\n\n"
    f.gets('') # => "Fourth line\nFifth line\n"
    f.close
    

Line Limit

Each of these methods uses a line limit, which specifies that the number of bytes returned may not be (much) longer than the given limit;

A multi-byte character will not be split, and so a line may be slightly longer than the given limit.

If limit is not given, the line is determined only by sep.

# Text with 1-byte characters.
File.open('t.txt') {|f| f.gets(1) }  # => "F"
File.open('t.txt') {|f| f.gets(2) }  # => "Fi"
File.open('t.txt') {|f| f.gets(3) }  # => "Fir"
File.open('t.txt') {|f| f.gets(4) }  # => "Firs"
# No more than one line.
File.open('t.txt') {|f| f.gets(10) } # => "First line"
File.open('t.txt') {|f| f.gets(11) } # => "First line\n"
File.open('t.txt') {|f| f.gets(12) } # => "First line\n"

# Text with 2-byte characters, which will not be split.
File.open('t.rus') {|f| f.gets(1).size } # => 1
File.open('t.rus') {|f| f.gets(2).size } # => 1
File.open('t.rus') {|f| f.gets(3).size } # => 2
File.open('t.rus') {|f| f.gets(4).size } # => 2

Line Separator and Line Limit

With arguments sep and limit given, combines the two behaviors:

  • Returns the next line as determined by line separator sep.

  • But returns no more bytes than are allowed by the limit.

Example:

File.open('t.txt') {|f| f.gets('li', 20) } # => "First li"
File.open('t.txt') {|f| f.gets('li', 2) }  # => "Fi"

Line Number

A readable IO stream has a non-negative integer line number.

The relevant methods:

Unless modified by a call to method IO#lineno=, the line number is the number of lines read by certain line-oriented methods, according to the given line separator sep:

  • IO.foreach: Increments the line number on each call to the block.

  • IO#each_line: Increments the line number on each call to the block.

  • IO#gets: Increments the line number.

  • IO#readline: Increments the line number.

  • IO#readlines: Increments the line number for each line read.

A new stream is initially has line number zero (and position zero); method rewind resets the line number (and position) to zero:

f = File.new('t.txt')
f.lineno # => 0
f.gets   # => "First line\n"
f.lineno # => 1
f.rewind
f.lineno # => 0
f.close

Reading lines from a stream usually changes its line number:

f = File.new('t.txt', 'r')
f.lineno   # => 0
f.readline # => "This is line one.\n"
f.lineno   # => 1
f.readline # => "This is the second line.\n"
f.lineno   # => 2
f.readline # => "Here's the third line.\n"
f.lineno   # => 3
f.eof?     # => true
f.close

Iterating over lines in a stream usually changes its line number:

File.open('t.txt') do |f|
  f.each_line do |line|
    p "position=#{f.pos} eof?=#{f.eof?} lineno=#{f.lineno}"
  end
end

Output:

"position=11 eof?=false lineno=1"
"position=23 eof?=false lineno=2"
"position=24 eof?=false lineno=3"
"position=36 eof?=false lineno=4"
"position=47 eof?=true lineno=5"

Unlike the stream’s position, the line number does not affect where the next read or write will occur:

f = File.new('t.txt')
f.lineno = 1000
f.lineno # => 1000
f.gets   # => "First line\n"
f.lineno # => 1001
f.close

Associated with the line number is the global variable $.:

  • When a stream is opened, $. is not set; its value is left over from previous activity in the process:

    $. = 41
    f = File.new('t.txt')
    $. = 41
    # => 41
    f.close
    
  • When a stream is read, #. is set to the line number for that stream:

    f0 = File.new('t.txt')
    f1 = File.new('t.dat')
    f0.readlines # => ["First line\n", "Second line\n", "\n", "Fourth line\n", "Fifth line\n"]
    $.           # => 5
    f1.readlines # => ["\xFE\xFF\x99\x90\x99\x91\x99\x92\x99\x93\x99\x94"]
    $.           # => 1
    f0.close
    f1.close
    
  • Methods IO#rewind and IO#seek do not affect $.:

    f = File.new('t.txt')
    f.readlines # => ["First line\n", "Second line\n", "\n", "Fourth line\n", "Fifth line\n"]
    $.          # => 5
    f.rewind
    f.seek(0, :SET)
    $.          # => 5
    f.close
    

Character IO

You can process an IO stream character-by-character using these methods:

  • IO#getc: Reads and returns the next character from the stream.

  • IO#readchar: Like getc, but raises an exception at end-of-stream.

  • IO#ungetc: Pushes back (“unshifts”) a character or integer onto the stream.

  • IO#putc: Writes a character to the stream.

  • IO#each_char: Reads each remaining character in the stream, passing the character to the given block.

Byte IO

You can process an IO stream byte-by-byte using these methods:

  • IO#getbyte: Returns the next 8-bit byte as an integer in range 0..255.

  • IO#readbyte: Like getbyte, but raises an exception if at end-of-stream.

  • IO#ungetbyte: Pushes back (“unshifts”) a byte back onto the stream.

  • IO#each_byte: Reads each remaining byte in the stream, passing the byte to the given block.

Codepoint IO

You can process an IO stream codepoint-by-codepoint:

What’s Here

First, what’s elsewhere. Class IO:

Here, class IO provides methods that are useful for:

Creating

  • ::new (aliased as ::for_fd): Creates and returns a new IO object for the given integer file descriptor.

  • ::open: Creates a new IO object.

  • ::pipe: Creates a connected pair of reader and writer IO objects.

  • ::popen: Creates an IO object to interact with a subprocess.

  • ::select: Selects which given IO instances are ready for reading, writing, or have pending exceptions.

Reading

  • ::binread: Returns a binary string with all or a subset of bytes from the given file.

  • ::read: Returns a string with all or a subset of bytes from the given file.

  • ::readlines: Returns an array of strings, which are the lines from the given file.

  • getbyte: Returns the next 8-bit byte read from self as an integer.

  • getc: Returns the next character read from self as a string.

  • gets: Returns the line read from self.

  • pread: Returns all or the next n bytes read from self, not updating the receiver’s offset.

  • read: Returns all remaining or the next n bytes read from self for a given n.

  • read_nonblock: the next n bytes read from self for a given n, in non-block mode.

  • readbyte: Returns the next byte read from self; same as getbyte, but raises an exception on end-of-stream.

  • readchar: Returns the next character read from self; same as getc, but raises an exception on end-of-stream.

  • readline: Returns the next line read from self; same as getline, but raises an exception of end-of-stream.

  • readlines: Returns an array of all lines read read from self.

  • readpartial: Returns up to the given number of bytes from self.

Writing

  • ::binwrite: Writes the given string to the file at the given filepath, in binary mode.

  • ::write: Writes the given string to self.

  • <<: Appends the given string to self.

  • print: Prints last read line or given objects to self.

  • printf: Writes to self based on the given format string and objects.

  • putc: Writes a character to self.

  • puts: Writes lines to self, making sure line ends with a newline.

  • pwrite: Writes the given string at the given offset, not updating the receiver’s offset.

  • write: Writes one or more given strings to self.

  • write_nonblock: Writes one or more given strings to self in non-blocking mode.

Positioning

  • lineno: Returns the current line number in self.

  • lineno=: Sets the line number is self.

  • pos (aliased as tell): Returns the current byte offset in self.

  • pos=: Sets the byte offset in self.

  • reopen: Reassociates self with a new or existing IO stream.

  • rewind: Positions self to the beginning of input.

  • seek: Sets the offset for self relative to given position.

Iterating

  • ::foreach: Yields each line of given file to the block.

  • each (aliased as each_line): Calls the given block with each successive line in self.

  • each_byte: Calls the given block with each successive byte in self as an integer.

  • each_char: Calls the given block with each successive character in self as a string.

  • each_codepoint: Calls the given block with each successive codepoint in self as an integer.

Settings

Querying

  • autoclose?: Returns whether self auto-closes.

  • binmode?: Returns whether self is in binary mode.

  • close_on_exec?: Returns the close-on-exec flag for self.

  • closed?: Returns whether self is closed.

  • eof? (aliased as eof): Returns whether self is at end-of-stream.

  • external_encoding: Returns the external encoding object for self.

  • fileno (aliased as to_i): Returns the integer file descriptor for self

  • internal_encoding: Returns the internal encoding object for self.

  • pid: Returns the process ID of a child process associated with self, if self was created by ::popen.

  • stat: Returns the File::Stat object containing status information for self.

  • sync: Returns whether self is in sync-mode.

  • tty? (aliased as isatty): Returns whether self is a terminal.

Buffering

  • fdatasync: Immediately writes all buffered data in self to disk.

  • flush: Flushes any buffered data within self to the underlying operating system.

  • fsync: Immediately writes all buffered data and attributes in self to disk.

  • ungetbyte: Prepends buffer for self with given integer byte or string.

  • ungetc: Prepends buffer for self with given string.

Low-Level Access

  • ::sysopen: Opens the file given by its path, returning the integer file descriptor.

  • advise: Announces the intention to access data from self in a specific way.

  • fcntl: Passes a low-level command to the file specified by the given file descriptor.

  • ioctl: Passes a low-level command to the device specified by the given file descriptor.

  • sysread: Returns up to the next n bytes read from self using a low-level read.

  • sysseek: Sets the offset for self.

  • syswrite: Writes the given string to self using a low-level write.

Other

  • ::copy_stream: Copies data from a source to a destination, each of which is a filepath or an IO-like object.

  • ::try_convert: Returns a new IO object resulting from converting the given object.

  • inspect: Returns the string representation of self.

Constants
No documentation available
No documentation available
No documentation available

Set I/O position from the beginning

Set I/O position from the current position

Set I/O position from the end

Set I/O position to the next location containing data

Set I/O position to the next hole

Class Methods

Behaves like IO.read, except that the stream is opened in binary mode with ASCII-8BIT encoding.

When called from class IO (but not subclasses of IO), this method has potential security vulnerabilities if called with untrusted input; see Command Injection.

Behaves like IO.write, except that the stream is opened in binary mode with ASCII-8BIT encoding.

When called from class IO (but not subclasses of IO), this method has potential security vulnerabilities if called with untrusted input; see Command Injection.

Returns an File instance opened console.

If sym is given, it will be sent to the opened console with args and the result will be returned instead of the console IO itself.

You must require ‘io/console’ to use this method.

Copies from the given src to the given dst, returning the number of bytes copied.

  • The given src must be one of the following:

    • The path to a readable file, from which source data is to be read.

    • An IO-like object, opened for reading and capable of responding to method :readpartial or method :read.

  • The given dst must be one of the following:

    • The path to a writable file, to which data is to be written.

    • An IO-like object, opened for writing and capable of responding to method :write.

The examples here use file t.txt as source:

File.read('t.txt')
# => "First line\nSecond line\n\nThird line\nFourth line\n"
File.read('t.txt').size # => 47

If only arguments src and dst are given, the entire source stream is copied:

# Paths.
IO.copy_stream('t.txt', 't.tmp')  # => 47

# IOs (recall that a File is also an IO).
src_io = File.open('t.txt', 'r') # => #<File:t.txt>
dst_io = File.open('t.tmp', 'w') # => #<File:t.tmp>
IO.copy_stream(src_io, dst_io)   # => 47
src_io.close
dst_io.close

With argument src_length a non-negative integer, no more than that many bytes are copied:

IO.copy_stream('t.txt', 't.tmp', 10) # => 10
File.read('t.tmp')                   # => "First line"

With argument src_offset also given, the source stream is read beginning at that offset:

IO.copy_stream('t.txt', 't.tmp', 11, 11) # => 11
IO.read('t.tmp')                         # => "Second line"

fallback to console window size

Synonym for IO.new.

Calls the block with each successive line read from the stream.

When called from class IO (but not subclasses of IO), this method has potential security vulnerabilities if called with untrusted input; see Command Injection.

The first argument must be a string that is one of the following:

  • Path: if self is a subclass of IO (File, for example), or if the string does not start with the pipe character ('|'), the string is the path to a file.

  • Command: if self is the class IO, and if the string starts with the pipe character, the rest of the string is a command to be executed as a subprocess. This usage has potential security vulnerabilities if called with untrusted input; see Command Injection.

With only argument path given, parses lines from the file at the given path, as determined by the default line separator, and calls the block with each successive line:

File.foreach('t.txt') {|line| p line }

Output: the same as above.

For both forms, command and path, the remaining arguments are the same.

With argument sep given, parses lines as determined by that line separator (see Line Separator):

File.foreach('t.txt', 'li') {|line| p line }

Output:

"First li"
"ne\nSecond li"
"ne\n\nThird li"
"ne\nFourth li"
"ne\n"

Each paragraph:

File.foreach('t.txt', '') {|paragraph| p paragraph }

Output:

"First line\nSecond line\n\n"
"Third line\nFourth line\n"

With argument limit given, parses lines as determined by the default line separator and the given line-length limit (see Line Limit):

File.foreach('t.txt', 7) {|line| p line }

Output:

"First l"
"ine\n"
"Second "
"line\n"
"\n"
"Third l"
"ine\n"
"Fourth l"
"line\n"

With arguments sep and limit given, parses lines as determined by the given line separator and the given line-length limit (see Line Separator and Line Limit):

Optional keyword arguments opts specify:

Returns an Enumerator if no block is given.

Creates and returns a new IO object (file stream) from a file descriptor.

IO.new may be useful for interaction with low-level libraries. For higher-level interactions, it may be simpler to create the file stream using File.open.

Argument fd must be a valid file descriptor (integer):

path = 't.tmp'
fd = IO.sysopen(path) # => 3
IO.new(fd)            # => #<IO:fd 3>

The new IO object does not inherit encoding (because the integer file descriptor does not have an encoding):

fd = IO.sysopen('t.rus', 'rb')
io = IO.new(fd)
io.external_encoding # => #<Encoding:UTF-8> # Not ASCII-8BIT.

Optional argument mode (defaults to ‘r’) must specify a valid mode; see Access Modes:

IO.new(fd, 'w')         # => #<IO:fd 3>
IO.new(fd, File::WRONLY) # => #<IO:fd 3>

Optional keyword arguments opts specify:

Examples:

IO.new(fd, internal_encoding: nil) # => #<IO:fd 3>
IO.new(fd, autoclose: true)        # => #<IO:fd 3>

Creates a new IO object, via IO.new with the given arguments.

With no block given, returns the IO object.

With a block given, calls the block with the IO object and returns the block’s value.

Creates a pair of pipe endpoints, read_io and write_io, connected to each other.

If argument enc_string is given, it must be a string containing one of:

  • The name of the encoding to be used as the external encoding.

  • The colon-separated names of two encodings to be used as the external and internal encodings.

If argument int_enc is given, it must be an Encoding object or encoding name string that specifies the internal encoding to be used; if argument ext_enc is also given, it must be an Encoding object or encoding name string that specifies the external encoding to be used.

The string read from read_io is tagged with the external encoding; if an internal encoding is also specified, the string is converted to, and tagged with, that encoding.

If any encoding is specified, optional hash arguments specify the conversion option.

Optional keyword arguments opts specify:

With no block given, returns the two endpoints in an array:

IO.pipe # => [#<IO:fd 4>, #<IO:fd 5>]

With a block given, calls the block with the two endpoints; closes both endpoints and returns the value of the block:

IO.pipe {|read_io, write_io| p read_io; p write_io }

Output:

#<IO:fd 6>
#<IO:fd 7>

Not available on all platforms.

In the example below, the two processes close the ends of the pipe that they are not using. This is not just a cosmetic nicety. The read end of a pipe will not generate an end of file condition if there are any writers with the pipe still open. In the case of the parent process, the rd.read will never return if it does not first issue a wr.close:

rd, wr = IO.pipe

if fork
  wr.close
  puts "Parent got: <#{rd.read}>"
  rd.close
  Process.wait
else
  rd.close
  puts 'Sending message to parent'
  wr.write "Hi Dad"
  wr.close
end

produces:

Sending message to parent
Parent got: <Hi Dad>

Executes the given command cmd as a subprocess whose $stdin and $stdout are connected to a new stream io.

This method has potential security vulnerabilities if called with untrusted input; see Command Injection.

If no block is given, returns the new stream, which depending on given mode may be open for reading, writing, or both. The stream should be explicitly closed (eventually) to avoid resource leaks.

If a block is given, the stream is passed to the block (again, open for reading, writing, or both); when the block exits, the stream is closed, and the block’s value is assigned to global variable $? and returned.

Optional argument mode may be any valid IO mode. See Access Modes.

Required argument cmd determines which of the following occurs:

  • The process forks.

  • A specified program runs in a shell.

  • A specified program runs with specified arguments.

  • A specified program runs with specified arguments and a specified argv0.

Each of these is detailed below.

The optional hash argument env specifies name/value pairs that are to be added to the environment variables for the subprocess:

IO.popen({'FOO' => 'bar'}, 'ruby', 'r+') do |pipe|
  pipe.puts 'puts ENV["FOO"]'
  pipe.close_write
  pipe.gets
end => "bar\n"

Optional keyword arguments opts specify:

Forked Process

When argument cmd is the 1-character string '-', causes the process to fork:

IO.popen('-') do |pipe|
  if pipe
    $stderr.puts "In parent, child pid is #{pipe.pid}\n"
  else
    $stderr.puts "In child, pid is #{$$}\n"
  end
end

Output:

In parent, child pid is 26253
In child, pid is 26253

Note that this is not supported on all platforms.

Shell Subprocess

When argument cmd is a single string (but not '-'), the program named cmd is run as a shell command:

IO.popen('uname') do |pipe|
  pipe.readlines
end

Output:

["Linux\n"]

Another example:

IO.popen('/bin/sh', 'r+') do |pipe|
  pipe.puts('ls')
  pipe.close_write
  $stderr.puts pipe.readlines.size
end

Output:

213

Program Subprocess

When argument cmd is an array of strings, the program named cmd[0] is run with all elements of cmd as its arguments:

IO.popen(['du', '..', '.']) do |pipe|
  $stderr.puts pipe.readlines.size
end

Output:

1111

Program Subprocess with argv0

When argument cmd is an array whose first element is a 2-element string array and whose remaining elements (if any) are strings:

  • cmd[0][0] (the first string in the nested array) is the name of a program that is run.

  • cmd[0][1] (the second string in the nested array) is set as the program’s argv[0].

  • cmd[1..-1] (the strings in the outer array) are the program’s arguments.

Example (sets $0 to ‘foo’):

IO.popen([['/bin/sh', 'foo'], '-c', 'echo $0']).read # => "foo\n"

Some Special Examples

# Set IO encoding.
IO.popen("nkf -e filename", :external_encoding=>"EUC-JP") {|nkf_io|
  euc_jp_string = nkf_io.read
}

# Merge standard output and standard error using Kernel#spawn option. See Kernel#spawn.
IO.popen(["ls", "/", :err=>[:child, :out]]) do |io|
  ls_result_with_error = io.read
end

# Use mixture of spawn options and IO options.
IO.popen(["ls", "/"], :err=>[:child, :out]) do |io|
  ls_result_with_error = io.read
end

 f = IO.popen("uname")
 p f.readlines
 f.close
 puts "Parent is #{Process.pid}"
 IO.popen("date") {|f| puts f.gets }
 IO.popen("-") {|f| $stderr.puts "#{Process.pid} is here, f is #{f.inspect}"}
 p $?
 IO.popen(%w"sed -e s|^|<foo>| -e s&$&;zot;&", "r+") {|f|
   f.puts "bar"; f.close_write; puts f.gets
 }

Output (from last section):

["Linux\n"]
Parent is 21346
Thu Jan 15 22:41:19 JST 2009
21346 is here, f is #<IO:fd 3>
21352 is here, f is nil
#<Process::Status: pid 21352 exit 0>
<foo>bar;zot;

Raises exceptions that IO.pipe and Kernel.spawn raise.

Opens the stream, reads and returns some or all of its content, and closes the stream; returns nil if no bytes were read.

When called from class IO (but not subclasses of IO), this method has potential security vulnerabilities if called with untrusted input; see Command Injection.

The first argument must be a string; its meaning depends on whether it starts with the pipe character ('|'):

  • If so (and if self is IO), the rest of the string is a command to be executed as a subprocess.

  • Otherwise, the string is the path to a file.

With only argument command given, executes the command in a shell, returns its entire $stdout:

IO.read('| cat t.txt')
# => "First line\nSecond line\n\nThird line\nFourth line\n"

With only argument path given, reads in text mode and returns the entire content of the file at the given path:

IO.read('t.txt')
# => "First line\nSecond line\n\nThird line\nFourth line\n"

On Windows, text mode can terminate reading and leave bytes in the file unread when encountering certain special bytes. Consider using IO.binread if all bytes in the file should be read.

For both forms, command and path, the remaining arguments are the same.

With argument length, returns length bytes if available:

IO.read('t.txt', 7) # => "First l"
IO.read('t.txt', 700)
# => "First line\r\nSecond line\r\n\r\nFourth line\r\nFifth line\r\n"

With arguments length and offset, returns length bytes if available, beginning at the given offset:

IO.read('t.txt', 10, 2)   # => "rst line\nS"
IO.read('t.txt', 10, 200) # => nil

Optional keyword arguments opts specify:

Returns an array of all lines read from the stream.

When called from class IO (but not subclasses of IO), this method has potential security vulnerabilities if called with untrusted input; see Command Injection.

The first argument must be a string; its meaning depends on whether it starts with the pipe character ('|'):

  • If so (and if self is IO), the rest of the string is a command to be executed as a subprocess.

  • Otherwise, the string is the path to a file.

With only argument command given, executes the command in a shell, parses its $stdout into lines, as determined by the default line separator, and returns those lines in an array:

IO.readlines('| cat t.txt')
# => ["First line\n", "Second line\n", "\n", "Third line\n", "Fourth line\n"]

With only argument path given, parses lines from the file at the given path, as determined by the default line separator, and returns those lines in an array:

IO.readlines('t.txt')
# => ["First line\n", "Second line\n", "\n", "Third line\n", "Fourth line\n"]

For both forms, command and path, the remaining arguments are the same.

With argument sep given, parses lines as determined by that line separator (see Line Separator):

# Ordinary separator.
IO.readlines('t.txt', 'li')
# =>["First li", "ne\nSecond li", "ne\n\nThird li", "ne\nFourth li", "ne\n"]
# Get-paragraphs separator.
IO.readlines('t.txt', '')
# => ["First line\nSecond line\n\n", "Third line\nFourth line\n"]
# Get-all separator.
IO.readlines('t.txt', nil)
# => ["First line\nSecond line\n\nThird line\nFourth line\n"]

With argument limit given, parses lines as determined by the default line separator and the given line-length limit (see Line Limit):

IO.readlines('t.txt', 7)
# => ["First l", "ine\n", "Second ", "line\n", "\n", "Third l", "ine\n", "Fourth ", "line\n"]

With arguments sep and limit given, parses lines as determined by the given line separator and the given line-length limit (see Line Separator and Line Limit):

Optional keyword arguments opts specify:

Invokes system call select(2), which monitors multiple file descriptors, waiting until one or more of the file descriptors becomes ready for some class of I/O operation.

Not implemented on all platforms.

Each of the arguments read_ios, write_ios, and error_ios is an array of IO objects.

Argument timeout is an integer timeout interval in seconds.

The method monitors the IO objects given in all three arrays, waiting for some to be ready; returns a 3-element array whose elements are:

  • An array of the objects in read_ios that are ready for reading.

  • An array of the objects in write_ios that are ready for writing.

  • An array of the objects in error_ios have pending exceptions.

If no object becomes ready within the given timeout, nil is returned.

IO.select peeks the buffer of IO objects for testing readability. If the IO buffer is not empty, IO.select immediately notifies readability. This “peek” only happens for IO objects. It does not happen for IO-like objects such as OpenSSL::SSL::SSLSocket.

The best way to use IO.select is invoking it after non-blocking methods such as read_nonblock, write_nonblock, etc. The methods raise an exception which is extended by IO::WaitReadable or IO::WaitWritable. The modules notify how the caller should wait with IO.select. If IO::WaitReadable is raised, the caller should wait for reading. If IO::WaitWritable is raised, the caller should wait for writing.

So, blocking read (readpartial) can be emulated using read_nonblock and IO.select as follows:

begin
  result = io_like.read_nonblock(maxlen)
rescue IO::WaitReadable
  IO.select([io_like])
  retry
rescue IO::WaitWritable
  IO.select(nil, [io_like])
  retry
end

Especially, the combination of non-blocking methods and IO.select is preferred for IO like objects such as OpenSSL::SSL::SSLSocket. It has to_io method to return underlying IO object. IO.select calls to_io to obtain the file descriptor to wait.

This means that readability notified by IO.select doesn’t mean readability from OpenSSL::SSL::SSLSocket object.

The most likely situation is that OpenSSL::SSL::SSLSocket buffers some data. IO.select doesn’t see the buffer. So IO.select can block when OpenSSL::SSL::SSLSocket#readpartial doesn’t block.

However, several more complicated situations exist.

SSL is a protocol which is sequence of records. The record consists of multiple bytes. So, the remote side of SSL sends a partial record, IO.select notifies readability but OpenSSL::SSL::SSLSocket cannot decrypt a byte and OpenSSL::SSL::SSLSocket#readpartial will block.

Also, the remote side can request SSL renegotiation which forces the local SSL engine to write some data. This means OpenSSL::SSL::SSLSocket#readpartial may invoke write system call and it can block. In such a situation, OpenSSL::SSL::SSLSocket#read_nonblock raises IO::WaitWritable instead of blocking. So, the caller should wait for ready for writability as above example.

The combination of non-blocking methods and IO.select is also useful for streams such as tty, pipe socket socket when multiple processes read from a stream.

Finally, Linux kernel developers don’t guarantee that readability of select(2) means readability of following read(2) even for a single process; see select(2)

Invoking IO.select before IO#readpartial works well as usual. However it is not the best way to use IO.select.

The writability notified by select(2) doesn’t show how many bytes are writable. IO#write method blocks until given whole string is written. So, IO#write(two or more bytes) can block after writability is notified by IO.select. IO#write_nonblock is required to avoid the blocking.

Blocking write (write) can be emulated using write_nonblock and IO.select as follows: IO::WaitReadable should also be rescued for SSL renegotiation in OpenSSL::SSL::SSLSocket.

while 0 < string.bytesize
  begin
    written = io_like.write_nonblock(string)
  rescue IO::WaitReadable
    IO.select([io_like])
    retry
  rescue IO::WaitWritable
    IO.select(nil, [io_like])
    retry
  end
  string = string.byteslice(written..-1)
end

Example:

rp, wp = IO.pipe
mesg = "ping "
100.times {
  # IO.select follows IO#read.  Not the best way to use IO.select.
  rs, ws, = IO.select([rp], [wp])
  if r = rs[0]
    ret = r.read(5)
    print ret
    case ret
    when /ping/
      mesg = "pong\n"
    when /pong/
      mesg = "ping "
    end
  end
  if w = ws[0]
    w.write(mesg)
  end
}

Output:

ping pong
ping pong
ping pong
(snipped)
ping

Opens the file at the given path with the given mode and permissions; returns the integer file descriptor.

If the file is to be readable, it must exist; if the file is to be writable and does not exist, it is created with the given permissions:

File.write('t.tmp', '')  # => 0
IO.sysopen('t.tmp')      # => 8
IO.sysopen('t.tmp', 'w') # => 9

Attempts to convert object into an IO object via method to_io; returns the new IO object if successful, or nil otherwise:

IO.try_convert(STDOUT)   # => #<IO:<STDOUT>>
IO.try_convert(ARGF)     # => #<IO:<STDIN>>
IO.try_convert('STDOUT') # => nil

Opens the stream, writes the given data to it, and closes the stream; returns the number of bytes written.

When called from class IO (but not subclasses of IO), this method has potential security vulnerabilities if called with untrusted input; see Command Injection.

The first argument must be a string; its meaning depends on whether it starts with the pipe character ('|'):

  • If so (and if self is IO), the rest of the string is a command to be executed as a subprocess.

  • Otherwise, the string is the path to a file.

With argument command given, executes the command in a shell, passes data through standard input, writes its output to $stdout, and returns the length of the given data:

IO.write('| cat', 'Hello World!') # => 12

Output:

Hello World!

With argument path given, writes the given data to the file at that path:

IO.write('t.tmp', 'abc')    # => 3
File.read('t.tmp')          # => "abc"

If offset is zero (the default), the file is overwritten:

IO.write('t.tmp', 'A')      # => 1
File.read('t.tmp')          # => "A"

If offset in within the file content, the file is partly overwritten:

IO.write('t.tmp', 'abcdef') # => 3
File.read('t.tmp')          # => "abcdef"
# Offset within content.
IO.write('t.tmp', '012', 2) # => 3
File.read('t.tmp')          # => "ab012f"

If offset is outside the file content, the file is padded with null characters "\u0000":

IO.write('t.tmp', 'xyz', 10) # => 3
File.read('t.tmp')           # => "ab012f\u0000\u0000\u0000\u0000xyz"

Optional keyword arguments opts specify:

Instance Methods

Writes the given object to self, which must be opened for writing (see Access Modes); returns self; if object is not a string, it is converted via method to_s:

$stdout << 'Hello' << ', ' << 'World!' << "\n"
$stdout << 'foo' << :bar << 2 << "\n"

Output:

Hello, World!
foobar2

Invokes Posix system call posix_fadvise(2), which announces an intention to access data from the current file in a particular manner.

The arguments and results are platform-dependent.

The relevant data is specified by:

  • offset: The offset of the first byte of data.

  • len: The number of bytes to be accessed; if len is zero, or is larger than the number of bytes remaining, all remaining bytes will be accessed.

Argument advice is one of the following symbols:

  • :normal: The application has no advice to give about its access pattern for the specified data. If no advice is given for an open file, this is the default assumption.

  • :sequential: The application expects to access the specified data sequentially (with lower offsets read before higher ones).

  • :random: The specified data will be accessed in random order.

  • :noreuse: The specified data will be accessed only once.

  • :willneed: The specified data will be accessed in the near future.

  • :dontneed: The specified data will not be accessed in the near future.

Not implemented on all platforms.

Sets auto-close flag.

f = open("/dev/null")
IO.for_fd(f.fileno)
# ...
f.gets # may cause Errno::EBADF

f = open("/dev/null")
IO.for_fd(f.fileno).autoclose = false
# ...
f.gets # won't cause Errno::EBADF

Returns true if the underlying file descriptor of ios will be closed automatically at its finalization, otherwise false.

No documentation available

Sets the stream’s data mode as binary (see Data Mode).

A stream’s data mode may not be changed from binary to text.

Returns true if the stream is on binary mode, false otherwise. See Data Mode.

No documentation available
No documentation available

Closes the stream for both reading and writing if open for either or both; returns nil. See Open and Closed Streams.

If the stream is open for writing, flushes any buffered writes to the operating system before closing.

If the stream was opened by IO.popen, sets global variable $? (child exit status).

Example:

IO.popen('ruby', 'r+') do |pipe|
  puts pipe.closed?
  pipe.close
  puts $?
  puts pipe.closed?
end

Output:

false
pid 13760 exit 0
true

Related: IO#close_read, IO#close_write, IO#closed?.

Sets a close-on-exec flag.

f = open("/dev/null")
f.close_on_exec = true
system("cat", "/proc/self/fd/#{f.fileno}") # cat: /proc/self/fd/3: No such file or directory
f.closed?                #=> false

Ruby sets close-on-exec flags of all file descriptors by default since Ruby 2.0.0. So you don’t need to set by yourself. Also, unsetting a close-on-exec flag can cause file descriptor leak if another thread use fork() and exec() (via system() method for example). If you really needs file descriptor inheritance to child process, use spawn()‘s argument such as fd=>fd.

Returns true if the stream will be closed on exec, false otherwise:

f = File.open('t.txt')
f.close_on_exec? # => true
f.close_on_exec = false
f.close_on_exec? # => false
f.close

Closes the stream for reading if open for reading; returns nil. See Open and Closed Streams.

If the stream was opened by IO.popen and is also closed for writing, sets global variable $? (child exit status).

Example:

IO.popen('ruby', 'r+') do |pipe|
  puts pipe.closed?
  pipe.close_write
  puts pipe.closed?
  pipe.close_read
  puts $?
  puts pipe.closed?
end

Output:

false
false
pid 14748 exit 0
true

Related: IO#close, IO#close_write, IO#closed?.

Closes the stream for writing if open for writing; returns nil. See Open and Closed Streams.

Flushes any buffered writes to the operating system before closing.

If the stream was opened by IO.popen and is also closed for reading, sets global variable $? (child exit status).

IO.popen('ruby', 'r+') do |pipe|
  puts pipe.closed?
  pipe.close_read
  puts pipe.closed?
  pipe.close_write
  puts $?
  puts pipe.closed?
end

Output:

false
false
pid 15044 exit 0
true

Related: IO#close, IO#close_read, IO#closed?.

Returns true if the stream is closed for both reading and writing, false otherwise. See Open and Closed Streams.

IO.popen('ruby', 'r+') do |pipe|
  puts pipe.closed?
  pipe.close_read
  puts pipe.closed?
  pipe.close_write
  puts pipe.closed?
end

Output:

false
false
true

Related: IO#close_read, IO#close_write, IO#close.

Returns a data represents the current console mode.

You must require ‘io/console’ to use this method.

Sets the console mode to mode.

You must require ‘io/console’ to use this method.

Yields self within cooked mode.

STDIN.cooked(&:gets)

will read and return a line with echo back and line editing.

You must require ‘io/console’ to use this method.

Enables cooked mode.

If the terminal mode needs to be back, use io.cooked { … }.

You must require ‘io/console’ to use this method.

No documentation available
No documentation available
No documentation available
No documentation available
No documentation available
No documentation available

Calls the block with each remaining line read from the stream; returns self. Does nothing if already at end-of-stream; See Line IO.

With no arguments given, reads lines as determined by line separator $/:

f = File.new('t.txt')
f.each_line {|line| p line }
f.each_line {|line| fail 'Cannot happen' }
f.close

Output:

"First line\n"
"Second line\n"
"\n"
"Fourth line\n"
"Fifth line\n"

With only string argument sep given, reads lines as determined by line separator sep; see Line Separator:

f = File.new('t.txt')
f.each_line('li') {|line| p line }
f.close

Output:

"First li"
"ne\nSecond li"
"ne\n\nFourth li"
"ne\nFifth li"
"ne\n"

The two special values for sep are honored:

f = File.new('t.txt')
# Get all into one string.
f.each_line(nil) {|line| p line }
f.close

Output:

"First line\nSecond line\n\nFourth line\nFifth line\n"

f.rewind
# Get paragraphs (up to two line separators).
f.each_line('') {|line| p line }

Output:

"First line\nSecond line\n\n"
"Fourth line\nFifth line\n"

With only integer argument limit given, limits the number of bytes in each line; see Line Limit:

f = File.new('t.txt')
f.each_line(8) {|line| p line }
f.close

Output:

"First li"
"ne\n"
"Second l"
"ine\n"
"\n"
"Fourth l"
"ine\n"
"Fifth li"
"ne\n"

With arguments sep and limit given, combines the two behaviors:

  • Calls with the next line as determined by line separator sep.

  • But returns no more bytes than are allowed by the limit.

Optional keyword argument chomp specifies whether line separators are to be omitted:

f = File.new('t.txt')
f.each_line(chomp: true) {|line| p line }
f.close

Output:

"First line"
"Second line"
""
"Fourth line"
"Fifth line"

Returns an Enumerator if no block is given.

IO#each is an alias for IO#each_line.

Calls the given block with each byte (0..255) in the stream; returns self. See Byte IO.

f = File.new('t.rus')
a = []
f.each_byte {|b| a << b }
a # => [209, 130, 208, 181, 209, 129, 209, 130]
f.close

Returns an Enumerator if no block is given.

Related: IO#each_char, IO#each_codepoint.

Calls the given block with each character in the stream; returns self. See Character IO.

f = File.new('t.rus')
a = []
f.each_char {|c| a << c.ord }
a # => [1090, 1077, 1089, 1090]
f.close

Returns an Enumerator if no block is given.

Related: IO#each_byte, IO#each_codepoint.

Calls the given block with each codepoint in the stream; returns self:

f = File.new('t.rus')
a = []
f.each_codepoint {|c| a << c }
a # => [1090, 1077, 1089, 1090]
f.close

Returns an Enumerator if no block is given.

Related: IO#each_byte, IO#each_char.

Enables/disables echo back. On some platforms, all combinations of this flags and raw/cooked mode may not be valid.

You must require ‘io/console’ to use this method.

Returns true if echo back is enabled.

You must require ‘io/console’ to use this method.

Returns true if the stream is positioned at its end, false otherwise; see Position:

f = File.open('t.txt')
f.eof           # => false
f.seek(0, :END) # => 0
f.eof           # => true
f.close

Raises an exception unless the stream is opened for reading; see Mode.

If self is a stream such as pipe or socket, this method blocks until the other end sends some data or closes it:

r, w = IO.pipe
Thread.new { sleep 1; w.close }
r.eof? # => true # After 1-second wait.

r, w = IO.pipe
Thread.new { sleep 1; w.puts "a" }
r.eof?  # => false # After 1-second wait.

r, w = IO.pipe
r.eof?  # blocks forever

Note that this method reads data to the input byte buffer. So IO#sysread may not behave as you intend with IO#eof?, unless you call IO#rewind first (which is not available for some streams).

IO#eof? is an alias for IO#eof.

An alias for eof
No documentation available
No documentation available

The expect library adds instance method IO#expect, which is similar to the TCL expect extension.

To use this method, you must require expect:

require 'expect'

Reads from the IO until the given pattern matches or the timeout is over.

It returns an array with the read buffer, followed by the matches. If a block is given, the result is yielded to the block and returns nil.

When called without a block, it waits until the input that matches the given pattern is obtained from the IO or the time specified as the timeout passes. An array is returned when the pattern is obtained from the IO. The first element of the array is the entire string obtained from the IO until the pattern matches, followed by elements indicating which the pattern which matched to the anchor in the regular expression.

The optional timeout parameter defines, in seconds, the total time to wait for the pattern. If the timeout expires or eof is found, nil is returned or yielded. However, the buffer in a timeout session is kept for the next expect call. The default timeout is 9999999 seconds.

Returns the Encoding object that represents the encoding of the stream, or nil if the stream is in write mode and no encoding is specified.

See Encodings.

Invokes Posix system call fcntl(2), which provides a mechanism for issuing low-level commands to control or query a file-oriented I/O stream. Arguments and results are platform dependent.

If +argument is a number, its value is passed directly; if it is a string, it is interpreted as a binary sequence of bytes. (Array#pack might be a useful way to build this string.)

Not implemented on all platforms.

Immediately writes to disk all data buffered in the stream, via the operating system’s: fdatasync(2), if supported, otherwise via fsync(2), if supported; otherwise raises an exception.

Returns the integer file descriptor for the stream:

$stdin.fileno             # => 0
$stdout.fileno            # => 1
$stderr.fileno            # => 2
File.open('t.txt').fileno # => 10
f.close

IO#to_i is an alias for IO#fileno.

Flushes data buffered in self to the operating system (but does not necessarily flush data buffered in the operating system):

$stdout.print 'no newline' # Not necessarily flushed.
$stdout.flush              # Flushed.

Immediately writes to disk all data buffered in the stream, via the operating system’s fsync(2).

Note this difference:

  • IO#sync=: Ensures that data is flushed from the stream’s internal buffers, but does not guarantee that the operating system actually writes the data to disk.

  • IO#fsync: Ensures both that data is flushed from internal buffers, and that data is written to disk.

Raises an exception if the operating system does not support fsync(2).

Reads and returns the next byte (in range 0..255) from the stream; returns nil if already at end-of-stream. See Byte IO.

f = File.open('t.txt')
f.getbyte # => 70
f.close
f = File.open('t.rus')
f.getbyte # => 209
f.close

Related: IO#readbyte (may raise EOFError).

Reads and returns the next 1-character string from the stream; returns nil if already at end-of-stream. See Character IO.

f = File.open('t.txt')
f.getc     # => "F"
f.close
f = File.open('t.rus')
f.getc.ord # => 1090
f.close

Related: IO#readchar (may raise EOFError).

Reads and returns a character in raw mode.

See IO#raw for details on the parameters.

You must require ‘io/console’ to use this method.

Reads and returns a line without echo back. Prints prompt unless it is nil.

The newline character that terminates the read line is removed from the returned string, see String#chomp!.

You must require ‘io/console’ to use this method.

Reads and returns a line from the stream; assigns the return value to $_. See Line IO.

With no arguments given, returns the next line as determined by line separator $/, or nil if none:

f = File.open('t.txt')
f.gets # => "First line\n"
$_     # => "First line\n"
f.gets # => "\n"
f.gets # => "Fourth line\n"
f.gets # => "Fifth line\n"
f.gets # => nil
f.close

With only string argument sep given, returns the next line as determined by line separator sep, or nil if none; see Line Separator:

f = File.new('t.txt')
f.gets('l')   # => "First l"
f.gets('li')  # => "ine\nSecond li"
f.gets('lin') # => "ne\n\nFourth lin"
f.gets        # => "e\n"
f.close

The two special values for sep are honored:

f = File.new('t.txt')
# Get all.
f.gets(nil) # => "First line\nSecond line\n\nFourth line\nFifth line\n"
f.rewind
# Get paragraph (up to two line separators).
f.gets('')  # => "First line\nSecond line\n\n"
f.close

With only integer argument limit given, limits the number of bytes in the line; see Line Limit:

# No more than one line.
File.open('t.txt') {|f| f.gets(10) } # => "First line"
File.open('t.txt') {|f| f.gets(11) } # => "First line\n"
File.open('t.txt') {|f| f.gets(12) } # => "First line\n"

With arguments sep and limit given, combines the two behaviors:

  • Returns the next line as determined by line separator sep, or nil if none.

  • But returns no more bytes than are allowed by the limit.

Optional keyword argument chomp specifies whether line separators are to be omitted:

f = File.open('t.txt')
# Chomp the lines.
f.gets(chomp: true) # => "First line"
f.gets(chomp: true) # => "Second line"
f.gets(chomp: true) # => ""
f.gets(chomp: true) # => "Fourth line"
f.gets(chomp: true) # => "Fifth line"
f.gets(chomp: true) # => nil
f.close
No documentation available
No documentation available

Flushes input buffer in kernel.

You must require ‘io/console’ to use this method.

Returns a string representation of self:

f = File.open('t.txt')
f.inspect # => "#<File:t.txt>"
f.close

Returns the Encoding object that represents the encoding of the internal string, if conversion is specified, or nil otherwise.

See Encodings.

Invokes Posix system call ioctl(2), which issues a low-level command to an I/O device.

Issues a low-level command to an I/O device. The arguments and returned value are platform-dependent. The effect of the call is platform-dependent.

If argument argument is an integer, it is passed directly; if it is a string, it is interpreted as a binary sequence of bytes.

Not implemented on all platforms.

Flushes input and output buffers in kernel.

You must require ‘io/console’ to use this method.

Returns true if the stream is associated with a terminal device (tty), false otherwise:

f = File.new('t.txt').isatty    #=> false
f.close
f = File.new('/dev/tty').isatty #=> true
f.close

IO#tty? is an alias for IO#isatty.

Returns the current line number for the stream; see Line Number.

Sets and returns the line number for the stream; see Line Number.

Yields self with disabling echo back.

STDIN.noecho(&:gets)

will read and return a line without echo back.

You must require ‘io/console’ to use this method.

Yields self in non-blocking mode.

When false is given as an argument, self is yielded in blocking mode. The original mode is restored after the block is executed.

Enables non-blocking mode on a stream when set to true, and blocking mode when set to false.

This method set or clear O_NONBLOCK flag for the file descriptor in ios.

The behavior of most IO methods is not affected by this flag because they retry system calls to complete their task after EAGAIN and partial read/write. (An exception is IO#syswrite which doesn’t retry.)

This method can be used to clear non-blocking mode of standard I/O. Since nonblocking methods (read_nonblock, etc.) set non-blocking mode but they doesn’t clear it, this method is usable as follows.

END { STDOUT.nonblock = false }
STDOUT.write_nonblock("foo")

Since the flag is shared across processes and many non-Ruby commands doesn’t expect standard I/O with non-blocking mode, it would be safe to clear the flag before Ruby program exits.

For example following Ruby program leaves STDIN/STDOUT/STDER non-blocking mode. (STDIN, STDOUT and STDERR are connected to a terminal. So making one of them nonblocking-mode effects other two.) Thus cat command try to read from standard input and it causes “Resource temporarily unavailable” error (EAGAIN).

% ruby -e '
STDOUT.write_nonblock("foo\n")'; cat
foo
cat: -: Resource temporarily unavailable

Clearing the flag makes the behavior of cat command normal. (cat command waits input from standard input.)

% ruby -rio/nonblock -e '
END { STDOUT.nonblock = false }
STDOUT.write_nonblock("foo")
'; cat
foo

Returns true if an IO object is in non-blocking mode.

Returns number of bytes that can be read without blocking. Returns zero if no information available.

You must require ‘io/wait’ to use this method.

Flushes output buffer in kernel.

You must require ‘io/console’ to use this method.

Returns the path associated with the IO, or nil if there is no path associated with the IO. It is not guaranteed that the path exists on the filesystem.

$stdin.path # => "<STDIN>"

File.open("testfile") {|f| f.path} # => "testfile"

Returns pathname configuration variable using fpathconf().

name should be a constant under Etc which begins with PC_.

The return value is an integer or nil. nil means indefinite limit. (fpathconf() returns -1 but errno is not set.)

require 'etc'
IO.pipe {|r, w|
  p w.pathconf(Etc::PC_PIPE_BUF) #=> 4096
}

Returns the process ID of a child process associated with the stream, which will have been set by IO#popen, or nil if the stream was not created by IO#popen:

pipe = IO.popen("-")
if pipe
  $stderr.puts "In parent, child pid is #{pipe.pid}"
else
  $stderr.puts "In child, pid is #{$$}"
end

Output:

In child, pid is 26209
In parent, child pid is 26209
An alias for tell

Seeks to the given new_position (in bytes); see Position:

f = File.open('t.txt')
f.tell     # => 0
f.pos = 20 # => 20
f.tell     # => 20
f.close

Related: IO#seek, IO#tell.

Behaves like IO#readpartial, except that it:

  • Reads at the given offset (in bytes).

  • Disregards, and does not modify, the stream’s position (see Position).

  • Bypasses any user space buffering in the stream.

Because this method does not disturb the stream’s state (its position, in particular), pread allows multiple threads and processes to use the same IO object for reading at various offsets.

f = File.open('t.txt')
f.read # => "First line\nSecond line\n\nFourth line\nFifth line\n"
f.pos  # => 52
# Read 12 bytes at offset 0.
f.pread(12, 0) # => "First line\n"
# Read 9 bytes at offset 8.
f.pread(9, 8)  # => "ne\nSecon"
f.close

Not available on some platforms.

No documentation available

Writes the given objects to the stream; returns nil. Appends the output record separator $OUTPUT_RECORD_SEPARATOR ($\), if it is not nil. See Line IO.

With argument objects given, for each object:

  • Converts via its method to_s if not a string.

  • Writes to the stream.

  • If not the last object, writes the output field separator $OUTPUT_FIELD_SEPARATOR ($,) if it is not nil.

With default separators:

f = File.open('t.tmp', 'w+')
objects = [0, 0.0, Rational(0, 1), Complex(0, 0), :zero, 'zero']
p $OUTPUT_RECORD_SEPARATOR
p $OUTPUT_FIELD_SEPARATOR
f.print(*objects)
f.rewind
p f.read
f.close

Output:

nil
nil
"00.00/10+0izerozero"

With specified separators:

$\ = "\n"
$, = ','
f.rewind
f.print(*objects)
f.rewind
p f.read

Output:

"0,0.0,0/1,0+0i,zero,zero\n"

With no argument given, writes the content of $_ (which is usually the most recent user input):

f = File.open('t.tmp', 'w+')
gets # Sets $_ to the most recent user input.
f.print
f.close

Formats and writes objects to the stream.

For details on format_string, see Format Specifications.

Writes a character to the stream. See Character IO.

If object is numeric, converts to integer if necessary, then writes the character whose code is the least significant byte; if object is a string, writes the first character:

$stdout.putc "A"
$stdout.putc 65

Output:

AA

Writes the given objects to the stream, which must be open for writing; returns nil.\ Writes a newline after each that does not already end with a newline sequence. If called without arguments, writes a newline. See Line IO.

Note that each added newline is the character "\n"<//tt>, not the output record separator (<tt>$\).

Treatment for each object:

  • String: writes the string.

  • Neither string nor array: writes object.to_s.

  • Array: writes each element of the array; arrays may be nested.

To keep these examples brief, we define this helper method:

def show(*objects)
  # Puts objects to file.
  f = File.new('t.tmp', 'w+')
  f.puts(objects)
  # Return file content.
  f.rewind
  p f.read
  f.close
end

# Strings without newlines.
show('foo', 'bar', 'baz')     # => "foo\nbar\nbaz\n"
# Strings, some with newlines.
show("foo\n", 'bar', "baz\n") # => "foo\nbar\nbaz\n"

# Neither strings nor arrays:
show(0, 0.0, Rational(0, 1), Complex(9, 0), :zero)
# => "0\n0.0\n0/1\n9+0i\nzero\n"

# Array of strings.
show(['foo', "bar\n", 'baz']) # => "foo\nbar\nbaz\n"
# Nested arrays.
show([[[0, 1], 2, 3], 4, 5])  # => "0\n1\n2\n3\n4\n5\n"

Behaves like IO#write, except that it:

  • Writes at the given offset (in bytes).

  • Disregards, and does not modify, the stream’s position (see Position).

  • Bypasses any user space buffering in the stream.

Because this method does not disturb the stream’s state (its position, in particular), pwrite allows multiple threads and processes to use the same IO object for writing at various offsets.

f = File.open('t.tmp', 'w+')
# Write 6 bytes at offset 3.
f.pwrite('ABCDEF', 3) # => 6
f.rewind
f.read # => "\u0000\u0000\u0000ABCDEF"
f.close

Not available on some platforms.

Yields self within raw mode, and returns the result of the block.

STDIN.raw(&:gets)

will read and return a line without echo back and line editing.

The parameter min specifies the minimum number of bytes that should be received when a read operation is performed. (default: 1)

The parameter time specifies the timeout in seconds with a precision of 1/10 of a second. (default: 0)

If the parameter intr is true, enables break, interrupt, quit, and suspend special characters.

Refer to the manual page of termios for further details.

You must require ‘io/console’ to use this method.

Enables raw mode, and returns io.

If the terminal mode needs to be back, use io.raw { ... }.

See IO#raw for details on the parameters.

You must require ‘io/console’ to use this method.

Reads bytes from the stream; the stream must be opened for reading (see Access Modes):

  • If maxlen is nil, reads all bytes using the stream’s data mode.

  • Otherwise reads up to maxlen bytes in binary mode.

Returns a string (either a new string or the given out_string) containing the bytes read. The encoding of the string depends on both maxLen and out_string:

  • maxlen is nil: uses internal encoding of self (regardless of whether out_string was given).

  • maxlen not nil:

    • out_string given: encoding of out_string not modified.

    • out_string not given: ASCII-8BIT is used.

Without Argument out_string

When argument out_string is omitted, the returned value is a new string:

f = File.new('t.txt')
f.read
# => "First line\nSecond line\n\nFourth line\nFifth line\n"
f.rewind
f.read(30) # => "First line\r\nSecond line\r\n\r\nFou"
f.read(30) # => "rth line\r\nFifth line\r\n"
f.read(30) # => nil
f.close

If maxlen is zero, returns an empty string.

With Argument out_string

When argument out_string is given, the returned value is out_string, whose content is replaced:

f = File.new('t.txt')
s = 'foo'      # => "foo"
f.read(nil, s) # => "First line\nSecond line\n\nFourth line\nFifth line\n"
s              # => "First line\nSecond line\n\nFourth line\nFifth line\n"
f.rewind
s = 'bar'
f.read(30, s)  # => "First line\r\nSecond line\r\n\r\nFou"
s              # => "First line\r\nSecond line\r\n\r\nFou"
s = 'baz'
f.read(30, s)  # => "rth line\r\nFifth line\r\n"
s              # => "rth line\r\nFifth line\r\n"
s = 'bat'
f.read(30, s)  # => nil
s              # => ""
f.close

Note that this method behaves like the fread() function in C. This means it retries to invoke read(2) system calls to read data with the specified maxlen (or until EOF).

This behavior is preserved even if the stream is in non-blocking mode. (This method is non-blocking-flag insensitive as other methods.)

If you need the behavior like a single read(2) system call, consider readpartial, read_nonblock, and sysread.

Related: IO#write.

Reads at most maxlen bytes from ios using the read(2) system call after O_NONBLOCK is set for the underlying file descriptor.

If the optional outbuf argument is present, it must reference a String, which will receive the data. The outbuf will contain only the received data after the method call even if it is not empty at the beginning.

read_nonblock just calls the read(2) system call. It causes all errors the read(2) system call causes: Errno::EWOULDBLOCK, Errno::EINTR, etc. The caller should care such errors.

If the exception is Errno::EWOULDBLOCK or Errno::EAGAIN, it is extended by IO::WaitReadable. So IO::WaitReadable can be used to rescue the exceptions for retrying read_nonblock.

read_nonblock causes EOFError on EOF.

On some platforms, such as Windows, non-blocking mode is not supported on IO objects other than sockets. In such cases, Errno::EBADF will be raised.

If the read byte buffer is not empty, read_nonblock reads from the buffer like readpartial. In this case, the read(2) system call is not called.

When read_nonblock raises an exception kind of IO::WaitReadable, read_nonblock should not be called until io is readable for avoiding busy loop. This can be done as follows.

# emulates blocking read (readpartial).
begin
  result = io.read_nonblock(maxlen)
rescue IO::WaitReadable
  IO.select([io])
  retry
end

Although IO#read_nonblock doesn’t raise IO::WaitWritable. OpenSSL::Buffering#read_nonblock can raise IO::WaitWritable. If IO and SSL should be used polymorphically, IO::WaitWritable should be rescued too. See the document of OpenSSL::Buffering#read_nonblock for sample code.

Note that this method is identical to readpartial except the non-blocking flag is set.

By specifying a keyword argument exception to false, you can indicate that read_nonblock should not raise an IO::WaitReadable exception, but return the symbol :wait_readable instead. At EOF, it will return nil instead of raising EOFError.

Reads and returns the next byte (in range 0..255) from the stream; raises EOFError if already at end-of-stream. See Byte IO.

f = File.open('t.txt')
f.readbyte # => 70
f.close
f = File.open('t.rus')
f.readbyte # => 209
f.close

Related: IO#getbyte (will not raise EOFError).

Reads and returns the next 1-character string from the stream; raises EOFError if already at end-of-stream. See Character IO.

f = File.open('t.txt')
f.readchar     # => "F"
f.close
f = File.open('t.rus')
f.readchar.ord # => 1090
f.close

Related: IO#getc (will not raise EOFError).

Reads a line as with IO#gets, but raises EOFError if already at end-of-stream.

Optional keyword argument chomp specifies whether line separators are to be omitted.

Reads and returns all remaining line from the stream; does not modify $_. See Line IO.

With no arguments given, returns lines as determined by line separator $/, or nil if none:

f = File.new('t.txt')
f.readlines
# => ["First line\n", "Second line\n", "\n", "Fourth line\n", "Fifth line\n"]
f.readlines # => []
f.close

With only string argument sep given, returns lines as determined by line separator sep, or nil if none; see Line Separator:

f = File.new('t.txt')
f.readlines('li')
# => ["First li", "ne\nSecond li", "ne\n\nFourth li", "ne\nFifth li", "ne\n"]
f.close

The two special values for sep are honored:

f = File.new('t.txt')
# Get all into one string.
f.readlines(nil)
# => ["First line\nSecond line\n\nFourth line\nFifth line\n"]
# Get paragraphs (up to two line separators).
f.rewind
f.readlines('')
# => ["First line\nSecond line\n\n", "Fourth line\nFifth line\n"]
f.close

With only integer argument limit given, limits the number of bytes in each line; see Line Limit:

f = File.new('t.txt')
f.readlines(8)
# => ["First li", "ne\n", "Second l", "ine\n", "\n", "Fourth l", "ine\n", "Fifth li", "ne\n"]
f.close

With arguments sep and limit given, combines the two behaviors:

  • Returns lines as determined by line separator sep.

  • But returns no more bytes in a line than are allowed by the limit.

Optional keyword argument chomp specifies whether line separators are to be omitted:

f = File.new('t.txt')
f.readlines(chomp: true)
# => ["First line", "Second line", "", "Fourth line", "Fifth line"]
f.close

Reads up to maxlen bytes from the stream; returns a string (either a new string or the given out_string). Its encoding is:

  • The unchanged encoding of out_string, if out_string is given.

  • ASCII-8BIT, otherwise.

  • Contains maxlen bytes from the stream, if available.

  • Otherwise contains all available bytes, if any available.

  • Otherwise is an empty string.

With the single non-negative integer argument maxlen given, returns a new string:

f = File.new('t.txt')
f.readpartial(20) # => "First line\nSecond l"
f.readpartial(20) # => "ine\n\nFourth line\n"
f.readpartial(20) # => "Fifth line\n"
f.readpartial(20) # Raises EOFError.
f.close

With both argument maxlen and string argument out_string given, returns modified out_string:

f = File.new('t.txt')
s = 'foo'
f.readpartial(20, s) # => "First line\nSecond l"
s = 'bar'
f.readpartial(0, s)  # => ""
f.close

This method is useful for a stream such as a pipe, a socket, or a tty. It blocks only when no data is immediately available. This means that it blocks only when all of the following are true:

  • The byte buffer in the stream is empty.

  • The content of the stream is empty.

  • The stream is not at EOF.

When blocked, the method waits for either more data or EOF on the stream:

  • If more data is read, the method returns the data.

  • If EOF is reached, the method raises EOFError.

When not blocked, the method responds immediately:

  • Returns data from the buffer if there is any.

  • Otherwise returns data from the stream if there is any.

  • Otherwise raises EOFError if the stream has reached EOF.

Note that this method is similar to sysread. The differences are:

  • If the byte buffer is not empty, read from the byte buffer instead of “sysread for buffered IO (IOError)”.

  • It doesn’t cause Errno::EWOULDBLOCK and Errno::EINTR. When readpartial meets EWOULDBLOCK and EINTR by read system call, readpartial retries the system call.

The latter means that readpartial is non-blocking-flag insensitive. It blocks on the situation IO#sysread causes Errno::EWOULDBLOCK as if the fd is blocking mode.

Examples:

#                        # Returned      Buffer Content    Pipe Content
r, w = IO.pipe           #
w << 'abc'               #               ""                "abc".
r.readpartial(4096)      # => "abc"      ""                ""
r.readpartial(4096)      # (Blocks because buffer and pipe are empty.)

#                        # Returned      Buffer Content    Pipe Content
r, w = IO.pipe           #
w << 'abc'               #               ""                "abc"
w.close                  #               ""                "abc" EOF
r.readpartial(4096)      # => "abc"      ""                 EOF
r.readpartial(4096)      # raises EOFError

#                        # Returned      Buffer Content    Pipe Content
r, w = IO.pipe           #
w << "abc\ndef\n"        #               ""                "abc\ndef\n"
r.gets                   # => "abc\n"    "def\n"           ""
w << "ghi\n"             #               "def\n"           "ghi\n"
r.readpartial(4096)      # => "def\n"    ""                "ghi\n"
r.readpartial(4096)      # => "ghi\n"    ""                ""

Returns a truthy value if input available without blocking, or a falsy value.

You must require ‘io/wait’ to use this method.

Reassociates the stream with another stream, which may be of a different class. This method may be used to redirect an existing stream to a new destination.

With argument other_io given, reassociates with that stream:

# Redirect $stdin from a file.
f = File.open('t.txt')
$stdin.reopen(f)
f.close

# Redirect $stdout to a file.
f = File.open('t.tmp', 'w')
$stdout.reopen(f)
f.close

With argument path given, reassociates with a new stream to that file path:

$stdin.reopen('t.txt')
$stdout.reopen('t.tmp', 'w')

Optional keyword arguments opts specify:

Repositions the stream to its beginning, setting both the position and the line number to zero; see Position and Line Number:

f = File.open('t.txt')
f.tell     # => 0
f.lineno   # => 0
f.gets     # => "First line\n"
f.tell     # => 12
f.lineno   # => 1
f.rewind   # => 0
f.tell     # => 0
f.lineno   # => 0
f.close

Note that this method cannot be used with streams such as pipes, ttys, and sockets.

No documentation available
No documentation available

Seeks to the position given by integer offset (see Position) and constant whence, which is one of:

  • :CUR or IO::SEEK_CUR: Repositions the stream to its current position plus the given offset:

    f = File.open('t.txt')
    f.tell            # => 0
    f.seek(20, :CUR)  # => 0
    f.tell            # => 20
    f.seek(-10, :CUR) # => 0
    f.tell            # => 10
    f.close
    
  • :END or IO::SEEK_END: Repositions the stream to its end plus the given offset:

    f = File.open('t.txt')
    f.tell            # => 0
    f.seek(0, :END)   # => 0  # Repositions to stream end.
    f.tell            # => 52
    f.seek(-20, :END) # => 0
    f.tell            # => 32
    f.seek(-40, :END) # => 0
    f.tell            # => 12
    f.close
    
  • :SET or IO:SEEK_SET: Repositions the stream to the given offset:

    f = File.open('t.txt')
    f.tell            # => 0
    f.seek(20, :SET) # => 0
    f.tell           # => 20
    f.seek(40, :SET) # => 0
    f.tell           # => 40
    f.close
    

Related: IO#pos=, IO#tell.

See Encodings.

Argument ext_enc, if given, must be an Encoding object; it is assigned as the encoding for the stream.

Argument int_enc, if given, must be an Encoding object; it is assigned as the encoding for the internal string.

Argument 'ext_enc:int_enc', if given, is a string containing two colon-separated encoding names; corresponding Encoding objects are assigned as the external and internal encodings for the stream.

Optional keyword arguments enc_opts specify Encoding options.

If the stream begins with a BOM (byte order marker), consumes the BOM and sets the external encoding accordingly; returns the result encoding if found, or nil otherwise:

File.write('t.tmp', "\u{FEFF}abc")
io = File.open('t.tmp', 'rb')
io.set_encoding_by_bom # => #<Encoding:UTF-8>
io.close

File.write('t.tmp', 'abc')
io = File.open('t.tmp', 'rb')
io.set_encoding_by_bom # => nil
io.close

Raises an exception if the stream is not binmode or its encoding has already been set.

Returns status information for ios as an object of type File::Stat.

f = File.new("testfile")
s = f.stat
"%o" % s.mode   #=> "100644"
s.blksize       #=> 4096
s.atime         #=> Wed Apr 09 08:53:54 CDT 2003

Returns the current sync mode of the stream. When sync mode is true, all output is immediately flushed to the underlying operating system and is not buffered by Ruby internally. See also fsync.

f = File.open('t.tmp', 'w')
f.sync # => false
f.sync = true
f.sync # => true
f.close

Sets the sync mode for the stream to the given value; returns the given value.

Values for the sync mode:

  • true: All output is immediately flushed to the underlying operating system and is not buffered internally.

  • false: Output may be buffered internally.

Example;

f = File.open('t.tmp', 'w')
f.sync # => false
f.sync = true
f.sync # => true
f.close

Related: IO#fsync.

Behaves like IO#readpartial, except that it uses low-level system functions.

This method should not be used with other stream-reader methods.

Behaves like IO#seek, except that it:

  • Uses low-level system functions.

  • Returns the new position.

Writes the given object to self, which must be opened for writing (see Modes); returns the number bytes written. If object is not a string is converted via method to_s:

f = File.new('t.tmp', 'w')
f.syswrite('foo') # => 3
f.syswrite(30)    # => 2
f.syswrite(:foo)  # => 3
f.close

This methods should not be used with other stream-writer methods.

Returns the current position (in bytes) in self (see Position):

f = File.open('t.txt')
f.tell # => 0
f.gets # => "First line\n"
f.tell # => 12
f.close

Related: IO#pos=, IO#seek.

IO#pos is an alias for IO#tell.

Get the internal timeout duration or nil if it was not set.

Set the internal timeout to the specified duration or nil. The timeout applies to all blocking operations where possible.

This affects the following methods (but is not limited to): gets, puts, read, write, wait_readable and wait_writable. This also affects blocking socket operations like Socket#accept and Socket#connect.

Some operations like File#open and IO#close are not affected by the timeout. A timeout during a write operation may leave the IO in an inconsistent state, e.g. data was partially written. Generally speaking, a timeout is a last ditch effort to prevent an application from hanging on slow I/O operations, such as those that occur during a slowloris attack.

An alias for fileno

Returns self.

An alias for path
An alias for isatty

Pushes back (“unshifts”) the given data onto the stream’s buffer, placing the data so that it is next to be read; returns nil. See Byte IO.

Note that:

  • Calling the method has no effect with unbuffered reads (such as IO#sysread).

  • Calling rewind on the stream discards the pushed-back data.

When argument integer is given, uses only its low-order byte:

File.write('t.tmp', '012')
f = File.open('t.tmp')
f.ungetbyte(0x41)   # => nil
f.read              # => "A012"
f.rewind
f.ungetbyte(0x4243) # => nil
f.read              # => "C012"
f.close

When argument string is given, uses all bytes:

File.write('t.tmp', '012')
f = File.open('t.tmp')
f.ungetbyte('A')    # => nil
f.read              # => "A012"
f.rewind
f.ungetbyte('BCDE') # => nil
f.read              # => "BCDE012"
f.close

Pushes back (“unshifts”) the given data onto the stream’s buffer, placing the data so that it is next to be read; returns nil. See Character IO.

Note that:

  • Calling the method has no effect with unbuffered reads (such as IO#sysread).

  • Calling rewind on the stream discards the pushed-back data.

When argument integer is given, interprets the integer as a character:

File.write('t.tmp', '012')
f = File.open('t.tmp')
f.ungetc(0x41)     # => nil
f.read             # => "A012"
f.rewind
f.ungetc(0x0442)   # => nil
f.getc.ord         # => 1090
f.close

When argument string is given, uses all characters:

File.write('t.tmp', '012')
f = File.open('t.tmp')
f.ungetc('A')      # => nil
f.read      # => "A012"
f.rewind
f.ungetc("\u0442\u0435\u0441\u0442") # => nil
f.getc.ord      # => 1090
f.getc.ord      # => 1077
f.getc.ord      # => 1089
f.getc.ord      # => 1090
f.close

Waits until the IO becomes ready for the specified events and returns the subset of events that become ready, or a falsy value when times out.

The events can be a bit mask of IO::READABLE, IO::WRITABLE or IO::PRIORITY.

Returns a truthy value immediately when buffered data is available.

Optional parameter mode is one of :read, :write, or :read_write.

You must require ‘io/wait’ to use this method.

Waits until IO is priority and returns a truthy value or a falsy value when times out. Priority data is sent and received using the Socket::MSG_OOB flag and is typically limited to streams.

You must require ‘io/wait’ to use this method.

Waits until IO is readable and returns a truthy value, or a falsy value when times out. Returns a truthy value immediately when buffered data is available.

You must require ‘io/wait’ to use this method.

Waits until IO is writable and returns a truthy value or a falsy value when times out.

You must require ‘io/wait’ to use this method.

Returns console size.

You must require ‘io/console’ to use this method.

Tries to set console size. The effect depends on the platform and the running environment.

You must require ‘io/console’ to use this method.

Writes each of the given objects to self, which must be opened for writing (see Access Modes); returns the total number bytes written; each of objects that is not a string is converted via method to_s:

$stdout.write('Hello', ', ', 'World!', "\n") # => 14
$stdout.write('foo', :bar, 2, "\n")          # => 8

Output:

Hello, World!
foobar2

Related: IO#read.

Writes the given string to ios using the write(2) system call after O_NONBLOCK is set for the underlying file descriptor.

It returns the number of bytes written.

write_nonblock just calls the write(2) system call. It causes all errors the write(2) system call causes: Errno::EWOULDBLOCK, Errno::EINTR, etc. The result may also be smaller than string.length (partial write). The caller should care such errors and partial write.

If the exception is Errno::EWOULDBLOCK or Errno::EAGAIN, it is extended by IO::WaitWritable. So IO::WaitWritable can be used to rescue the exceptions for retrying write_nonblock.

# Creates a pipe.
r, w = IO.pipe

# write_nonblock writes only 65536 bytes and return 65536.
# (The pipe size is 65536 bytes on this environment.)
s = "a" * 100000
p w.write_nonblock(s)     #=> 65536

# write_nonblock cannot write a byte and raise EWOULDBLOCK (EAGAIN).
p w.write_nonblock("b")   # Resource temporarily unavailable (Errno::EAGAIN)

If the write buffer is not empty, it is flushed at first.

When write_nonblock raises an exception kind of IO::WaitWritable, write_nonblock should not be called until io is writable for avoiding busy loop. This can be done as follows.

begin
  result = io.write_nonblock(string)
rescue IO::WaitWritable, Errno::EINTR
  IO.select(nil, [io])
  retry
end

Note that this doesn’t guarantee to write all data in string. The length written is reported as result and it should be checked later.

On some platforms such as Windows, write_nonblock is not supported according to the kind of the IO object. In such cases, write_nonblock raises Errno::EBADF.

By specifying a keyword argument exception to false, you can indicate that write_nonblock should not raise an IO::WaitWritable exception, but return the symbol :wait_writable instead.