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.
$ HGENCODING=utf-8
$ export HGENCODING
$ try() {
> hg debugrevspec --debug $@
> }
$ log() {
> hg log --template '{rev}\n' -r "$1"
> }
$ hg init repo
$ cd repo
$ try 'p1()'
(func
(symbol 'p1')
None)
* set:
<baseset []>
$ try 'p2()'
(func
(symbol 'p2')
None)
* set:
<baseset []>
$ try 'parents()'
(func
(symbol 'parents')
None)
* set:
<baseset+ []>
null revision
$ log 'p1()'
$ log 'p2()'
$ log 'parents()'
working dir with a single parent
$ echo a > a
$ hg ci -Aqm0
$ log 'p1()'
0
$ log 'tag() and p1()'
$ log 'p2()'
$ log 'parents()'
0
$ log 'tag() and parents()'
merge in progress
$ echo b > b
$ hg ci -Aqm1
$ hg up -q 0
$ echo c > c
$ hg ci -Aqm2
$ hg merge -q
$ log 'p1()'
2
$ log 'p2()'
1
$ log 'tag() and p2()'
$ log 'parents()'
1
2
$ cd ..