phases: avoid N² behavior in `advanceboundary`
We allowed duplicated entries in the deque, which each entry could potentially
insert all its ancestors. So advancing boundary for the full repository would
mean each revision would walk all its ancestors, resulting in O(N²) iteration.
For repository of any decent size, N² is quickly insane.
We introduce a simple set to avoid this and get back to reasonable performance.
#require no-windows
$ . "$TESTDIR/remotefilelog-library.sh"
$ hg init master
$ cd master
$ cat >> .hg/hgrc <<EOF
> [remotefilelog]
> server=True
> EOF
$ echo x > x
$ hg commit -qAm x
$ echo y >> x
$ hg commit -qAm y
$ echo z >> x
$ hg commit -qAm z
$ hg update 1
1 files updated, 0 files merged, 0 files removed, 0 files unresolved
$ echo w >> x
$ hg commit -qAm w
$ cd ..
Shallow clone and activate getflogheads testing extension
$ hgcloneshallow ssh://user@dummy/master shallow --noupdate
streaming all changes
3 files to transfer, 908 bytes of data (no-rust !)
transferred 908 bytes in * seconds (*/sec) (glob) (no-rust !)
5 files to transfer, 1.01 KB of data (rust !)
transferred 1.01 KB in * seconds (* */sec) (glob) (rust !)
searching for changes
no changes found
$ cd shallow
$ cat >> .hg/hgrc <<EOF
> [extensions]
> getflogheads=$TESTDIR/remotefilelog-getflogheads.py
> EOF
Get heads of a remotefilelog
$ hg getflogheads x
2797809ca5e9c2f307d82b1345e832f655fb99a2
ca758b402ddc91e37e3113e1a97791b537e1b7bb
Get heads of a non-existing remotefilelog
$ hg getflogheads y
EMPTY