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
fixheaders()
{
sed -e 's/\(Message-Id:.*@\).*/\1/' \
-e 's/\(In-Reply-To:.*@\).*/\1/' \
-e 's/\(References:.*@\).*/\1/' \
-e 's/\(User-Agent:.*\)\/.*/\1/' \
-e 's/===.*/===/'
}
echo "[extensions]" >> $HGRCPATH
echo "patchbomb=" >> $HGRCPATH
hg init t
cd t
echo a > a
hg commit -Ama -d '1 0'
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar -r tip | \
fixheaders
echo b > b
hg commit -Amb -d '2 0'
hg email --date '1970-1-1 0:2' -n -f quux -t foo -c bar -s test -r 0:tip | \
fixheaders
hg email -m test.mbox -f quux -t foo -c bar -s test 0:tip
cd ..
hg clone -q t t2
cd t2
echo c > c
hg commit -Amc -d '3 0'
cat > description <<EOF
a multiline
description
EOF
echo "% test bundle and description"
hg email --date '1970-1-1 0:3' -n -f quux -t foo \
-c bar -s test -r tip -b --desc description | \
fixheaders
echo "% utf-8 patch"
python -c 'fp = open("utf", "wb"); fp.write("h\xC3\xB6mma!\n"); fp.close();'
hg commit -A -d '4 0' -m 'charset=utf-8; content-transfer-encoding: base64'
echo "% no mime encoding for email --test"
hg email --date '1970-1-1 0:4' -f quux -t foo -c bar -r tip -n | \
fixheaders > mailtest
echo "% md5sum of 8-bit output"
$TESTDIR/md5sum.py mailtest
rm mailtest
echo "% mime encoded mbox (base64)"
hg email --date '1970-1-1 0:4' -f quux -t foo -c bar -r tip -m mbox
cat mbox | fixheaders
rm mbox
echo "% mime encoded mbox (quoted-printable)"
python -c 'fp = open("qp", "wb"); fp.write("%s\nfoo\n\nbar\n" % \
("x" * 1024)); fp.close();'
hg commit -A -d '4 0' -m \
'charset=utf-8; content-transfer-encoding: quoted-printable'
echo "% no mime encoding for email --test"
hg email --date '1970-1-1 0:4' -f quux -t foo -c bar -r tip -n | \
fixheaders > mailtest
echo "% md5sum of qp output"
$TESTDIR/md5sum.py mailtest
rm mailtest
echo "% mime encoded mbox (quoted-printable)"
hg email --date '1970-1-1 0:4' -f quux -t foo -c bar -r tip -m mbox
cat mbox | fixheaders
rm mbox
echo "% iso-8859-1 patch"
python -c 'fp = open("isolatin", "wb"); fp.write("h\xF6mma!\n"); fp.close();'
hg commit -A -d '5 0' -m 'charset=us-ascii; content-transfer-encoding: 8bit'
echo "% fake ascii mbox"
hg email --date '1970-1-1 0:5' -f quux -t foo -c bar -r tip -m mbox
fixheaders < mbox > mboxfix
echo "% md5sum of 8-bit output"
$TESTDIR/md5sum.py mboxfix
echo "% test diffstat for single patch"
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar -s test -d -y -r 2 | \
fixheaders
echo "% test diffstat for multiple patches"
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar -s test -d -y \
-r 0:1 | fixheaders
echo "% test inline for single patch"
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar -s test -i -r 2 | \
fixheaders
echo "% test inline for single patch (quoted-printable)"
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar -s test -i -r 4 | \
fixheaders
echo "% test inline for multiple patches"
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar -s test -i \
-r 0:1 -r 4 | fixheaders
echo "% test attach for single patch"
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar -s test -a -r 2 | \
fixheaders
echo "% test attach for single patch (quoted-printable)"
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar -s test -a -r 4 | \
fixheaders
echo "% test attach for multiple patches"
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar -s test -a \
-r 0:1 -r 4 | fixheaders
echo "% test intro for single patch"
hg email --date '1970-1-1 0:1' -n --intro -f quux -t foo -c bar -s test \
-r 2 | fixheaders
echo "% test --desc without --intro for a single patch"
echo foo > intro.text
hg email --date '1970-1-1 0:1' -n --desc intro.text -f quux -t foo -c bar \
-s test -r 2 | fixheaders
echo "% test intro for multiple patches"
hg email --date '1970-1-1 0:1' -n --intro -f quux -t foo -c bar -s test \
-r 0:1 | fixheaders
echo "% tagging csets"
hg tag -r0 zero zero.foo
hg tag -r1 one one.patch
hg tag -r2 two two.diff
echo "% test inline for single named patch"
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar -s test -i -r 2 | \
fixheaders
echo "% test inline for multiple named/unnamed patches"
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar -s test -i -r 0:1 | \
fixheaders
echo "% test inreplyto"
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar --in-reply-to baz \
-r tip | fixheaders
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar --in-reply-to baz \
-r 0:1 | fixheaders
hg email --date '1970-1-1 0:1' -n -f quux -t foo -c bar --in-reply-to baz \
-s test -r 0:1 | fixheaders
echo "% test single flag for single patch"
hg email --date '1970-1-1 0:1' -n --flag fooFlag -f quux -t foo -c bar -s test \
-r 2 | fixheaders
echo "% test single flag for multiple patches"
hg email --date '1970-1-1 0:1' -n --flag fooFlag -f quux -t foo -c bar -s test \
-r 0:1 | fixheaders
echo "% test mutiple flags for single patch"
hg email --date '1970-1-1 0:1' -n --flag fooFlag --flag barFlag -f quux -t foo \
-c bar -s test -r 2 | fixheaders
echo "% test multiple flags for multiple patches"
hg email --date '1970-1-1 0:1' -n --flag fooFlag --flag barFlag -f quux -t foo \
-c bar -s test -r 0:1 | fixheaders
echo "% test multi-address parsing"
hg email --date '1980-1-1 0:1' -m tmp.mbox -f quux -t 'spam<spam><eggs>' \
-t toast -c 'foo,bar@example.com' -c '"A, B <>" <a@example.com>' -s test -r 0 \
--config email.bcc='"Quux, A." <quux>'
cat tmp.mbox | fixheaders
echo "% test multi-byte domain parsing"
UUML=`python -c 'import sys; sys.stdout.write("\374")'`
HGENCODING=iso-8859-1
export HGENCODING
hg email --date '1980-1-1 0:1' -m tmp.mbox -f quux -t "bar@${UUML}nicode.com" \
-s test -r 0
cat tmp.mbox | fixheaders
echo "% test outgoing"
hg up 1
hg branch test
echo d > d
hg add d
hg ci -md -d '4 0'
hg email --date '1980-1-1 0:1' -n -t foo -s test -o ../t | fixheaders
echo "% dest#branch URIs"
hg email --date '1980-1-1 0:1' -n -t foo -s test -o ../t#test | fixheaders