[Pythonmac-SIG] Bundlebuilder--why remove it?

[Ronald Oussoren]

On 10 Dec, 2009, at 0:26, Christopher Barker wrote:

> Ronald Oussoren wrote:
> 
>> What really should be done is restart py2app development,
> 
> Unless PyInstaller is looking good -- there may be an advantage to a code base with more people working on it. Mac folks would still need to do the Mac-specific stuff, but not the rest of it. bbFreeze isn't bad, either, but needs more work -- I don't think it ever could produce an App bundle.

PyInstaller is GPL licensed and I am therefore not interested.


> 
>> starting  with automated tests
> 
> What did you have in mind here? I can't quite image what unit-tests to write, but maybe a suite of test cases fro various packages that could as least be auto run on various platforms.

There probably should be unittests at some point, but I'm more interested in functional/acceptence test style tests where every testcase builds an application bundle (or plugin bundle) and verifies that the result is correct. The verification step should idealy run the application to check that it works instead of only checking if files are present in the bundle.

> 
> There is a start to this in the samples dir -- so maybe we "just" need a test runner.
> 
> > and improved egg support.
> 
> by the way, I think bbFreeze has extended modulegraph -- partly for better egg support.
> 
> and I suppose quad architecture support -- Kevin, isn't that your show-stopper?

Quad architecture is pretty low on my list, I have the files from Kevin to reproduce the issue but little time to work on py2app. 

Ronald

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3567 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/pythonmac-sig/attachments/20091210/fc67a1e0/attachment-0001.bin>