view tests/test-convert-bzr-merges.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

#require bzr

N.B. bzr 1.13 has a bug that breaks this test.  If you see this
test fail, check your bzr version.  Upgrading to bzr 1.13.1
should fix it.

  $ . "$TESTDIR/bzr-definitions"

test multiple merges at once

  $ mkdir test-multimerge
  $ cd test-multimerge
  $ bzr init -q source
  $ cd source
  $ echo content > file
  $ echo text > rename_me
  $ bzr add -q file rename_me
  $ bzr commit -q -m 'Initial add' '--commit-time=2009-10-10 08:00:00 +0100'
  $ cd ..
  $ bzr branch -q source source-branch1
  $ cd source-branch1
  $ echo morecontent >> file
  $ echo evenmorecontent > file-branch1
  $ bzr add -q file-branch1
  $ bzr commit -q -m 'Added branch1 file' '--commit-time=2009-10-10 08:00:01 +0100'
  $ cd ../source
  $ sleep 1
  $ echo content > file-parent
  $ bzr add -q file-parent
  $ bzr commit -q -m 'Added parent file' '--commit-time=2009-10-10 08:00:02 +0100'
  $ cd ..
  $ bzr branch -q source source-branch2
  $ cd source-branch2
  $ echo somecontent > file-branch2
  $ bzr add -q file-branch2
  $ bzr mv -q rename_me renamed
  $ echo change > renamed
  $ bzr commit -q -m 'Added brach2 file' '--commit-time=2009-10-10 08:00:03 +0100'
  $ sleep 1
  $ cd ../source
  $ bzr merge -q ../source-branch1
  $ bzr merge -q --force ../source-branch2
  $ bzr commit -q -m 'Merged branches' '--commit-time=2009-10-10 08:00:04 +0100'
  $ cd ..

BUG: file-branch2 should not be added in rev 4, and the rename_me -> renamed
move should be recorded in the fixup merge.
  $ hg convert --datesort --config convert.bzr.saverev=False source source-hg
  initializing destination source-hg repository
  scanning source...
  sorting...
  converting...
  4 Initial add
  3 Added branch1 file
  2 Added parent file
  1 Added brach2 file
  0 Merged branches
  warning: can't find ancestor for 'renamed' copied from 'rename_me'!
  $ glog -R source-hg
  o    5@source "(octopus merge fixup)" files+: [], files-: [], files: [renamed]
  |\
  | o    4@source "Merged branches" files+: [file-branch1 file-branch2 renamed], files-: [rename_me], files: [file]
  | |\
  o---+  3@source-branch2 "Added brach2 file" files+: [file-branch2 renamed], files-: [rename_me], files: []
   / /
  | o  2@source "Added parent file" files+: [file-parent], files-: [], files: []
  | |
  o |  1@source-branch1 "Added branch1 file" files+: [file-branch1], files-: [], files: [file]
  |/
  o  0@source "Initial add" files+: [file rename_me], files-: [], files: []
  
  $ manifest source-hg tip
  % manifest of tip
  644   file
  644   file-branch1
  644   file-branch2
  644   file-parent
  644   renamed

  $ hg convert source-hg hg2hg
  initializing destination hg2hg repository
  scanning source...
  sorting...
  converting...
  5 Initial add
  4 Added branch1 file
  3 Added parent file
  2 Added brach2 file
  1 Merged branches
  0 (octopus merge fixup)

BUG: The manifest entries should be the same for matching revisions, and
nothing should be outgoing

  $ hg -R source-hg manifest --debug -r tip | grep renamed
  67109fdebf6c556eb0a9d5696dd98c8420520405 644   renamed
  $ hg -R hg2hg manifest --debug -r tip | grep renamed
  27c968376d7c3afd095ecb9c7697919b933448c8 644   renamed
  $ hg -R source-hg manifest --debug -r 'tip^' | grep renamed
  27c968376d7c3afd095ecb9c7697919b933448c8 644   renamed
  $ hg -R hg2hg manifest --debug -r 'tip^' | grep renamed
  27c968376d7c3afd095ecb9c7697919b933448c8 644   renamed

BUG: The revisions found should be the same in both repos

  $ hg --cwd source-hg log -r 'file("renamed")' -G -Tcompact
  o    5[tip]:4,3   6652429c300a   2009-10-10 08:00 +0100   foo
  |\     (octopus merge fixup)
  | |
  | o    4:2,1   e0ae8af3503a   2009-10-10 08:00 +0100   foo
  | |\     Merged branches
  | ~ ~
  o  3   138bed2e14be   2009-10-10 08:00 +0100   foo
  |    Added brach2 file
  ~
  $ hg --cwd hg2hg log -r 'file("renamed")' -G -Tcompact
  o    4:2,1   e0ae8af3503a   2009-10-10 08:00 +0100   foo
  |\     Merged branches
  ~ ~
  o  3   138bed2e14be   2009-10-10 08:00 +0100   foo
  |    Added brach2 file
  ~

BUG(?): The move seems to be recorded in rev 4, so it should probably show up
there.  It's not recorded as a move in rev 5, even in source-hg.

  $ hg -R source-hg up -q tip
  $ hg -R hg2hg up -q tip
  $ hg --cwd source-hg log -r 'follow("renamed")' -G -Tcompact
  @    5[tip]:4,3   6652429c300a   2009-10-10 08:00 +0100   foo
  |\     (octopus merge fixup)
  | :
  o :  3   138bed2e14be   2009-10-10 08:00 +0100   foo
  :/     Added brach2 file
  :
  o  0   18b86f5df51b   2009-10-10 08:00 +0100   foo
       Initial add
  
  $ hg --cwd hg2hg log -r 'follow("renamed")' -G -Tcompact
  o  3   138bed2e14be   2009-10-10 08:00 +0100   foo
  :    Added brach2 file
  :
  o  0   18b86f5df51b   2009-10-10 08:00 +0100   foo
       Initial add
  

  $ hg -R hg2hg out source-hg -T compact
  comparing with source-hg
  searching for changes
  5[tip]:4,3   3be2299ccd31   2009-10-10 08:00 +0100   foo
    (octopus merge fixup)
  

  $ glog -R hg2hg
  @    5@source "(octopus merge fixup)" files+: [], files-: [], files: []
  |\
  | o    4@source "Merged branches" files+: [file-branch1 file-branch2 renamed], files-: [rename_me], files: [file]
  | |\
  o---+  3@source-branch2 "Added brach2 file" files+: [file-branch2 renamed], files-: [rename_me], files: []
   / /
  | o  2@source "Added parent file" files+: [file-parent], files-: [], files: []
  | |
  o |  1@source-branch1 "Added branch1 file" files+: [file-branch1], files-: [], files: [file]
  |/
  o  0@source "Initial add" files+: [file rename_me], files-: [], files: []
  

  $ hg -R source-hg log --debug -r tip
  changeset:   5:6652429c300ab66fdeaf2e730945676a00b53231
  branch:      source
  tag:         tip
  phase:       draft
  parent:      4:e0ae8af3503af9bbffb0b29268a02744cc61a561
  parent:      3:138bed2e14be415a2692b02e41405b2864f758b4
  manifest:    5:1eabd5f5d4b985784cf2c45c717ff053eca14b0d
  user:        Foo Bar <[e-mail]>
  date:        Sat Oct 10 08:00:04 2009 +0100
  files:       renamed
  extra:       branch=source
  description:
  (octopus merge fixup)
  
  
  $ hg -R hg2hg log --debug -r tip
  changeset:   5:3be2299ccd315ff9aab2b49bdb0d14e3244435e8
  branch:      source
  tag:         tip
  phase:       draft
  parent:      4:e0ae8af3503af9bbffb0b29268a02744cc61a561
  parent:      3:138bed2e14be415a2692b02e41405b2864f758b4
  manifest:    4:3ece3c7f2cc6df15b3cbbf3273c69869fc7c3ab0
  user:        Foo Bar <[e-mail]>
  date:        Sat Oct 10 08:00:04 2009 +0100
  extra:       branch=source
  description:
  (octopus merge fixup)
  
  
  $ hg -R source-hg manifest --debug -r tip
  cdf31ed9242b209cd94697112160e2c5b37a667d 644   file
  5108144f585149b29779d7c7e51d61dd22303ffe 644   file-branch1
  80753c4a9ac3806858405b96b24a907b309e3616 644   file-branch2
  7108421418404a937c684d2479a34a24d2ce4757 644   file-parent
  67109fdebf6c556eb0a9d5696dd98c8420520405 644   renamed
  $ hg -R source-hg manifest --debug -r 'tip^'
  cdf31ed9242b209cd94697112160e2c5b37a667d 644   file
  5108144f585149b29779d7c7e51d61dd22303ffe 644   file-branch1
  80753c4a9ac3806858405b96b24a907b309e3616 644   file-branch2
  7108421418404a937c684d2479a34a24d2ce4757 644   file-parent
  27c968376d7c3afd095ecb9c7697919b933448c8 644   renamed

  $ hg -R hg2hg manifest --debug -r tip
  cdf31ed9242b209cd94697112160e2c5b37a667d 644   file
  5108144f585149b29779d7c7e51d61dd22303ffe 644   file-branch1
  80753c4a9ac3806858405b96b24a907b309e3616 644   file-branch2
  7108421418404a937c684d2479a34a24d2ce4757 644   file-parent
  27c968376d7c3afd095ecb9c7697919b933448c8 644   renamed
  $ hg -R hg2hg manifest --debug -r 'tip^'
  cdf31ed9242b209cd94697112160e2c5b37a667d 644   file
  5108144f585149b29779d7c7e51d61dd22303ffe 644   file-branch1
  80753c4a9ac3806858405b96b24a907b309e3616 644   file-branch2
  7108421418404a937c684d2479a34a24d2ce4757 644   file-parent
  27c968376d7c3afd095ecb9c7697919b933448c8 644   renamed

  $ cd ..