view tests/test-convert-git @ 12252:4481f8a93c7a stable

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

"$TESTDIR/hghave" git || exit 80

echo "[extensions]" >> $HGRCPATH
echo "convert=" >> $HGRCPATH
echo 'hgext.graphlog =' >> $HGRCPATH

GIT_AUTHOR_NAME='test'; export GIT_AUTHOR_NAME
GIT_AUTHOR_EMAIL='[e-mail]'; export GIT_AUTHOR_EMAIL
GIT_AUTHOR_DATE="2007-01-01 00:00:00 +0000"; export GIT_AUTHOR_DATE
GIT_COMMITTER_NAME="$GIT_AUTHOR_NAME"; export GIT_COMMITTER_NAME
GIT_COMMITTER_EMAIL="$GIT_AUTHOR_EMAIL"; export GIT_COMMITTER_EMAIL
GIT_COMMITTER_DATE="$GIT_AUTHOR_DATE"; export GIT_COMMITTER_DATE

count=10
commit()
{
    GIT_AUTHOR_DATE="2007-01-01 00:00:$count +0000"
    GIT_COMMITTER_DATE="$GIT_AUTHOR_DATE"
    git commit "$@" >/dev/null 2>/dev/null || echo "git commit error"
    count=`expr $count + 1`
}

mkdir git-repo
cd git-repo
git init-db >/dev/null 2>/dev/null
echo a > a
mkdir d
echo b > d/b
git add a d
commit -a -m t1

# Remove the directory, then try to replace it with a file
# (issue 754)
git rm -f d/b
commit -m t2
echo d > d
git add d
commit -m t3

echo b >> a
commit -a -m t4.1

git checkout -b other HEAD~ >/dev/null 2>/dev/null
echo c > a
echo a >> a
commit -a -m t4.2

git checkout master >/dev/null 2>/dev/null
git pull --no-commit . other > /dev/null 2>/dev/null
commit -m 'Merge branch other'
cd ..

hg convert --datesort git-repo
hg up -q -R git-repo-hg
hg -R git-repo-hg tip -v

count=10
mkdir git-repo2
cd git-repo2
git init-db >/dev/null 2>/dev/null

echo foo > foo
git add foo
commit -a -m 'add foo'

echo >> foo
commit -a -m 'change foo'

git checkout -b Bar HEAD~ >/dev/null 2>/dev/null
echo quux >> quux
git add quux
commit -a -m 'add quux'

echo bar > bar
git add bar
commit -a -m 'add bar'

git checkout -b Baz HEAD~ >/dev/null 2>/dev/null
echo baz > baz
git add baz
commit -a -m 'add baz'

git checkout master >/dev/null 2>/dev/null
git pull --no-commit . Bar Baz > /dev/null 2>/dev/null
commit -m 'Octopus merge'

echo bar >> bar
commit -a -m 'change bar'

git checkout -b Foo HEAD~ >/dev/null 2>/dev/null
echo >> foo
commit -a -m 'change foo'

git checkout master >/dev/null 2>/dev/null
git pull --no-commit -s ours . Foo > /dev/null 2>/dev/null
commit -m 'Discard change to foo'

cd ..

glog()
{
    hg glog --template '{rev} "{desc|firstline}" files: {files}\n' "$@"
}

splitrepo()
{
    msg="$1"
    files="$2"
    opts=$3
    echo "% $files: $msg"
    prefix=`echo "$files" | sed -e 's/ /-/g'`
    fmap="$prefix.fmap"
    repo="$prefix.repo"
    for i in $files; do
	echo "include $i" >> "$fmap"
    done
    hg -q convert $opts --filemap "$fmap" --datesort git-repo2 "$repo"
    hg up -q -R "$repo"
    glog -R "$repo"
    hg -R "$repo" manifest --debug
}

echo '% full conversion'
hg -q convert --datesort git-repo2 fullrepo
hg up -q -R fullrepo
glog -R fullrepo
hg -R fullrepo manifest --debug

splitrepo 'octopus merge' 'foo bar baz'

splitrepo 'only some parents of an octopus merge; "discard" a head' 'foo baz quux'

echo
echo '% test binary conversion (issue 1359)'
mkdir git-repo3
cd git-repo3
git init-db >/dev/null 2>/dev/null
python -c 'file("b", "wb").write("".join([chr(i) for i in range(256)])*16)'
git add b
commit -a -m addbinary
cd ..

echo '% convert binary file'
hg convert git-repo3 git-repo3-hg

cd git-repo3-hg
hg up -C
python -c 'print len(file("b", "rb").read())'
cd ..

echo
echo '% test author vs committer'
mkdir git-repo4
cd git-repo4
git init-db >/dev/null 2>/dev/null
echo >> foo
git add foo
commit -a -m addfoo
echo >> foo
GIT_AUTHOR_NAME="nottest"
commit -a -m addfoo2
cd ..

echo '% convert author committer'
hg convert git-repo4 git-repo4-hg
hg -R git-repo4-hg log -v

echo '% --sourceorder should fail'
hg convert --sourcesort git-repo4 git-repo4-sourcesort-hg

echo '% damage git repository and convert again'
cat > damage.py <<EOF
import os
for root, dirs, files in os.walk('git-repo4/.git/objects'):
    if files:
        path = os.path.join(root, files[0])
        os.remove(path)
        break
EOF
python damage.py
hg convert git-repo4 git-repo4-broken-hg 2>&1 | \
    grep 'abort:' | sed 's/abort:.*/abort:/g'

true