Merge lp://staging/~gary/z3c.recipe.filetemplate/relative-paths into lp://staging/z3c.recipe.filetemplate
Status: | Needs review |
---|---|
Proposed branch: | lp://staging/~gary/z3c.recipe.filetemplate/relative-paths |
Merge into: | lp://staging/z3c.recipe.filetemplate |
Prerequisite: | lp://staging/~gary/z3c.recipe.filetemplate/cleanup |
Diff against target: |
1409 lines (+954/-192) 7 files modified
.bzrignore (+7/-0) CHANGES.txt (+37/-4) MANIFEST.in (+3/-0) setup.py (+1/-1) z3c/recipe/filetemplate/README.txt (+531/-123) z3c/recipe/filetemplate/__init__.py (+300/-56) z3c/recipe/filetemplate/tests.txt (+75/-8) |
To merge this branch: | bzr merge lp://staging/~gary/z3c.recipe.filetemplate/relative-paths |
Related bugs: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Francis J. Lacoste (community) | Approve | ||
Review via email:
|
Description of the change
This branch adds support for the buildout relative-paths option to z3c.recipe.
The approach chosen adds two features to the recipe: ``path extensions`` and ``filters``. With these features, and some magic variables, I was able to provide a workable solution to the problem of relative paths.
As noted in the MP fields, this builds on the "cleanup" branch.
Unmerged revisions
- 22. By Gary Poster
-
add comments to generated code; only log a problem in filters and dynamic options if they are not stopped because of SystemExit and KeyboardInterrupt.
- 21. By Gary Poster
-
remove comment that turned out not to be true.
- 20. By Gary Poster
-
tweak based on usage
- 19. By Gary Poster
-
add support for relative paths
- 18. By Gary Poster
-
re-commit the pertinent work from lp:~gary/z3c.recipe.filetemplate/support-system-python
For reference, this is a patch to Launchpad that takes advantage of this new feature.
http:// pastebin. ubuntu. com/418823/
After this patch and lp~gary/ zc.buildout/ python- support- 9-relative- paths, absolute paths based on the build are only found in Launchpad in three locations:
- in scripts generated by z3c.recipe.tag (would be easy to fix)
- in scripts generated by z3c.recipe.i18n (would be easy to fix)
- in code generated by our Mailman integration (would be hard to fix, but after we move to Python 2.6 Barry Warsaw may be willing and able to help us move to the new Mailman code base).