blob: 4824301ff4e83bbc5560119dbdda0cfc7b3156ed [file] [log] [blame]
* * * * * * * * * * * * * * * * * * * * * * * * * * * *
* *
* THIS RELEASE STREAM IS OPEN TO BUG FIXES. *
* *
* * * * * * * * * * * * * * * * * * * * * * * * * * * *
This file tracks the status of releases in the 1.7.x line.
See http://subversion.apache.org/docs/community-guide/releasing.html#release-stabilization
for details on how release lines and voting work, what kinds of bugs can
delay a release, etc.
Status of 1.7.0:
Candidate changes:
==================
* r1146131 and followups
Add svn_fs_verify() and implement rep-cache verification in FSFS using it.
Branch:
1.7.x-fs-verify
Justification:
Would be nice to extend verify already in 1.7.x.
Notes:
This does not include progress reporting; that has been postponed to 1.8.
Conflicts:
On fs_fs.c. Resolve with 'theirs-conflict'.
Votes:
+1: danielsh, rhuijben
+0: gstein (progress would be nice) (without r1161164, r1161180)
-0: cmpilato (For a function with this name, the implementation is far
too anemic. I'd prefer this be a private function in 1.7.x, leaving
room for a future public API that really does full verification.)
(without r1161164, r1161180)
* r1147540, r1147541
Remove unused variables in build system.
Justification:
Backporting of future changes will be easier.
Votes:
+1: arfrever
-0: stsp (we can backport these when we need to)
-0: gstein
* r1154278, r1154379, r1154382
Make 'svn ls' capable of listing Subversion 1.0-1.3 repositories using
ra_serf.
Justification:
We promised not to break backwards compatibility and it is not nice
that serf doesn't want to list the neon repository.
(Check svn ls http://svn.webdav.org/repos/projects/neon/)
Notes:
r1154278 is a quick hack to detect old servers and to at least return
some information.
r1154379 is the real fix: fallback to a more compatible request.
r1154382 makes sure the right result is used directly after falling
back from svn_ra_serf__get_dir.
(Most likely the problem is detected earlier than this function)
Votes:
+1: rhuijben, cmpilato
* r1147299, r1162974, r1162995
Fix issue #3867 'reintegrate merges create mergeinfo for non-existent
paths' and issue #3961 'reintegrate merge of replaced source produces
inaccurate mergeinfo'.
Justification:
Avoid potential confusion which can occur when a reintegrate merge
sets mergeinfo (to describe the merge) which contains non-existent
path-revs, other lines of history, and/or valid path-revs which were
never merged.
Notes:
r1147299 and r1162995 is the addition and tweak of a test for
issue #3961. A test already existed for issue #3867 prior to branching
1.7.x. r1162974 is the fix for both issues.
Votes:
+1: pburba, stsp
* r1163243, r1163383
Fix issue #3791 'svn mergeinfo shows natural history of added subtrees
as eligible'.
Justification:
The 'svn mergeinfo [ -R | --depth [ files | immediates | infinity ] ]'
subcommand gives incorrect answers in some use cases.
Notes:
r1163243 is a test for the issue, r1163383 is the fix.
Votes:
+1: pburba
* r1163557
Relax wcng consistency checks for file externals.
Justification:
Debug build claims wcng DB is inconsistent when it is OK.
Branch:
1.7.x-r1163557
Votes:
+1: philip, rhuijben
+0: neels (Much other code still dismally fails at file externals in
unversioned dirs, e.g. 'status', 'commit'. This apparently
only fixes maintainer-mode 'svn cleanup', IMHO low prio.)
* r1164386
Avoid an expensive database operation in 'svn info'.
Justification:
Improves performance of 'svn info' on a single unmodified node by 20-30%, by
moving an in most cases unneeded database call in the error handling.
Votes:
+1: rhuijben
+1: gstein (for 1.7.1)
* r1164426, r1164614
Fix STMT_SELECT_NODE_CHILDREN_WALKER_INFO for correctness and speed.
Justification:
Recursive info and propset don't scale well with large working copies
and may possibly act on the wrong nodes.
On a Linux machine with a Subversion trunk working copy on a local disk
running 'svn info --depth infinity' shows it to be 2 or 3 times faster.
On a larger working copy it's orders of magnitude faster.
Votes:
+1: philip
+0: peters, gstein (looks correct, but I didn't test, and while
performance should be better, I wonder if that's
been tested at all)
Veto-blocked changes:
=====================
Approved changes:
=================