path: root/lib/dialyzer/doc/src/dialyzer.xml

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

<erlref>
  <header>
    <copyright>
      <year>2006</year><year>2019</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>dialyzer</title>
    <prepared></prepared>
    <docno></docno>
    <date>2016-09-20</date>
    <rev></rev>
    <file>dialyzer.xml</file>
  </header>
  <module since="">dialyzer</module>
  <modulesummary>Dialyzer, a DIscrepancy AnaLYZer for ERlang programs.
  </modulesummary>
  <description>
    <p>Dialyzer is a static analysis tool that identifies software
      discrepancies, such as definite type errors, code that has become dead
      or unreachable because of programming error, and unnecessary tests,
      in single Erlang modules or entire (sets of) applications.</p>

    <p>Dialyzer starts its analysis from either
      debug-compiled BEAM bytecode or from Erlang source code. The file
      and line number of a discrepancy is reported along with an
      indication of what the discrepancy is about. Dialyzer bases its
      analysis on the concept of success typings, which allows for sound
      warnings (no false positives).</p>
  </description>

  <section>
    <marker id="command_line"></marker>
    <title>Using Dialyzer from the Command Line</title>
    <p>Dialyzer has a command-line version for automated use. This
      section provides a brief description of the options. The same information
      can be obtained by writing the following in a shell:</p>

    <code type="none">
dialyzer --help</code>

    <p>For more details about the operation of Dialyzer, see section
      <seealso marker="dialyzer_chapter#dialyzer_gui">
      Using Dialyzer from the GUI</seealso> in the User's Guide.</p>

    <p><em>Exit status of the command-line version:</em></p>

    <taglist>
      <tag><c>0</c></tag>
      <item>
        <p>No problems were found during the analysis and no warnings were
          emitted.</p>
      </item>
      <tag><c>1</c></tag>
      <item>
        <p>Problems were found during the analysis.</p>
      </item>
      <tag><c>2</c></tag>
      <item>
        <p>No problems were found during the analysis, but warnings were
          emitted.</p>
      </item>
    </taglist>

    <p><em>Usage:</em></p>

    <code type="none">
dialyzer [--add_to_plt] [--apps applications] [--build_plt]
         [--check_plt] [-Ddefine]* [-Dname] [--dump_callgraph file]
         [files_or_dirs] [--fullpath] [--get_warnings] [--gui] [--help]
         [-I include_dir]* [--no_check_plt] [--no_indentation] [--no_native]
         [--no_native_cache] [-o outfile] [--output_plt file] [-pa dir]*
         [--plt plt] [--plt_info] [--plts plt*] [--quiet] [-r dirs]
         [--raw] [--remove_from_plt] [--shell] [--src] [--statistics]
         [--verbose] [--version] [-Wwarn]*</code>

    <note>
      <p>* denotes that multiple occurrences of the option are possible.</p>
    </note>

    <p><em>Options:</em></p>

    <taglist>
      <tag><c>--add_to_plt</c></tag>
      <item>
        <p>The PLT is extended to also include the files specified with
          <c>-c</c> and <c>-r</c>. Use
          <c>--plt</c> to specify which PLT to start from,
          and <c>--output_plt</c> to specify where to put the PLT.
          Notice that the analysis possibly can include files from the PLT if
          they depend on the new files. This option only works for BEAM
          files.</p>
      </item>
      <tag><c>--apps applications</c></tag>
      <item>
        <p>This option is typically used when building or modifying a PLT as
          in:</p>
        <code type="none">
dialyzer --build_plt --apps erts kernel stdlib mnesia ...</code>
        <p>to refer conveniently to library applications corresponding to the
          Erlang/OTP installation. However, this option is general and can also
          be used during analysis to refer to Erlang/OTP applications.
          File or directory names can also be included, as in:</p>
        <code type="none">
dialyzer --apps inets ssl ./ebin ../other_lib/ebin/my_module.beam</code>
      </item>
      <tag><c>--build_plt</c></tag>
      <item>
        <p>The analysis starts from an empty PLT and creates a new one from
          the files specified with <c>-c</c> and
          <c>-r</c>. This option only works for BEAM files.
          To override the default PLT location, use
          <c>--plt</c> or <c>--output_plt</c>.</p>
      </item>
      <tag><c>--check_plt</c></tag>
      <item>
        <p>Check the PLT for consistency and rebuild it if it is not
          up-to-date.</p>
      </item>
      <tag><c>-Dname</c> (or <c>-Dname=value</c>)</tag>
      <item>
        <p>When analyzing from source, pass the define to Dialyzer.
          (**)</p>
      </item>
      <tag><c>--dump_callgraph file</c></tag>
      <item>
        <p>Dump the call graph into the specified file whose format is
          determined by the filename extension. Supported extensions are:
          <c>raw</c>, <c>dot</c>, and <c>ps</c>. If something else is used as
          filename extension, default format <c>.raw</c> is used.</p>
      </item>
      <tag><c>files_or_dirs</c> (for backward compatibility also
        as <c>-c files_or_dirs</c>)</tag>
      <item>
        <p>Use Dialyzer from the command line to detect defects in the
          specified files or directories containing <c>.erl</c> or
          <c>.beam</c> files, depending on the type of the
          analysis.</p>
      </item>
      <tag><c>--fullpath</c></tag>
      <item>
        <p>Display the full path names of files for which warnings are
          emitted.</p>
      </item>
      <tag><c>--get_warnings</c></tag>
      <item>
        <p>Make Dialyzer emit warnings even when manipulating the PLT.
          Warnings are only emitted for files that are analyzed.</p>
      </item>
      <tag><c>--gui</c></tag>
      <item>
        <p>Use the GUI.</p></item>
      <tag><c>--help</c> (or <c>-h</c>)</tag>
      <item>
        <p>Print this message and exit.</p>
      </item>
      <tag><c>-I include_dir</c></tag>
      <item>
        <p>When analyzing from source, pass the <c>include_dir</c>
          to Dialyzer. (**)</p>
      </item>
      <tag><c>--no_check_plt</c></tag>
      <item>
        <p>Skip the PLT check when running Dialyzer. This is useful when
          working with installed PLTs that never change.</p>
      </item>
      <tag><c>--no_indentation</c></tag>
      <item>
	<p>Do not insert line breaks in types, contracts, and Erlang
	  Code when formatting warnings.</p>
      </item>
      <tag><c>--no_native</c> (or <c>-nn</c>)</tag>
      <item>
        <p>Bypass the native code compilation of some key files that
          Dialyzer heuristically performs when dialyzing many files.
          This avoids the compilation time, but can result in (much) longer
          analysis time.</p>
      </item>
      <tag><c>--no_native_cache</c></tag>
      <item>
        <p>By default, Dialyzer caches the results of native compilation
          in directory <c>$XDG_CACHE_HOME/erlang/dialyzer_hipe_cache</c>.
          <c>XDG_CACHE_HOME</c> defaults to <c>$HOME/.cache</c>.
          Use this option to disable caching.</p>
      </item>
      <tag><c>-o outfile</c> (or
        <c>--output outfile</c>)</tag>
      <item>
        <p>When using Dialyzer from the command line, send the analysis
          results to the specified outfile rather than to <c>stdout</c>.</p>
      </item>
      <tag><c>--output_plt file</c></tag>
      <item>
        <p>Store the PLT at the specified file after building it.</p>
      </item>
      <tag><c>-pa dir</c></tag>
      <item>
        <p>Include <c>dir</c> in the path for Erlang. This is useful
          when analyzing files that have <c>-include_lib()</c>
          directives.</p>
      </item>
      <tag><c>--plt plt</c></tag>
      <item>
        <p>Use the specified PLT as the initial PLT. If the PLT was built
          during setup, the files are checked for consistency.</p>
      </item>
      <tag><c>--plt_info</c></tag>
      <item>
        <p>Make Dialyzer print information about the PLT and then quit.
          The PLT can be specified with <c>--plt(s)</c>.</p>
      </item>
      <tag><c>--plts plt*</c></tag>
      <item>
        <p>Merge the specified PLTs to create the initial PLT. This requires
          that the PLTs are disjoint (that is, do not have any module
          appearing in more than one PLT).
          The PLTs are created in the usual way:</p>
        <code type="none">
dialyzer --build_plt --output_plt plt_1 files_to_include
...
dialyzer --build_plt --output_plt plt_n files_to_include</code>
        <p>They can then be used in either of the following ways:</p>
        <code type="none">
dialyzer files_to_analyze --plts plt_1 ... plt_n</code>
        <p>or</p>
        <code type="none">
dialyzer --plts plt_1 ... plt_n -- files_to_analyze</code>
        <p>Notice the <c>--</c> delimiter in the second case.</p>
      </item>
      <tag><c>--quiet</c> (or <c>-q</c>)</tag>
      <item>
        <p>Make Dialyzer a bit more quiet.</p>
      </item>
      <tag><c>-r dirs</c></tag>
      <item>
        <p>Same as <c>files_or_dirs</c>, but the specified
          directories are searched
          recursively for subdirectories containing <c>.erl</c> or
          <c>.beam</c> files in them, depending on the type of
          analysis.</p>
      </item>
      <tag><c>--raw</c></tag>
      <item>
        <p>When using Dialyzer from the command line, output the raw
          analysis results (Erlang terms) instead of the formatted result.
          The raw format
          is easier to post-process (for example, to filter warnings or to
          output HTML pages).</p>
      </item>
      <tag><c>--remove_from_plt</c></tag>
      <item>
        <p>The information from the files specified with
          <c>-c</c> and <c>-r</c> is removed from
          the PLT. Notice that this can cause a reanalysis of the remaining
          dependent files.</p>
      </item>
      <tag><c>--shell</c></tag>
      <item>
        <p>Do not disable the Erlang shell while running the GUI.</p>
      </item>
      <tag><c>--src</c></tag>
      <item>
        <p>Override the default, which is to analyze BEAM files, and
          analyze starting from Erlang source code instead.</p>
      </item>
      <tag><c>--statistics</c></tag>
      <item>
        <p>Print information about the progress of execution (analysis phases,
          time spent in each, and size of the relative input).</p>
      </item>
      <tag><c>--verbose</c></tag>
      <item>
        <p>Make Dialyzer a bit more verbose.</p>
      </item>
      <tag><c>--version</c> (or <c>-v</c>)</tag>
      <item>
        <p>Print the Dialyzer version and some more information and
          exit.</p>
      </item>
      <tag><c>-Wwarn</c></tag>
      <item>
        <p>A family of options that selectively turn on/off warnings.
          (For help on the names of warnings, use
          <c>dialyzer -Whelp</c>.)
          Notice that the options can also be specified in the file with a
          <c>-dialyzer()</c> attribute. For details, see section <seealso
          marker="#suppression">Requesting or Suppressing Warnings in
          Source Files</seealso>.</p>
      </item>
    </taglist>

    <note>
      <p>** options <c>-D</c> and <c>-I</c> work both
        from the command line and in the Dialyzer GUI; the syntax of
        defines and includes is the same as that used by
        <seealso marker="erts:erlc">erlc(1)</seealso>.</p>
    </note>

    <p><em>Warning options:</em></p>

    <taglist>
      <tag><c>-Werror_handling</c> (***)</tag>
      <item>
        <p>Include warnings for functions that only return by an exception.</p>
      </item>
      <tag><c>-Wno_behaviours</c></tag>
      <item>
        <p>Suppress warnings about behavior callbacks that drift from the
          published recommended interfaces.</p>
      </item>
      <tag><c>-Wno_contracts</c></tag>
      <item>
        <p>Suppress warnings about invalid contracts.</p>
      </item>
      <tag><c>-Wno_fail_call</c></tag>
      <item>
        <p>Suppress warnings for failing calls.</p>
      </item>
      <tag><c>-Wno_fun_app</c></tag>
      <item>
        <p>Suppress warnings for fun applications that will fail.</p>
      </item>
      <tag><c>-Wno_improper_lists</c></tag>
      <item>
        <p>Suppress warnings for construction of improper lists.</p>
      </item>
      <tag><c>-Wno_match</c></tag>
      <item>
        <p>Suppress warnings for patterns that are unused or cannot match.</p>
      </item>
      <tag><c>-Wno_missing_calls</c></tag>
      <item>
        <p>Suppress warnings about calls to missing functions.</p>
      </item>
      <tag><c>-Wno_opaque</c></tag>
      <item>
        <p>Suppress warnings for violations of opacity of data types.</p>
      </item>
      <tag><c>-Wno_return</c></tag>
      <item>
        <p>Suppress warnings for functions that will never return a value.</p>
      </item>
      <tag><c>-Wno_undefined_callbacks</c></tag>
      <item>
        <p>Suppress warnings about behaviors that have no
          <c>-callback</c> attributes for their callbacks.</p>
      </item>
      <tag><c>-Wno_unused</c></tag>
      <item>
        <p>Suppress warnings for unused functions.</p>
      </item>
      <tag><c>-Wrace_conditions</c> (***)</tag>
      <item>
        <p>Include warnings for possible race conditions. Notice that the
          analysis that finds data races performs intra-procedural data flow
          analysis and can sometimes explode in time. Enable it at your own
          risk.</p>
      </item>
      <tag><c>-Wunderspecs</c> (***)</tag>
      <item>
        <p>Warn about underspecified functions (the specification is strictly
          more allowing than the success typing).</p>
      </item>
      <tag><c>-Wunknown</c> (***)</tag>
      <item>
        <p>Let warnings about unknown functions and types affect the
          exit status of the command-line version. The default is to ignore
          warnings about unknown functions and types when setting the exit
          status. When using Dialyzer from Erlang, warnings about unknown
          functions and types are returned; the default is not to return
          these warnings.</p>
      </item>
      <tag><c>-Wunmatched_returns</c> (***)</tag>
      <item>
        <p>Include warnings for function calls that ignore a structured return
          value or do not match against one of many possible return
          value(s).</p>
      </item>
    </taglist>

    <p>The following options are also available, but their use is not
      recommended (they are mostly for Dialyzer developers and internal
      debugging):</p>

    <taglist>
      <tag><c>-Woverspecs</c> (***)</tag>
      <item>
        <p>Warn about overspecified functions (the specification is strictly
          less allowing than the success typing).</p>
      </item>
      <tag><c>-Wspecdiffs</c> (***)</tag>
      <item>
        <p>Warn when the specification is different than the success typing.</p>
      </item>
    </taglist>

    <note>
      <p>*** denotes options that turn on warnings rather than
        turning them off.</p>
    </note>
  </section>

  <section>
    <title>Using Dialyzer from Erlang</title>
    <p>Dialyzer can be used directly from Erlang. Both the GUI and the
      command-line versions are also available. The options are similar to the
      ones given from the command line, see section
      <seealso marker="#command_line">
      Using Dialyzer from the Command Line</seealso>.</p>
  </section>

  <section>
    <marker id="suppression"></marker>
    <title>Requesting or Suppressing Warnings in Source Files</title>
    <p>Attribute <c>-dialyzer()</c> can be used for turning off
      warnings in a module by specifying functions or warning options.
      For example, to turn off all warnings for the function
      <c>f/0</c>, include the following line:</p>

    <code type="none">
-dialyzer({nowarn_function, f/0}).</code>

    <p>To turn off warnings for improper lists, add the following line
      to the source file:</p>

    <code type="none">
-dialyzer(no_improper_lists).</code>

    <p>Attribute <c>-dialyzer()</c> is allowed after function
      declarations. Lists of warning options or functions are allowed:</p>

    <code type="none">
-dialyzer([{nowarn_function, [f/0]}, no_improper_lists]).</code>

    <p>Warning options can be restricted to functions:</p>

    <code type="none">
-dialyzer({no_improper_lists, g/0}).</code>

    <code type="none">
-dialyzer({[no_return, no_match], [g/0, h/0]}).</code>

    <p>For help on the warning options, use <c>dialyzer -Whelp</c>. The
      options are also enumerated, see function <seealso marker="#gui/1">
      <c>gui/1</c></seealso> below (<c>WarnOpts</c>).</p>

    <note>
      <p>Warning option <c>-Wrace_conditions</c> has no effect when
        set in source files.</p>
    </note>

    <p>Attribute <c>-dialyzer()</c> can also be used for turning on
      warnings. For example, if a module has been fixed regarding
      unmatched returns, adding the following line can help in assuring
      that no new unmatched return warnings are introduced:</p>

    <code type="none">
-dialyzer(unmatched_returns).</code>
  </section>

  <funcs>
    <func>
      <name since="">format_warning(Msg) -> string()</name>
      <fsummary>Get the string version of a warning message.</fsummary>
      <type>
        <v>Msg = {Tag, Id, msg()}</v>
        <d>See <c>run/1</c>.</d>
      </type>
      <desc>
        <p>Get a string from warnings as returned by
          <seealso marker="#run/1"><c>run/1</c></seealso>.</p>
      </desc>
    </func>

    <func>
      <name since="">format_warning(Msg, Options) -> string()</name>
      <fsummary>Get the string version of a warning message.</fsummary>
      <type>
        <v>Msg = {Tag, Id, msg()}</v>
        <d>See <c>run/1</c>.</d>
        <v>Options = [{indent_opt, boolean()}]</v>
      </type>
      <desc>
        <p>Get a string from warnings as returned by
          <seealso marker="#run/1"><c>run/1</c></seealso>.</p>
	<p>If <c>indent_opt</c> is set to <c>true</c> (default),
          line breaks are inserted in types, contracts, and Erlang
	  code to improve readability.</p>
      </desc>
    </func>

    <func>
      <name since="">gui() -> ok | {error, Msg}</name>
      <name since="">gui(OptList) -> ok | {error, Msg}</name>
      <fsummary>Dialyzer GUI version.</fsummary>
      <type>
        <v>OptList</v>
        <d>See below.</d>
      </type>
      <desc>
        <p>Dialyzer GUI version.</p>
        <code type="none">
OptList  :: [Option]
Option   :: {files,          [Filename :: string()]}
          | {files_rec,      [DirName :: string()]}
          | {defines,        [{Macro :: atom(), Value :: term()}]}
          | {from,           src_code | byte_code}
                               %% Defaults to byte_code
          | {init_plt,       FileName :: string()}
                               %% If changed from default
          | {plts,           [FileName :: string()]}
                               %% If changed from default
          | {include_dirs,   [DirName :: string()]}
          | {output_file,    FileName :: string()}
          | {output_plt,     FileName :: string()}
          | {check_plt,      boolean()},
          | {analysis_type,  'succ_typings' |
                             'plt_add' |
                             'plt_build' |
                             'plt_check' |
                             'plt_remove'}
          | {warnings,       [WarnOpts]}
          | {get_warnings,   bool()}

WarnOpts :: error_handling
          | no_behaviours
          | no_contracts
          | no_fail_call
          | no_fun_app
          | no_improper_lists
          | no_match
	  | no_missing_calls
          | no_opaque
	  | no_return
          | no_undefined_callbacks
          | no_unused
          | race_conditions
          | underspecs
          | unknown
          | unmatched_returns
          | overspecs
          | specdiffs</code>
      </desc>
    </func>

    <func>
      <name since="">plt_info(string()) -> {'ok', [{atom(), any()}]} | {'error', atom()}</name>
      <fsummary>Return information about the specified PLT.</fsummary>
      <desc>
        <p>Returns information about the specified PLT.</p>
      </desc>
    </func>

    <func>
      <name since="">run(OptList) -> Warnings</name>
      <fsummary>Dialyzer command-line version.</fsummary>
      <type>
        <v>OptList</v>
        <d>See <c>gui/0,1</c>.</d>
        <v>Warnings</v>
        <d>See below.</d>
      </type>
      <desc>
        <p>Dialyzer command-line version.</p>
        <code type="none">
Warnings :: [{Tag, Id, Msg}]
Tag      :: 'warn_behaviour'
          | 'warn_bin_construction'
          | 'warn_callgraph'
          | 'warn_contract_not_equal'
          | 'warn_contract_range'
          | 'warn_contract_subtype'
          | 'warn_contract_supertype'
          | 'warn_contract_syntax'
          | 'warn_contract_types'
          | 'warn_failing_call'
          | 'warn_fun_app'
          | 'warn_matching'
          | 'warn_non_proper_list'
          | 'warn_not_called'
          | 'warn_opaque'
          | 'warn_race_condition'
          | 'warn_return_no_exit'
          | 'warn_return_only_exit'
          | 'warn_umatched_return'
          | 'warn_undefined_callbacks'
          | 'warn_unknown'
Id  = {File :: string(), Line :: integer()}
Msg = msg() -- Undefined</code>
      </desc>
    </func>
  </funcs>
</erlref>