Module::Dependency::Changes - List of significant changes to Module::Dependency
No longer use umask(0) before creating dir for data file. Move db file store function from Indexer.pm to Info.pm Use $ENV{PERL_PMD_DB} to override location of data file.
Fixed blank lines in output. Fixed parent to mean what item depends_on and child to mean what item is depended_on_by Added sorting of final top-level output order with -sort Added -Unique option (global version of -unique) Added -All option to dump everything using Data::Dumper Added != and !~ support to object selector patterns Non-scripts always have a / in the key and filename (./ is prepended if there's no / already) Improved duplicate file/package warning. The -h option doesn't remove indent, use -i='' for that.
Replaced pmd_dumper.plx with much more flexible pmd_dump.pl Unresolved dependencies now get entries in allobjects, with their reverse dependencies listed so they can be queried. Because of that the grapher html output changed but the tests haven't yet been updated so they're skipped. All entries have a 'key' field matching the key of the allobjects hash.
Fixed VERSION numbers. Simplified README docs, INSTALL and added my name to LICENCE.
Stored data changes: - 'filename' is now a relative file path, not absolute. - the "key" name for scripts is now the relative file path not basename. - new 'filerootdir' holds absolute prefix of filename. - 'scripts' now holds relative file paths Ignore ~ and .bak files. Don't decend into ignored directories (.svn, .cpan etc) Attempt to recognise .t test files. Improve robustness of parsing. Give more, and more useful, warning messages. More compact output for pmd_dumper -a
First release under new maintainer Tim Bunce.
Fixed parsing of multi-line use statements. Fixed parsing of unusual =cut statements. Fixed tests to not depend on hash order. Renamed scripts to have a pmd_ prefix. Moved examples into new examples/ subdirectory. Moced pmd_cgidepend.plx and pmd_cgiform.html into examples. Unified script and module parsing code.
Date: 2002/09/25 23:09:31
Fixed minor typoes and niggling things that should have been corrected in the distribution - thanks to Mike Castle & Ron Savage
Indexer.pm converts paths to absolute paths before running File::Find, giving the correct, expected, behaviour when you try to index relative folders
Added support for the 'base' pragma.
Added the 'relationship' method (and hence the -r option to dumper.plx) which tells you how two items are related.
Improved POD-skipping algorithm - i.e. it now has one.
There should be no platform-specific code left, and in theory this will run on any OS.
Added a little POD-only module called Module::Dependency.
We 'use' as little as possible and 'require' the rest.
The indexer can optionally (and does, by default) look at a file's #! line to see if it's a perl program. .pl, .plx and .pm files are always indexed.
Added accessor routine for the 'check_shebang' switch.
cgidepend.plx, the CGI program, slightly tweaked to be even more mod_perl friendly.
Some code and ideas kindly contributed by S Rezic, S Heffner and C Sampson, among others.
To install Module::Dependency::Info, copy and paste the appropriate command in to your terminal.
cpanm
cpanm Module::Dependency::Info
CPAN shell
perl -MCPAN -e shell install Module::Dependency::Info
For more information on module installation, please visit the detailed CPAN module installation guide.