amend: query the wdir parent after taking lock (issue5266)
If we query wdir parent without taking a lock, that can lead to bugs because the
wdir parent can change is another process has changed the wdir parent.
One such example of this was issue 5266. When a user is running amend and that
amend is waiting for commit message, the user runs another amend which waits for
lock. The second amend is waiting for lock, but has already read the description
from the parent of working directory to use. Once the first amend completes the
wdir parent changes but we still have the description from an old wdir parent.
This patch fixes the bug by querying the description after taking lock.
Attempts were made to add test for this but the results were unstable because
they depend on time in which lock is released.
Testing `hg stack` on complex cases when we have multiple successors because of
divergence, split etc.
$ . "$TESTDIR/testlib/topic_setup.sh"
Setup
$ cat << EOF >> $HGRCPATH
> [experimental]
> evolution = all
> [ui]
> interactive = True
> [extensions]
> show =
> EOF
$ echo "evolve=$(echo $(dirname $TESTDIR))/hgext3rd/evolve/" >> $HGRCPATH
$ hg init test
$ cd test
$ echo foo > foo
$ hg add foo
$ hg ci -m "Added foo"
$ hg phase -r . --public
$ hg topic foo
marked working directory as topic: foo
$ echo a > a
$ echo b > b
$ hg ci -Aqm "Added a and b"
$ echo c > c
$ echo d > d
$ hg ci -Aqm "Added c and d"
$ echo e > e
$ echo f > f
$ hg ci -Aqm "Added e and f"
$ hg show work
@ f1d3 (foo) Added e and f
o 8e82 (foo) Added c and d
o 002b (foo) Added a and b
o f360 Added foo
Testing in case of split within the topic
$ hg stack
### topic: foo
### target: default (branch)
t3@ Added e and f (current)
t2: Added c and d
t1: Added a and b
t0^ Added foo (base)
$ hg prev
0 files updated, 0 files merged, 2 files removed, 0 files unresolved
[2] Added c and d
$ echo 0 > num
$ cat > editor.sh << '__EOF__'
> NUM=$(cat num)
> NUM=`expr "$NUM" + 1`
> echo "$NUM" > num
> echo "split$NUM" > "$1"
> __EOF__
$ export HGEDITOR="\"sh\" \"editor.sh\""
$ hg split << EOF
> y
> y
> n
> y
> EOF
0 files updated, 0 files merged, 2 files removed, 0 files unresolved
adding c
adding d
diff --git a/c b/c
new file mode 100644
examine changes to 'c'? [Ynesfdaq?] y
@@ -0,0 +1,1 @@
+c
record change 1/2 to 'c'? [Ynesfdaq?] y
diff --git a/d b/d
new file mode 100644
examine changes to 'd'? [Ynesfdaq?] n
Done splitting? [yN] y
1 new orphan changesets
$ hg stack
### topic: foo
### target: default (branch)
t4$ Added e and f (unstable)
t3@ split2 (current)
t2: split1
t1: Added a and b
t0^ Added foo (base)
$ hg show work
@ 5cce (foo) split2
o f26c (foo) split1
| * f1d3 (foo) Added e and f
| x 8e82 (foo) Added c and d
|/
o 002b (foo) Added a and b
o f360 Added foo
$ hg prev
0 files updated, 0 files merged, 1 files removed, 0 files unresolved
[4] split1
$ echo foo > c
$ hg diff
diff -r f26c1b9addde c
--- a/c Thu Jan 01 00:00:00 1970 +0000
+++ b/c Thu Jan 01 00:00:00 1970 +0000
@@ -1,1 +1,1 @@
-c
+foo
$ hg amend
1 new orphan changesets
$ hg show work
@ 7d94 (foo) split1
| * 5cce (foo) split2
| x f26c (foo) split1
|/
| * f1d3 (foo) Added e and f
| x 8e82 (foo) Added c and d
|/
o 002b (foo) Added a and b
o f360 Added foo
$ hg stack
### topic: foo (2 heads)
### target: default (branch), 2 behind
t4$ Added e and f (unstable)
t3$ split2 (unstable)
t2@ split1 (current)
t1: Added a and b
t0^ Added foo (base)