convert/darcs: handle non-ASCII metadata in darcs changelog (issue2354)
Given a commit author or message with non-ASCII characters in a darcs
repo, convert would raise a UnicodeEncodeError when adding changesets
to the hg changelog.
This happened because etree returns back unicode objects for any text
it can't encode into ASCII. convert was passing these objects to
changelog.add(), which would then attempt encoding.fromlocal() on
them.
This patch ensures converter_source.recode() is called on each piece
of commit data returned by etree.
(Also note that darcs is currently encoding agnostic and will print
out whatever is in a patch's metadata byte-for-byte, even in the XML
changelog.)
#!/bin/sh
hg init test
cd test
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
cd ..
for i in 0 1 2 3 4 5 6 7 8; do
hg clone -r "$i" test test-"$i"
cd test-"$i"
hg verify
cd ..
done
cd test-8
hg pull ../test-7
hg verify