tests/test-mq-caches.t

store: implement fncache basic path encoding in C

(This is not yet enabled; it will be turned on in a followup patch.)

The path encoding performed by fncache is complex and (perhaps
surprisingly) slow enough to negatively affect the overall performance
of Mercurial.

For a short path (< 120 bytes), the Python code can be reduced to a fairly
tractable state machine that either determines that nothing needs to be
done in a single pass, or performs the encoding in a second pass.

For longer paths, we avoid the more complicated hashed encoding scheme
for now, and fall back to Python.

Raw performance: I measured in a repo containing 150,000 files in its tip
manifest, with a median path name length of 57 bytes, and 95th percentile
of 96 bytes.

In this repo, the Python code takes 3.1 seconds to encode all path
names, while the hybrid C-and-Python code (called from Python) takes
0.21 seconds, for a speedup of about 14.

Across several other large repositories, I've measured the speedup from
the C code at between 26x and 40x.

For path names above 120 bytes where we must fall back to Python for
hashed encoding, the speedup is about 1.7x.  Thus absolute performance
will depend strongly on the characteristics of a particular repository.

  $ branches=.hg/cache/branchheads
  $ echo '[extensions]' >> $HGRCPATH
  $ echo 'mq =' >> $HGRCPATH

  $ show_branch_cache()
  > {
  >     # force cache (re)generation
  >     hg log -r does-not-exist 2> /dev/null
  >     hg log -r tip --template 'tip: {rev}\n'
  >     if [ -f $branches ]; then
  >       sort $branches
  >     else
  >       echo No branch cache
  >     fi
  >     if [ "$1" = 1 ]; then
  >       for b in foo bar; do
  >         hg log -r $b --template "branch $b: "'{rev}\n'
  >       done
  >     fi
  > }

  $ hg init a
  $ cd a
  $ hg qinit -c


mq patch on an empty repo

  $ hg qnew -d '0 0' p1
  $ show_branch_cache
  tip: 0
  No branch cache

  $ echo > pfile
  $ hg add pfile
  $ hg qrefresh -m 'patch 1'
  $ show_branch_cache
  tip: 0
  d986d5caac23a7d44a46efc0ddaf5eb9665844cf 0
  d986d5caac23a7d44a46efc0ddaf5eb9665844cf default

some regular revisions

  $ hg qpop
  popping p1
  patch queue now empty
  $ echo foo > foo
  $ hg add foo
  $ echo foo > .hg/branch
  $ hg ci -m 'branch foo'

  $ echo bar > bar
  $ hg add bar
  $ echo bar > .hg/branch
  $ hg ci -m 'branch bar'
  $ show_branch_cache
  tip: 1
  c229711f16da3d7591f89b1b8d963b79bda22714 1
  c229711f16da3d7591f89b1b8d963b79bda22714 bar
  dc25e3827021582e979f600811852e36cbe57341 foo

add some mq patches

  $ hg qpush
  applying p1
  now at: p1
  $ show_branch_cache
  tip: 2
  c229711f16da3d7591f89b1b8d963b79bda22714 1
  c229711f16da3d7591f89b1b8d963b79bda22714 bar
  dc25e3827021582e979f600811852e36cbe57341 foo

  $ hg qnew -d '0 0' p2
  $ echo foo > .hg/branch
  $ echo foo2 >> foo
  $ hg qrefresh -m 'patch 2'
  $ show_branch_cache 1
  tip: 3
  982611f6955f9c48d3365decea203217c945ef0d 2
  982611f6955f9c48d3365decea203217c945ef0d bar
  dc25e3827021582e979f600811852e36cbe57341 foo
  branch foo: 3
  branch bar: 2

removing the cache

  $ rm $branches
  $ show_branch_cache 1
  tip: 3
  c229711f16da3d7591f89b1b8d963b79bda22714 1
  c229711f16da3d7591f89b1b8d963b79bda22714 bar
  dc25e3827021582e979f600811852e36cbe57341 foo
  branch foo: 3
  branch bar: 2

importing rev 1 (the cache now ends in one of the patches)

  $ hg qimport -r 1 -n p0
  $ show_branch_cache 1
  tip: 3
  c229711f16da3d7591f89b1b8d963b79bda22714 1
  c229711f16da3d7591f89b1b8d963b79bda22714 bar
  dc25e3827021582e979f600811852e36cbe57341 foo
  branch foo: 3
  branch bar: 2
  $ hg log -r qbase --template 'qbase: {rev}\n'
  qbase: 1

detect an invalid cache

  $ hg qpop -a
  popping p2
  popping p1
  popping p0
  patch queue now empty
  $ hg qpush -a
  applying p0
  applying p1
  applying p2
  now at: p2
  $ show_branch_cache
  tip: 3
  dc25e3827021582e979f600811852e36cbe57341 0
  dc25e3827021582e979f600811852e36cbe57341 foo

  $ cd ..