path: root/lib/kernel/doc/src/os.xml

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

<erlref>
  <header>
    <copyright>
      <year>1997</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>os</title>
    <prepared></prepared>
    <docno></docno>
    <date></date>
    <rev></rev>
  </header>
  <module>os</module>
  <modulesummary>Operating system-specific functions.</modulesummary>
  <description>
    <p>The functions in this module are operating system-specific.
      Careless use of these functions results in programs that will
      only run on a specific platform. On the other hand, with careful
      use, these functions can be of help in enabling a program to run on
      most platforms.</p>

    <note>
      <p>
	File operations used to accept filenames containing
	null characters (integer value zero). This caused
	the name to be truncated and in some cases arguments
	to primitive operations to be mixed up. Filenames
	containing null characters inside the filename
	are now <em>rejected</em> and will cause primitive
	file operations to fail.
      </p>
      <p>
	Also environment variable operations used to accept
	names and values of environment variables containing
	null characters (integer value zero). This caused
	operations to silently produce erroneous results.
	Environment variable names and values containing
	null characters inside the name or value are now
	<em>rejected</em> and will cause environment variable
	operations to fail.
      </p>
    </note>
  </description>

  <datatypes>
    <datatype>
      <name name="env_var_name"/>
      <desc>
        <p>A string containing valid characters on the specific
	OS for environment variable names using
	<seealso marker="file#native_name_encoding/0"><c>file:native_name_encoding()</c></seealso>
	encoding. Note that specifically null characters (integer
	value zero) and <c>$=</c> characters are not allowed.
	However, note that not all invalid characters necessarily
	will cause the primitiv operations to fail, but may instead
	produce invalid results.
	</p>
      </desc>
    </datatype>
    <datatype>
      <name name="env_var_value"/>
      <desc>
        <p>A string containing valid characters on the specific
	OS for environment variable values using
	<seealso marker="file#native_name_encoding/0"><c>file:native_name_encoding()</c></seealso>
	encoding. Note that specifically null characters (integer
	value zero) are not allowed. However, note that not all
	invalid characters necessarily will cause the primitiv
	operations to fail, but may instead produce invalid results.
	</p>
      </desc>
    </datatype>
    <datatype>
      <name name="env_var_name_value"/>
      <desc>
        <p>
	  Assuming that environment variables has been correctly
	  set, a strings containing valid characters on the specific
	  OS for environment variable names and values using
	  <seealso marker="file#native_name_encoding/0"><c>file:native_name_encoding()</c></seealso>
	  encoding. The first <c>$=</c> characters appearing in
	  the string separates environment variable name (on the
	  left) from environment variable value (on the right).
	</p>
      </desc>
    </datatype>
    <datatype>
      <name name="os_command"/>
      <desc>
        <p>All characters needs to be valid characters on the
	specific OS using
	<seealso marker="file#native_name_encoding/0"><c>file:native_name_encoding()</c></seealso>
	encoding. Note that specifically null characters (integer
	value zero) are not allowed. However, note that not all
	invalid characters not necessarily will cause
	<seealso marker="#cmd/1"><c>os:cmd/1</c></seealso>
	to fail, but may instead produce invalid results.
	</p>
      </desc>
    </datatype>
    <datatype>
      <name name="os_command_opts"/>
      <desc>
        <p>Options for <seealso marker="#cmd/2"><c>os:cmd/2</c></seealso></p>
        <taglist>
          <tag><c>max_size</c></tag>
          <item>
            <p>The maximum size of the data returned by the <c>os:cmd</c> call.
            See the <seealso marker="#cmd/2"><c>os:cmd/2</c></seealso>
            documentation for more details.</p>
          </item>
        </taglist>
      </desc>
    </datatype>
  </datatypes>

  <funcs>
    <func>
      <name name="cmd" arity="1"/>
      <name name="cmd" arity="2"/>
      <fsummary>Execute a command in a shell of the target OS.</fsummary>
      <desc>
        <p>Executes <c><anno>Command</anno></c> in a command shell of the
	  target OS, captures the standard output of the command,
          and returns this result as a string.</p>
	  <warning><p>Previous implementation used to allow all characters
	  as long as they were integer values greater than or equal to zero.
	  This sometimes lead to unwanted results since null characters
	  (integer value zero) often are interpreted as string termination. The
	  current implementation rejects these.</p></warning>
        <p><em>Examples:</em></p>
        <code type="none">
LsOut = os:cmd("ls"), % on unix platform
DirOut = os:cmd("dir"), % on Win32 platform</code>
        <p>Notice that in some cases, standard output of a command when
          called from another program (for example, <c>os:cmd/1</c>)
          can differ, compared with the standard output of the command
          when called directly from an OS command shell.</p>
        <p><c>os:cmd/2</c> was added in kernel-5.5 (OTP-20.2.1). It makes it
          possible to pass an options map as the second argument in order to
          control the behaviour of <c>os:cmd</c>. The possible options are:
        </p>
        <taglist>
          <tag><c>max_size</c></tag>
          <item>
            <p>The maximum size of the data returned by the <c>os:cmd</c> call.
              This option is a safety feature that should be used when the command
              executed can return a very large, possibly infinite, result.</p>
            <code type="none">
> os:cmd("cat /dev/zero", #{ max_size => 20 }).
[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]</code>
          </item>
        </taglist>
      </desc>
    </func>

    <func>
      <name name="find_executable" arity="1"/>
      <name name="find_executable" arity="2"/>
      <fsummary>Absolute filename of a program.</fsummary>
      <desc>
        <p>These two functions look up an executable program, with the
          specified name and a search path, in the same way as the underlying
          OS. <c>find_executable/1</c> uses the current
          execution path (that is, the environment variable <c>PATH</c> on
          Unix and Windows).</p>
        <p><c><anno>Path</anno></c>, if specified, is to conform to the syntax
          of execution paths on the OS. Returns the absolute filename of the
          executable program <c><anno>Name</anno></c>,
          or <c>false</c> if the program is not found.</p>
      </desc>
    </func>

    <func>
      <name name="getenv" arity="0"/>
      <fsummary>List all environment variables.</fsummary>
      <desc>
        <p>Returns a list of all environment variables.
          Each environment variable is expressed as a single string on
          the format <c>"VarName=Value"</c>, where <c>VarName</c> is
          the name of the variable and <c>Value</c> its value.</p>
	<p>If Unicode filename encoding is in effect (see the
          <seealso marker="erts:erl#file_name_encoding"><c>erl</c> manual
	  page</seealso>), the strings can contain characters with
	  codepoints &gt; 255.</p>
      </desc>
    </func>

    <func>
      <name name="getenv" arity="1"/>
      <fsummary>Get the value of an environment variable.</fsummary>
      <desc>
        <p>Returns the <c><anno>Value</anno></c> of the environment variable
          <c><anno>VarName</anno></c>, or <c>false</c> if the environment
          variable is undefined.</p>
	<p>If Unicode filename encoding is in effect (see the
          <seealso marker="erts:erl#file_name_encoding"><c>erl</c> manual
	  page</seealso>), the strings <c><anno>VarName</anno></c> and
	  <c><anno>Value</anno></c> can contain characters with
          codepoints &gt; 255.</p>
      </desc>
    </func>

    <func>
      <name name="getenv" arity="2"/>
      <fsummary>Get the value of an environment variable.</fsummary>
      <desc>
        <p>Returns the <c><anno>Value</anno></c> of the environment variable
          <c><anno>VarName</anno></c>, or <c>DefaultValue</c> if the
          environment variable is undefined.</p>
	<p>If Unicode filename encoding is in effect (see the
          <seealso marker="erts:erl#file_name_encoding"><c>erl</c> manual
	page</seealso>), the strings <c><anno>VarName</anno></c> and
	<c><anno>Value</anno></c> can contain characters with
        codepoints &gt; 255.</p>
      </desc>
    </func>

    <func>
      <name name="getpid" arity="0"/>
      <fsummary>Return the process identifier of the emulator
        process.</fsummary>
      <desc>
        <p>Returns the process identifier of the current Erlang emulator
          in the format most commonly used by the OS environment.
          Returns <c><anno>Value</anno></c> as a string containing
          the (usually) numerical identifier for a process. On Unix,
          this is typically the return value of the <c>getpid()</c>
          system call. On Windows,
          the process id as returned by the <c>GetCurrentProcessId()</c>
          system call is used.</p>
      </desc>
    </func>

    <func>
      <name name="putenv" arity="2"/>
      <fsummary>Set a new value for an environment variable.</fsummary>
      <desc>
        <p>Sets a new <c><anno>Value</anno></c> for environment variable
          <c><anno>VarName</anno></c>.</p>
	<p>If Unicode filename encoding is in effect (see the
          <seealso marker="erts:erl#file_name_encoding"><c>erl</c> manual
	  page</seealso>), the strings <c><anno>VarName</anno></c> and
	  <c><anno>Value</anno></c> can contain characters with
          codepoints &gt; 255.</p>
	<p>On Unix platforms, the environment is set using UTF-8 encoding
	  if Unicode filename translation is in effect. On Windows, the
	  environment is set using wide character interfaces.</p>
	  <note>
	    <p>
	      <c><anno>VarName</anno></c> is not allowed to contain
	      an <c>$=</c> character. Previous implementations used
	      to just let the <c>$=</c> character through which
	      silently caused erroneous results. Current implementation
	      will instead throw a <c>badarg</c> exception.
	    </p>
	  </note>
      </desc>
    </func>

    <func>
      <name name="set_signal" arity="2"/>
      <fsummary>Enables or disables handling of OS signals.</fsummary>
      <desc>
          <p>Enables or disables OS signals.</p>
          <p>Each signal my be set to one of the following options:</p>
          <taglist>
              <tag><c>ignore</c></tag>
              <item>
                  This signal will be ignored.
              </item>

              <tag><c>default</c></tag>
              <item>
                  This signal will use the default signal handler for the operating system.
              </item>

              <tag><c>handle</c></tag>
              <item>
                  This signal will notify
                  <seealso marker="kernel_app#erl_signal_server"><c>erl_signal_server</c></seealso>
                  when it is received by the Erlang runtime system.
              </item>
          </taglist>
      </desc>
    </func>

    <func>
      <name name="system_time" arity="0"/>
      <fsummary>Current OS system time.</fsummary>
      <desc>
	<p>Returns the current
	<seealso marker="erts:time_correction#OS_System_Time">OS system time</seealso>
	in <c>native</c>
	<seealso marker="erts:erlang#type_time_unit">time unit</seealso>.</p>
	<note><p>This time is <em>not</em> a monotonically increasing time.</p>
        </note>
      </desc>
    </func>

    <func>
      <name name="system_time" arity="1"/>
      <fsummary>Current OS system time.</fsummary>
      <desc>
	<p>Returns the current
	<seealso marker="erts:time_correction#OS_System_Time">OS system time</seealso>
	converted into the <c><anno>Unit</anno></c> passed as argument.</p>
	<p>Calling <c>os:system_time(<anno>Unit</anno>)</c> is equivalent to
	<seealso marker="erts:erlang#convert_time_unit/3"><c>erlang:convert_time_unit</c></seealso>(<seealso marker="#system_time/0"><c>os:system_time()</c></seealso><c>,
	native, <anno>Unit</anno>)</c>.</p>
	<note><p>This time is <em>not</em> a monotonically increasing time.</p>
        </note>
      </desc>
    </func>

    <func>
      <name name="timestamp" arity="0"/>
      <fsummary>Current OS system time on the <c>erlang:timestamp/0</c> format.</fsummary>
      <type_desc variable="Timestamp">Timestamp = {MegaSecs, Secs, MicroSecs}</type_desc>
      <desc>
      <p>Returns the current
	<seealso marker="erts:time_correction#OS_System_Time">OS system time</seealso>
        in the same format as
        <seealso marker="erts:erlang#timestamp/0"><c>erlang:timestamp/0</c></seealso>.
	The tuple can be used together with function
	<seealso marker="stdlib:calendar#now_to_universal_time/1"><c>calendar:now_to_universal_time/1</c></seealso>
        or <seealso marker="stdlib:calendar#now_to_local_time/1"><c>calendar:now_to_local_time/1</c></seealso>
        to get calendar time. Using the calendar time, together with the
        <c>MicroSecs</c> part of the return tuple from this function, allows
        you to log time stamps in high resolution and consistent with the
        time in the rest of the OS.</p>
      <p>Example of code formatting a string in format
        &quot;DD Mon YYYY HH:MM:SS.mmmmmm&quot;, where DD is the day of month,
        Mon is the textual month name, YYYY is the year, HH:MM:SS is the time,
        and mmmmmm is the microseconds in six positions:</p>
      <code>
-module(print_time).
-export([format_utc_timestamp/0]).
format_utc_timestamp() ->
    TS = {_,_,Micro} = os:timestamp(),
    {{Year,Month,Day},{Hour,Minute,Second}} =
calendar:now_to_universal_time(TS),
    Mstr = element(Month,{"Jan","Feb","Mar","Apr","May","Jun","Jul",
    "Aug","Sep","Oct","Nov","Dec"}),
    io_lib:format("~2w ~s ~4w ~2w:~2..0w:~2..0w.~6..0w",
    [Day,Mstr,Year,Hour,Minute,Second,Micro]).</code>
       <p>This module can be used as follows:</p>
       <pre>
1> <input>io:format("~s~n",[print_time:format_utc_timestamp()]).</input>
29 Apr 2009  9:55:30.051711</pre>
       <p>OS system time can also be retreived by
	<seealso marker="#system_time/0"><c>system_time/0</c></seealso> and
	<seealso marker="#system_time/1"><c>system_time/1</c></seealso>.</p>
      </desc>
    </func>

    <func>
      <name name="perf_counter" arity="0"/>
      <fsummary>Returns a performance counter</fsummary>
      <desc>
        <p>Returns the current performance counter value in <c>perf_counter</c>
	<seealso marker="erts:erlang#type_time_unit">time unit</seealso>.
        This is a highly optimized call that might not be traceable.
        </p>
      </desc>
    </func>
    <func>
      <name name="perf_counter" arity="1"/>
      <fsummary>Returns a performance counter</fsummary>
      <desc><p>Returns a performance counter that can be used as a very fast and
      high resolution timestamp. This counter is read directly from the hardware or operating
      system with the same guarantees. This means that two consecutive calls
      to the function are not guaranteed to be monotonic, though it most likely will be.
      The performance counter will be converted to the resolution passed as an argument.</p>
      <pre>1> <input>T1 = os:perf_counter(1000),receive after 10000 -> ok end,T2 = os:perf_counter(1000).</input>
176525861
2> <input>T2 - T1.</input>
10004</pre>
      </desc>
    </func>
    <func>
      <name name="type" arity="0"/>
      <fsummary>Return the OS family and, in some cases, the OS name of the
        current OS.</fsummary>
      <desc>
        <p>Returns the <c><anno>Osfamily</anno></c> and, in some cases, the
          <c><anno>Osname</anno></c> of the current OS.</p>
        <p>On Unix, <c><anno>Osname</anno></c> has the same value as
          <c>uname -s</c> returns, but in lower case. For example, on
          Solaris 1 and 2, it is <c>sunos</c>.</p>
        <p>On Windows, <c><anno>Osname</anno></c> is <c>nt</c>.</p>
        <note>
          <p>Think twice before using this function. Use module
            <seealso marker="stdlib:filename"><c>filename</c></seealso>
            if you want to inspect or build filenames in a portable way.
            Avoid matching on atom <c><anno>Osname</anno></c>.</p>
        </note>
      </desc>
    </func>

    <func>
      <name name="unsetenv" arity="1"/>
      <fsummary>Delete an environment variable.</fsummary>
      <desc>
	<p>Deletes the environment variable <c><anno>VarName</anno></c>.</p>
	<p>If Unicode filename encoding is in effect (see the
        <seealso marker="erts:erl#file_name_encoding"><c>erl</c> manual
	page</seealso>), the string <c><anno>VarName</anno></c> can
	contain characters with codepoints &gt; 255.</p>
      </desc>
    </func>

    <func>
      <name name="version" arity="0"/>
      <fsummary>Return the OS versions.</fsummary>
      <desc>
        <p>Returns the OS version.
          On most systems, this function returns a tuple, but a string
          is returned instead if the system has versions that
          cannot be expressed as three numbers.</p>
        <note>
          <p>Think twice before using this function. If you still need
            to use it, always <c>call os:type()</c> first.</p>
        </note>
      </desc>
    </func>
  </funcs>
</erlref>