view tests/test-rebase-abort.t @ 49777:e1953a34c110

bundle: emit full snapshot as is, without doing a redelta

With the new `forced` delta-reused policy, it become important to be able to
send full snapshot where full snapshot are needed. Otherwise, the fallback delta
will simply be used on the client side… creating monstrous delta chain, since
revision that are meant as a reset of delta-chain chain becoming too complex are
simply adding a new full delta-tree on the leaf of another one.

In the `non-forced` cases, client process full snapshot from the bundle
differently from deltas, so client will still try to convert the full snapshot
into a delta if possible. So this will no lead to pathological storage
explosion.

I have considered making this configurable, but the impact seems limited enough
that it does not seems to be worth it. Especially with the current
sparse-revlog format that use "delta-tree" with multiple level snapshots, full
snapshot are much less frequent and not that different from other intermediate
snapshot that we are already sending over the wire anyway.

CPU wise, this will help the bundling side a little as it will not need to
reconstruct revisions and compute deltas. The unbundling side might save a tiny
amount of CPU as it won't need to reconstruct the delta-base to reconstruct the
revision full text. This only slightly visible in some of the benchmarks. And
have no real impact on most of them.

### data-env-vars.name      = pypy-2018-08-01-zstd-sparse-revlog
  # benchmark.name          = perf-bundle
  # benchmark.variants.revs = last-40000
before:          11.467186 seconds
just-emit-full:  11.190576 seconds   (-2.41%)
with-pull-force: 11.041091 seconds   (-3.72%)
  # benchmark.name          = perf-unbundle
  # benchmark.variants.revs = last-40000
before:     16.744862
just-emit-full:: 16.561036 seconds   (-1.10%)
with-pull-force: 16.389344 seconds   (-2.12%)
  # benchmark.name          = pull
  # benchmark.variants.revs = last-40000
before:     26.870569
just-emit-full:  26.391188 seconds   (-1.78%)
with-pull-force: 25.633184 seconds   (-4.60%)


Space wise (so network-wise) the impact is fairly small. When taking compression into
account.

Below are tests the size of `hg bundle --all` for a handful of benchmark repositories
(with bzip, zstd compression and without it)

This show a small increase in the bundle size, but nothing really significant
except maybe for mozilla-try (+12%) that nobody really pulls large chunk of anyway.
Mozilla-try is also the repository that benefit the most for not having to
recompute deltas client size.

### mercurial:
bzip-before:     26 406 342 bytes
bzip-after:      26 691 543 bytes  +1.08%

zstd-before:     27 918 645 bytes
zstd-after:      28 075 896 bytes  +0.56%

none-before:     98 675 601 bytes
none-after:     100 411 237 bytes  +1.76%


### pypy
bzip-before:    201 295 752 bytes
bzip-after:     209 780 282 bytes  +4.21%

zstd-before:    202 974 795 bytes
zstd-after:     205 165 780 bytes  +1.08%

none-before:    871 070 261 bytes
none-after:     993 595 057 bytes +14.07%


### netbeans
bzip-before:    601 314 330 bytes
bzip-after:     614 246 241 bytes  +2.15%

zstd-before:    604 745 136 bytes
zstd-after:     615 497 705 bytes  +1.78%

none-before:  3 338 238 571 bytes
none-after:   3 439 422 535 bytes  +3.03%


### mozilla-central
bzip-before:  1 493 006 921 bytes
bzip-after:   1 549 650 570 bytes  +3.79%

zstd-before:  1 481 910 102 bytes
zstd-after:   1 513 052 415 bytes  +2.10%

none-before:  6 535 929 910 bytes
none-after:   7 010 191 342 bytes  +7.26%


### mozilla-try
bzip-before:  6 583 425 999 bytes
bzip-after:   7 423 536 928 bytes +12.76%

zstd-before:  6 021 009 212 bytes
zstd-after:   6 674 922 420 bytes +10.86%

none-before: 22 954 739 558 bytes
none-after:  26 013 854 771 bytes +13.32%

#testcases abortcommand abortflag
#testcases continuecommand continueflag

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

#if abortflag
  $ cat >> $HGRCPATH <<EOF
  > [alias]
  > abort = rebase --abort
  > EOF
#endif

#if continueflag
  $ cat >> $HGRCPATH <<EOF
  > [alias]
  > continue = rebase --continue
  > EOF
#endif

  $ 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 tip "L2"
  merging common
  warning: conflicts while merging common! (edit, then use 'hg resolve --mark')
  unresolved conflicts (see 'hg resolve', then 'hg rebase --continue')
  [240]

Insert unsupported advisory merge record:

  $ hg --config extensions.fakemergerecord=$TESTDIR/fakemergerecord.py fakemergerecord -x
  $ hg debugmergestate
  local (dest): 3e046f2ecedb793b97ed32108086edd1a162f8bc
  other (source): 46f0b057b5c061d276b91491c22151f78698abd2
  file: common (state "u")
    local path: common (hash 94c8c21d08740f5da9eaa38d1f175c592692f0d1, flags "")
    ancestor path: common (node de0a666fdd9c1a0b0698b90d85064d8bd34f74b6)
    other path: common (node 2f6411de53677f6f1048fef5bf888d67a342e0a5)
    extra: ancestorlinknode = 3163e20567cc93074fbb7a53c8b93312e59dbf2c
    extra: merged = yes
  $ hg resolve -l
  U common

Insert unsupported mandatory merge record:

  $ hg --config extensions.fakemergerecord=$TESTDIR/fakemergerecord.py fakemergerecord -X
  $ hg debugmergestate
  abort: unsupported merge state records: X
  (see https://mercurial-scm.org/wiki/MergeStateRecords for more information)
  [255]
  $ hg resolve -l
  abort: unsupported merge state records: X
  (see https://mercurial-scm.org/wiki/MergeStateRecords for more information)
  [255]
  $ hg resolve -ma
  abort: unsupported merge state records: X
  (see https://mercurial-scm.org/wiki/MergeStateRecords for more information)
  [255]

Abort (should clear out unsupported merge state):

#if abortcommand
when in dry-run mode
  $ hg abort --dry-run
  rebase in progress, will be aborted
#endif

  $ hg abort
  saved backup bundle to $TESTTMP/a/.hg/strip-backup/3e046f2ecedb-6beef7d5-backup.hg
  rebase aborted
  $ hg debugmergestate
  no merge state found

  $ 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 tip "L2"
  merging common
  warning: conflicts while merging common! (edit, then use 'hg resolve --mark')
  unresolved conflicts (see 'hg resolve', then 'hg rebase --continue')
  [240]

  $ 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 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 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: not rebasing 3:a6484957d6b9 "B bis", its destination already has all its changes
  rebasing 4:145842775fec tip "C1"
  merging c
  warning: conflicts while merging c! (edit, then use 'hg resolve --mark')
  unresolved conflicts (see 'hg resolve', then 'hg rebase --continue')
  [240]

  $ 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 foo tip "C"
  merging c
  warning: conflicts while merging c! (edit, then use 'hg resolve --mark')
  unresolved conflicts (see 'hg resolve', then 'hg rebase --continue')
  [240]
  $ hg 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 foo tip "b2"
  unresolved conflicts (see 'hg resolve', then 'hg rebase --continue')
  [240]

  $ 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', 'hg rebase --abort', or 'hg rebase --stop')
  [20]

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

  $ cd ..

test aborting an interrupted series (issue5084)
  $ hg init interrupted
  $ cd interrupted
  $ touch base
  $ hg add base
  $ hg commit -m base
  $ touch a
  $ hg add a
  $ hg commit -m a
  $ echo 1 > a
  $ hg commit -m 1
  $ touch b
  $ hg add b
  $ hg commit -m b
  $ echo 2 >> a
  $ hg commit -m c
  $ touch d
  $ hg add d
  $ hg commit -m d
  $ hg co -q 1
  $ hg rm a
  $ hg commit -m no-a
  created new head
  $ hg co 0
  0 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg log -G --template "{rev} {desc} {bookmarks}"
  o  6 no-a
  |
  | o  5 d
  | |
  | o  4 c
  | |
  | o  3 b
  | |
  | o  2 1
  |/
  o  1 a
  |
  @  0 base
  
  $ hg --config extensions.n=$TESTDIR/failfilemerge.py rebase -s 3 -d tip
  rebasing 3:3a71550954f1 "b"
  rebasing 4:e80b69427d80 "c"
  abort: ^C
  [255]

New operations are blocked with the correct state message

  $ find .hg -name '*state' -prune | sort
  .hg/dirstate
  .hg/merge/state
  .hg/rebasestate
  .hg/undo.backup.dirstate
  .hg/undo.dirstate
  .hg/updatestate

  $ hg rebase -s 3 -d tip
  abort: rebase in progress
  (use 'hg rebase --continue', 'hg rebase --abort', or 'hg rebase --stop')
  [20]
  $ hg up .
  abort: rebase in progress
  (use 'hg rebase --continue', 'hg rebase --abort', or 'hg rebase --stop')
  [20]
  $ hg up -C .
  abort: rebase in progress
  (use 'hg rebase --continue', 'hg rebase --abort', or 'hg rebase --stop')
  [20]

  $ hg graft 3
  abort: rebase in progress
  (use 'hg rebase --continue', 'hg rebase --abort', or 'hg rebase --stop')
  [20]

  $ hg abort
  saved backup bundle to $TESTTMP/interrupted/.hg/strip-backup/3d8812cf300d-93041a90-backup.hg
  rebase aborted
  $ hg log -G --template "{rev} {desc} {bookmarks}"
  o  6 no-a
  |
  | o  5 d
  | |
  | o  4 c
  | |
  | o  3 b
  | |
  | o  2 1
  |/
  o  1 a
  |
  @  0 base
  
  $ hg summary
  parent: 0:df4f53cec30a 
   base
  branch: default
  commit: (clean)
  update: 6 new changesets (update)
  phases: 7 draft

  $ 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')
  [240]
  $ hg 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
  $ cd ..

test aborting a rebase succeeds after rebasing with skipped commits onto a
public changeset (issue4896)

  $ hg init succeedonpublic
  $ cd succeedonpublic
  $ echo 'content' > root
  $ hg commit -A -m 'root' -q

set up public branch
  $ echo 'content' > disappear
  $ hg commit -A -m 'disappear public' -q
commit will cause merge conflict on rebase
  $ echo '' > root
  $ hg commit -m 'remove content public' -q
  $ hg phase --public

setup the draft branch that will be rebased onto public commit
  $ hg up -r 0 -q
  $ echo 'content' > disappear
commit will disappear
  $ hg commit -A -m 'disappear draft' -q
  $ echo 'addedcontADDEDentadded' > root
commit will cause merge conflict on rebase
  $ hg commit -m 'add content draft' -q

  $ hg rebase -d 'public()' --tool :merge -q
  note: not rebasing 3:0682fd3dabf5 "disappear draft", its destination already has all its changes
  warning: conflicts while merging root! (edit, then use 'hg resolve --mark')
  unresolved conflicts (see 'hg resolve', then 'hg rebase --continue')
  [240]
  $ hg abort
  rebase aborted
  $ cd ..