On 19 Aug, 2010, at 20:04, Virgil Dupras wrote: > On Thu, Aug 19, 2010 at 12:02 PM, Ronald Oussoren > <ronaldousso...@mac.com> wrote: >> >> Option to use the built-in modulefinder instead of modulegraph: I was >> getting tired of having to write workaround imports to modulegraph >> (which doesn't like relative imports very well, even the latest >> version) and I didn't want to dig in modulegraph itself, so as an >> experiment, I added a "--use-modulefinder" option to py2app. I haven't >> tested it much, but so far, I'm able to build one of my apps *without* >> the workarounds I had for modulegraph, so it my case, it works better. >> >> >> This won't get merged, I'd much rather fix modulegraph. If that doesn't work >> out for some reason modulegraph should get ripped out, choosing between the >> two is unnecessary complexity for both the maintainers and users of py2app. >> One of the things on my todolist for py2app and related modules is to add >> proper tests, which should make it easier to ensure that bugs get fixed and >> stay fixed. >> And then there are the mac-related bugs in CPython itself, which also need >> my attention. >> Ronald >> > > Sure. I didn't implement this in the hope that it was merged, it was > more of a way to experiment how modulegraph and modulefinder compared > when I ran it on my apps.
There is something that would help trying to fix the problem your having with modulegraph: Create simple programs that reproduce the problem(s). Ronald
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ Pythonmac-SIG maillist - Pythonmac-SIG@python.org http://mail.python.org/mailman/listinfo/pythonmac-sig unsubscribe: http://mail.python.org/mailman/options/Pythonmac-SIG