view tests/test-encoding-align @ 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

########################################

HGENCODING=utf-8
export HGENCODING

hg init t
cd t

python << EOF
# (byte, width) = (6, 4)
s = "\xe7\x9f\xad\xe5\x90\x8d"
# (byte, width) = (7, 7): odd width is good for alignment test
m = "MIDDLE_"
# (byte, width) = (18, 12)
l = "\xe9\x95\xb7\xe3\x81\x84\xe9\x95\xb7\xe3\x81\x84\xe5\x90\x8d\xe5\x89\x8d"

f = file('s', 'w'); f.write(s); f.close()
f = file('m', 'w'); f.write(m); f.close()
f = file('l', 'w'); f.write(l); f.close()

# instant extension to show list of options
f = file('showoptlist.py', 'w'); f.write("""# encoding: utf-8
def showoptlist(ui, repo, *pats, **opts):
    '''dummy command to show option descriptions'''
    return 0

cmdtable = {
    'showoptlist':
        (showoptlist,
         [('s', 'opt1', '', 'short width',  '""" + s + """'),
          ('m', 'opt2', '', 'middle width', '""" + m + """'),
          ('l', 'opt3', '', 'long width',   '""" + l + """')
         ],
         ""
        )
}
""")
f.close()
EOF

S=`cat s`
M=`cat m`
L=`cat l`

########################################
#### alignment of:
####     - option descriptions in help

cat <<EOF > .hg/hgrc
[extensions]
ja_ext = `pwd`/showoptlist.py
EOF
echo '% check alignment of option descriptions in help'
hg help showoptlist

########################################
#### alignment of:
####     - user names in annotate
####     - file names in diffstat

rm -f s; touch s
rm -f m; touch m
rm -f l; touch l

#### add files

cp s $S
hg add $S
cp m $M
hg add $M
cp l $L
hg add $L

#### commit(1)

echo 'first line(1)' >> s; cp s $S
echo 'first line(2)' >> m; cp m $M
echo 'first line(3)' >> l; cp l $L
hg commit -m 'first commit' -u $S -d "1000000 0"

#### commit(2)

echo 'second line(1)' >> s; cp s $S
echo 'second line(2)' >> m; cp m $M
echo 'second line(3)' >> l; cp l $L
hg commit -m 'second commit' -u $M -d "1000000 0"

#### commit(3)

echo 'third line(1)' >> s; cp s $S
echo 'third line(2)' >> m; cp m $M
echo 'third line(3)' >> l; cp l $L
hg commit -m 'third commit' -u $L -d "1000000 0"

#### check

echo '% check alignment of user names in annotate'
hg annotate -u $M
echo '% check alignment of filenames in diffstat'
hg diff -c tip --stat

########################################
#### alignment of:
####     - branch names in list
####     - tag names in list

#### add branches/tags

hg branch $S
hg tag -d "1000000 0" $S
hg branch $M
hg tag -d "1000000 0" $M
hg branch $L
hg tag -d "1000000 0" $L

#### check

echo '% check alignment of branches'
hg tags
echo '% check alignment of tags'
hg tags

########################################

exit 0