My program can currently: * read Portage's configuration files and environment variables * parse an ebuild (file that describes the way a package is built and installed) to get the dependency information * parse the dependency string taking into account USE flags (determine which optional components get built for certain packages) * take a dependency string such as 'x11-base/xfree', '>=net-fs/samba-2.3', etc. and return a list of candidate versions in the Portage tree (directory structure containing ~80,000 ebuilds taking into account masked packages * determine the highest version of a particular package from a list of available package versions * determine if a certain version of a certain package is already installed
I plan on writing a function that will get the dependency information for a certain package, check the dependencies for those dependencies, and so on down the line. I'd have a structure like:
mainpackage-1.0 |--dependency1-1.3 | ---dep1_of_dep1 |--dependency2-2.3 | ---dep1_of_dep2 | ---dep2_of_dep2 | ---dep3_of_dep2 |--dependency3
where each dependency is a string in the form of 'x11-base/xfree', '>=net-fs/samba-2.3', etc. What I can't figure out is how to take that structure and turn it into a list that looks something like:
dep1_of_dep1 dependency1-1.3 dep1_of_dep2 dep2_of_dep2 dep3_of_dep2 dependency2-2.3 dependency3 mainpackage-1.0
Generating the above would be pretty simple, but it isn't so simple to take into account packages that have dependencies on different version ranges than another package (one package having '>somepackage-3.4' and another '<somepackage-4.5') and all that other fun dependency stuff. Can anyone give me a few pointers?
-- Andrew Gaffney Network Administrator Skyline Aeronautics, LLC. 636-357-1548
-- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] <http://learn.perl.org/> <http://learn.perl.org/first-response>
