Page MenuHomePhabricator

mailman - convert all language templates to UTF-8 and remove illegal chars
Closed, ResolvedPublic

Description

convert all mailman global language templates for listinfo pages (back) to UTF-8, remove all illegal characters (that break puppet), add correct characters instead, re-declare them as UTF-8 files

The fix described on T114289#1691038 for German needs to be applied to multiple other languages as well.


affected were Korean, German (linked tickets) and the following:

~/wmf/puppet/modules/mailman/files/templates$ find . -name listinfo.html -exec file {} \; | grep -v UTF-8
./ro/listinfo.html: HTML document, ASCII text
./ca/listinfo.html: HTML document, ISO-8859 text
./nl/listinfo.html: HTML document, ASCII text
./es/listinfo.html: HTML document, ISO-8859 text
./it/listinfo.html: HTML document, ASCII text
./fi/listinfo.html: HTML document, ISO-8859 text
./id/listinfo.html: HTML document, ASCII text
./en/listinfo.html: HTML document, ASCII text

Event Timeline

Dzahn raised the priority of this task from to Medium.
Dzahn updated the task description. (Show Details)
Dzahn added a project: Wikimedia-Mailing-lists.
Dzahn added subscribers: Dzahn, JohnLewis, Platonides.

Change 242749 had a related patch set uploaded (by Dzahn):
mailman: convert language templates, ca,es,fi,fr

https://gerrit.wikimedia.org/r/242749

Dzahn set Security to None.

I say convert "back" because this is partially from https://gerrit.wikimedia.org/r/#/c/157708/ but that was a fix before the mailman upgrade back in 2014.

Change 242749 merged by Dzahn:
mailman: convert language templates, ca,es,fi,fr

https://gerrit.wikimedia.org/r/242749

Change 242771 had a related patch set uploaded (by Dzahn):
mailman: convert language templates, pt.2

https://gerrit.wikimedia.org/r/242771

Change 242771 merged by Dzahn:
mailman: convert language templates, pt.2

https://gerrit.wikimedia.org/r/242771

done.

./fr/listinfo.html: HTML document, UTF-8 Unicode text
./de/listinfo.html: HTML document, UTF-8 Unicode text
./hu/listinfo.html: HTML document, UTF-8 Unicode text
./zh/listinfo.html: HTML document, UTF-8 Unicode text
./pt_BR/listinfo.html: HTML document, UTF-8 Unicode text
./ta/listinfo.html: HTML document, UTF-8 Unicode text
./ro/listinfo.html: HTML document, ASCII text
./ca/listinfo.html: HTML document, UTF-8 Unicode text
./ja/listinfo.html: HTML document, UTF-8 Unicode text
./sv/listinfo.html: HTML document, UTF-8 Unicode text
./pl/listinfo.html: HTML document, UTF-8 Unicode text
./ar/listinfo.html: HTML document, UTF-8 Unicode text
./nl/listinfo.html: HTML document, ASCII text
./sk/listinfo.html: HTML document, UTF-8 Unicode text
./uk/listinfo.html: HTML document, UTF-8 Unicode text
./ko/listinfo.html: HTML document, UTF-8 Unicode text
./cs/listinfo.html: HTML document, UTF-8 Unicode text
./he/listinfo.html: HTML document, UTF-8 Unicode text
./lt/listinfo.html: HTML document, UTF-8 Unicode text
./vi/listinfo.html: HTML document, UTF-8 Unicode text
./ru/listinfo.html: HTML document, UTF-8 Unicode text
./no/listinfo.html: HTML document, UTF-8 Unicode text
./hi/listinfo.html: HTML document, UTF-8 Unicode text
./es/listinfo.html: HTML document, UTF-8 Unicode text
./it/listinfo.html: HTML document, ASCII text
./fi/listinfo.html: HTML document, UTF-8 Unicode text
./id/listinfo.html: HTML document, ASCII text
./hr/listinfo.html: HTML document, UTF-8 Unicode text
./tr/listinfo.html: HTML document, UTF-8 Unicode text
./fa/listinfo.html: HTML document, UTF-8 Unicode text
./pt/listinfo.html: HTML document, UTF-8 Unicode text
./en/listinfo.html: HTML document, ASCII text

Where it says ASCII text it's also UTF-8 but doesn't use any characters outside ASCII. I ran a loop across this to convert them all with iconv -t utf8 and there is no more change here.

Also, after doing T114289#1691038 on the server side, now puppet runs complete without crying about invalid byte sequence in UTF-8.

Dzahn claimed this task.