diff options
Diffstat (limited to 'PLAN.txt')
| -rw-r--r-- | PLAN.txt | 28 |
1 files changed, 23 insertions, 5 deletions
@@ -184,7 +184,7 @@ features. Run the tests. Fix bugs. Repeat until satisfied. Note: this may interact with the branch integration task. -Project: integration with main branch +Project: integration with main branch *** This is done - tim *** ************************************* Tasks: @@ -217,7 +217,7 @@ Done: To undo or rename before final merge: Modules/spam.c has worked its way into the branch Unix and Windows builds (pythoncore.dsp and PC/config.c); also imported by test_descr.py. How about renaming to -xxsubtype.c (whatever) now? +xxsubtype.c (whatever) now? *** this is done - tim *** Project: performance tuning @@ -280,17 +280,35 @@ Resolved. Rebuilt from scratch; ran all tests; checked into branch about 22:40. -Merged descr-branch back into trunk: +Merged descr-branch back into trunk (SEE BELOW -- this specific way of +doing it was a bad idea): + cvs -q -z3 up -j descr-branch python 34 conflicts. Hmm! OK, looks like every file in the project with an embedded RCS Id is "a conflict". Others make no sense, e.g., a dozen conflicts in dictobject.c, sometimes enclosing identical(!) blobs of source code. And CVS remains utterly baffled by Python type object decls. -Every line of ceval.c's generator code si in conflict blocks ... OK, +Every line of ceval.c's generator code is in conflict blocks ... OK, there's no pattern or sense here, I'll just deal with it. -Conflicts resolved; rebuilt from scratch; test_weakref fails. +Conflicts resolved; rebuilt from scratch; test_weakref fails. Didn't find +an obvious reason and it was late, so committed it anyway. Tagged the +trunk then with tag: + + after-descr-branch-merge + +Tracked the test_weakref failure to a botched conflict resolution in +classobject.c; checked in a fix. + +LATER: The merge should have been done via: + + upd -j date2001-08-01 -j descr-branch python + +instead. This would have caused only one conflict, a baffler in +bltinmodule.c. It would have avoided the classobject.c error I made. +Luckily, except for that one, we got to the same place in the end anyway, +apart from a few curious tabs-vs-spaces differences. ---------------------------------------------------------------------------- 2001-07-30 |