view tests/test-rebase-abort.t @ 25757:4d1382fd96ff

context: write dirstate out explicitly at the end of markcommitted

To detect change of a file without redundant comparison of file
content, dirstate recognizes a file as certainly clean, if:

  (1) it is already known as "normal",
  (2) dirstate entry for it has valid (= not "-1") timestamp, and
  (3) mode, size and timestamp of it on the filesystem are as same as
      ones expected in dirstate

This works as expected in many cases, but doesn't in the corner case
that changing a file keeps mode, size and timestamp of it on the
filesystem.

The timetable below shows steps in one of typical such situations:

  ---- ----------------------------------- ----------------
                                           timestamp of "f"
                                           ----------------
                                           dirstate   file-
  time          action                     mem  file  system
  ---- ----------------------------------- ---- ----- -----
   *                                             ***    ***
       - 'hg transplant REV1 REV2 ...'
         - transplanting REV1
           ....
   N
           - change "f", but keep size                   N
             (via 'patch.patch()')
           - 'dirstate.normal("f")'          N   ***
             (via 'repo.commit()')

         - transplanting REV2
           - change "f", but keep size                   N
             (via 'patch.patch()')
           - aborted while patching
  N+1
         - release wlock
           - 'dirstate.write()'              N    N      N

       - 'hg status' shows "r1" as "clean"   N    N      N
  ---- ----------------------------------- ---- ----- -----

The most important point is that 'dirstate.write()' is executed at N+1
or later. This causes writing dirstate timestamp N of "f" out
successfully. If it is executed at N, 'parsers.pack_dirstate()'
replaces timestamp N with "-1" before actual writing dirstate out.

This issue can occur when 'hg transplant' satisfies conditions below:

  - multiple revisions to be transplanted change the same file
  - those revisions don't change mode and size of the file, and
  - the 2nd or later revision of them fails after changing the file

The root cause of this issue is that files are changed without
flushing in-memory dirstate changes via 'repo.commit()' (even though
omitting 'dirstate.normallookup()' on files changed by 'patch.patch()'
for efficiency also causes this issue).

To detect changes of files correctly, this patch writes in-memory
dirstate changes out explicitly after marking files as clean in
'committablectx.markcommitted()', which is invoked via
'repo.commit()'.

After this change, timetable is changed as below:

  ---- ----------------------------------- ----------------
                                           timestamp of "f"
                                           ----------------
                                           dirstate   file-
  time          action                     mem  file  system
  ---- ----------------------------------- ---- ----- -----
   *                                             ***    ***
       - 'hg transplant REV1 REV2 ...'
         - transplanting REV1
           ....
   N
           - change "f", but keep size                   N
             (via 'patch.patch()')
           - 'dirstate.normal("f")'          N    ***
             (via 'repo.commit()')
       ----------------------------------- ---- ----- -----
           - 'dirsttate.write()'            -1    -1
       ----------------------------------- ---- ----- -----
         - transplanting REV2
           - change "f", but keep size                   N
             (via 'patch.patch()')
           - aborted while patching
  N+1
         - release wlock
           - 'dirstate.write()'              -1   -1     N

       - 'hg status' shows "r1" as "clean"   -1   -1     N
  ---- ----------------------------------- ---- ----- -----

To reproduce this issue in tests certainly, this patch emulates some
timing critical actions as below:

  - change "f" at N

    'patch.patch()' with 'fakepatchtime.py' explicitly changes mtime
    of patched files to "2000-01-01 00:00" (= N).

  - 'dirstate.write()' via 'repo.commit()' at N

    'fakedirstatewritetime.py' forces 'pack_dirstate()' to use
    "2000-01-01 00:00" as "now", only if 'pack_dirstate()' is invoked
    via 'committablectx.markcommitted()'.

  - 'dirstate.write()' via releasing wlock at N+1 (or "not at N")

    'pack_dirstate()' via releasing wlock uses actual timestamp at
    runtime as "now", and it should be different from the "2000-01-01
    00:00" of "f".

BTW, this patch doesn't test cases below, even though 'patch.patch()'
is used similarly in these cases:

  1. failure of 'hg import' or 'hg qpush'

  2. success of 'hg import', 'hg qpush' or 'hg transplant'

Case (1) above doesn't cause this kind of issue, because:

  - if patching is aborted by conflicts, changed files are committed

    changed files are marked as CLEAN, even though they are partially
    patched.

  - otherwise, dirstate are fully restored by 'dirstateguard'

    For example in timetable above, timestamp of "f" in .hg/dirstate
    is restored to -1 (or less than N), and subsequent 'hg status' can
    detect changes correctly.

Case (2) always causes 'repo.status()' invocation via 'repo.commit()'
just after changing files inside same wlock scope.

  ---- ----------------------------------- ----------------
                                           timestamp of "f"
                                           ----------------
                                           dirstate   file-
  time          action                     mem  file  system
  ---- ----------------------------------- ---- ----- -----
  N                                              ***   ***
       - make file "f" clean                            N

       - execute 'hg foobar'
         ....
         - 'dirstate.normal("f")'           N    ***
           (e.g. via dirty check
            or previous 'repo.commit()')

         - change "f", but keep size                    N
         - 'repo.status()' (*1)
           (via 'repo.commit()')
  ---- ----------------------------------- ---- ----- -----

At a glance, 'repo.status()' at (*1) seems to cause similar issue (=
"changed files are treated as clean"), but actually doesn't.

'dirstate._lastnormaltime' should be N at (*1) above, because
'dirstate.normal()' via dirty check is finished at N.

Therefore, "f" changed at N (= 'dirstate._lastnormaltime') is forcibly
treated as "unsure" at (*1), and changes are detected as expected (see
'dirstate.status()' for detail).

If 'hg import' is executed with '--no-commit', 'repo.status()' isn't
invoked just after changing files inside same wlock scope.

But preceding 'dirstate.normal()' is invoked inside another wlock
scope via 'cmdutil.bailifchanged()', and in-memory changes should be
flushed at the end of that scope.

Therefore, timestamp N of clean "f" should be replaced by -1, if
'dirstate.write()' is invoked at N. It means that condition of this
issue isn't satisfied.

  $ cat >> $HGRCPATH <<EOF
  > [extensions]
  > rebase=
  > 
  > [phases]
  > publish=False
  > 
  > [alias]
  > tglog = log -G --template "{rev}:{phase} '{desc}' {branches}\n"
  > EOF


  $ hg init a
  $ cd a

  $ touch .hg/rebasestate
  $ hg sum
  parent: -1:000000000000 tip (empty repository)
  branch: default
  commit: (clean)
  update: (current)
  abort: .hg/rebasestate is incomplete
  [255]
  $ rm .hg/rebasestate

  $ echo c1 > common
  $ hg add common
  $ hg ci -m C1

  $ echo c2 >> common
  $ hg ci -m C2

  $ echo c3 >> common
  $ hg ci -m C3

  $ hg up -q -C 1

  $ echo l1 >> extra
  $ hg add extra
  $ hg ci -m L1
  created new head

  $ sed -e 's/c2/l2/' common > common.new
  $ mv common.new common
  $ hg ci -m L2

  $ hg phase --force --secret 2

  $ hg tglog
  @  4:draft 'L2'
  |
  o  3:draft 'L1'
  |
  | o  2:secret 'C3'
  |/
  o  1:draft 'C2'
  |
  o  0:draft 'C1'
  

Conflicting rebase:

  $ hg rebase -s 3 -d 2
  rebasing 3:3163e20567cc "L1"
  rebasing 4:46f0b057b5c0 "L2" (tip)
  merging common
  warning: conflicts during merge.
  merging common incomplete! (edit conflicts, then use 'hg resolve --mark')
  unresolved conflicts (see hg resolve, then hg rebase --continue)
  [1]

Abort:

  $ hg rebase --abort
  saved backup bundle to $TESTTMP/a/.hg/strip-backup/3e046f2ecedb-6beef7d5-backup.hg (glob)
  rebase aborted

  $ hg tglog
  @  4:draft 'L2'
  |
  o  3:draft 'L1'
  |
  | o  2:secret 'C3'
  |/
  o  1:draft 'C2'
  |
  o  0:draft 'C1'
  
Test safety for inconsistent rebase state, which may be created (and
forgotten) by Mercurial earlier than 2.7. This emulates Mercurial
earlier than 2.7 by renaming ".hg/rebasestate" temporarily.

  $ hg rebase -s 3 -d 2
  rebasing 3:3163e20567cc "L1"
  rebasing 4:46f0b057b5c0 "L2" (tip)
  merging common
  warning: conflicts during merge.
  merging common incomplete! (edit conflicts, then use 'hg resolve --mark')
  unresolved conflicts (see hg resolve, then hg rebase --continue)
  [1]

  $ mv .hg/rebasestate .hg/rebasestate.back
  $ hg update --quiet --clean 2
  $ hg --config extensions.mq= strip --quiet "destination()"
  $ mv .hg/rebasestate.back .hg/rebasestate

  $ hg rebase --continue
  abort: cannot continue inconsistent rebase
  (use "hg rebase --abort" to clear broken state)
  [255]
  $ hg summary | grep '^rebase: '
  rebase: (use "hg rebase --abort" to clear broken state)
  $ hg rebase --abort
  rebase aborted (no revision is removed, only broken state is cleared)

  $ cd ..


Construct new repo:

  $ hg init b
  $ cd b

  $ echo a > a
  $ hg ci -Am A
  adding a

  $ echo b > b
  $ hg ci -Am B
  adding b

  $ echo c > c
  $ hg ci -Am C
  adding c

  $ hg up -q 0

  $ echo b > b
  $ hg ci -Am 'B bis'
  adding b
  created new head

  $ echo c1 > c
  $ hg ci -Am C1
  adding c

  $ hg phase --force --secret 1
  $ hg phase --public 1

Rebase and abort without generating new changesets:

  $ hg tglog
  @  4:draft 'C1'
  |
  o  3:draft 'B bis'
  |
  | o  2:secret 'C'
  | |
  | o  1:public 'B'
  |/
  o  0:public 'A'
  
  $ hg rebase -b 4 -d 2
  rebasing 3:a6484957d6b9 "B bis"
  note: rebase of 3:a6484957d6b9 created no changes to commit
  rebasing 4:145842775fec "C1" (tip)
  merging c
  warning: conflicts during merge.
  merging c incomplete! (edit conflicts, then use 'hg resolve --mark')
  unresolved conflicts (see hg resolve, then hg rebase --continue)
  [1]

  $ hg tglog
  @  4:draft 'C1'
  |
  o  3:draft 'B bis'
  |
  | @  2:secret 'C'
  | |
  | o  1:public 'B'
  |/
  o  0:public 'A'
  
  $ hg rebase -a
  rebase aborted

  $ hg tglog
  @  4:draft 'C1'
  |
  o  3:draft 'B bis'
  |
  | o  2:secret 'C'
  | |
  | o  1:public 'B'
  |/
  o  0:public 'A'
  

  $ cd ..

rebase abort should not leave working copy in a merge state if tip-1 is public
(issue4082)

  $ hg init abortpublic
  $ cd abortpublic
  $ echo a > a && hg ci -Aqm a
  $ hg book master
  $ hg book foo
  $ echo b > b && hg ci -Aqm b
  $ hg up -q master
  $ echo c > c && hg ci -Aqm c
  $ hg phase -p -r .
  $ hg up -q foo
  $ echo C > c && hg ci -Aqm C
  $ hg log -G --template "{rev} {desc} {bookmarks}"
  @  3 C foo
  |
  | o  2 c master
  | |
  o |  1 b
  |/
  o  0 a
  

  $ hg rebase -d master -r foo
  rebasing 3:6c0f977a22d8 "C" (tip foo)
  merging c
  warning: conflicts during merge.
  merging c incomplete! (edit conflicts, then use 'hg resolve --mark')
  unresolved conflicts (see hg resolve, then hg rebase --continue)
  [1]
  $ hg rebase --abort
  rebase aborted
  $ hg log -G --template "{rev} {desc} {bookmarks}"
  @  3 C foo
  |
  | o  2 c master
  | |
  o |  1 b
  |/
  o  0 a
  
  $ cd ..

Make sure we don't clobber changes in the working directory when the
user has somehow managed to update to a different revision (issue4009)

  $ hg init noupdate
  $ cd noupdate
  $ hg book @
  $ echo original > a
  $ hg add a
  $ hg commit -m a
  $ echo x > b
  $ hg add b
  $ hg commit -m b1
  $ hg up 0
  0 files updated, 0 files merged, 1 files removed, 0 files unresolved
  (leaving bookmark @)
  $ hg book foo
  $ echo y > b
  $ hg add b
  $ hg commit -m b2
  created new head

  $ hg rebase -d @ -b foo --tool=internal:fail
  rebasing 2:070cf4580bb5 "b2" (tip foo)
  unresolved conflicts (see hg resolve, then hg rebase --continue)
  [1]

  $ mv .hg/rebasestate ./ # so we're allowed to hg up like in mercurial <2.6.3
  $ hg up -C 0            # user does other stuff in the repo
  0 files updated, 0 files merged, 1 files removed, 0 files unresolved

  $ mv rebasestate .hg/   # user upgrades to 2.7

  $ echo new > a
  $ hg up 1               # user gets an error saying to run hg rebase --abort
  abort: rebase in progress
  (use 'hg rebase --continue' or 'hg rebase --abort')
  [255]

  $ cat a
  new
  $ hg rebase --abort
  rebase aborted
  $ cat a
  new

  $ cd ..

On the other hand, make sure we *do* clobber changes whenever we
haven't somehow managed to update the repo to a different revision
during a rebase (issue4661)

  $ hg ini yesupdate
  $ cd yesupdate
  $ echo "initial data" > foo.txt
  $ hg add
  adding foo.txt
  $ hg ci -m "initial checkin"
  $ echo "change 1" > foo.txt
  $ hg ci -m "change 1"
  $ hg up 0
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ echo "conflicting change 1" > foo.txt
  $ hg ci -m "conflicting 1"
  created new head
  $ echo "conflicting change 2" > foo.txt
  $ hg ci -m "conflicting 2"

  $ hg rebase -d 1 --tool 'internal:fail'
  rebasing 2:e4ea5cdc9789 "conflicting 1"
  unresolved conflicts (see hg resolve, then hg rebase --continue)
  [1]
  $ hg rebase --abort
  rebase aborted
  $ hg summary
  parent: 3:b16646383533 tip
   conflicting 2
  branch: default
  commit: (clean)
  update: 1 new changesets, 2 branch heads (merge)
  phases: 4 draft