path: root/lib/stdlib/doc/src/io.xml

<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE erlref SYSTEM "erlref.dtd">

<erlref>
  <header>
    <copyright>
      <year>1996</year><year>2017</year>
      <holder>Ericsson AB. All Rights Reserved.</holder>
    </copyright>
    <legalnotice>
      Licensed under the Apache License, Version 2.0 (the "License");
      you may not use this file except in compliance with the License.
      You may obtain a copy of the License at
 
          http://www.apache.org/licenses/LICENSE-2.0

      Unless required by applicable law or agreed to in writing, software
      distributed under the License is distributed on an "AS IS" BASIS,
      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
      See the License for the specific language governing permissions and
      limitations under the License.

    </legalnotice>

    <title>io</title>
    <prepared></prepared>
    <docno></docno>
    <date></date>
    <rev></rev>
  </header>
  <module>io</module>
  <modulesummary>Standard I/O server interface functions.</modulesummary>
  <description>
    <p>This module provides an interface to standard Erlang I/O servers.
      The output functions all return <c>ok</c> if they are successful,
      or exit if they are not.</p>
 
   <p>All functions in this module have an optional
      parameter <c>IoDevice</c>. If included, it must be the pid of a
      process that handles the I/O protocols. Normally, it is the
      <c>IoDevice</c> returned by
      <seealso marker="kernel:file#open/2"><c>file:open/2</c></seealso>.</p>

    <p>For a description of the I/O protocols, see section
      <seealso marker="io_protocol">The Erlang I/O Protocol</seealso>
      in the User's Guide.</p>

    <warning>
    <p>As from Erlang/OTP R13A, data supplied to function
      <seealso marker="#put_chars/2"><c>put_chars/2</c></seealso>
      is to be in the <seealso marker="unicode#type-chardata">
      <c>unicode:chardata()</c></seealso> format. This means that programs
      supplying binaries to this function must convert them to UTF-8
      before trying to output the data on an I/O device.</p>
    <p>If an I/O device is set in binary mode, functions
      <seealso marker="#get_chars/2"><c>get_chars/2,3</c></seealso> and
      <seealso marker="#get_line/1"><c>get_line/1,2</c></seealso>
      can return binaries instead of lists.
      The binaries are, as from Erlang/OTP R13A,
      encoded in UTF-8.</p>
    <p>To work with binaries in ISO Latin-1 encoding, use the
      <seealso marker="kernel:file"><c>file</c></seealso> module instead.</p>
    <p>For conversion functions between character encodings, see the
      <seealso marker="stdlib:unicode"><c>unicode</c></seealso> module.</p>
    </warning>
  </description>

  <datatypes>
    <datatype>
      <name name="device"/>
      <desc>
        <p>An I/O device, either <c>standard_io</c>, <c>standard_error</c>, a
          registered name, or a pid handling I/O protocols (returned from
          <seealso marker="kernel:file#open/2"><c>file:open/2</c></seealso>).
        </p>
      </desc>
    </datatype>
    <datatype>
      <name name="opt_pair"/>
    </datatype>
    <datatype>
      <name name="expand_fun"/>
    </datatype>
    <datatype>
      <name name="encoding"/>
    </datatype>
    <datatype>
      <name name="setopt"/>
    </datatype>
    <datatype>
      <name name="format"/>
    </datatype>
    <datatype>
      <name name="location"/>
    </datatype>
    <datatype>
      <name name="prompt"/>
    </datatype>
    <datatype>
      <name name="server_no_data"/>
      <desc><p>What the I/O server sends when there is no data.</p></desc>
    </datatype>
  </datatypes>

  <funcs>
    <func>
      <name name="columns" arity="0"/>
      <name name="columns" arity="1"/>
      <fsummary>Get the number of columns of an I/O device.</fsummary>
      <desc>
          <p>Retrieves the number of columns of the
          <c><anno>IoDevice</anno></c> (that is, the width of a terminal).
          The function succeeds for terminal devices and returns
          <c>{error, enotsup}</c> for all other I/O devices.</p>
      </desc>
    </func>

    <func>
      <name name="format" arity="1"/>
      <name name="format" arity="2"/>
      <name name="format" arity="3"/>
      <name name="fwrite" arity="1"/>
      <name name="fwrite" arity="2"/>
      <name name="fwrite" arity="3"/>
      <fsummary>Write formatted output.</fsummary>
      <desc>
        <p>Writes the items in <c><anno>Data</anno></c> (<c>[]</c>) on the
          standard output (<c><anno>IoDevice</anno></c>) in accordance with
          <c><anno>Format</anno></c>. <c><anno>Format</anno></c> contains
          plain characters that are copied to
          the output device, and control sequences for formatting, see
          below. If <c><anno>Format</anno></c> is an atom or a binary, it is
          first converted to a list with the aid of <c>atom_to_list/1</c> or
          <c>binary_to_list/1</c>. Example:</p>
        <pre>
1> <input>io:fwrite("Hello world!~n", []).</input>
Hello world!
ok</pre>
        <p>The general format of a control sequence is <c>~F.P.PadModC</c>.</p>
        <p>Character <c>C</c> determines the type of control sequence
          to be used, <c>F</c> and <c>P</c> are optional numeric
          arguments. If <c>F</c>, <c>P</c>, or <c>Pad</c> is <c>*</c>,
          the next argument in <c>Data</c> is used as the numeric value
          of <c>F</c> or <c>P</c>.</p>
        <list type="bulleted">
        <item>
          <p><c>F</c> is the <c>field width</c> of the printed argument. A
            negative value means that the argument is left-justified
            within the field, otherwise right-justified. If no
            field width is specified, the required print width is
            used. If the field width specified is too small, the
            whole field is filled with <c>*</c> characters.</p>
        </item>
        <item>
          <p><c>P</c> is the <c>precision</c> of the printed argument. A
            default value is used if no precision is specified. The
            interpretation of precision depends on the control sequences.
            Unless otherwise specified, argument <c>within</c> is used
            to determine print width.</p>
        </item>
        <item>
          <p><c>Pad</c> is the padding character. This is the character
            used to pad the printed representation of the argument so that
            it conforms to the specified field width and precision. Only
            one padding character can be specified and, whenever
            applicable, it is used for both the field width and precision.
            The default padding character is <c>' '</c> (space).</p>
        </item>
        <item>
          <p><c>Mod</c> is the control sequence modifier. It is either a
            single character (<c>t</c>, for Unicode
            translation, and <c>l</c>, for stopping <c>p</c> and
            <c>P</c> from detecting printable characters)
            that changes the interpretation of <c>Data</c>.</p>
        </item>
      </list>
        <p><em>Available control sequences:</em></p>
        <taglist>
          <tag><c>~</c></tag>
          <item>
            <p>Character <c>~</c> is written.</p>
          </item>
          <tag><c>c</c></tag>
          <item>
            <p>The argument is a number that is interpreted as an
              ASCII code. The precision is the number of times the
              character is printed and defaults to the field width,
              which in turn defaults to 1. Example:</p>
            <pre>
1> <input>io:fwrite("|~10.5c|~-10.5c|~5c|~n", [$a, $b, $c]).</input>
|     aaaaa|bbbbb     |ccccc|
ok</pre>
            <p>If the Unicode translation modifier (<c>t</c>) is in effect,
	    the integer argument can be any number representing a
	    valid Unicode codepoint, otherwise it is to be an integer
	    less than or equal to 255, otherwise it is masked with 16#FF:</p>
<pre>
2> <input>io:fwrite("~tc~n",[1024]).</input>
\x{400}
ok
3> <input>io:fwrite("~c~n",[1024]).</input>
^@
ok</pre>
          </item>
          <tag><c>f</c></tag>
          <item>
            <p>The argument is a float that is written as
              <c>[-]ddd.ddd</c>, where the precision is the number of
              digits after the decimal point. The default precision is 6
              and it cannot be &lt; 1.</p>
          </item>
          <tag><c>e</c></tag>
          <item>
            <p>The argument is a float that is written as
              <c>[-]d.ddde+-ddd</c>, where the precision is the number
              of digits written. The default precision is 6 and it
              cannot be &lt; 2.</p>
          </item>
          <tag><c>g</c></tag>
          <item>
            <p>The argument is a float that is written as <c>f</c>, if
              it is &gt;= 0.1 and &lt; 10000.0. Otherwise, it is written
              in the <c>e</c> format. The precision is the number of
              significant digits. It defaults to 6 and is not to be
              &lt; 2. If the absolute value of the float does not
              allow it to be written in the <c>f</c> format with the
              desired number of significant digits, it is also written
              in the <c>e</c> format.</p>
          </item>
          <tag><c>s</c></tag>
          <item>
            <p>Prints the argument with the string syntax. The
              argument is, if no Unicode translation modifier is present, an 
              <c>iolist()</c>, a <c>binary()</c>, or an <c>atom()</c>.
              If the Unicode translation modifier (<c>t</c>) is in effect,
              the argument is <c>unicode:chardata()</c>, meaning that
              binaries are in UTF-8. The characters
              are printed without quotes. The string is first truncated
              by the specified precision and then padded and justified to the
              specified field width. The default precision is the field width.
            </p>
            <p>This format can be used for printing any object and
              truncating the output so it fits a specified field:</p>
            <pre>
1> <input>io:fwrite("|~10w|~n", [{hey, hey, hey}]).</input>
|**********|
ok
2> <input>io:fwrite("|~10s|~n", [io_lib:write({hey, hey, hey})]).</input>
|{hey,hey,h|
3> <input>io:fwrite("|~-10.8s|~n", [io_lib:write({hey, hey, hey})]).</input>
|{hey,hey  |
ok</pre>
          <p>A list with integers &gt; 255 is considered an error if the
            Unicode translation modifier is not specified:</p>
<pre>
4> <input>io:fwrite("~ts~n",[[1024]]).</input>
\x{400}
ok
5> <input>io:fwrite("~s~n",[[1024]]).</input>
** exception exit: {badarg,[{io,format,[&lt;0.26.0&gt;,"~s~n",[[1024]]]},
   ...</pre>
          </item>
          <tag><c>w</c></tag>
          <item>
            <p>Writes data with the standard syntax. This is used to
              output Erlang terms. Atoms are printed within quotes if
              they contain embedded non-printable characters.
	      Atom characters &gt; 255 are escaped unless the
	      Unicode translation modifier (<c>t</c>) is used.
              Floats are printed accurately as the shortest, correctly
              rounded string.</p>
          </item>
          <tag><c>p</c></tag>
          <item>
            <p>Writes the data with standard syntax in the same way as
              <c>~w</c>, but breaks terms whose printed representation
              is longer than one line into many lines and indents each
              line sensibly. Left-justification is not supported.
              It also tries to detect lists of
              printable characters and to output these as strings. The
              Unicode translation modifier is used for determining
              what characters are printable, for example:</p>
            <pre>
1> <input>T = [{attributes,[[{id,age,1.50000},{mode,explicit},</input>
<input>{typename,"INTEGER"}], [{id,cho},{mode,explicit},{typename,'Cho'}]]},</input>
<input>{typename,'Person'},{tag,{'PRIVATE',3}},{mode,implicit}].</input>
...
2> <input>io:fwrite("~w~n", [T]).</input>
[{attributes,[[{id,age,1.5},{mode,explicit},{typename,
[73,78,84,69,71,69,82]}],[{id,cho},{mode,explicit},{typena
me,'Cho'}]]},{typename,'Person'},{tag,{'PRIVATE',3}},{mode
,implicit}]
ok
3> <input>io:fwrite("~62p~n", [T]).</input>
[{attributes,[[{id,age,1.5},
               {mode,explicit},
               {typename,"INTEGER"}],
              [{id,cho},{mode,explicit},{typename,'Cho'}]]},
 {typename,'Person'},
 {tag,{'PRIVATE',3}},
 {mode,implicit}]
ok</pre>
            <p>The field width specifies the maximum line length.
              Defaults to 80. The precision specifies the initial
              indentation of the term. It defaults to the number of
              characters printed on this line in the <em>same</em> call to
              <seealso marker="#write/1"><c>write/1</c></seealso> or
              <seealso marker="#format/1"><c>format/1,2,3</c></seealso>.
              For example, using <c>T</c> above:</p>
            <pre>
4> <input>io:fwrite("Here T = ~62p~n", [T]).</input>
Here T = [{attributes,[[{id,age,1.5},
                        {mode,explicit},
                        {typename,"INTEGER"}],
                       [{id,cho},
                        {mode,explicit},
                        {typename,'Cho'}]]},
          {typename,'Person'},
          {tag,{'PRIVATE',3}},
          {mode,implicit}]
ok</pre>
            <p>When the modifier <c>l</c> is specified, no detection of
              printable character lists takes place, for example:</p>
            <pre>
5> <input>S = [{a,"a"}, {b, "b"}].</input>
6> <input>io:fwrite("~15p~n", [S]).</input>
[{a,"a"},
 {b,"b"}]
ok
7> <input>io:fwrite("~15lp~n", [S]).</input>
[{a,[97]},
 {b,[98]}]
ok</pre>
            <p>Binaries that look like UTF-8 encoded strings are
              output with the string syntax if the Unicode translation
              modifier is specified:</p>
            <pre>
9> <input>io:fwrite("~p~n",[[1024]]).</input>
[1024]
10> <input>io:fwrite("~tp~n",[[1024]]).</input>
"\x{400}"
11> <input>io:fwrite("~tp~n", [&lt;&lt;128,128&gt;&gt;]).</input>
&lt;&lt;128,128&gt;&gt;
12> <input>io:fwrite("~tp~n", [&lt;&lt;208,128&gt;&gt;]).</input>
&lt;&lt;"\x{400}"/utf8&gt;&gt;
ok</pre>
          </item>
          <tag><c>W</c></tag>
          <item>
            <p>Writes data in the same way as <c>~w</c>, but takes an
              extra argument that is the maximum depth to which terms
              are printed. Anything below this depth is replaced with
              <c>...</c>. For example, using <c>T</c> above:</p>
            <pre>
8> <input>io:fwrite("~W~n", [T,9]).</input>
[{attributes,[[{id,age,1.5},{mode,explicit},{typename,...}],
[{id,cho},{mode,...},{...}]]},{typename,'Person'},
{tag,{'PRIVATE',3}},{mode,implicit}]
ok</pre>
            <p>If the maximum depth is reached, it cannot
              be read in the resultant output. Also, the
              <c>,...</c> form in a tuple denotes that there are more
              elements in the tuple but these are below the print depth.</p>
          </item>
          <tag><c>P</c></tag>
          <item>
            <p>Writes data in the same way as <c>~p</c>, but takes an
              extra argument that is the maximum depth to which terms
              are printed. Anything below this depth is replaced with
              <c>...</c>, for example:</p>
            <pre>
9> <input>io:fwrite("~62P~n", [T,9]).</input>
[{attributes,[[{id,age,1.5},{mode,explicit},{typename,...}],
              [{id,cho},{mode,...},{...}]]},
 {typename,'Person'},
 {tag,{'PRIVATE',3}},
 {mode,implicit}]
ok</pre>
          </item>
          <tag><c>B</c></tag>
          <item>
            <p>Writes an integer in base 2-36, the default base is
              10. A leading dash is printed for negative integers.</p>
            <p>The precision field selects base, for example:</p>
            <pre>
1> <input>io:fwrite("~.16B~n", [31]).</input>
1F
ok
2> <input>io:fwrite("~.2B~n", [-19]).</input>
-10011
ok
3> <input>io:fwrite("~.36B~n", [5*36+35]).</input>
5Z
ok</pre>
          </item>
          <tag><c>X</c></tag>
          <item>
            <p>Like <c>B</c>, but takes an extra argument that is a
              prefix to insert before the number, but after the leading
              dash, if any.</p>
            <p>The prefix can be a possibly deep list of characters or
              an atom. Example:</p>
            <pre>
1> <input>io:fwrite("~X~n", [31,"10#"]).</input>
10#31
ok
2> <input>io:fwrite("~.16X~n", [-31,"0x"]).</input>
-0x1F
ok</pre>
          </item>
          <tag><c>#</c></tag>
          <item>
            <p>Like <c>B</c>, but prints the number with an Erlang style
              <c>#</c>-separated base prefix. Example:</p>
            <pre>
1> <input>io:fwrite("~.10#~n", [31]).</input>
10#31
ok
2> <input>io:fwrite("~.16#~n", [-31]).</input>
-16#1F
ok</pre>
          </item>
          <tag><c>b</c></tag>
          <item>
            <p>Like <c>B</c>, but prints lowercase letters.</p>
          </item>
          <tag><c>x</c></tag>
          <item>
            <p>Like <c>X</c>, but prints lowercase letters.</p>
          </item>
          <tag><c>+</c></tag>
          <item>
            <p>Like <c>#</c>, but prints lowercase letters.</p>
          </item>
          <tag><c>n</c></tag>
          <item>
            <p>Writes a new line.</p>
          </item>
          <tag><c>i</c></tag>
          <item>
            <p>Ignores the next term.</p>
          </item>
        </taglist>
        <p>The function returns:</p>
        <taglist>
          <tag><c>ok</c></tag>
          <item>
            <p>The formatting succeeded.</p>
          </item>
        </taglist>
        <p>If an error occurs, there is no output. Example:</p>
        <pre>
1> <input>io:fwrite("~s ~w ~i ~w ~c ~n",['abc def', 'abc def', {foo, 1},{foo, 1}, 65]).</input>
abc def 'abc def'  {foo,1} A
ok
2> <input>io:fwrite("~s", [65]).</input>
** exception exit: {badarg,[{io,format,[&lt;0.22.0>,"~s","A"]},
                            {erl_eval,do_apply,5},
                            {shell,exprs,6},
                            {shell,eval_exprs,6},
                            {shell,eval_loop,3}]}
     in function  io:o_request/2</pre>
        <p>In this example, an attempt was made to output the single
          character 65 with the aid of the string formatting directive
          <c>"~s"</c>.</p>
      </desc>
    </func>

    <func>
      <name name="fread" arity="2"/>
      <name name="fread" arity="3"/>
      <fsummary>Read formatted input.</fsummary>
      <type name="server_no_data"/>
      <desc>
        <p>Reads characters from the standard input
          (<c><anno>IoDevice</anno></c>), prompting it with
          <c><anno>Prompt</anno></c>. Interprets the characters in accordance
          with <c><anno>Format</anno></c>. <c><anno>Format</anno></c> contains
          control sequences that directs the interpretation of the input.</p>
        <p><c><anno>Format</anno></c> can contain the following:</p>
        <list type="bulleted">
          <item>
            <p>Whitespace characters (<em>Space</em>, <em>Tab</em>, and
              <em>Newline</em>) that cause input to be read to the next
              non-whitespace character.</p>
          </item>
          <item>
            <p>Ordinary characters that must match the next input
              character.</p>
          </item>
          <item>
            <p>Control sequences, which have the general format
              <c>~*FMC</c>, where:</p>
              <list type="bulleted">
                <item>
                  <p>Character <c>*</c> is an optional return suppression
                    character. It provides a method to specify a field that
                    is to be omitted.</p>
                </item>
                <item>
                  <p><c>F</c> is the <c>field width</c> of the input field.</p>
                </item>
                <item>
                  <p><c>M</c> is an optional translation modifier (of which
                    <c>t</c> is the only supported, meaning Unicode
                    translation).</p>
                </item>
                <item>
                  <p><c>C</c> determines the type of control sequence.</p>
                </item>
              </list>
            <p>Unless otherwise specified, leading whitespace is
              ignored for all control sequences. An input field cannot
              be more than one line wide.</p>
            <p><em>Available control sequences:</em></p>
            <taglist>
              <tag><c>~</c></tag>
              <item>
                <p>A single <c>~</c> is expected in the input.</p>
              </item>
              <tag><c>d</c></tag>
              <item>
                <p>A decimal integer is expected.</p>
              </item>
              <tag><c>u</c></tag>
              <item>
                <p>An unsigned integer in base 2-36 is expected. The
                  field width parameter is used to specify base. Leading
                  whitespace characters are not skipped.</p>
              </item>
              <tag><c>-</c></tag>
              <item>
                <p>An optional sign character is expected. A sign
                  character <c>-</c> gives return value <c>-1</c>. Sign
                  character <c>+</c> or none gives <c>1</c>. The field width
                  parameter is ignored. Leading whitespace characters
                  are not skipped.</p>
              </item>
              <tag><c>#</c></tag>
              <item>
                <p>An integer in base 2-36 with Erlang-style base
                  prefix (for example, <c>"16#ffff"</c>) is expected.</p>
              </item>
              <tag><c>f</c></tag>
              <item>
                <p>A floating point number is expected. It must follow
                  the Erlang floating point number syntax.</p>
              </item>
              <tag><c>s</c></tag>
              <item>
                <p>A string of non-whitespace characters is read. If a
                  field width has been specified, this number of
                  characters are read and all trailing whitespace
                  characters are stripped. An Erlang string (list of
                  characters) is returned.</p>
                <p>If Unicode translation is in effect (<c>~ts</c>),
                  characters &gt; 255 are accepted, otherwise
                  not. With the translation modifier, the returned
                  list can as a consequence also contain integers &gt; 255:</p>
<pre>
1> <input>io:fread("Prompt> ","~s").</input>
Prompt> <input>&lt;Characters beyond latin1 range not printable in this medium&gt;</input>
{error,{fread,string}}
2> <input>io:fread("Prompt> ","~ts").</input>
Prompt> <input>&lt;Characters beyond latin1 range not printable in this medium&gt;</input>
{ok,[[1091,1085,1080,1094,1086,1076,1077]]}</pre>
              </item>
              <tag><c>a</c></tag>
              <item>
                <p>Similar to <c>s</c>, but the resulting string is
                  converted into an atom.</p>
                <p>The Unicode translation modifier is not allowed (atoms
                  cannot contain characters beyond the <c>latin1</c> range).</p>
              </item>
              <tag><c>c</c></tag>
              <item>
                <p>The number of characters equal to the field width are
                  read (default is 1) and returned as an Erlang string.
                  However, leading and trailing whitespace characters
                  are not omitted as they are with <c>s</c>. All
                  characters are returned.</p>
                <p>The Unicode translation modifier works as with <c>s</c>:
                </p>
<pre>
1> <input>io:fread("Prompt> ","~c").</input>
Prompt> <input>&lt;Character beyond latin1 range not printable in this medium&gt;</input>
{error,{fread,string}}
2> <input>io:fread("Prompt> ","~tc").</input>
Prompt> <input>&lt;Character beyond latin1 range not printable in this medium&gt;</input>
{ok,[[1091]]}</pre>
              </item>
              <tag><c>l</c></tag>
              <item>
                <p>Returns the number of characters that have been
                  scanned up to that point, including whitespace
                  characters.</p>
              </item>
            </taglist>
            <p>The function returns:</p>
            <taglist>
              <tag><c>{ok, <anno>Terms</anno>}</c></tag>
              <item>
                <p>The read was successful and <c><anno>Terms</anno></c> is
                  the list of successfully matched and read items.</p>
              </item>
              <tag><c>eof</c></tag>
              <item>
                <p>End of file was encountered.</p>
              </item>
              <tag><c>{error, <anno>FreadError</anno>}</c></tag>
              <item>
                <p>The reading failed and <c>FreadError</c> gives a
                  hint about the error.</p>
              </item>
              <tag><c>{error, <anno>ErrorDescription</anno>}</c></tag>
              <item>
                <p>The read operation failed and parameter
                  <c><anno>ErrorDescription</anno></c> gives a hint about
                  the error.</p>
              </item>
            </taglist>
          </item>
        </list>
        <p><em>Examples:</em></p>
        <pre>
20> <input>io:fread('enter>', "~f~f~f").</input>
enter><input>1.9 35.5e3 15.0</input>
{ok,[1.9,3.55e4,15.0]}
21> <input>io:fread('enter>', "~10f~d").</input>
enter>     <input>5.67899</input>
{ok,[5.678,99]}
22> <input>io:fread('enter>', ":~10s:~10c:").</input>
enter><input>:</input>   <input>alan</input>   <input>:</input>   <input>joe</input>    <input>:</input>
{ok, ["alan", "   joe    "]}</pre>
      </desc>
    </func>

    <func>
      <name name="get_chars" arity="2"/>
      <name name="get_chars" arity="3"/>
      <fsummary>Read a specified number of characters.</fsummary>
      <type name="server_no_data"/>
      <desc>
        <p>Reads <c><anno>Count</anno></c> characters from standard input
          (<c><anno>IoDevice</anno></c>), prompting it with
          <c><anno>Prompt</anno></c>.</p>
        <p>The function returns:</p>
        <taglist>
          <tag><c><anno>Data</anno></c></tag>
          <item>
            <p>The input characters. If the I/O device supports Unicode,
              the data can represent codepoints &gt; 255 (the
              <c>latin1</c> range). If the I/O server is set to deliver
              binaries, they are encoded in UTF-8 (regardless of whether
              the I/O device supports Unicode).</p>
          </item>
          <tag><c>eof</c></tag>
          <item>
            <p>End of file was encountered.</p>
          </item>
          <tag><c>{error, <anno>ErrorDescription</anno>}</c></tag>
          <item>
            <p>Other (rare) error condition, such as <c>{error, estale}</c>
             if reading from an NFS file system.</p>
	  </item>
        </taglist>
      </desc>
    </func>

    <func>
      <name name="get_line" arity="1"/>
      <name name="get_line" arity="2"/>
      <fsummary>Read a line.</fsummary>
      <type name="server_no_data"/>
      <desc>
        <p>Reads a line from the standard input (<c><anno>IoDevice</anno></c>),
          prompting it with <c><anno>Prompt</anno></c>.</p>
        <p>The function returns:</p>
        <taglist>
          <tag><c><anno>Data</anno></c></tag>
          <item>
            <p>The characters in the line terminated by a line feed (or end of
              file). If the I/O device supports Unicode,
              the data can represent codepoints &gt; 255 (the
              <c>latin1</c> range). If the I/O server is set to deliver
              binaries, they are encoded in UTF-8 (regardless of if
              the I/O device supports Unicode).</p>
          </item>
          <tag><c>eof</c></tag>
          <item>
            <p>End of file was encountered.</p>
          </item>
          <tag><c>{error, <anno>ErrorDescription</anno>}</c></tag>
          <item>
            <p>Other (rare) error condition, such as <c>{error, estale}</c>
              if reading from an NFS file system.</p>
	  </item>
        </taglist>
      </desc>
    </func>

    <func>
      <name name="getopts" arity="0"/>
      <name name="getopts" arity="1"/>
      <fsummary>Get the supported options and values from an I/O server.
      </fsummary>
      <desc>
      <p>Requests all available options and their current
        values for a specific I/O device, for example:</p>
<pre>
1> <input>{ok,F} = file:open("/dev/null",[read]).</input>
{ok,&lt;0.42.0&gt;}
2> <input>io:getopts(F).</input>
[{binary,false},{encoding,latin1}]</pre>
      <p>Here the file I/O server returns all available options for a file,
        which are the expected ones, <c>encoding</c> and <c>binary</c>.
        However, the standard shell has some more options:</p>
<pre>
3> io:getopts().
[{expand_fun,#Fun&lt;group.0.120017273&gt;},
 {echo,true},
 {binary,false},
 {encoding,unicode}]</pre>
      <p>This example is, as can be seen, run in an environment where the
        terminal supports Unicode input and output.</p>
      </desc>
    </func>

    <func>
      <name name="nl" arity="0"/>
      <name name="nl" arity="1"/>
      <fsummary>Write a newline.</fsummary>
      <desc>
        <p>Writes new line to the standard output
          (<c><anno>IoDevice</anno></c>).</p>
      </desc>
    </func>

    <func>
      <name name="parse_erl_exprs" arity="1"/>
      <name name="parse_erl_exprs" arity="2"/>
      <name name="parse_erl_exprs" arity="3"/>
      <name name="parse_erl_exprs" arity="4"/>
      <fsummary>Read, tokenize, and parse Erlang expressions.</fsummary>
      <type name="parse_ret"/>
      <type name="server_no_data"/>
      <desc>
        <p>Reads data from the standard input
          (<c><anno>IoDevice</anno></c>), prompting it with
          <c><anno>Prompt</anno></c>. Starts reading at location
          <c><anno>StartLocation</anno></c> (<c>1</c>). Argument
          <c><anno>Options</anno></c> is passed on as argument
          <c>Options</c> of function <seealso marker="erl_scan#tokens/4">
          <c>erl_scan:tokens/4</c></seealso>. The data is tokenized and parsed
          as if it was a sequence of Erlang expressions until a final dot
          (<c>.</c>) is reached.</p>
        <p>The function returns:</p>
        <taglist>
          <tag><c>{ok, ExprList, EndLocation}</c></tag>
          <item>
            <p>The parsing was successful.</p>
          </item>
          <tag><c>{eof, EndLocation}</c></tag>
          <item>
            <p>End of file was encountered by the tokenizer.</p>
          </item>
          <tag><c>eof</c></tag>
          <item>
            <p>End of file was encountered by the I/O server.</p>
          </item>
          <tag><c>{error, ErrorInfo, ErrorLocation}</c></tag>
          <item>
            <p>An error occurred while tokenizing or parsing.</p>
          </item>
          <tag><c>{error, <anno>ErrorDescription</anno>}</c></tag>
          <item>
            <p>Other (rare) error condition, such as <c>{error, estale}</c>
              if reading from an NFS file system.</p>
          </item>
        </taglist>
        <p>Example:</p>
        <pre>
25> <input>io:parse_erl_exprs('enter>').</input>
enter><input>abc(), "hey".</input>
{ok, [{call,1,{atom,1,abc},[]},{string,1,"hey"}],2}
26> <input>io:parse_erl_exprs ('enter>').</input>
enter><input>abc("hey".</input>
{error,{1,erl_parse,["syntax error before: ",["'.'"]]},2}</pre>
      </desc>
    </func>

    <func>
      <name name="parse_erl_form" arity="1"/>
      <name name="parse_erl_form" arity="2"/>
      <name name="parse_erl_form" arity="3"/>
      <name name="parse_erl_form" arity="4"/>
      <fsummary>Read, tokenize, and parse an Erlang form.</fsummary>
      <type name="parse_form_ret"/>
      <type name="server_no_data"/>
      <desc>
        <p>Reads data from the standard input (<c><anno>IoDevice</anno></c>),
          prompting it with <c><anno>Prompt</anno></c>. Starts reading at
          location <c><anno>StartLocation</anno></c> (<c>1</c>). Argument
          <c><anno>Options</anno></c> is passed on as argument
          <c>Options</c> of function <seealso marker="erl_scan#tokens/4">
          <c>erl_scan:tokens/4</c></seealso>. The data is tokenized and parsed
          as if it was an Erlang form (one of the valid Erlang expressions
          in an Erlang source file) until a final dot (<c>.</c>) is reached.</p>
        <p>The function returns:</p>
        <taglist>
          <tag><c>{ok, AbsForm, EndLocation}</c></tag>
          <item>
            <p>The parsing was successful.</p>
          </item>
          <tag><c>{eof, EndLocation}</c></tag>
          <item>
            <p>End of file was encountered by the tokenizer.</p>
          </item>
          <tag><c>eof</c></tag>
          <item>
            <p>End of file was encountered by the I/O server.</p>
          </item>
          <tag><c>{error, ErrorInfo, ErrorLocation}</c></tag>
          <item>
            <p>An error occurred while tokenizing or parsing.</p>
          </item>
          <tag><c>{error, <anno>ErrorDescription</anno>}</c></tag>
          <item>
            <p>Other (rare) error condition, such as <c>{error, estale}</c>
              if reading from an NFS file system.</p>
          </item>
        </taglist>
      </desc>
    </func>

    <func>
      <name name="printable_range" arity="0"/>
      <fsummary>Get user-requested printable character range.</fsummary>
      <desc>
        <p>Returns the user-requested range of printable Unicode characters.</p>
        <p>The user can request a range of characters that are to be considered
          printable in heuristic detection of strings by the shell and by the
          formatting functions. This is done by supplying
          <c>+pc &lt;range&gt;</c> when starting Erlang.</p>
        <p>The only valid values for <c>&lt;range&gt;</c> are
          <c>latin1</c> and <c>unicode</c>. <c>latin1</c> means that only code
          points &lt; 256 (except control characters, and so on)
          are considered printable. <c>unicode</c> means that all printable
          characters in all Unicode character ranges are considered printable
          by the I/O functions.</p>
        <p>By default, Erlang is started so that only the <c>latin1</c> range
          of characters indicate that a list of integers is a string.</p>
        <p>The simplest way to use the setting is to call
          <seealso marker="io_lib#printable_list/1">
          <c>io_lib:printable_list/1</c></seealso>, which uses the return
          value of this function to decide if a list is a string of printable
          characters.</p>
        <note>
          <p>In a future release, this function may return more values and
            ranges. To avoid compatibility problems, it is recommended to use
            function <seealso marker="io_lib#printable_list/1">
            <c>io_lib:printable_list/1</c></seealso>.</p></note>
      </desc>
    </func>

    <func>
      <name name="put_chars" arity="1"/>
      <name name="put_chars" arity="2"/>
      <fsummary>Write a list of characters.</fsummary>
      <desc>
        <p>Writes the characters of <c><anno>CharData</anno></c> to the I/O
          server (<c><anno>IoDevice</anno></c>).</p>
      </desc>
    </func>

    <func>
      <name name="read" arity="1"/>
      <name name="read" arity="2"/>
      <fsummary>Read a term.</fsummary>
      <type name="server_no_data"/>
      <desc>
        <p>Reads a term <c><anno>Term</anno></c> from the standard input
          (<c><anno>IoDevice</anno></c>), prompting it with
          <c><anno>Prompt</anno></c>.</p>
        <p>The function returns:</p>
        <taglist>
          <tag><c>{ok, <anno>Term</anno>}</c></tag>
          <item>
            <p>The parsing was successful.</p>
          </item>
          <tag><c>eof</c></tag>
          <item>
            <p>End of file was encountered.</p>
          </item>
          <tag><c>{error, <anno>ErrorInfo</anno>}</c></tag>
          <item>
            <p>The parsing failed.</p>
          </item>
          <tag><c>{error, <anno>ErrorDescription</anno>}</c></tag>
          <item>
            <p>Other (rare) error condition, such as <c>{error, estale}</c>
              if reading from an NFS file system.</p>
          </item>
        </taglist>
      </desc>
    </func>

    <func>
      <name name="read" arity="3"/>
      <name name="read" arity="4"/>
      <fsummary>Read a term.</fsummary>
      <type name="server_no_data"/>
      <desc>
        <p>Reads a term <c><anno>Term</anno></c> from
          <c><anno>IoDevice</anno></c>, prompting it
          with <c><anno>Prompt</anno></c>. Reading starts at location
          <c><anno>StartLocation</anno></c>. Argument
          <c><anno>Options</anno></c> is passed on as argument <c>Options</c>
          of function <seealso marker="erl_scan#tokens/4">
          <c>erl_scan:tokens/4</c></seealso>.</p>
        <p>The function returns:</p>
        <taglist>
          <tag><c>{ok, Term, <anno>EndLocation</anno>}</c></tag>
          <item>
            <p>The parsing was successful.</p>
          </item>
          <tag><c>{eof, <anno>EndLocation</anno>}</c></tag>
          <item>
            <p>End of file was encountered.</p>
          </item>
          <tag><c>{error, <anno>ErrorInfo</anno>,
            <anno>ErrorLocation</anno>}</c></tag>
          <item>
            <p>The parsing failed.</p>
          </item>
          <tag><c>{error, <anno>ErrorDescription</anno>}</c></tag>
          <item>
            <p>Other (rare) error condition, such as <c>{error, estale}</c>
              if reading from an NFS file system.</p>
          </item>
        </taglist>
      </desc>
    </func>

    <func>
      <name name="rows" arity="0"/>
      <name name="rows" arity="1"/>
      <fsummary>Get the number of rows of an I/O device.</fsummary>
      <desc>
          <p>Retrieves the number of rows of <c><anno>IoDevice</anno></c>
          (that is, the height of a terminal). The function
          only succeeds for terminal devices, for all other I/O devices
          the function returns <c>{error, enotsup}</c>.</p>
      </desc>
    </func>

    <func>
      <name name="scan_erl_exprs" arity="1"/>
      <name name="scan_erl_exprs" arity="2"/>
      <name name="scan_erl_exprs" arity="3"/>
      <name name="scan_erl_exprs" arity="4"/>
      <fsummary>Read and tokenize Erlang expressions.</fsummary>
      <type name="server_no_data"/>
      <desc>
        <p>Reads data from the standard input (<c>IoDevice</c>),
          prompting it with <c>Prompt</c>. Reading starts at location
          <c>StartLocation</c> (<c>1</c>). Argument <c><anno>Options</anno></c>
          is passed on as argument <c>Options</c> of function
          <seealso marker="erl_scan#tokens/4">
          <c>erl_scan:tokens/4</c></seealso>. The data is tokenized as if it
          were a sequence of Erlang expressions until a final dot (<c>.</c>) is
          reached. This token is also returned.</p>
        <p>The function returns:</p>
        <taglist>
          <tag><c>{ok, Tokens, EndLocation}</c></tag>
          <item>
            <p>The tokenization succeeded.</p>
          </item>
          <tag><c>{eof, EndLocation}</c></tag>
          <item>
            <p>End of file was encountered by the tokenizer.</p>
          </item>
          <tag><c>eof</c></tag>
          <item>
            <p>End of file was encountered by the I/O server.</p>
          </item>
          <tag><c>{error, ErrorInfo, ErrorLocation}</c></tag>
          <item>
            <p>An error occurred while tokenizing.</p>
          </item>
          <tag><c>{error, <anno>ErrorDescription</anno>}</c></tag>
          <item>
            <p>Other (rare) error condition, such as <c>{error, estale}</c>
              if reading from an NFS file system.</p>
          </item>
        </taglist>
        <p><em>Example:</em></p>
        <pre>
23> <input>io:scan_erl_exprs('enter>').</input>
enter><input>abc(), "hey".</input>
{ok,[{atom,1,abc},{'(',1},{')',1},{',',1},{string,1,"hey"},{dot,1}],2}
24> <input>io:scan_erl_exprs('enter>').</input>
enter><input>1.0er.</input>
{error,{1,erl_scan,{illegal,float}},2}</pre>
      </desc>
    </func>

    <func>
      <name name="scan_erl_form" arity="1"/>
      <name name="scan_erl_form" arity="2"/>
      <name name="scan_erl_form" arity="3"/>
      <name name="scan_erl_form" arity="4"/>
      <fsummary>Read and tokenize an Erlang form.</fsummary>
      <type name="server_no_data"/>
      <desc>
        <p>Reads data from the standard input (<c><anno>IoDevice</anno></c>),
          prompting it with <c><anno>Prompt</anno></c>. Starts reading
          at location <c><anno>StartLocation</anno></c> (<c>1</c>).
          Argument <c><anno>Options</anno></c> is passed on as argument
          <c>Options</c> of function <seealso marker="erl_scan#tokens/4">
          <c>erl_scan:tokens/4</c></seealso>. The data is tokenized as if it
          was an Erlang form (one of the valid Erlang expressions in an
          Erlang source file) until a final dot (<c>.</c>) is reached.
          This last token is also returned.</p>
        <p>The return values are the same as for
          <seealso marker="#scan_erl_exprs/1">
          <c>scan_erl_exprs/1,2,3,4</c></seealso>.</p>
      </desc>
    </func>

    <func>
      <name name="setopts" arity="1"/>
      <name name="setopts" arity="2"/>
      <fsummary>Set options.</fsummary>
      <desc>
        <p>Set options for the standard I/O device
          (<c><anno>IoDevice</anno></c>).</p>
        <p>Possible options and values vary depending on the
          I/O device. For a list of supported options and their current values
          on a specific I/O device, use function
          <seealso marker="#getopts/1"><c>getopts/1</c></seealso>.</p>
        <p>The options and values supported by the OTP I/O devices
          are as follows:</p>
        <taglist>
          <tag><c>binary</c>, <c>list</c>, or <c>{binary, boolean()}</c></tag>
          <item>
            <p>If set in binary mode (<c>binary</c> or <c>{binary, true}</c>),
              the I/O server sends binary data (encoded in UTF-8) as answers
              to the <c>get_line</c>, <c>get_chars</c>, and, if possible,
              <c>get_until</c> requests (for details, see section
              <seealso marker="io_protocol">The Erlang I/O Protocol</seealso>)
              in the User's Guide). The immediate effect is that
              <seealso marker="#get_chars/2"><c>get_chars/2,3</c></seealso> and
              <seealso marker="#get_line/1"><c>get_line/1,2</c></seealso>
              return UTF-8 binaries instead of lists of characters
              for the affected I/O device.</p>
            <p>By default, all I/O devices in OTP are set in <c>list</c> mode.
              However, the I/O functions can handle any of these modes and so
              should other, user-written, modules behaving as clients to I/O
              servers.</p>
            <p>This option is supported by the standard shell
              (<c>group.erl</c>), the 'oldshell' (<c>user.erl</c>), and the
              file I/O servers.</p>
          </item>
          <tag><c>{echo, boolean()}</c></tag>
          <item>
            <p>Denotes if the terminal is to echo input. Only supported for
              the standard shell I/O server (<c>group.erl</c>)</p>
          </item>
          <tag><c>{expand_fun, expand_fun()}</c></tag>
          <item>
            <p>Provides a function for tab-completion (expansion)
              like the Erlang shell. This function is called
              when the user presses the <em>Tab</em> key. The expansion is
              active when calling line-reading functions, such as
              <seealso marker="#get_line/1"><c>get_line/1,2</c></seealso>.</p>
            <p>The function is called with the current line, up to
              the cursor, as a reversed string. It is to return a
              three-tuple: <c>{yes|no, string(), [string(), ...]}</c>. The
              first element gives a beep if <c>no</c>, otherwise the
              expansion is silent; the second is a string that will be
              entered at the cursor position; the third is a list of
              possible expansions. If this list is not empty,
              it is printed and the current input line is written
              once again.</p>
            <p>Trivial example (beep on anything except empty line, which
              is expanded to <c>"quit"</c>):</p>
            <code type="none">
fun("") -> {yes, "quit", []};
   (_) -> {no, "", ["quit"]} end</code>
            <p>This option is only supported by the standard shell
              (<c>group.erl</c>).</p>
          </item>
          <tag><c>{encoding, latin1 | unicode}</c></tag>
          <item>
            <p>Specifies how characters are input or output from or to the I/O
              device, implying that, for example, a terminal is set to handle
              Unicode input and output or a file is set to handle UTF-8 data
              encoding.</p>
            <p>The option <em>does not</em> affect how data is returned from the
              I/O functions or how it is sent in the I/O protocol, it only
              affects how the I/O device is to handle Unicode characters to the
              &quot;physical&quot; device.</p>
            <p>The standard shell is set for <c>unicode</c> or <c>latin1</c>
              encoding when
              the system is started. The encoding is set with the help of the
              <c>LANG</c> or <c>LC_CTYPE</c> environment variables on Unix-like
              system or by other means on other systems.
              So, the user can input Unicode characters and the I/O device
              is in <c>{encoding, unicode}</c> mode if the I/O device supports
              it. The mode can be changed, if the assumption of the runtime
              system is wrong, by setting this option.</p>
            <p>The I/O device used when Erlang is started with the "-oldshell"
              or "-noshell" flags is by default set to <c>latin1</c> encoding,
              meaning that any characters &gt; codepoint 255 are escaped
              and that input is expected to be plain 8-bit ISO Latin-1.
              If the encoding is changed to Unicode, input and output from
              the standard file descriptors are in UTF-8 (regardless of
              operating system).</p>
            <p>Files can also be set in <c>{encoding, unicode}</c>, meaning
              that data is written and read as UTF-8. More encodings are
              possible for files, see below.</p>
            <p><c>{encoding, unicode | latin1}</c> is supported by both the
              standard shell (<c>group.erl</c> including <c>werl</c> on
              Windows), the 'oldshell' (<c>user.erl</c>), and the file I/O
              servers.</p>
          </item>
          <tag><c>{encoding, utf8 | utf16 | utf32 | {utf16,big} |
            {utf16,little} | {utf32,big} | {utf32,little}}</c></tag>
          <item>
            <p>For disk files, the encoding can be set to various UTF variants.
              This has the effect that data is expected to be read as the
              specified encoding from the file, and the data is written in the
              specified encoding to the disk file.</p>
            <p><c>{encoding, utf8}</c> has the same effect as
              <c>{encoding, unicode}</c> on files.</p>
            <p>The extended encodings are only supported on disk files
              (opened by function
              <seealso marker="kernel:file#open/2">
              <c>file:open/2</c></seealso>).</p>
          </item>
        </taglist>
      </desc>
    </func>

    <func>
      <name name="write" arity="1"/>
      <name name="write" arity="2"/>
      <fsummary>Write a term.</fsummary>
      <desc>
        <p>Writes term <c><anno>Term</anno></c> to the standard output
          (<c><anno>IoDevice</anno></c>).</p>
      </desc>
    </func>
  </funcs>

  <section>
    <title>Standard Input/Output</title>
    <p>All Erlang processes have a default standard I/O device. This
      device is used when no <c>IoDevice</c> argument is specified in
      the function calls in this module. However, it is sometimes desirable to
      use an explicit <c>IoDevice</c> argument that refers to the
      default I/O device. This is the case with functions that can
      access either a file or the default I/O device. The atom
      <c>standard_io</c> has this special meaning. The following example
      illustrates this:</p>

    <pre>
27> <input>io:read('enter>').</input>
enter><input>foo.</input>
{ok,foo}
28> <input>io:read(standard_io, 'enter>').</input>
enter><input>bar.</input>
{ok,bar}</pre>

    <p>There is always a process registered under the name of
      <c>user</c>. This can be used for sending output to the user.</p>
  </section>

 <section>
    <title>Standard Error</title>
    <p>In certain situations, especially when the standard output is
      redirected, access to an I/O server specific for error messages can be
      convenient. The I/O device <c>standard_error</c> can be used to direct
      output to whatever the current operating system considers a suitable
      I/O device for error output. Example on a Unix-like operating system:</p>

<pre>
$ <input>erl -noshell -noinput -eval 'io:format(standard_error,"Error: ~s~n",["error 11"]),'\</input>
<input>'init:stop().' > /dev/null</input>
Error: error 11</pre>
  </section>

  <section>
    <title>Error Information</title>
    <p>The <c>ErrorInfo</c> mentioned in this module is the standard
      <c>ErrorInfo</c> structure that is returned from all I/O modules.
      It has the following format:</p>

    <code type="none">
{ErrorLocation, Module, ErrorDescriptor}</code>

    <p>A string that describes the error is obtained with the following
      call:</p>

    <code type="none">
Module:format_error(ErrorDescriptor)</code>
  </section>
</erlref>