- SOURCE AVAILABILITY
- TO DO
- COPYRIGHT AND LICENSE
- SEE ALSO
Mac::PropertyList - work with Mac plists at a low level
use Mac::PropertyList; my $data = parse_plist( $text ); # == OR == my $data = parse_plist_file( $filename ); # == OR == open my( $fh ), $filename or die "..."; my $data = parse_plist_fh( $fh ); my $text = plist_as_string( $data ); my $plist = create_from_hash( \%hash ); my $plist = create_from_array( \@array ); my $plist = Mac::PropertyList::dict->new( \%hash );
This module is a low-level interface to the Mac OS X Property List (plist) format. You probably shouldn't use this in applications---build interfaces on top of this so you don't have to put all the heinous multi-level object stuff where people have to look at it.
You can parse a plist file and get back a data structure. You can take that data structure and get back the plist as XML. If you want to change the structure inbetween that's your business. :)
You don't need to be on Mac OS X to use this. It simply parses and manipulates a text format that Mac OS X uses.
The MacOS X Property List format is simple XML. You can read the DTD to get the details.
One big problem exists---its dict type uses a flat structure to list keys and values so that values are only associated with their keys by their position in the file rather than by the structure of the DTD. This problem is the major design hinderance in this module. A smart XML format would have made things much easier.
If the parse_plist encounters an empty key tag in a dict structure (i.e.
<key></key> ) the function croaks.
A plist can have one or more of any of the plist objects, and we have to remember the type of thing so we can go back to the XML format. Perl treats numbers and strings the same, but the plist format doesn't.
Therefore, everything Mac::PropertyList creates is an object of some sort. Container objects like Mac::PropertyList::array and Mac::PropertyList::dict hold other objects.
There are several types of objects:
Mac::PropertyList::string Mac::PropertyList::data Mac::PropertyList::real Mac::PropertyList::integer Mac::PropertyList::date Mac::PropertyList::array Mac::PropertyList::dict
- new( VALUE )
Create the object.
Access the value of the object. At the moment you cannot change the value
Access the type of the object (string, data, etc)
Create a string version of the object, recursively if necessary.
These functions are available for individual or group import. Nothing will be imported unless you ask for it.
use Mac::PropertyList qw( parse_plist ); use Mac::PropertyList qw( :all );
- parse_plist( TEXT )
Parse the XML plist in TEXT and return the Mac::PropertyList object.
- parse_plist_fh( FILEHANDLE )
Parse the XML plist from FILEHANDLE and return the Mac::PropertyList data structure. Returns false if the arguments is not a reference.
You can do this in a couple of ways. You can open the file with a lexical filehandle (since Perl 5.6).
open my( $fh ), $file or die "..."; parse_plist_fh( $fh );
Or, you can use a bareword filehandle and pass a reference to its typeglob. I don't recommmend this unless you are using an older Perl.
open FILE, $file or die "..."; parse_plist_fh( \*FILE );
- parse_plist_file( FILE_PATH )
Parse the XML plist in FILE_PATH and return the Mac::PropertyList data structure. Returns false if the file does not exist.
Alternately, you can pass a filehandle reference, but that just calls
- create_from_hash( HASH_REF )
Create a plist dictionary from the hash reference.
The values of the hash can only be simple scalars---not references. Reference values are silently ignored.
Returns a string representing the hash in the plist format.
- create_from_array( ARRAY_REF )
Create a plist array from the array reference.
The values of the array can only be simple scalars---not references. Reference values are silently ignored.
Returns a string representing the array in the plist format.
Turn the plist data structure into the plist string, which is the form the Mac likes to see it in.
This source is part of a SourceForge project which always has the latest sources in SVN, as well as all of the previous releases.
If, for some reason, I disappear from the world, one of the other members of the project can shepherd this module appropriately.
Thanks to Chris Nandor for general Mac kung fu and Chad Walker for help figuring out the recursion for nested structures.
Mike Ciul provided some classes for the different input modes, and these allow us to optimize the parsing code for each of those.
Ricardo Signes added the as_basic_types() methods so you can all the plist junk and just play with the data.
* change the value of an object
* validate the values of objects (date, integer)
* methods to add to containers (dict, array)
* do this from a filehandle or a scalar reference instead of a scalar + generate closures to handle the work.
brian d foy,
Copyright (c) 2004-2008 brian d foy. All rights reserved.
This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.