tests/test-git-export

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 a
cd a

echo start > start
hg ci -Amstart -d '0 0'
echo new > new
hg ci -Amnew -d '0 0'
echo '% new file'
hg diff --git -r 0

hg cp new copy
hg ci -mcopy -d '0 0'
echo '% copy'
hg diff --git -r 1:tip

hg mv copy rename
hg ci -mrename -d '0 0'
echo '% rename'
hg diff --git -r 2:tip

hg rm rename
hg ci -mdelete -d '0 0'
echo '% delete'
hg diff --git -r 3:tip

cat > src <<EOF
1
2
3
4
5
EOF
hg ci -Amsrc -d '0 0'
chmod +x src
hg ci -munexec -d '0 0'
echo '% chmod 644'
hg diff --git -r 5:tip

hg mv src dst
chmod -x dst
echo a >> dst
hg ci -mrenamemod -d '0 0'
echo '% rename+mod+chmod'
hg diff --git -r 6:tip

echo '% nonexistent in tip+chmod'
hg diff --git -r 5:6

echo '% binary diff'
cp $TESTDIR/binfile.bin .
hg add binfile.bin
hg diff --git > b.diff
cat b.diff

echo '% import binary diff'
hg revert binfile.bin
rm binfile.bin
hg import -mfoo b.diff
cmp binfile.bin $TESTDIR/binfile.bin

echo
echo '% rename binary file'
hg mv binfile.bin renamed.bin
hg diff --git

echo
echo '% diff across many revisions'
hg mv dst dst2
hg ci -m 'mv dst dst2' -d '0 0'

echo >> start
hg ci -m 'change start' -d '0 0'

hg revert -r -2 start
hg mv dst2 dst3
hg ci -m 'mv dst2 dst3; revert start' -d '0 0'

hg diff --git -r 9:11
echo '%  reversed'
hg diff --git -r 11:9

echo a >> foo
hg add foo
hg ci -m 'add foo'
echo b >> foo
hg ci -m 'change foo'
hg mv foo bar
hg ci -m 'mv foo bar'
echo c >> bar
hg ci -m 'change bar'

echo
echo '% file created before r1 and renamed before r2'
hg diff --git -r -3:-1
echo '%  reversed'
hg diff --git -r -1:-3
echo
echo '% file created in r1 and renamed before r2'
hg diff --git -r -4:-1
echo '%  reversed'
hg diff --git -r -1:-4
echo
echo '% file created after r1 and renamed before r2'
hg diff --git -r -5:-1
echo '%  reversed'
hg diff --git -r -1:-5

echo
echo '% comparing with the working dir'
echo >> start
hg ci -m 'change start again' -d '0 0'

echo > created
hg add created
hg ci -m 'add created'

hg mv created created2
hg ci -m 'mv created created2'

hg mv created2 created3
echo "% there's a copy in the working dir..."
hg diff --git
echo
echo "% ...but there's another copy between the original rev and the wd"
hg diff --git -r -2
echo
echo "% ...but the source of the copy was created after the original rev"
hg diff --git -r -3
hg ci -m 'mv created2 created3'

echo > brand-new
hg add brand-new
hg ci -m 'add brand-new'
hg mv brand-new brand-new2
echo '% created in parent of wd; renamed in the wd'
hg diff --git

echo
echo '% created between r1 and parent of wd; renamed in the wd'
hg diff --git -r -2
hg ci -m 'mv brand-new brand-new2'

echo '% one file is copied to many destinations and removed'
hg cp brand-new2 brand-new3
hg mv brand-new2 brand-new3-2
hg ci -m 'multiple renames/copies'
hg diff --git -r -2 -r -1
echo '%  reversed'
hg diff --git -r -1 -r -2

echo '% there should be a trailing TAB if there are spaces in the file name'
echo foo > 'with spaces'
hg add 'with spaces'
hg diff --git
hg ci -m 'add filename with spaces'