revlog: break up compression of large deltas
Python's zlib apparently makes an internal copy of strings passed to
compress(). To avoid this, compress strings 1M at a time, then join
them at the end if the result would be smaller than the original.
For initial commits of large but compressible files, this cuts peak
memory usage nearly in half.
#!/bin/sh
hg init remote
cd remote
echo "# creating 'remote'"
cat >>afile <<EOF
0
EOF
hg add afile
hg commit -m "0.0"
cat >>afile <<EOF
1
EOF
hg commit -m "0.1"
cat >>afile <<EOF
2
EOF
hg commit -m "0.2"
cat >>afile <<EOF
3
EOF
hg commit -m "0.3"
hg update -C 0
cat >>afile <<EOF
1
EOF
hg commit -m "1.1"
cat >>afile <<EOF
2
EOF
hg commit -m "1.2"
cat >fred <<EOF
a line
EOF
cat >>afile <<EOF
3
EOF
hg add fred
hg commit -m "1.3"
hg mv afile adifferentfile
hg commit -m "1.3m"
hg update -C 3
hg mv afile anotherfile
hg commit -m "0.3m"
hg debugindex .hg/store/data/afile.i
hg debugindex .hg/store/data/adifferentfile.i
hg debugindex .hg/store/data/anotherfile.i
hg debugindex .hg/store/data/fred.i
hg debugindex .hg/store/00manifest.i
hg verify
echo "# Starting server"
hg serve -p $HGPORT -d --pid-file=../hg1.pid
cd ..
cat hg1.pid >> $DAEMON_PIDS
echo "# clone remote via stream"
for i in 0 1 2 3 4 5 6 7 8; do
hg clone -r "$i" http://localhost:$HGPORT/ test-"$i" 2>&1
if cd test-"$i"; then
hg verify
cd ..
fi
done
cd test-8
hg pull ../test-7
hg verify
cd ..
cd test-1
hg pull -r 4 http://localhost:$HGPORT/ 2>&1 | sed -e 's,:[0-9][0-9]*/,/,'
hg verify
hg pull http://localhost:$HGPORT/ 2>&1 | sed -e 's,:[0-9][0-9]*/,/,'
cd ..
cd test-2
hg pull -r 5 http://localhost:$HGPORT/ 2>&1 | sed -e 's,:[0-9][0-9]*/,/,'
hg verify
hg pull http://localhost:$HGPORT/ 2>&1 | sed -e 's,:[0-9][0-9]*/,/,'
hg verify
cd ..