merge: do early copy to deal with issue636
Without copies/renames, merges source names are 1:1 with their
targets. Copies and renames introduce the possibility that there will
be two merges with the same input but different output. By doing the
copy to the destination name before the merge, the actual merge
becomes 1:1 again, and no source is the input to two different merges.
- add a preliminary scan to applyupdates to do copies
- for the merge action, pass the old name (for finding ancestors) and
the new name (for input to the merge) to filemerge
- eliminate the old post-merge copy
- lookup file contents from new name in filemerge
- pass new name to external merge helper
- report merge failure at new name
- add a test
#!/bin/sh
HGMERGE=true; export HGMERGE
set -e
mkdir r1
cd r1
hg init
echo a > a
hg addremove
hg commit -m "1" -d "1000000 0"
hg clone . ../r2
cd ../r2
hg up
echo abc > a
hg diff --nodates
cd ../r1
echo b > b
echo a2 > a
hg addremove
hg commit -m "2" -d "1000000 0"
cd ../r2
hg -q pull ../r1
hg status
hg parents
hg --debug up
hg parents
hg --debug up 0
hg parents
hg --debug merge || echo failed
hg parents
hg --debug up
hg parents
hg -v history
hg diff --nodates
# create a second head
cd ../r1
hg up 0
echo b2 > b
echo a3 > a
hg addremove
hg commit -m "3" -d "1000000 0"
cd ../r2
hg -q pull ../r1
hg status
hg parents
hg --debug up || echo failed
hg --debug merge || echo failed
hg --debug merge -f
hg parents
hg diff --nodates
# test a local add
cd ..
hg init a
hg init b
echo a > a/a
echo a > b/a
hg --cwd a commit -A -m a
cd b
hg add a
hg pull -u ../a
hg st