=head1 Module::Dependency README Overview -------- This is a suite of modules (and a set of helper programs that you may use if you wish) that allow you to investigate the dependencies between perl programs and modules. This is primarily aimed at the programmer/ developer audience. Not just for modules --- ---- --- ------- The system is split into two halves - gathering the data, and examining the data - so you are free to create a database (mentioned below) however you want based on whatever metrics you want and query that using the ::Info and ::Grapher modules. Or you may choose to build your own tools for querying the database. In any case these modules work with heirarchies of many-to-many parent-child relationships, so if your data can be viewed like that maybe some of the code here could be useful. What are dependencies? ---- --- ------------- When your foo.pl program says 'use CGI;' that program /depends/ upon CGI.pm - I'll refer to this as 'CGI is a child dependency of foo.pl', and similarly 'foo.pl is a parent dependency of CGI'. These modules work with all the parent and child dependencies between a collection of perl files that you specify. How does it work? --- ---- -- ----- Basically the first thing you do is create a database by feeding a list of directories to Module::Dependency::Indexer::makeIndex() These directories are recursively searched using File::Find to examine all the .pl, .plx and .pm files. Parent and child dependencies are stored, as are a few other bits of information. One database is created for all the files found in that run. Making the database is quite expensive so the idea is to create the database once, and use it many times. Typical values are 5-10 seconds to index a reasonably large site_perl directory. You can create many different database files should you need to. You use functions in Module::Dependency::Info to get information out of the database. You use functions in Module::Dependency::Grapher to visualize the links between items. It can output in plain text, HTML fragment, PostScript, Encapsulated PostScript, SVG or any format your copy of GD supports. In using GD for output, the available output formats depend upon your local copy of GD, but PNG should be available on all installations. GIF and JPEG are also possible. This module can also produce basic plain text output and HTML fragment output. See the README.EXAMPLES file. The Grapher output is fast enough to use in a CGI application, even with a database containing many hundreds of items. It was written with the intention of going on an intranet software engineers support site. A working CGI is included. See README.EXAMPLES. Installation ------------ Read the INSTALL file. You can probably get started using the programs bundled with this package - see README.COMMANDS - so you don't have to write any code. For more advanced or specialist usages you'll want to know how to use the modules - see README.MODULES for that. For examples see the README.EXAMPLES file. Improvements ------------ Please tell me about any bugs you find, or any suggestions for improvements: http://rt.cpan.org/Public/Dist/Display.html?Name=Module-Dependency Copyright and License --------- --- ------- This software is copyright 2001-2002 P Kent, and copyright 2006 Tim Bunce. This is free software; you can redistribute it and/or modify it under the same terms as Perl itself. =head1 Bundled Commands This readme should give you an overview of the commands bundled with this distribution. Please see the README.MODULES file for more behind-the-scenes information. See the README for an overview. pmd_indexer.plx --------------- Creates a database file holding dependency information for the source files in one or more directories. pmd_dump.pl -------------- Extracts information from the database and displays it on standard output. E.g. parent dependencies, child dependencies, actual filesystem path, all items indexed, etc. pmd_grapher.plx --------------- Generates a graphical 'dependency tree'. It can map parent, child or both directions of dependency, and it can be told which items to plot dependencies for (e.g. all the perl scripts, or a particular script or module). You may wish to edit the bit near the top of the source to make it create the right format of image. All these programs may be run with the -h option to get a usage message, and with the -t option to get all the tracing messages so that you can see what's going on under the bonnet. examples/pmd_cgidepend.plx -------------------------- An example CGI interface. You'll need to create a database before this will do anything useful - use pmd_indexer.plx for that. You'll need to edit a few lines in this CGI - they are clearly marked near the top. Run it with no query string to get the help message. Takes these CGI params: go, embed, format, datafile, seed, allscripts, kind. See its POD for more detail. An example installation is at: http://www.selsyn.co.uk/depend.shtml =head1 Module::Dependency Modules This readme should give you an idea of what the modules actually do in this distribution. Module::Dependency::Indexer --------------------------- This module takes care of creating the database file. Module::Dependency::Grapher --------------------------- Uses GD to plot graphical dependency information. Can also produce simple plain text output, PostScript, SVG, or HTML fragment output. Module::Dependency::Info ------------------------ This module is used as the interface to query the database and get information out. =head1 TODO - make items be real objects with methods etc - use overloading to stringify as key - move some pmd_dump.pl subs into object methods - abstract the Module::Dependency::Indexer more - Look at using Graph::Easy =cut