Author: cmpilato
Date: Tue May 17 09:43:52 2011
New Revision: 1104086
URL: http://svn.apache.org/viewvc?rev=1104086&view=rev
Log:
Cast some votes.
Modified:
subversion/branches/1.6.x/STATUS
Modified: subversion/branches/1.6.x/STATUS
URL:
http://svn.apache.org/viewvc/subversion/branches/1.6.x/STATUS?rev=1104086&r1=1104085&r2=1104086&view=diff
==============================================================================
--- subversion/branches/1.6.x/STATUS (original)
+++ subversion/branches/1.6.x/STATUS Tue May 17 09:43:52 2011
@@ -45,7 +45,7 @@ Candidate changes:
Branch:
^/subversion/branches/1.6.x-r917523
Votes:
- +1: kameshj (with r878590 and r916286 groups)
+ +1: kameshj, cmpilato
* r1028084, r1028125, r1041638
Fix JVM recognition on OS X Snow Leopard (10.6).
@@ -64,64 +64,12 @@ Candidate changes:
didn't do a line-by-line review of the changes
+0: danielsh (only touches java.m4)
- * 1.6.x-svn_fs_commit_txn branch
- Have all of server-side Subversion, from svn_fs_commit_txn(),
- through to mod_dav_svn, implement the documented (since 1.0.x)
- behavior that svn_fs_commit_txn() and svn_repos_fs_commit_txn()
- should indicate a successful commit by a valid returned revision
- number, not by any returned error. Now, regardless if
- svn_fs_commit_txn()'s post commit FS processing (that's its new
- official name) fails, svn_repos_fs_commit_txn() will run the
- post-commit hook. All code that uses svn_fs_commit_txn() and/or
- svn_repos_fs_commit_txn() now uses the revision number to test if
- the commit succeeded.
- Justification:
- Fixes bug reports on [email protected] where if post commit FS
- processing fails, the client reports an error but the commit
- succeeded, which is confusing for users.
- Implementing the documented and correct behavior is a good thing.
- Notes:
- For a successful unit test pass, it requires that r1051632,
- r1051638 and r1051751 listed above be merged previously.
- The branch has many commits in it from trunk, which includes some
- churn from reviews on trunk. I recommend reviewing a cumulative
- diff:
- svn diff -r r1053420:r1053500
^/subversion/branches/1.6.x-svn_fs_commit_txn
- This merge doesn't address the client side behavior.
- Branch:
- ^/subversion/branches/1.6.x-svn_fs_commit_txn
- Note:
- Adds a new private API, which would break 1.6.16 mod_dav_svn with
- 1.6.15 libsvn_repos; thread:
- http://mid.gmane.org/[email protected]
- http://mid.gmane.org/[email protected]
- Note:
- Found a potential issue in close_edit() during review, I would prefer to
- see it fixed before merge.
- Votes:
- +1: blair, danielsh
-
* ^/branches/1.6.x-r1053984
Fix bug where committing a changelist was prevented by a file outside that
changelist having unexpectedly changed special status.
Justification:
Local fix. users@ reports.
Votes:
- +0: danielsh
-
- * 1.6.x-fsfs-begin-txn-deadlock
- Fix a deadlock that can occur when two or more multithreaded
- Subversion servers on the same system serve two or more fsfs
- repositories.
- Justification:
- Avoiding deadlocks is a good thing.
- Branch:
- ^/subversion/branches/1.6.x-fsfs-begin-txn-deadlock
- Thread:
- http://mid.gmane.org/[email protected]
- http://mid.gmane.org/[email protected]
- Votes:
- +1: blair, danielsh
* r962377, r962378, r1036978, r1037762, r1063572, r1063573, r1063592
Fix issue #3641 svnsync handling of directory copyfrom.
@@ -138,9 +86,8 @@ Candidate changes:
Notes:
See also the r1036429 group.
Votes:
- +1: cmpilato (r962377, r962378 only)
- +1: philip (r962377, r962378, r1063572, r1063573, r1063592 only)
- +1: pburba
+ +1: philip (962377, r962378, r1063572, r1063573, r1063592 only)
+ +1: pburba, cmpilato
-0: danielsh (authz concerns on the original fix)
* r1074572
@@ -423,3 +370,56 @@ Approved changes:
that are currently nominated...so my +1 is conditional on
that fix being applied first.)
+1: danielsh (pburba: a fix to #3641 was committed to trunk)
+
+
+ * 1.6.x-svn_fs_commit_txn branch
+ Have all of server-side Subversion, from svn_fs_commit_txn(),
+ through to mod_dav_svn, implement the documented (since 1.0.x)
+ behavior that svn_fs_commit_txn() and svn_repos_fs_commit_txn()
+ should indicate a successful commit by a valid returned revision
+ number, not by any returned error. Now, regardless if
+ svn_fs_commit_txn()'s post commit FS processing (that's its new
+ official name) fails, svn_repos_fs_commit_txn() will run the
+ post-commit hook. All code that uses svn_fs_commit_txn() and/or
+ svn_repos_fs_commit_txn() now uses the revision number to test if
+ the commit succeeded.
+ Justification:
+ Fixes bug reports on [email protected] where if post commit FS
+ processing fails, the client reports an error but the commit
+ succeeded, which is confusing for users.
+ Implementing the documented and correct behavior is a good thing.
+ Notes:
+ For a successful unit test pass, it requires that r1051632,
+ r1051638 and r1051751 listed above be merged previously.
+ The branch has many commits in it from trunk, which includes some
+ churn from reviews on trunk. I recommend reviewing a cumulative
+ diff:
+ svn diff -r r1053420:r1053500
^/subversion/branches/1.6.x-svn_fs_commit_txn
+ This merge doesn't address the client side behavior.
+ Branch:
+ ^/subversion/branches/1.6.x-svn_fs_commit_txn
+ Note:
+ Adds a new private API, which would break 1.6.16 mod_dav_svn with
+ 1.6.15 libsvn_repos; thread:
+ http://mid.gmane.org/[email protected]
+ http://mid.gmane.org/[email protected]
+ Note:
+ Found a potential issue in close_edit() during review, I would prefer to
+ see it fixed before merge.
+ Votes:
+ +1: blair, danielsh, cmpilato
+
+ * 1.6.x-fsfs-begin-txn-deadlock
+ Fix a deadlock that can occur when two or more multithreaded
+ Subversion servers on the same system serve two or more fsfs
+ repositories.
+ Justification:
+ Avoiding deadlocks is a good thing.
+ Branch:
+ ^/subversion/branches/1.6.x-fsfs-begin-txn-deadlock
+ Thread:
+ http://mid.gmane.org/[email protected]
+ http://mid.gmane.org/[email protected]
+ Votes:
+ +1: blair, danielsh, cmpilato
+
