Correct optimization from 3464f5e77f34; add a test.
If there are two (or more) heads that point to the same .hgtags
node, we can safely skip parsing the file in all but the last head.
(In 3464f5e77f34, we were parsing the file in the first head and
skipping all the others.)
#!/bin/sh
hg init a
cd a
echo a > foo
hg rm foo
hg add foo
hg commit -m 1 -d "1000000 0"
hg remove
rm foo
hg remove foo
hg revert --all
rm foo
hg remove --after
hg commit -m 2 -d "1000000 0"
hg export --nodates 0
hg export --nodates 1
hg log -p -r 0
hg log -p -r 1
echo a > a
hg add a
hg rm a
hg rm -f a
echo b > b
hg ci -A -m 3 -d "1000001 0"
echo c >> b
hg rm b
hg rm -f b
cd ..
hg clone a b