SYNOPSIS

     use 5.010;
     use Org::Parser;
     my $orgp = Org::Parser->new();
    
     # parse a file
     my $doc = $orgp->parse_file("$ENV{HOME}/todo.org");
    
     # parse a string
     $doc = $orgp->parse(<<EOF);
     #+TODO: TODO | DONE CANCELLED
     <<<radio target>>>
     * heading1a
     ** TODO heading2a
     SCHEDULED: <2011-03-31 Thu>
     [[some][link]]
     ** DONE heading2b
     [2011-03-18 ]
     this will become a link: radio target
     * TODO heading1b *bold*
     - some
     - plain
     - list
     - [ ] with /checkbox/
       * and
       * sublist
     * CANCELLED heading1c
     + definition :: list
     + another :: def
     EOF
    
     # walk the document tree
     $doc->walk(sub {
         my ($el) = @_;
         return unless $el->isa('Org::Element::Headline');
         say "heading level ", $el->level, ": ", $el->title->as_string;
     });

    will print something like:

     heading level 1: heading1a
     heading level 2: heading2a
     heading level 2: heading2b *bold*
     heading level 1: heading1b
     heading level 1: heading1c

    A command-line utility (in a separate distribution: App::OrgUtils) is
    available for debugging:

     % dump-org-structure ~/todo.org
     Document:
       Setting: "#+TODO: TODO | DONE CANCELLED\n"
       RadioTarget: "<<<radio target>>>"
       Text: "\n"
       Headline: l=1
         (title)
         Text: "heading1a"
         (children)
         Headline: l=2 todo=TODO
           (title)
           Text: "heading2a"
           (children)
           Text: "SCHEDULED: "
     ...

DESCRIPTION

    This module parses Org documents. See http://orgmode.org/ for more
    details on Org documents.

    See todo.org in the distribution for the list of already- and not yet
    implemented stuffs.

ATTRIBUTES

METHODS

 new()

    Create a new parser instance.

 $orgp->parse($str | $arrayref | $coderef | $filehandle, \%opts) => $doc

    Parse document (which can be contained in a scalar $str, an arrayref of
    lines $arrayref, a subroutine which will be called for chunks until it
    returns undef, or a filehandle).

    Returns Org::Document object.

    If 'handler' attribute is specified, will call handler repeatedly
    during parsing. See the 'handler' attribute for more details.

    Will die if there are syntax errors in documents.

    Known options:

      * time_zone => STR

      Will be passed to Org::Document's constructor.

 $orgp->parse_file($filename, \%opts) => $doc

    Just like parse(), but will load document from file instead.

    Known options (aside from those known by parse()):

      * cache => bool (default: from PERL_ORG_PARSER_CACHE, or 0)

      Since Org::Parser can spend some time to parse largish Org files,
      this is an option to store the parse result (using Storable). If
      caching is turned on, then after the first parse, the result will be
      stored in:

       ~/.cache/perl-org-parser/<filename>.<md5-digest-of-file-absolute-path>.storable

      and subsequent calls to this function can directly use this cache, as
      long as the cache is not stale.

FAQ

 Why? Just as only perl can parse Perl, only org-mode can parse Org anyway!

    True. I'm only targetting good enough. As long as I can parse/process
    all my Org notes and todo files, I have no complaints.

 It's too slow!

    Parser is completely regex-based at the moment (I plan to use Marpa
    someday). Performance is quite lousy but I'm not annoyed enough at the
    moment to overhaul it.

ENVIRONMENT

 PERL_ORG_PARSER_CACHE => bool

    Set default for cache option in parse_file().

SEE ALSO

    Org::Document