findrenames: Optimise "addremove -s100" by matching files by their SHA1 hashes.
We speed up 'findrenames' for the usecase when a user specifies they
want a similarity of 100% by matching files by their exact SHA1 hash
value. This reduces the number of comparisons required to find exact
matches from O(n^2) to O(n).
While it would be nice if we could just use mercurial's pre-calculated
SHA1 hash for existing files, this hash includes the file's ancestor
information making it unsuitable for our purposes. Instead, we calculate
the hash of old content from scratch.
The following benchmarks were taken on the current head of crew:
addremove 100% similarity:
rm -rf *; hg up -C; mv tests tests.new
hg --time addremove -s100 --dry-run
before: real 176.350 secs (user 128.890+0.000 sys 47.430+0.000)
after: real 2.130 secs (user 1.890+0.000 sys 0.240+0.000)
addremove 75% similarity:
rm -rf *; hg up -C; mv tests tests.new; \
for i in tests.new/*; do echo x >> $i; done
hg --time addremove -s75 --dry-run
before: real 264.560 secs (user 215.130+0.000 sys 49.410+0.000)
after: real 218.710 secs (user 172.790+0.000 sys 45.870+0.000)
#!/bin/sh
cat >> $HGRCPATH <<EOF
[extensions]
convert=
[convert]
hg.saverev=False
EOF
hg init orig
cd orig
echo foo > foo
echo bar > bar
hg ci -qAm 'add foo bar' -d '0 0'
echo >> foo
hg ci -m 'change foo' -d '1 0'
hg up -qC 0
hg copy --after --force foo bar
hg copy foo baz
hg ci -m 'make bar and baz copies of foo' -d '2 0'
hg merge
hg ci -m 'merge local copy' -d '3 0'
hg up -C 1
hg merge 2
hg ci -m 'merge remote copy' -d '4 0'
chmod +x baz
hg ci -m 'mark baz executable' -d '5 0'
cd ..
hg convert --datesort orig new 2>&1 | grep -v 'subversion python bindings could not be loaded'
cd new
hg out ../orig
cd ..
echo '% check shamap LF and CRLF handling'
cat > rewrite.py <<EOF
import sys
# Interlace LF and CRLF
lines = [(l.rstrip() + ((i % 2) and '\n' or '\r\n'))
for i, l in enumerate(file(sys.argv[1]))]
file(sys.argv[1], 'wb').write(''.join(lines))
EOF
python rewrite.py new/.hg/shamap
cd orig
hg up -qC 1
echo foo >> foo
hg ci -qm 'change foo again'
hg up -qC 2
echo foo >> foo
hg ci -qm 'change foo again again'
cd ..
hg convert --datesort orig new 2>&1 | grep -v 'subversion python bindings could not be loaded'
echo % init broken repository
hg init broken
cd broken
echo a >> a
echo b >> b
hg ci -qAm init
echo a >> a
echo b >> b
hg copy b c
hg ci -qAm changeall
hg up -qC 0
echo bc >> b
hg ci -m changebagain
HGMERGE=internal:local hg -q merge
hg ci -m merge
hg mv b d
hg ci -m moveb
echo % break it
rm .hg/store/data/b.*
cd ..
hg --config convert.hg.ignoreerrors=True convert broken fixed
hg -R fixed verify
echo '% manifest -r 0'
hg -R fixed manifest -r 0
echo '% manifest -r tip'
hg -R fixed manifest -r tip
true