tests/test-convert-clonebranches.t

obsolete: order of magnitude speedup in _computebumpedset

Reminder: a changeset is said "bumped" if it tries to obsolete a immutable
          changeset.


The previous algorithm for computing bumped changeset was:

    1) Get all public changesets
    2) Find all they successors
    3) Search for stuff that are eligible for being "bumped"
       (mutable and non obsolete)

The entry size of this algorithm is `O(len(public))` which is mostly the same as
`O(len(repo))`. Even this this approach mean fewer obsolescence marker are
traveled, this is not very scalable.

The new algorithm is:

    1) For each potential bumped changesets (non obsolete mutable)
    2) iterate over precursors
    3) if a precursors is public. changeset is bumped

We travel more obsolescence marker, but the entry size is much smaller since
the amount of potential bumped should remains mostly stable with time `O(1)`.

On some confidential gigantic repo this move bumped computation from 15.19s to
0.46s (×33 speedup…). On "smaller" repo (mercurial, cubicweb's review) no
significant gain were seen. The additional traversal of obsolescence marker is
probably probably counter balance the advantage of it.

Other optimisation could be done in the future (eg: sharing precursors cache
for divergence detection)

  $ echo "[extensions]" >> $HGRCPATH
  $ echo "convert = " >> $HGRCPATH
  $ echo "[convert]" >> $HGRCPATH
  $ echo "hg.tagsbranch=0" >> $HGRCPATH
  $ hg init source
  $ cd source
  $ echo a > a
  $ hg ci -qAm adda

Add a merge with one parent in the same branch

  $ echo a >> a
  $ hg ci -qAm changea
  $ hg up -qC 0
  $ hg branch branch0
  marked working directory as branch branch0
  (branches are permanent and global, did you want a bookmark?)
  $ echo b > b
  $ hg ci -qAm addb
  $ hg up -qC
  $ hg merge default
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg ci -qm mergeab
  $ hg tag -ql mergeab
  $ cd ..

Miss perl... sometimes

  $ cat > filter.py <<EOF
  > import sys, re
  > 
  > r = re.compile(r'^(?:\d+|pulling from)')
  > sys.stdout.writelines([l for l in sys.stdin if r.search(l)])
  > EOF

convert

  $ hg convert -v --config convert.hg.clonebranches=1 source dest |
  >     python filter.py
  3 adda
  2 changea
  1 addb
  pulling from default into branch0
  1 changesets found
  0 mergeab
  pulling from default into branch0
  1 changesets found

Add a merge with both parents and child in different branches

  $ cd source
  $ hg branch branch1
  marked working directory as branch branch1
  (branches are permanent and global, did you want a bookmark?)
  $ echo a > file1
  $ hg ci -qAm c1
  $ hg up -qC mergeab
  $ hg branch branch2
  marked working directory as branch branch2
  (branches are permanent and global, did you want a bookmark?)
  $ echo a > file2
  $ hg ci -qAm c2
  $ hg merge branch1
  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
  (branch merge, don't forget to commit)
  $ hg branch branch3
  marked working directory as branch branch3
  (branches are permanent and global, did you want a bookmark?)
  $ hg ci -qAm c3
  $ cd ..

incremental conversion

  $ hg convert -v --config convert.hg.clonebranches=1 source dest |
  >     python filter.py
  2 c1
  pulling from branch0 into branch1
  4 changesets found
  1 c2
  pulling from branch0 into branch2
  4 changesets found
  0 c3
  pulling from branch1 into branch3
  5 changesets found
  pulling from branch2 into branch3
  1 changesets found