revset: fix order of nested 'or' expression (BC)
This fixes the order of 'x & (y + z)' where 'y' and 'z' are not trivial.
The follow-order 'or' operation is slower than the ordered operation if
an input set is large:
#0 #1 #2 #3
0) 0.002968 0.002980 0.002982 0.073042
1) 0.004513 0.004485 0.012029 0.075261
#0: 0:4000 & (0:1099 + 1000:2099 + 2000:3099)
#1: 4000:0 & (0:1099 + 1000:2099 + 2000:3099)
#2: 10000:0 & (0:1099 + 1000:2099 + 2000:3099)
#3: file("path:hg") & (0:1099 + 1000:2099 + 2000:3099)
I've tried another implementation, but which appeared to be slower than
this version.
ss = [getset(repo, fullreposet(repo), x) for x in xs]
return subset.filter(lambda r: any(r in s for s in ss), cache=False)
$ hg init outer
$ cd outer
$ echo '[paths]' >> .hg/hgrc
$ echo 'default = http://example.net/' >> .hg/hgrc
hg debugsub with no remapping
$ echo 'sub = libfoo' > .hgsub
$ hg add .hgsub
$ hg debugsub
path sub
source libfoo
revision
hg debugsub with remapping
$ echo '[subpaths]' >> .hg/hgrc
$ printf 'http://example.net/lib(.*) = C:\\libs\\\\1-lib\\\n' >> .hg/hgrc
$ hg debugsub
path sub
source C:\libs\foo-lib\
revision
test cumulative remapping, the $HGRCPATH file is loaded first
$ echo '[subpaths]' >> $HGRCPATH
$ echo 'libfoo = libbar' >> $HGRCPATH
$ hg debugsub
path sub
source C:\libs\bar-lib\
revision
test absolute source path -- testing with a URL is important since
standard os.path.join wont treat that as an absolute path
$ echo 'abs = http://example.net/abs' > .hgsub
$ hg debugsub
path abs
source http://example.net/abs
revision
$ echo 'abs = /abs' > .hgsub
$ hg debugsub
path abs
source /abs
revision
test bad subpaths pattern
$ cat > .hg/hgrc <<EOF
> [subpaths]
> .* = \1
> EOF
$ hg debugsub
abort: bad subrepository pattern in $TESTTMP/outer/.hg/hgrc:2: invalid group reference (glob)
[255]
$ cd ..