view tests/test-rebase-conflicts.t @ 40326:fed697fa1734

sqlitestore: file storage backend using SQLite

This commit provides an extension which uses SQLite to store file
data (as opposed to revlogs).

As the inline documentation describes, there are still several
aspects to the extension that are incomplete. But it's a start.
The extension does support basic clone, checkout, and commit
workflows, which makes it suitable for simple use cases.

One notable missing feature is support for "bundlerepos." This is
probably responsible for the most test failures when the extension
is activated as part of the test suite.

All revision data is stored in SQLite. Data is stored as zstd
compressed chunks (default if zstd is available), zlib compressed
chunks (default if zstd is not available), or raw chunks (if
configured or if a compressed delta is not smaller than the raw
delta). This makes things very similar to revlogs.

Unlike revlogs, the extension doesn't yet enforce a limit on delta
chain length. This is an obvious limitation and should be addressed.
This is somewhat mitigated by the use of zstd, which is much faster
than zlib to decompress.

There is a dedicated table for storing deltas. Deltas are stored
by the SHA-1 hash of their uncompressed content. The "fileindex" table
has columns that reference the delta for each revision and the base
delta that delta should be applied against. A recursive SQL query
is used to resolve the delta chain along with the delta data.

By storing deltas by hash, we are able to de-duplicate delta storage!
With revlogs, the same deltas in different revlogs would result in
duplicate storage of that delta. In this scheme, inserting the
duplicate delta is a no-op and delta chains simply reference the
existing delta.

When initially implementing this extension, I did not have
content-indexed deltas and deltas could be duplicated across files
(just like revlogs). When I implemented content-indexed deltas, the
size of the SQLite database for a full clone of mozilla-unified
dropped:

before: 2,554,261,504 bytes
after:  2,488,754,176 bytes

Surprisingly, this is still larger than the bytes size of revlog
files:

revlog files: 2,104,861,230 bytes
du -b:        2,254,381,614

I would have expected storage to be smaller since we're not limiting
delta chain length and since we're using zstd instead of zlib. I
suspect the SQLite indexes and per-column overhead account for the
bulk of the differences. (Keep in mind that revlog uses a 64-byte
packed struct for revision index data and deltas are stored without
padding. Aside from the 12 unused bytes in the 32 byte node field,
revlogs are pretty efficient.) Another source of overhead is file
name storage. With revlogs, file names are stored in the filesystem.
But with SQLite, we need to store file names in the database. This is
roughly equivalent to the size of the fncache file, which for the
mozilla-unified repository is ~34MB.

Since the SQLite database isn't append-only and since delta chains
can reference any delta, this opens some interesting possibilities.
For example, we could store deltas in reverse, such that fulltexts
are stored for newer revisions and deltas are applied to reconstruct
older revisions. This is likely a more optimal storage strategy for
version control, as new data tends to be more frequently accessed
than old data. We would obviously need wire protocol support for
transferring revision data from newest to oldest. And we would
probably need some kind of mechanism for "re-encoding" stores. But
it should be doable.

This extension is very much experimental quality. There are a handful
of features that don't work. It probably isn't suitable for day-to-day
use. But it could be used in limited cases (e.g. read-only checkouts
like in CI). And it is also a good proving ground for alternate
storage backends. As we continue to define interfaces for all things
storage, it will be useful to have a viable alternate storage backend
to see how things shake out in practice.

test-storage.py passes on Python 2 and introduces no new test failures on
Python 3. Having the storage-level unit tests has proved to be insanely
useful when developing this extension. Those tests caught numerous bugs
during development and I'm convinced this style of testing is the way
forward for ensuring alternate storage backends work as intended. Of
course, test coverage isn't close to what it needs to be. But it is
a start. And what coverage we have gives me confidence that basic store
functionality is implemented properly.

Differential Revision: https://phab.mercurial-scm.org/D4928

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

  $ hg init a
  $ cd a
  $ 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

  $ echo l3 >> extra2
  $ hg add extra2
  $ hg ci -m L3
  $ hg bookmark mybook

  $ hg phase --force --secret 4

  $ hg tglog
  @  5:secret 'L3'  mybook
  |
  o  4:secret 'L2'
  |
  o  3:draft 'L1'
  |
  | o  2:draft 'C3'
  |/
  o  1:draft 'C2'
  |
  o  0:draft 'C1'
  
Try to call --continue:

  $ hg rebase --continue
  abort: no rebase in progress
  [255]

Conflicting rebase:

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

  $ hg status --config commands.status.verbose=1
  M common
  ? common.orig
  # The repository is in an unfinished *rebase* state.
  
  # Unresolved merge conflicts:
  # 
  #     common
  # 
  # To mark files as resolved:  hg resolve --mark FILE
  
  # To continue:    hg rebase --continue
  # To abort:       hg rebase --abort
  

Try to continue without solving the conflict:

  $ hg rebase --continue
  abort: unresolved merge conflicts (see 'hg help resolve')
  [255]

Conclude rebase:

  $ echo 'resolved merge' >common
  $ hg resolve -m common
  (no more unresolved files)
  continue: hg rebase --continue
  $ hg rebase --continue
  already rebased 3:3163e20567cc "L1" as 3e046f2ecedb
  rebasing 4:46f0b057b5c0 "L2"
  rebasing 5:8029388f38dc "L3" (mybook)
  saved backup bundle to $TESTTMP/a/.hg/strip-backup/3163e20567cc-5ca4656e-rebase.hg

  $ hg tglog
  @  5:secret 'L3'  mybook
  |
  o  4:secret 'L2'
  |
  o  3:draft 'L1'
  |
  o  2:draft 'C3'
  |
  o  1:draft 'C2'
  |
  o  0:draft 'C1'
  
Check correctness:

  $ hg cat -r 0 common
  c1

  $ hg cat -r 1 common
  c1
  c2

  $ hg cat -r 2 common
  c1
  c2
  c3

  $ hg cat -r 3 common
  c1
  c2
  c3

  $ hg cat -r 4 common
  resolved merge

  $ hg cat -r 5 common
  resolved merge

Bookmark stays active after --continue
  $ hg bookmarks
   * mybook                    5:d67b21408fc0

  $ cd ..

Check that the right ancestors is used while rebasing a merge (issue4041)

  $ hg init issue4041
  $ cd issue4041
  $ hg unbundle "$TESTDIR/bundles/issue4041.hg"
  adding changesets
  adding manifests
  adding file changes
  added 11 changesets with 8 changes to 3 files (+1 heads)
  new changesets 24797d4f68de:2f2496ddf49d (11 drafts)
  (run 'hg heads' to see heads)
  $ hg up default
  2 files updated, 0 files merged, 0 files removed, 0 files unresolved
  $ hg log -G
  o    changeset:   10:2f2496ddf49d
  |\   branch:      f1
  | |  tag:         tip
  | |  parent:      7:4c9fbe56a16f
  | |  parent:      9:e31216eec445
  | |  user:        szhang
  | |  date:        Thu Sep 05 12:59:39 2013 -0400
  | |  summary:     merge
  | |
  | o  changeset:   9:e31216eec445
  | |  branch:      f1
  | |  user:        szhang
  | |  date:        Thu Sep 05 12:59:10 2013 -0400
  | |  summary:     more changes to f1
  | |
  | o    changeset:   8:8e4e2c1a07ae
  | |\   branch:      f1
  | | |  parent:      2:4bc80088dc6b
  | | |  parent:      6:400110238667
  | | |  user:        szhang
  | | |  date:        Thu Sep 05 12:57:59 2013 -0400
  | | |  summary:     bad merge
  | | |
  o | |  changeset:   7:4c9fbe56a16f
  |/ /   branch:      f1
  | |    parent:      2:4bc80088dc6b
  | |    user:        szhang
  | |    date:        Thu Sep 05 12:54:00 2013 -0400
  | |    summary:     changed f1
  | |
  | o  changeset:   6:400110238667
  | |  branch:      f2
  | |  parent:      4:12e8ec6bb010
  | |  user:        szhang
  | |  date:        Tue Sep 03 13:58:02 2013 -0400
  | |  summary:     changed f2 on f2
  | |
  | | @  changeset:   5:d79e2059b5c0
  | | |  parent:      3:8a951942e016
  | | |  user:        szhang
  | | |  date:        Tue Sep 03 13:57:39 2013 -0400
  | | |  summary:     changed f2 on default
  | | |
  | o |  changeset:   4:12e8ec6bb010
  | |/   branch:      f2
  | |    user:        szhang
  | |    date:        Tue Sep 03 13:57:18 2013 -0400
  | |    summary:     created f2 branch
  | |
  | o  changeset:   3:8a951942e016
  | |  parent:      0:24797d4f68de
  | |  user:        szhang
  | |  date:        Tue Sep 03 13:57:11 2013 -0400
  | |  summary:     added f2.txt
  | |
  o |  changeset:   2:4bc80088dc6b
  | |  branch:      f1
  | |  user:        szhang
  | |  date:        Tue Sep 03 13:56:20 2013 -0400
  | |  summary:     added f1.txt
  | |
  o |  changeset:   1:ef53c9e6b608
  |/   branch:      f1
  |    user:        szhang
  |    date:        Tue Sep 03 13:55:26 2013 -0400
  |    summary:     created f1 branch
  |
  o  changeset:   0:24797d4f68de
     user:        szhang
     date:        Tue Sep 03 13:55:08 2013 -0400
     summary:     added default.txt
  
  $ hg rebase -s9 -d2 --debug # use debug to really check merge base used
  rebase onto 4bc80088dc6b starting from e31216eec445
  rebasing on disk
  rebase status stored
  rebasing 9:e31216eec445 "more changes to f1"
   future parents are 2 and -1
   update to 2:4bc80088dc6b
  resolving manifests
   branchmerge: False, force: True, partial: False
   ancestor: d79e2059b5c0+, local: d79e2059b5c0+, remote: 4bc80088dc6b
   f2.txt: other deleted -> r
  removing f2.txt
   f1.txt: remote created -> g
  getting f1.txt
   merge against 9:e31216eec445
     detach base 8:8e4e2c1a07ae
    searching for copies back to rev 3
    unmatched files in other (from topological common ancestor):
     f2.txt
  resolving manifests
   branchmerge: True, force: True, partial: False
   ancestor: 8e4e2c1a07ae, local: 4bc80088dc6b+, remote: e31216eec445
   f1.txt: remote is newer -> g
  getting f1.txt
  committing files:
  f1.txt
  committing manifest
  committing changelog
  updating the branch cache
  rebased as 19c888675e13
  rebase status stored
  rebasing 10:2f2496ddf49d "merge" (tip)
   future parents are 11 and 7
   already in destination
   merge against 10:2f2496ddf49d
     detach base 9:e31216eec445
    searching for copies back to rev 3
    unmatched files in other (from topological common ancestor):
     f2.txt
  resolving manifests
   branchmerge: True, force: True, partial: False
   ancestor: e31216eec445, local: 19c888675e13+, remote: 2f2496ddf49d
   f1.txt: remote is newer -> g
  getting f1.txt
  committing files:
  f1.txt
  committing manifest
  committing changelog
  updating the branch cache
  rebased as 2a7f09cac94c
  rebase status stored
  rebase merging completed
  update back to initial working directory parent
  resolving manifests
   branchmerge: False, force: False, partial: False
   ancestor: 2a7f09cac94c, local: 2a7f09cac94c+, remote: d79e2059b5c0
   f1.txt: other deleted -> r
  removing f1.txt
   f2.txt: remote created -> g
  getting f2.txt
  2 changesets found
  list of changesets:
  e31216eec445e44352c5f01588856059466a24c9
  2f2496ddf49d69b5ef23ad8cf9fb2e0e4faf0ac2
  bundle2-output-bundle: "HG20", (1 params) 3 parts total
  bundle2-output-part: "changegroup" (params: 1 mandatory 1 advisory) streamed payload
  bundle2-output-part: "cache:rev-branch-cache" (advisory) streamed payload
  bundle2-output-part: "phase-heads" 24 bytes payload
  saved backup bundle to $TESTTMP/issue4041/.hg/strip-backup/e31216eec445-15f7a814-rebase.hg
  3 changesets found
  list of changesets:
  4c9fbe56a16f30c0d5dcc40ec1a97bbe3325209c
  19c888675e133ab5dff84516926a65672eaf04d9
  2a7f09cac94c7f4b73ebd5cd1a62d3b2e8e336bf
  bundle2-output-bundle: "HG20", 3 parts total
  bundle2-output-part: "changegroup" (params: 1 mandatory 1 advisory) streamed payload
  bundle2-output-part: "cache:rev-branch-cache" (advisory) streamed payload
  bundle2-output-part: "phase-heads" 24 bytes payload
  adding branch
  bundle2-input-bundle: with-transaction
  bundle2-input-part: "changegroup" (params: 1 mandatory 1 advisory) supported
  adding changesets
  add changeset 4c9fbe56a16f
  add changeset 19c888675e13
  add changeset 2a7f09cac94c
  adding manifests
  adding file changes
  adding f1.txt revisions
  added 2 changesets with 2 changes to 1 files
  bundle2-input-part: total payload size 1686
  bundle2-input-part: "cache:rev-branch-cache" (advisory) supported
  bundle2-input-part: total payload size 74
  truncating cache/rbc-revs-v1 to 56
  bundle2-input-part: "phase-heads" supported
  bundle2-input-part: total payload size 24
  bundle2-input-bundle: 2 parts total
  updating the branch cache
  invalid branchheads cache (served): tip differs
  rebase completed

Test minimization of merge conflicts
  $ hg up -q null
  $ echo a > a
  $ hg add a
  $ hg commit -q -m 'a'
  $ echo b >> a
  $ hg commit -q -m 'ab'
  $ hg bookmark ab
  $ hg up -q '.^'
  $ echo b >> a
  $ echo c >> a
  $ hg commit -q -m 'abc'
  $ hg rebase -s 7bc217434fc1 -d ab --keep
  rebasing 13:7bc217434fc1 "abc" (tip)
  merging a
  warning: conflicts while merging a! (edit, then use 'hg resolve --mark')
  unresolved conflicts (see hg resolve, then hg rebase --continue)
  [1]
  $ hg diff
  diff -r 328e4ab1f7cc a
  --- a/a	Thu Jan 01 00:00:00 1970 +0000
  +++ b/a	* (glob)
  @@ -1,2 +1,6 @@
   a
   b
  +<<<<<<< dest:   328e4ab1f7cc ab - test: ab
  +=======
  +c
  +>>>>>>> source: 7bc217434fc1 - test: abc
  $ hg rebase --abort
  rebase aborted
  $ hg up -q -C 7bc217434fc1
  $ hg rebase -s . -d ab --keep -t internal:merge3
  rebasing 13:7bc217434fc1 "abc" (tip)
  merging a
  warning: conflicts while merging a! (edit, then use 'hg resolve --mark')
  unresolved conflicts (see hg resolve, then hg rebase --continue)
  [1]
  $ hg diff
  diff -r 328e4ab1f7cc a
  --- a/a	Thu Jan 01 00:00:00 1970 +0000
  +++ b/a	* (glob)
  @@ -1,2 +1,8 @@
   a
  +<<<<<<< dest:   328e4ab1f7cc ab - test: ab
   b
  +||||||| base
  +=======
  +b
  +c
  +>>>>>>> source: 7bc217434fc1 - test: abc

Test rebase with obsstore turned on and off (issue5606)

  $ cd $TESTTMP
  $ hg init b
  $ cd b
  $ hg debugdrawdag <<'EOS'
  > D
  > |
  > C
  > |
  > B E
  > |/
  > A
  > EOS

  $ hg update E -q
  $ echo 3 > B
  $ hg commit --amend -m E -A B -q
  $ hg rebase -r B+D -d . --config experimental.evolution=true
  rebasing 1:112478962961 "B" (B)
  merging B
  warning: conflicts while merging B! (edit, then use 'hg resolve --mark')
  unresolved conflicts (see hg resolve, then hg rebase --continue)
  [1]

  $ echo 4 > B
  $ hg resolve -m
  (no more unresolved files)
  continue: hg rebase --continue
  $ hg rebase --continue --config experimental.evolution=none
  rebasing 1:112478962961 "B" (B)
  rebasing 3:f585351a92f8 "D" (D)
  warning: orphaned descendants detected, not stripping 112478962961
  saved backup bundle to $TESTTMP/b/.hg/strip-backup/f585351a92f8-e536a9e4-rebase.hg

  $ rm .hg/localtags
  $ hg tglog
  o  5:draft 'D'
  |
  o  4:draft 'B'
  |
  @  3:draft 'E'
  |
  | o  2:draft 'C'
  | |
  | o  1:draft 'B'
  |/
  o  0:draft 'A'