path: root/erts/doc/src/escript.xml

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

<comref>
  <header>
    <copyright>
      <year>2007</year><year>2015</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>escript</title>
    <prepared></prepared>
    <docno></docno>
    <date></date>
    <rev></rev>
    <file>escript.xml</file>
  </header>
  <com>escript</com>
  <comsummary>Erlang scripting support</comsummary>
  <description>
    <p><c>escript</c> provides support for running short Erlang programs
      without having to compile them first and an easy way to retrieve the
      command line arguments.</p>
  </description>
  <funcs>
    <func>
      <name>script-name script-arg1 script-arg2...</name>
      <name>escript escript-flags script-name script-arg1 script-arg2...</name>
      <fsummary>Run a script written in Erlang</fsummary>
      <desc>
        <p><c>escript</c> runs a script written in Erlang.</p>
        <p>Here follows an example.</p>
        <pre>
$ <input>chmod u+x factorial</input>
$ <input>cat factorial</input>
#!/usr/bin/env escript
%% -*- erlang -*-
%%! -smp enable -sname factorial -mnesia debug verbose
main([String]) ->
    try
        N = list_to_integer(String),
        F = fac(N),
        io:format("factorial ~w = ~w\n", [N,F])
    catch
        _:_ ->
            usage()
    end;
main(_) ->
    usage().

usage() ->
    io:format("usage: factorial integer\n"),
    halt(1).

fac(0) -> 1;
fac(N) -> N * fac(N-1).
$ <input>./factorial 5</input>
factorial 5 = 120
$ <input>./factorial</input>
usage: factorial integer
$ <input>./factorial five</input>
usage: factorial integer
        </pre>
        <p>The header of the Erlang script in the example differs from
        a normal Erlang module.  The first line is intended to be the
        interpreter line, which invokes <c>escript</c>. However if you
        invoke the <c>escript</c> like this</p>
        <pre>
$ <input>escript factorial 5</input>        </pre>
        <p>the contents of the first line does not matter, but it
        cannot contain Erlang code as it will be ignored.</p>
        <p>The second line in the example, contains an optional
          directive to the <c>Emacs</c> editor which causes it to
          enter the major mode for editing Erlang source files. If the
          directive is present it must be located on the second
          line.</p>

        <p>If there is a comment selecting the <seealso
          marker="stdlib:epp#encoding">encoding</seealso> it can be
          located on the second line.</p>

        <note><p>
          The encoding specified by the above mentioned comment
          applies to the script itself. The encoding of the
          I/O-server, however, has to be set explicitly like this:</p>
<code>io:setopts([{encoding, unicode}])</code>
          <p>The default encoding of the I/O-server for <c>standard_io</c>
          is <c>latin1</c>
          since the script runs in a non-interactive terminal
          (see <seealso marker="stdlib:unicode_usage#unicode_options_summary">
          Using Unicode in Erlang</seealso>).
        </p></note>

	<p>On the third line (or second line depending on the presence
          of the Emacs directive), it is possible to give arguments to
          the emulator, such as </p>
        <pre>
%%! -smp enable -sname factorial -mnesia debug verbose</pre>
        <p>Such an argument line must start with <c>%%!</c> and the
          rest of the line will interpreted as arguments to the emulator.</p>
        <p>If you know the location of the <c>escript</c> executable, the first
          line can directly give the path to <c>escript</c>. For instance:</p>
        <pre>
#!/usr/local/bin/escript        </pre>
        <p>As any other kind of scripts, Erlang scripts will not work on
          Unix platforms if the execution bit for the script file is not set.
          (Use <c>chmod +x script-name</c> to turn on the execution bit.)
          </p>

        <p>The rest of the Erlang script file may either contain
        Erlang <c>source code</c>, an <c>inlined beam file</c> or an
        <c>inlined archive file</c>.</p>

        <p>An Erlang script file must always contain the function
        <em>main/1</em>.  When the script is run, the
        <c>main/1</c> function will be called with a list
        of strings representing the arguments given to the script (not
        changed or interpreted in any way).</p>

        <p>If the <c>main/1</c> function in the script returns successfully,
          the exit status for the script will be 0. If an exception is generated
          during execution, a short message will be printed and the script terminated
          with exit status 127.</p>

        <p>To return your own non-zero exit code, call <c>halt(ExitCode)</c>;
          for instance:</p>
        <pre>
halt(1).</pre>

        <p>Call <seealso marker="#script_name_0">escript:script_name()</seealso>
	from your to script to retrieve the pathname of the script
	(the pathname is usually, but not always, absolute).</p>

        <p>If the file contains source code (as in the example above),
          it will be processed by the preprocessor <c>epp</c>. This
          means that you for example may use pre-defined macros (such as
          <c>?MODULE</c>) as well as include directives like
          the <c>-include_lib</c> directive. For instance, use</p>
        <pre>
-include_lib("kernel/include/file.hrl").</pre>
        <p>to include the record definitions for the records used by the
          <c>file:read_link_info/1</c> function. You can also select
          encoding by including a encoding comment here, but if there
          is a valid encoding comment on the second line it takes
          precedence.</p>

        <p>The script will be checked for syntactic and semantic
          correctness before being run. If there are warnings (such as
          unused variables), they will be printed and the script will
          still be run. If there are errors, they will be printed and
          the script will not be run and its exit status will be
          127.</p>

        <p>Both the module declaration and the export declaration of
          the <c>main/1</c> function are optional.</p>

        <p>By default, the script will be interpreted. You can force
          it to be compiled by including the following line somewhere
          in the script file:</p><pre>
-mode(compile).</pre>

        <p>Execution of interpreted code is slower than compiled code.
          If much of the execution takes place in interpreted code it
          may be worthwhile to compile it, even though the compilation
          itself will take a little while. It is also possible to supply
	  <c>native</c> instead of <c>compile</c>, this will compile the script
	  using the native flag, again depending on the characteristics 
	  of the escript this could or could not be worth while.</p>

        <p>As mentioned earlier, it is possible to have a script which
          contains precompiled <c>beam</c> code. In a precompiled
          script, the interpretation of the script header is exactly
          the same as in a script containing source code. That means
          that you can make a <c>beam</c> file executable by
          prepending the file with the lines starting with <c>#!</c>
          and <c>%%!</c> mentioned above. In a precompiled script, the
          function
          <c>main/1</c> must be exported.</p>

        <p>As yet another option it is possible to have an entire
          Erlang archive in the script. In a archive script, the
          interpretation of the script header is exactly the same as
          in a script containing source code. That means that you can
          make an archive file executable by prepending the file with
          the lines starting with <c>#!</c> and <c>%%!</c> mentioned
          above. In an archive script, the function <c>main/1</c> must
          be exported. By default the <c>main/1</c> function in the
          module with the same name as the basename of the
          <c>escript</c> file will be invoked. This behavior can be
          overridden by setting the flag <c>-escript main Module</c>
          as one of the emulator flags. The <c>Module</c> must be the
          name of a module which has an exported <c>main/1</c>
          function. See <seealso marker="kernel:code">code(3)</seealso>
          for more information about archives and code loading.</p>

	  <p>In many cases it is very convenient to have a header in
	  the escript, especially on Unix platforms. But the header is
	  in fact optional. This means that you directly can "execute"
	  an Erlang module, beam file or archive file without adding
	  any header to them. But then you have to invoke the script
	  like this:</p>
	  <pre>
$ <input>escript factorial.erl 5</input>
factorial 5 = 120
$ <input>escript factorial.beam 5</input>
factorial 5 = 120
$ <input>escript factorial.zip 5</input>
factorial 5 = 120
</pre>
      </desc>
    </func>
    <func>
      <name>escript:create(FileOrBin, Sections) -> ok | {ok, binary()} | {error, term()}</name>
      <fsummary>Create an escript</fsummary>
      <type>
        <v>FileOrBin = filename() | 'binary'</v>
        <v>Sections = [Header] Body | Body</v>
        <v>Header = shebang | {shebang, Shebang}
                    | comment | {comment, Comment}
                    | {emu_args, EmuArgs}</v>
        <v>Shebang = string() | 'default' | 'undefined'</v>
        <v>Comment = string() | 'default' | 'undefined'</v>
        <v>EmuArgs = string() | 'undefined'</v>
        <v>Body = {source, SourceCode}
                | {beam, BeamCode}
                | {archive, ZipArchive}
                | {archive, ZipFiles, ZipOptions}</v>
        <v>SourceCode = BeamCode = file:filename() | binary()</v>
        <v>ZipArchive = <seealso marker="stdlib:zip#type-filename">zip:filename()</seealso> | binary()</v>
        <v>ZipFiles = [ZipFile]</v>
        <v>ZipFile = file:filename() | {file:filename(), binary()} | {file:filename(), binary(), file:file_info()}</v>
        <v>ZipOptions = [<seealso marker="stdlib:zip#type-create_option">zip:create_option()</seealso>]</v>
      </type>
      <desc>
        <p>The <marker id="create_2"></marker> <c>create/2</c>
        function creates an escript from a list of sections. The
        sections can be given in any order. An escript begins with an
        optional <c>Header</c> followed by a mandatory <c>Body</c>. If
        the header is present, it does always begin with a
        <c>shebang</c>, possibly followed by a <c>comment</c> and
        <c>emu_args</c>. The <c>shebang</c> defaults to
        <c>"/usr/bin/env escript"</c>. The comment defaults to
        <c>"This is an -*- erlang -*- file"</c>. The created escript
        can either be returned as a binary or written to file.</p>

	<p>As an example of how the function can be used, we create an
	interpreted escript which uses <c>emu_args</c> to set some emulator
	flag. In this case it happens to disable the smp_support. We
	do also extract the different sections from the newly created
	script:</p>
        <pre>
&gt; <input>Source = "%% Demo\nmain(_Args) ->\n    io:format(erlang:system_info(smp_support)).\n".</input>
"%% Demo\nmain(_Args) ->\n    io:format(erlang:system_info(smp_support)).\n"
&gt; <input>io:format("~s\n", [Source]).</input>
%% Demo
main(_Args) ->
    io:format(erlang:system_info(smp_support)).

ok
&gt; <input>{ok, Bin} = escript:create(binary, [shebang, comment, {emu_args, "-smp disable"},
                                      {source, list_to_binary(Source)}]).</input>
{ok,&lt;&lt;"#!/usr/bin/env escript\n%% This is an -*- erlang -*- file\n%%!-smp disabl"...&gt;&gt;}
&gt; <input>file:write_file("demo.escript", Bin).</input>
ok
&gt; <input>os:cmd("escript demo.escript").</input>
"false"
&gt; <input>escript:extract("demo.escript", []).</input>
{ok,[{shebang,default}, {comment,default}, {emu_args,"-smp disable"},
     {source,&lt;&lt;"%% Demo\nmain(_Args) ->\n    io:format(erlang:system_info(smp_su"...&gt;&gt;}]}
	</pre>

	<p>An escript without header can be created like this:</p>
<pre>
&gt; <input>file:write_file("demo.erl",
                  ["%% demo.erl\n-module(demo).\n-export([main/1]).\n\n", Source]).</input>
ok
&gt; <input>{ok, _, BeamCode} = compile:file("demo.erl", [binary, debug_info]).</input>
{ok,demo,
    &lt;&lt;70,79,82,49,0,0,2,208,66,69,65,77,65,116,111,109,0,0,0,
      79,0,0,0,9,4,100,...&gt;&gt;}
&gt; <input>escript:create("demo.beam", [{beam, BeamCode}]).</input>
ok
&gt; <input>escript:extract("demo.beam", []).</input>
{ok,[{shebang,undefined}, {comment,undefined}, {emu_args,undefined},
     {beam,&lt;&lt;70,79,82,49,0,0,3,68,66,69,65,77,65,116,
             111,109,0,0,0,83,0,0,0,9,...&gt;&gt;}]}
&gt; <input>os:cmd("escript demo.beam").</input>
"true"
</pre>
	<p>Here we create an archive script containing both Erlang
	code as well as beam code. Then we iterate over all files in
	the archive and collect their contents and some info about
	them.
	</p>
<pre>
&gt; <input>{ok, SourceCode} = file:read_file("demo.erl").</input>
{ok,&lt;&lt;"%% demo.erl\n-module(demo).\n-export([main/1]).\n\n%% Demo\nmain(_Arg"...&gt;&gt;}
&gt; <input>escript:create("demo.escript",
                 [shebang,
                  {archive, [{"demo.erl", SourceCode},
                             {"demo.beam", BeamCode}], []}]).</input>
ok
&gt; <input>{ok, [{shebang,default}, {comment,undefined}, {emu_args,undefined},
     {archive, ArchiveBin}]} = escript:extract("demo.escript", []).</input>
{ok,[{shebang,default}, {comment,undefined}, {emu_args,undefined},
     {{archive,&lt;&lt;80,75,3,4,20,0,0,0,8,0,118,7,98,60,105,
                152,61,93,107,0,0,0,118,0,...&gt;&gt;}]}
&gt; <input>file:write_file("demo.zip", ArchiveBin).</input>
ok
&gt; <input>zip:foldl(fun(N, I, B, A) -> [{N, I(), B()} | A] end, [], "demo.zip").</input>
{ok,[{"demo.beam",
      {file_info,748,regular,read_write,
                 {{2010,3,2},{0,59,22}},
                 {{2010,3,2},{0,59,22}},
                 {{2010,3,2},{0,59,22}},
                 54,1,0,0,0,0,0},
      &lt;&lt;70,79,82,49,0,0,2,228,66,69,65,77,65,116,111,109,0,0,0,
        83,0,0,...&gt;&gt;},
     {"demo.erl",
      {file_info,118,regular,read_write,
                 {{2010,3,2},{0,59,22}},
                 {{2010,3,2},{0,59,22}},
                 {{2010,3,2},{0,59,22}},
                 54,1,0,0,0,0,0},
      &lt;&lt;"%% demo.erl\n-module(demo).\n-export([main/1]).\n\n%% Demo\nmain(_Arg"...&gt;&gt;}]}</pre>
      </desc>
    </func>
    <func>
      <name>escript:extract(File, Options) -> {ok, Sections} | {error, term()}</name>
      <fsummary>Parses an escript and extracts its sections</fsummary>
      <type>
        <v>File = filename()</v>
        <v>Options = [] | [compile_source]</v>
        <v>Sections = Headers Body</v>
        <v>Headers =  {shebang, Shebang}
                      {comment, Comment}
                      {emu_args, EmuArgs}</v>
        <v>Shebang = string() | 'default' | 'undefined'</v>
        <v>Comment = string() | 'default' | 'undefined'</v>
        <v>EmuArgs = string() | 'undefined'</v>
        <v>Body = {source, SourceCode}
                | {source, BeamCode}
                | {beam, BeamCode}
	        | {archive, ZipArchive}</v>
        <v>SourceCode = BeamCode = ZipArchive = binary()</v>
      </type>
      <desc>
        <p>The <marker id="extract_2"></marker> <c>extract/2</c>
        function parses an escript and extracts its sections. This is
        the reverse of <c>create/2</c>.</p>

	<p>All sections are returned even if they do not exist in the
	escript. If a particular section happens to have the same
	value as the default value, the extracted value is set to the
	atom <c>default</c>. If a section is missing, the extracted
	value is set to the atom <c>undefined</c>.  </p>

	<p>The <c>compile_source</c> option only affects the result if
	the escript contains <c>source</c> code. In that case the
	Erlang code is automatically compiled and <c>{source,
	BeamCode}</c> is returned instead of <c>{source,
	SourceCode}</c>.</p>

	<pre>
&gt; <input>escript:create("demo.escript",
                 [shebang, {archive, [{"demo.erl", SourceCode},
                                      {"demo.beam", BeamCode}], []}]).</input>
ok
&gt; <input>{ok, [{shebang,default}, {comment,undefined}, {emu_args,undefined},
     {archive, ArchiveBin}]} =
              escript:extract("demo.escript", []).</input>
{ok,[{{archive,&lt;&lt;80,75,3,4,20,0,0,0,8,0,118,7,98,60,105,
                152,61,93,107,0,0,0,118,0,...&gt;&gt;}
     {emu_args,undefined}]}
	</pre>
      </desc>
    </func>
    <func>
      <name>escript:script_name() -> File</name>
      <fsummary>Returns the name of an escript</fsummary>
      <type>
        <v>File = filename()</v>
      </type>
      <desc>
        <p>The <marker id="script_name_0"></marker>
        <c>script_name/0</c> function returns the name of the escript
        being executed. If the function is invoked outside the context
        of an escript, the behavior is undefined.</p>
      </desc>
    </func>
  </funcs>

  <section>
    <title>Options accepted by escript</title>
    <taglist>
      <tag>-c</tag>
      <item>Compile the escript regardless of the value of the mode attribute.
      </item>

      <tag>-d</tag>
      <item>Debug the escript. Starts the debugger, loads the module
      containing the <c>main/1</c> function into the debugger, sets a
      breakpoint in <c>main/1</c> and invokes <c>main/1</c>. If the
      module is precompiled, it must be explicitly compiled with the
      <c>debug_info</c> option.
      </item>

      <tag>-i</tag>
      <item>Interpret the escript regardless of the value of the mode attribute.
      </item>

      <tag>-s</tag>
      <item>Only perform a syntactic and semantic check of the script file.
       Warnings and errors (if any) are written to the standard output, but
       the script will not be run. The exit status will be 0 if there were
       no errors, and 127 otherwise.</item>

       <tag>-n</tag>
      <item>Compile the escript using the +native flag.</item>
    </taglist>
  </section>
</comref>