pick: remove transaction on the whole command (issue6037)
At its core, pick is a pretty straightforward and well-behaving command, it
uses functions already in core hg, it checks that wdir is clean and that
changeset to pick is not public, it checks if there happen to be merge
conflicts and can be --continue'd later, etc.
It is very similar to graft in core (it also uses mergemod.graft function), but
it obsoletes the original changeset. However, graft does not experience this
incorrect behavior from issue 6037.
What happens in the test case for this issue when we pick a revision that
touches both "a" and "b": mergemod.graft() takes the original changeset and
tries to apply it to the wdir, which results in "b" being marked as newly added
and ready to be committed, "a" updated with the new content and being marked as
modified, but "a" also has conflicts. Pick correctly notices this and saves its
state before asking for user intervention. So far so good. However, when the
command raises InterventionRequired to print a user-facing message and exit
while being wrapped in repo.transaction() context manager, the latter partially
undoes what mergemod.graft() did: it unmarks "b" as added. And when user
continues pick, "b" is therefore not tracked and is not included in the
resulting commit.
The transaction is not useful here, because it doesn't touch wdir (it's still
dirty), it doesn't remove pickstate (and other commands will refuse to work
until pick --abort or --continue), it just makes "b" untracked.
The solution is to use repo.transaction() only to wrap code that writes data to
hg store in the final stages of the command after all checks have passed and is
not expected to fail on trivial cases like merge conflicts. For example,
committing the picked changeset. But since pick uses repo.commit() for that,
and because that function already uses a transaction, wrapping it in another
transaction doesn't make sense.
This test file test the various messages when accessing obsolete
revisions.
Global setup
============
$ . $TESTDIR/testlib/obshistory_setup.sh
Test output with lots of split commit
=====================================
Test setup
----------
$ hg init $TESTTMP/local-lots-split
$ cd $TESTTMP/local-lots-split
$ mkcommit ROOT
$ echo 42 >> a
$ echo 43 >> b
$ echo 44 >> c
$ echo 45 >> d
$ hg commit -A -m "A0"
adding a
adding b
adding c
adding d
$ hg log --hidden -G
@ changeset: 1:de7290d8b885
| tag: tip
| user: test
| date: Thu Jan 01 00:00:00 1970 +0000
| summary: A0
|
o changeset: 0:ea207398892e
user: test
date: Thu Jan 01 00:00:00 1970 +0000
summary: ROOT
$ hg split -r 'desc(A0)' -d "0 0" << EOF
> y
> y
> n
> n
> n
> y
> y
> y
> n
> n
> y
> y
> y
> n
> y
> y
> y
> EOF
0 files updated, 0 files merged, 4 files removed, 0 files unresolved
adding a
adding b
adding c
adding d
diff --git a/a b/a
new file mode 100644
examine changes to 'a'? [Ynesfdaq?] y
@@ -0,0 +1,1 @@
+42
record change 1/4 to 'a'? [Ynesfdaq?] y
diff --git a/b b/b
new file mode 100644
examine changes to 'b'? [Ynesfdaq?] n
diff --git a/c b/c
new file mode 100644
examine changes to 'c'? [Ynesfdaq?] n
diff --git a/d b/d
new file mode 100644
examine changes to 'd'? [Ynesfdaq?] n
created new head
continue splitting? [Ycdq?] y
diff --git a/b b/b
new file mode 100644
examine changes to 'b'? [Ynesfdaq?] y
@@ -0,0 +1,1 @@
+43
record change 1/3 to 'b'? [Ynesfdaq?] y
diff --git a/c b/c
new file mode 100644
examine changes to 'c'? [Ynesfdaq?] n
diff --git a/d b/d
new file mode 100644
examine changes to 'd'? [Ynesfdaq?] n
continue splitting? [Ycdq?] y
diff --git a/c b/c
new file mode 100644
examine changes to 'c'? [Ynesfdaq?] y
@@ -0,0 +1,1 @@
+44
record change 1/2 to 'c'? [Ynesfdaq?] y
diff --git a/d b/d
new file mode 100644
examine changes to 'd'? [Ynesfdaq?] n
continue splitting? [Ycdq?] y
diff --git a/d b/d
new file mode 100644
examine changes to 'd'? [Ynesfdaq?] y
@@ -0,0 +1,1 @@
+45
record this change to 'd'? [Ynesfdaq?] y
no more change to split
$ hg log --hidden -G
@ changeset: 5:c7f044602e9b
| tag: tip
| user: test
| date: Thu Jan 01 00:00:00 1970 +0000
| summary: A0
|
o changeset: 4:1ae8bc733a14
| user: test
| date: Thu Jan 01 00:00:00 1970 +0000
| summary: A0
|
o changeset: 3:f257fde29c7a
| user: test
| date: Thu Jan 01 00:00:00 1970 +0000
| summary: A0
|
o changeset: 2:337fec4d2edc
| parent: 0:ea207398892e
| user: test
| date: Thu Jan 01 00:00:00 1970 +0000
| summary: A0
|
| x changeset: 1:de7290d8b885
|/ user: test
| date: Thu Jan 01 00:00:00 1970 +0000
| obsolete: split using split as 2:337fec4d2edc, 3:f257fde29c7a, 4:1ae8bc733a14, 5:c7f044602e9b
| summary: A0
|
o changeset: 0:ea207398892e
user: test
date: Thu Jan 01 00:00:00 1970 +0000
summary: ROOT
Actual test
-----------
$ hg obslog de7290d8b885 --hidden --patch
x de7290d8b885 (1) A0
rewritten(parent, content) as 1ae8bc733a14, 337fec4d2edc, c7f044602e9b, f257fde29c7a using split by test (Thu Jan 01 00:00:00 1970 +0000)
(No patch available, too many successors (4))
$ hg obslog de7290d8b885 --hidden --all --patch
o 1ae8bc733a14 (4) A0
|
| o 337fec4d2edc (2) A0
|/
| @ c7f044602e9b (5) A0
|/
| o f257fde29c7a (3) A0
|/
x de7290d8b885 (1) A0
rewritten(parent, content) as 1ae8bc733a14, 337fec4d2edc, c7f044602e9b, f257fde29c7a using split by test (Thu Jan 01 00:00:00 1970 +0000)
(No patch available, too many successors (4))
$ hg obslog de7290d8b885 --hidden --no-graph -Tjson | python -m json.tool
[
{
"markers": [
{
"date": [
*, (glob)
0 (glob)
],
"effect": [
"parent",
"content"
],
"operation": "split",
"succnodes": [
"1ae8bc733a14",
"337fec4d2edc",
"c7f044602e9b",
"f257fde29c7a"
],
"user": "test",
"verb": "rewritten"
}
],
"node": "de7290d8b885",
"rev": 1,
"shortdescription": "A0"
}
]
$ hg obslog c7f044602e9b --patch
@ c7f044602e9b (5) A0
|
x de7290d8b885 (1) A0
rewritten(parent, content) as 1ae8bc733a14, 337fec4d2edc, c7f044602e9b, f257fde29c7a using split by test (Thu Jan 01 00:00:00 1970 +0000)
(No patch available, too many successors (4))
$ hg obslog c7f044602e9b --no-graph -Tjson | python -m json.tool
[
{
"markers": [],
"node": "c7f044602e9b",
"rev": 5,
"shortdescription": "A0"
},
{
"markers": [
{
"date": [
*, (glob)
0 (glob)
],
"effect": [
"parent",
"content"
],
"operation": "split",
"succnodes": [
"1ae8bc733a14",
"337fec4d2edc",
"c7f044602e9b",
"f257fde29c7a"
],
"user": "test",
"verb": "rewritten"
}
],
"node": "de7290d8b885",
"rev": 1,
"shortdescription": "A0"
}
]
Check that debugobshistory on all heads show a coherent graph
$ hg obslog 2::5 --patch
o 1ae8bc733a14 (4) A0
|
| o 337fec4d2edc (2) A0
|/
| @ c7f044602e9b (5) A0
|/
| o f257fde29c7a (3) A0
|/
x de7290d8b885 (1) A0
rewritten(parent, content) as 1ae8bc733a14, 337fec4d2edc, c7f044602e9b, f257fde29c7a using split by test (Thu Jan 01 00:00:00 1970 +0000)
(No patch available, too many successors (4))
$ hg obslog 5 --all --patch
o 1ae8bc733a14 (4) A0
|
| o 337fec4d2edc (2) A0
|/
| @ c7f044602e9b (5) A0
|/
| o f257fde29c7a (3) A0
|/
x de7290d8b885 (1) A0
rewritten(parent, content) as 1ae8bc733a14, 337fec4d2edc, c7f044602e9b, f257fde29c7a using split by test (Thu Jan 01 00:00:00 1970 +0000)
(No patch available, too many successors (4))
$ hg update de7290d8b885
abort: hidden revision 'de7290d8b885' was split as: 337fec4d2edc, f257fde29c7a and 2 more!
(use --hidden to access hidden revisions)
[255]
$ hg update --hidden 'min(desc(A0))'
0 files updated, 0 files merged, 0 files removed, 0 files unresolved
updated to hidden changeset de7290d8b885
(hidden revision 'de7290d8b885' was split as: 337fec4d2edc, f257fde29c7a and 2 more)
working directory parent is obsolete! (de7290d8b885)
(use 'hg evolve' to update to its tipmost successor: 337fec4d2edc, f257fde29c7a and 2 more)