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.
=============================================================
Check that we can compute and exchange revision rank properly
=============================================================
$ cat << EOF >> $HGRCPATH
> [format]
> exp-use-changelog-v2=enable-unstable-format-and-corrupt-my-data
> EOF
Test minimal rank computation with merge
$ hg init rank-repo-minimal
$ cd rank-repo-minimal
$ touch 0
$ hg commit -Aqm 0
$ touch 1
$ hg commit -Aqm 1
$ hg update -qr 0
$ touch 2
$ hg commit -Aqm 2
$ hg merge -qr 1
$ hg commit -m 3
$ touch 4
$ hg commit -Aqm 4
$ hg log --graph --template '{rev} {_fast_rank}\n'
@ 4 5
|
o 3 4
|\
| o 2 2
| |
o | 1 2
|/
o 0 1
$ cd ..
Build a bigger example repo
$ hg init rank-repo-generated
$ cd rank-repo-generated
$ hg debugbuilddag '.:root1+5:mp1<root1+10:mp2/mp1+3<mp1+2:mp3/mp2$+15/mp1+4'
$ hg log -G -T '{desc}'
o r42
|
o r41
|
o r40
|
o r39
|
o r38
|\
| o r37
| |
| o r36
| |
| o r35
| |
| o r34
| |
| o r33
| |
| o r32
| |
| o r31
| |
| o r30
| |
| o r29
| |
| o r28
| |
| o r27
| |
| o r26
| |
| o r25
| |
| o r24
| |
| o r23
|
| o r22
| |\
| | o r21
| | |
+---o r20
| |
| | o r19
| | |
| | o r18
| | |
| | o r17
| | |
+---o r16
| |/
| o r15
| |
| o r14
| |
| o r13
| |
| o r12
| |
| o r11
| |
| o r10
| |
| o r9
| |
| o r8
| |
| o r7
| |
| o r6
| |
o | r5
| |
o | r4
| |
o | r3
| |
o | r2
| |
o | r1
|/
o r0
Check the rank
--------------
$ hg log -G -T '{_fast_rank}'
o 26
|
o 25
|
o 24
|
o 23
|
o 22
|\
| o 15
| |
| o 14
| |
| o 13
| |
| o 12
| |
| o 11
| |
| o 10
| |
| o 9
| |
| o 8
| |
| o 7
| |
| o 6
| |
| o 5
| |
| o 4
| |
| o 3
| |
| o 2
| |
| o 1
|
| o 19
| |\
| | o 8
| | |
+---o 7
| |
| | o 20
| | |
| | o 19
| | |
| | o 18
| | |
+---o 17
| |/
| o 11
| |
| o 10
| |
| o 9
| |
| o 8
| |
| o 7
| |
| o 6
| |
| o 5
| |
| o 4
| |
| o 3
| |
| o 2
| |
o | 6
| |
o | 5
| |
o | 4
| |
o | 3
| |
o | 2
|/
o 1