-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hanno Schlichting wrote:
Tres Seaver wrote:
Hanno Schlichting wrote:
In general all tests pass now, when run from the eggified Zope in: svn://svn.zope.org/repos/main/Zope2.buildout/trunk I just vendor imported mechanize 0.1.9 into the Zope SVN, and have changed the svn:external in my Z2 trunk checkout to point to it. I then made a separate tree and vendor imported ClientForm 0.2.9, including hacking on the faked-out package we are using to support svn:externals, and switched the external to point to it.
At this point, the 'aqlegacy_ftest' doctest still fails in my trunk sandbox, as does the 'zope.testborwser.README' doctest.
Hhm, the zope.testbrowser tests aren't run in the buildout version. It doesn't run tests for egg dependencies.
I tried this again and now that aqlegacy_ftest fails for me again as well. Either a test interdependency or a timing issue I'd guess.
My hypothesis is that one or more of the 'svn:externals' changes Sidnei made to bring in 2.6-compatible packages is the culprit (maybe an interaction between several of them). I am attaching readable diff for each of 'lib/python', 'lib/python/zope', and 'lib/python/zope/app'. As an aside / vent: the reason for the now-removed EXTERNALS.txt files was to keep the canonical information about the externals in a diffable file: why subversion can't do a proper diff on its own line-oriented property is beyond me. Another benefit of an EXTERNALS.txt file was that it could be inspected in the web view of a directory, which isn't true for the svn:externals property itself. Tres. - -- =================================================================== Tres Seaver +1 540-429-0999 tseaver@palladion.com Palladion Software "Excellence by Design" http://palladion.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFJN/NL+gerLs4ltQ4RApouAJ0WUYuUh3btoXQaLNNBLISqXcVFJACgyhsB YP0uIear89z8b5qP2x6F6Gs= =slqV -----END PGP SIGNATURE-----