IO

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"

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 the path to a file.

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:

• Open Options.

• Encoding options.

• Line Options.

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:

• Open Options.

• Encoding options.

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:

• Open Options.

• Encoding Options.

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:

• Open options.

• Encoding options.

• Options for Kernel#spawn.

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 that is the path to a file.

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.

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:

• Open Options.

• Encoding options.

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 that is the path to a file.

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"]

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:

• Open Options.

• Encoding options.

• Line Options.

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 that is the path to a file.

With only 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:

• Open Options.

• Encoding options.

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 = File.open(File::NULL)
IO.for_fd(f.fileno).close
f.gets # raises Errno::EBADF

f = File.open(File::NULL)
g = IO.for_fd(f.fileno)
g.autoclose = false
g.close
f.gets # won't cause Errno::EBADF

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

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.

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 = File.open(File::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?.