copies: do full filtering at end of _changesetforwardcopies()
As mentioned earlier, pathcopies() is very slow when copies are stored
in the changeset. Most of the cost comes from calling _chain() for
every changeset, which is slow because it needs to read manifests. It
needs to read manifests to be able to filter out copies that are were
created in one commit and then deleted. (It also filters out copies
that were created from a file that didn't exist in the starting
revision, but that's a fixed revision across calls to _chain(), so
it's much cheaper.)
This patch changes from _chainandfilter() to just _chain() in the main
loop in _changesetforwardcopies(). It instead removes copies that have
subsequently been removed by using ctx.filesremoved(). We thus rely on
that to be fast.
It timed this command in mozilla-unified:
hg debugpathcopies FIREFOX_59_0b3_BUILD2 FIREFOX_BETA_59_END
It took 18s before and 1.1s after. It's still faster when copy
information is stored in filelogs: 0.70s. It also still gets slow when
there are merge commits involved, because we read manifests there
too. We'll deal with that later.
Differential Revision: https://phab.mercurial-scm.org/D6419
$ echo "[extensions]" >> $HGRCPATH
$ echo "mq=" >> $HGRCPATH
$ hg init
$ echo qqq>qqq.txt
rollback dry run without rollback information
$ hg rollback
no rollback information available
[1]
add file
$ hg add
adding qqq.txt
commit first revision
$ hg ci -m 1
set bookmark
$ hg book test
$ echo www>>qqq.txt
commit second revision
$ hg ci -m 2
set bookmark
$ hg book test2
update to -2 (deactivates the active bookmark)
$ hg update -r -2
1 files updated, 0 files merged, 0 files removed, 0 files unresolved
(leaving bookmark test2)
$ echo eee>>qqq.txt
commit new head
$ hg ci -m 3
created new head
bookmarks updated?
$ hg book
test 1:25e1ee7a0081
test2 1:25e1ee7a0081
strip to revision 1
$ hg strip 1
saved backup bundle to $TESTTMP/.hg/strip-backup/*-backup.hg (glob)
list bookmarks
$ hg book
test 0:5c9ad3787638
test2 0:5c9ad3787638