NAME

       "LaTeXML::Mouth" - tokenize the input.

DESCRIPTION

       A "LaTeXML::Mouth" (and subclasses) is responsible for tokenizing, ie.  converting plain
       text and strings into LaTeXML::Tokens according to the current category codes (catcodes)
       stored in the "LaTeXML::State".

       "LaTeXML::FileMouth"
           specializes "LaTeXML::Mouth" to tokenize from a file.

       "LaTeXML::StyleMouth"
           further specializes "LaTeXML::FileMouth" for processing style files, setting the
           catcode for "@" and ignoring comments.

       "LaTeXML::PerlMouth"
           is not really a Mouth in the above sense, but is used to definitions from perl modules
           with exensions ".ltxml" and ".latexml".

   Creating Mouths
       "$mouth = LaTeXML::Mouth->new($string);"
           Creates a new Mouth reading from $string.

       "$mouth = LaTeXML::FileMouth->new($pathname);"
           Creates a new FileMouth to read from the given file.

       "$mouth = LaTeXML::StyleMouth->new($pathname);"
           Creates a new StyleMouth to read from the given style file.

   Methods
       "$token = $mouth->readToken;"
           Returns the next LaTeXML::Token from the source.

       "$boole = $mouth->hasMoreInput;"
           Returns whether there is more data to read.

       "$string = $mouth->getLocator($long);"
           Return a description of current position in the source, for reporting errors.

       "$tokens = $mouth->readTokens($until);"
           Reads tokens until one matches $until (comparing the character, but not catcode).
           This is useful for the "\verb" command.

       "$lines = $mouth->readRawLines($endline,$exact);"
           Reads raw (untokenized) lines from $mouth until a line matching $endline is found.  If
           $exact is true, $endline is matched exactly, with no leading or trailing data (like in
           the c<comment> package).  Otherwise, the match is done like with the c<verbatim>
           environment; any text preceding $endline is returned as the last line, and any
           characters after $endline remains in the mouth to be tokenized.

AUTHOR

       Bruce Miller <bruce.miller@nist.gov>

COPYRIGHT

       Public domain software, produced as part of work done by the United States Government &
       not subject to copyright in the US.