FUJIWARA Katsunori <foozy@lares.dti.ne.jp> [Thu, 08 Oct 2015 01:41:30 +0900] rev 26520
bookmarks: use recordchange instead of writing if transaction is active
Before this patch, 'bmstore.write()' always write in-memory bookmark
changes into '.hg/bookmarks' regardless of transaction activity.
If 'bmstore.write()' is invoked inside a transaction and it writes
changes into '.hg/bookmarks', then:
- original bookmarks aren't restored at failure of that transaction
This breaks "all or nothing" policy of the transaction.
BTW, "hg rollback" can restore bookmarks successfully even before
this patch, because original bookmarks are saved into
'.hg/journal.bookmarks' at the beginning of the transaction, and
it (actually renamed as '.hg/undo.bookmarks') is used by "hg
rollback".
- uncommitted bookmark changes are visible to other processes
This is a kind of "dirty read"
For example, 'rebase.rebase()' implies 'bmstore.write()', and it may
be executed inside the transaction of "hg unshelve". Then, intentional
aborting at the end of "hg unshelve" transaction doesn't restore
original bookmarks (this is obviously a bug).
This patch uses 'bmstore.recordchange()' instead of actual writing by
'bmstore._writerepo()', if any transaction is active
This patch also removes meaningless restoring bmstore explicitly at
the end of "hg shelve".
This patch doesn't choose fixing each 'bmstore.write()' callers as
like below, because writing similar code here and there is very
redundant.
before:
bmstore.write()
after:
tr = repo.currenttransaction()
if tr:
bmstore.recordchange(tr)
else:
bmstore.write()
Even though 'bmstore.write()' itself may have to be discarded by
putting bookmark operations into transaction scope, this patch chose
fixing it to implement "transactional dirstate" at first.
Siddharth Agarwal <sid0@fb.com> [Wed, 07 Oct 2015 00:27:23 -0700] rev 26519
filemerge: run symlink check for :merge3
Just like :merge, :merge3 doesn't support merging symlinks.
Siddharth Agarwal <sid0@fb.com> [Wed, 07 Oct 2015 00:24:44 -0700] rev 26518
filemerge: print correct name of tool for symlink checks
Earlier we'd print ':merge' even if the tool was something else like ':union'.
That's clearly a bug.
Siddharth Agarwal <sid0@fb.com> [Wed, 07 Oct 2015 00:01:16 -0700] rev 26517
filemerge: normalize 'internal:foo' names to ':foo'
In upcoming patches we're going to present these names in the UI -- it would be
good not to present deprecated names.
Siddharth Agarwal <sid0@fb.com> [Tue, 06 Oct 2015 22:57:21 -0700] rev 26516
filemerge: use symlinkcheck for :merge and :union
This exposes a couple of bugs, both of which will be fixed in upcoming patches.
Siddharth Agarwal <sid0@fb.com> [Tue, 06 Oct 2015 22:56:33 -0700] rev 26515
filemerge: add a precheck for symlinks
This will be used by internal merge tools.
Siddharth Agarwal <sid0@fb.com> [Tue, 06 Oct 2015 22:55:21 -0700] rev 26514
filemerge: call precheck if available
In upcoming patches we'll define a precheck function for some merge tools.
Siddharth Agarwal <sid0@fb.com> [Tue, 06 Oct 2015 22:54:14 -0700] rev 26513
filemerge: add a before-merge callback to internal merge tools
We're going to separate the pre-merge and merge steps for merge tools. The
merge step will be specific to the tool, but the pre-merge step will be common
to all merge tools that need it.
However, some merge tools run checks *before* the pre-merge step. This callback
will allow that to continue to work.
Siddharth Agarwal <sid0@fb.com> [Tue, 06 Oct 2015 17:39:13 -0700] rev 26512
filemerge: indent filemerge.filemerge
This will make upcoming patches much easier to review.
Pierre-Yves David <pierre-yves.david@fb.com> [Thu, 01 Oct 2015 20:15:00 -0700] rev 26511
test-bundle-type: replace unbundle with debugbundle
We now have a convenient command to look at bundle contents, let's use
it.