localization - Tool in the gettext suite to unify source strings with fuzzy match? -
is there way leverage tools in gettext suite fuzzy match source strings within 1 po file find strings identical? seem useful quality check improve sources. example:
#: my_file msgid "sorry, went wrong" msgstr "" #: some_other_file msgid "sorry, went wrong." msgstr "" #: yet_another_file msgid "sorry, wrong" msgstr ""
these strings virtually identical , source code possibly changed use same message in each instance. reduce l10n work , make ui more coherent. seem me fuzzy match algorithm in msgmerge
should pretty suited identify these instances. yet not find obvious way this.
you don't want kind of folding without human supervision. translation tools have feature, human should validate such folding. can't identical strings because of context.
why:
- buttons ("commands") translated differently labels , titles ("descriptions)")
example: "print" translated french "imprimer" (buttons) or "impression" (titles) - gender, number, case, change translation.
example: translating "new" button spanish can give "nuovo" (masculine, singular), "nuevos" (masculine, plural), "nueva" (feminine, singular) "nuevas" (feminine, plural) - the same word can translated differently if has different meaning.
example: "scan" have different translations if scanning disk (for virus) or scanning piece of paper.
so, don't want "magically merge strings" save few cents, if price lower translations quality.
Comments
Post a Comment