path: root/lib/parsetools/doc/src/leex.xml

<?xml version="1.0" encoding="latin1" ?>
<!DOCTYPE erlref SYSTEM "erlref.dtd">

<erlref>
  <header>
    <copyright>
      <year>2009</year><year>2009</year>
      <holder>Ericsson AB. All Rights Reserved.</holder>
    </copyright>
    <legalnotice>
      Copyright (c) 2008,2009 Robert Virding. All rights reserved.
    </legalnotice>

    <title>leex</title>
    <prepared>Robert Virding</prepared>
    <responsible>nobody</responsible>
    <docno></docno>
    <approved>nobody</approved>
    <checked></checked>
    <date>2009-05-07</date>
    <rev>A</rev>
    <file>leex.xml</file>
  </header>
  <module>leex</module>
  <modulesummary>Lexical analyzer generator for Erlang</modulesummary>
  <description>
    <p>A regular expression based lexical analyzer generator for
      Erlang, similar to lex or flex.</p>
    <note><p>The Leex module should be considered experimental
      as it will be subject to changes in future releases.</p></note>
  </description>
  <section>
    <title>DATA TYPES</title>
    <code type="none">
ErrorInfo = {ErrorLine,module(),error_descriptor()}
ErrorLine = integer()
Token = tuple()</code>
  </section>
  <funcs>
    <func>
      <name>file(FileName) -> ok | error</name>
      <name>file(FileName, Options) -> ok | error</name>
      <fsummary>Generate a lexical analyzer</fsummary>
      <type>
        <v>FileName = filename()</v>
        <v>Options = Option | [Option]</v>
        <v>Option =&nbsp;-&nbsp;see below&nbsp;-</v>
        <v>FileReturn = {ok, Scannerfile} 
                      | {ok, Scannerfile, Warnings}
                      | error
                      | {error, Warnings, Errors}</v>
        <v>Scannerfile = filename()</v>
        <v>Warnings = Errors = [{filename(), [ErrorInfo]}]</v>
      </type>
      <desc>
        <p>Generates a lexical analyzer from the definition in the input
          file. The input file has the extension <c>.xrl</c>. This is
          added to the filename if it is not given. The resulting module
          is the Xrl filename without the <c>.xrl</c> extension.</p>

        <p>The current options are:</p>
          <taglist>
            <tag><c>dfa_graph</c></tag>
            <item><p>Generates a <c>.dot</c> file which contains a
              description of the DFA in a format which can be viewed with
              Graphviz, <c>www.graphviz.com</c>.</p>
            </item>
            <tag><c>{includefile,Includefile}</c></tag>
            <item><p>Uses a specific or customised prologue file
              instead of default
              <c>lib/parsetools/include/leexinc.hrl</c> which is
              otherwise included.</p>
            </item>
            <tag><c>{report_errors, bool()}</c></tag>
            <item><p>Causes errors to be printed as they occur. Default is
              <c>true</c>.</p>
            </item>
            <tag><c>{report_warnings, bool()}</c></tag>
            <item><p>Causes warnings to be printed as they occur. Default is
              <c>true</c>.</p>
            </item>
            <tag><c>{report, bool()}</c></tag>
            <item><p>This is a short form for both <c>report_errors</c> and
              <c>report_warnings</c>.</p>
            </item>
            <tag><c>{return_errors, bool()}</c></tag>
            <item><p>If this flag is set, <c>{error, Errors, Warnings}</c>
              is returned when there are errors. Default is <c>false</c>.</p>
            </item>
            <tag><c>{return_warnings, bool()}</c></tag>
            <item><p>If this flag is set, an extra field containing
              <c>Warnings</c> is added to the tuple returned upon
               success. Default is <c>false</c>.</p>
            </item>
            <tag><c>{return, bool()}</c></tag>
            <item><p>This is a short form for both <c>return_errors</c> and
              <c>return_warnings</c>.</p>
            </item>
            <tag><c>{scannerfile, Scannerfile}</c></tag>
            <item><p><c>Scannerfile</c> is the name of the file that
             will contain the Erlang scanner code that is generated.
             The default (<c>""</c>) is to add the extension
             <c>.erl</c> to <c>FileName</c> stripped of the
             <c>.xrl</c> extension.</p>
            </item>
            <tag><c>{verbose, bool()}</c></tag>
            <item><p>Outputs information from parsing the input file and
              generating the internal tables.</p>
            </item>
          </taglist>
        <p>Any of the Boolean options can be set to <c>true</c> by 
          stating the name of the option. For example, <c>verbose</c>
          is equivalent to <c>{verbose, true}</c>.</p>
        <p>Leex will add the extension <c>.hrl</c> to the 
          <c>Includefile</c> name and the extension <c>.erl</c> to the
          <c>Scannerfile</c> name, unless the extension is already
          there.</p>
      </desc>
    </func>
    <func>
      <name>format_error(ErrorInfo) -> Chars</name>
      <fsummary>Return an English description of a an error tuple.</fsummary>
      <type>
        <v>Chars = [char() | Chars]</v>
      </type>
      <desc>
        <p>Returns a string which describes the error
          <c>ErrorInfo</c> returned when there is an error in a
          regular expression.</p>
      </desc>
    </func>
  </funcs>
  

  <section>
    <title>GENERATED SCANNER EXPORTS</title>
    <p>The following functions are exported by the generated scanner.</p>
  </section>

  <funcs>
    <func>
      <name>string(String) -> StringRet</name>
      <name>string(String, StartLine) -> StringRet</name>
      <fsummary>Generated by Leex</fsummary>
      <type>
        <v>String = string()</v>
        <v>StringRet = {ok,Tokens,EndLine} | ErrorInfo</v>
        <v>Tokens = [Token]</v>
        <v>EndLine = StartLine = integer()</v>
      </type>
      <desc>
        <p>Scans <c>String</c> and returns all the tokens in it, or an
          error.</p>
        <note><p>It is an error if not all of the characters in
          <c>String</c> are consumed.</p></note>
      </desc>
    </func>

    <func>
      <name>token(Cont, Chars) -> {more,Cont1} | {done,TokenRet,RestChars}
      </name>
      <name>token(Cont, Chars, StartLine) -> {more,Cont1} 
             | {done,TokenRet,RestChars}
      </name>
      <fsummary>Generated by Leex</fsummary>
      <type>
        <v>Cont = [] | Cont1</v>
        <v>Cont1 = tuple()</v>
        <v>Chars = RestChars = string() | eof</v>
        <v>TokenRet = {ok, Token, EndLine} 
                    | {eof, EndLine}
                    | ErrorInfo</v>
        <v>StartLine = EndLine = integer()</v>
      </type>
      <desc>
        <p>This is a re-entrant call to try and scan one token from
          <c>Chars</c>. If there are enough characters in <c>Chars</c>
          to either scan a token or detect an error then this will be
          returned with <c>{done,...}</c>. Otherwise
          <c>{cont,Cont}</c> will be returned where <c>Cont</c> is
          used in the next call to <c>token()</c> with more characters
          to try an scan the token. This is continued until a token
          has been scanned. <c>Cont</c> is initially <c>[]</c>.</p>
 
        <p>It is not designed to be called directly by an application
          but used through the i/o system where it can typically be
          called in an application by:</p>
        <code>
io:request(InFile, {get_until,Prompt,Module,token,[Line]})
  -> TokenRet</code>
      </desc>
    </func>

    <func>
      <name>tokens(Cont, Chars) -> {more,Cont1} | {done,TokensRet,RestChars}
      </name>
      <name>tokens(Cont, Chars, StartLine) -> 
               {more,Cont1} | {done,TokensRet,RestChars}
      </name>
      <fsummary>Generated by Leex</fsummary>
      <type>
        <v>Cont = [] | Cont1</v>
        <v>Cont1 = tuple()</v>
        <v>Chars = RestChars = string() | eof</v>
        <v>TokensRet = {ok, Tokens, EndLine} 
                     | {eof, EndLine}
                     | ErrorInfo</v>
        <v>Tokens = [Token]</v>
        <v>StartLine = EndLine = integer()</v>
      </type>
      <desc>
        <p>This is a re-entrant call to try and scan tokens from
          <c>Chars</c>. If there are enough characters in <c>Chars</c>
          to either scan tokens or detect an error then this will be
          returned with <c>{done,...}</c>. Otherwise
          <c>{cont,Cont}</c> will be returned where <c>Cont</c> is
          used in the next call to <c>tokens()</c> with more
          characters to try an scan the tokens. This is continued
          until all tokens have been scanned. <c>Cont</c> is initially
          <c>[]</c>.</p>
 
        <p>This functions differs from <c>token</c> in that it will
          continue to scan tokens upto and including an
          <c>{end_token,Token}</c> has been scanned (see next
          section). It will then return all the tokens. This is
          typically used for scanning grammars like Erlang where there
          is an explicit end token, <c>'.'</c>. If no end token is
          found then the whole file will be scanned and returned. If
          an error occurs then all tokens upto and including the next
          end token will be skipped.</p>
 
        <p>It is not designed to be called directly by an application
          but used through the i/o system where it can typically be
          called in an application by:</p>
        <code>
io:request(InFile, {get_until,Prompt,Module,tokens,[Line]})
  -> TokensRet</code>
      </desc>
    </func>
  </funcs>

  <section>
    <title>Input File Format</title>
    <p>Erlang style comments starting with a <c>%</c> are allowed in
      scanner files. A definition file has the following format:</p>
    <code>
&lt;Header>

Definitions.

&lt;Macro Definitions>

Rules.

&lt;Token Rules>

Erlang code.

&lt;Erlang code></code>
 
    <p>The "Definitions.", "Rules." and "Erlang code." headings are
      mandatory and must occur at the beginning of a source line. The
      &lt;Header>, &lt;Macro Definitions> and &lt;Erlang code>
      sections may be empty but there must be at least one rule.</p>
 
    <p>Macro definitions have the following format:</p>

    <code>
NAME = VALUE</code>

    <p>and there must be spaces around <c>=</c>. Macros can be used in
      the regular expressions of rules by writing <c>{NAME}</c>.</p>

    <note><p>When macros are expanded in expressions the macro calls
      are replaced by the macro value without any form of quoting or
      enclosing in parentheses.</p></note>
 
    <p>Rules have the following format:</p>

    <code>
&lt;Regexp> : &lt;Erlang code>.</code>
 
    <p>The &lt;Regexp> must occur at the start of a line and not
      include any blanks; use <c>\\t</c> and <c>\\s</c> to include TAB
      and SPACE characters in the regular expression. If &lt;Regexp>
      matches then the corresponding &lt;Erlang code> is evaluated to
      generate a token. With the Erlang code the following predefined
      variables are available:</p>
 
    <taglist>
      <tag><c>TokenChars</c></tag>
      <item><p>A list of the characters in the matched token.</p>
      </item>
      <tag><c>TokenLen</c></tag>
      <item><p>The number of characters in the matched token.</p>
      </item>
      <tag><c>TokenLine</c></tag>
      <item><p>The line number where the token occurred.</p>
      </item>
    </taglist>
 
    <p>The code must return:</p>
 
    <taglist>
      <tag><c>{token,Token}</c></tag>
      <item><p>Return <c>Token</c> to the caller.</p>
      </item>
      <tag><c>{end_token,Token}</c></tag>
      <item><p>Return <c>Token</c> and is last token in a tokens call.</p>
      </item>
      <tag><c>skip_token</c></tag>
      <item><p>Skip this token completely.</p>
      </item>
      <tag><c>{error,ErrString}</c></tag>
      <item><p>An error in the token, <c>ErrString</c> is a string
         describing the error.</p>
      </item>
    </taglist>
 
    <p>It is also possible to push back characters into the input
      characters with the following returns:</p>
 
    <list>
      <item><c>{token,Token,PushBackList}</c></item>
      <item><c>{end_token,Token,PushBackList}</c></item>
      <item><c>{skip_token,PushBackList}</c></item>
    </list>
 
    <p>These have the same meanings as the normal returns but the
      characters in <c>PushBackList</c> will be prepended to the input
      characters and scanned for the next token. Note that pushing
      back a newline will mean the line numbering will no longer be
      correct.</p>

    <note><p>Pushing back characters gives you unexpected
      possibilities to cause the scanner to loop!</p></note>
 
    <p>The following example would match a simple Erlang integer or
      float and return a token which could be sent to the Erlang
      parser:</p>
    <code>
D = [0-9]

{D}+ :
  {token,{integer,TokenLine,list_to_integer(TokenChars)}}.

{D}+\\.{D}+((E|e)(\\+|\\-)?{D}+)? :
  {token,{float,TokenLine,list_to_float(TokenChars)}}.</code>
 
    <p>The Erlang code in the "Erlang code." section is written into
      the output file directly after the module declaration and
      predefined exports declaration so it is possible to add extra
      exports, define imports and other attributes which are then
      visible in the whole file.</p>
  </section>
  
  <section>
    <title>Regular Expressions</title>
 
    <p>The regular expressions allowed here is a subset of the set
      found in <c>egrep</c> and in the AWK programming language, as
      defined in the book, The AWK Programming Language, by A. V. Aho,
      B. W. Kernighan, P. J. Weinberger. They are composed of the
      following characters:</p>
 
    <taglist>
      <tag><c>c</c></tag>
      <item><p>Matches the non-metacharacter c.</p>
      </item>
      <tag><c>\\c</c></tag>
      <item><p>Matches the escape sequence or literal character c.</p>
      </item>
      <tag><c>.</c></tag>
      <item><p>Matches any character.</p>
      </item>
      <tag><c>^</c></tag>
      <item><p>Matches the beginning of a string.</p>
      </item>
      <tag><c>$</c></tag>
      <item><p>Matches the end of a string.</p></item>
      <tag><c>[abc...]</c></tag>
      <item><p>Character class, which matches any of the characters
        <c>abc...</c>. Character ranges are specified by a pair of
        characters separated by a <c>-</c>.</p>
      </item>
      <tag><c>[^abc...]</c></tag>
      <item><p>Negated character class, which matches any character
        except <c>abc...</c>.</p>
      </item>
      <tag><c>r1 | r2</c></tag>
      <item><p>Alternation. It matches either <c>r1</c> or <c>r2</c>.</p>
      </item>
      <tag><c>r1r2</c></tag>
      <item><p>Concatenation. It matches <c>r1</c> and then <c>r2</c>.</p>
      </item>
      <tag><c>r+</c></tag>
      <item><p>Matches one or more <c>rs</c>.</p>
      </item>
      <tag><c>r*</c></tag>
      <item><p>Matches zero or more <c>rs</c>.</p>
      </item>
      <tag><c>r?</c></tag>
      <item><p>Matches zero or one <c>rs</c>.</p>
      </item>
      <tag><c>(r)</c></tag>
      <item><p>Grouping. It matches <c>r</c>.</p>
      </item>
    </taglist>
 
    <p>The escape sequences allowed are the same as for Erlang strings:</p>

    <taglist>
      <tag><c>\\b</c></tag>
      <item><p>Backspace.</p></item>
      <tag><c>\\f</c></tag>
      <item><p>Form feed.</p></item>
      <tag><c>\\n</c></tag>
      <item><p>Newline (line feed).</p></item>
      <tag><c>\\r</c></tag>
      <item><p>Carriage return.</p></item>
      <tag><c>\\t</c></tag>
      <item><p>Tab.</p></item>
      <tag><c>\\e</c></tag>
      <item><p>Escape.</p></item>
      <tag><c>\\v</c></tag>
      <item><p>Vertical tab.</p></item>
      <tag><c>\\s</c></tag>
      <item><p>Space.</p></item>
      <tag><c>\\d</c></tag>
      <item><p>Delete.</p></item>
      <tag><c>\\ddd</c></tag>
      <item><p>The octal value <c>ddd</c>.</p></item>
      <tag><c>\\xhh</c></tag>
      <item><p>The hexadecimal value <c>hh</c>.</p></item>
      <tag><c>\\x{h...}</c></tag>
      <item><p>The hexadecimal value <c>h...</c>.</p></item>
      <tag><c>\\c</c></tag>
      <item><p>Any other character literally, for example <c>\\\\</c> for
        backslash, <c>\\"</c> for <c>"</c>.</p>
      </item>
    </taglist>
 
    <p>The following examples define Erlang data types:</p>
    <code> 
Atoms [a-z][0-9a-zA-Z_]*

Variables [A-Z_][0-9a-zA-Z_]*

Floats (\\+|-)?[0-9]+\\.[0-9]+((E|e)(\\+|-)?[0-9]+)?</code>
 
    <note><p>Anchoring a regular expression with <c>^</c> and <c>$</c>
      is not implemented in the current version of Leex and just
      generates a parse error.</p></note>
  </section>
</erlref>