| Age | Commit message (Collapse) | Author |
|---|
|
Signed-off-by: Lukas Fleischer <archlinux@cryptocrack.de>
|
|
Signed-off-by: Lukas Fleischer <archlinux@cryptocrack.de>
|
|
Signed-off-by: Lukas Fleischer <archlinux@cryptocrack.de>
|
|
Signed-off-by: Lukas Fleischer <archlinux@cryptocrack.de>
|
|
Signed-off-by: Lukas Fleischer <archlinux@cryptocrack.de>
|
|
Signed-off-by: Lukas Fleischer <archlinux@cryptocrack.de>
|
|
This ensures we have proper native portable objects instead of those
created by sed(1)'ing the former translation files.
Signed-off-by: Lukas Fleischer <archlinux@cryptocrack.de>
|
|
Those are just dummy headers created by ripping them off from "aur.pot"
and changing the "Content-Type" charset to UTF-8.
Needed to make msgmerge(1) operate on our converted ".po" files -
otherwise it would fail to detect the files' encodings and do nothing
but throw a pile of "invalid multibyte sequence" errors.
Signed-off-by: Lukas Fleischer <archlinux@cryptocrack.de>
|
|
Put new gettext compatible files into "po/" while keeping old
translations in "web/lang/".
Conversion was done using following hacky shell script:
----
for f in *.po; do
# Remove "<?php" shebang and "global" statements.
sed -i '1d; /^global/d' "$f"
# Convert former translation strings into real PO statements.
sed -i 's/^$_t\[\(.*\)\] = \(.*\);.*/msgid \1\nmsgstr \2/' "$f"
# Ensure there always is an empty line after each msgstr.
sed -i '/^msgstr/!b; n; /^$/!i\
' "$f"
done
----
Signed-off-by: Lukas Fleischer <archlinux@cryptocrack.de>
|
