1
0
mirror of https://github.com/matrix-org/matrix-react-sdk.git synced 2025-07-28 15:22:05 +03:00
Commit Graph

206 Commits

Author SHA1 Message Date
8447d15013 Don't include the |other in the translation value 2017-10-24 14:26:30 +01:00
93fca8439e Merge pull request #1521 from matrix-org/dbkr/geni18n-addplural
Fix bug with gen-i18n/js when adding new plurals
2017-10-24 11:15:51 +01:00
46e9d4197a Use the 'and' feature! 2017-10-24 10:21:41 +01:00
2340c1308e Fix bug with gen-i18n/js when adding new plurals 2017-10-23 19:53:17 +01:00
54458f1438 Actually even out brackets 2017-10-23 14:02:58 +01:00
8d1aea5b2e Misc PR review fixes 2017-10-23 10:18:29 +01:00
317ad64ae6 Make the gen-i18n script validate _t calls
And throw a massive tantrum if you've messed up your format strings.

Because broken format strings making their way into the app cause it
to throw exceptions.
2017-10-20 18:38:22 +01:00
d397858e81 Don't remove other plural variants 2017-10-19 14:39:23 +01:00
8a9de7343c Add script to prune unused translations 2017-10-19 12:11:21 +01:00
027a70ffe7 Add gen-i18n as a 'binary'
and add appropriate shebang
2017-10-19 10:51:54 +01:00
1bada070ca Make gen-i18n support 'HTML'
Where by 'HTML' I mean just run the same regex as riot does over
some text.

Also make it walk multiple paths.

This mostly means it can be used for riot-web.
2017-10-18 19:36:07 +01:00
2bbac903f1 'base file' is a bit overloaded by weblate 2017-10-18 14:45:04 +01:00
caa4979e37 Comment ignoring getTKey returning null 2017-10-18 14:43:18 +01:00
63a7d37f5a commented logging 2017-10-18 14:41:54 +01:00
a2bfdd8925 missing saemi 2017-10-18 14:33:21 +01:00
f759467193 Get English plural strings from an input file 2017-10-18 14:30:55 +01:00
11761f535d Copyright 2017-10-18 13:39:04 +01:00
8370c823b6 Script to generate the translations base file
Also replaces the translations base file with the output of the
script.
2017-10-17 19:19:01 +01:00
07b6040c39 Update travis script to use lintwithexclusions 2017-09-28 11:33:36 +01:00
2de4dd7b71 Filter out unsupported emoji from the emoji meta data
Because apparently emojione provides meta data for more than the emojis it supports itself.

Fixes https://github.com/vector-im/riot-web/issues/4504
2017-07-05 17:09:16 +01:00
982b009b90 Implement ascii emoji tab completion
When a fully plaintext, ascii emoji is typed like ";-)", pressing tab will suggest emojione to replace it with based off of the meta data provided by emojione.

e.g. the aliases_ascii for `😃` are [":D",":-D","=D"] so typing ":D *tab*" will insert a real 😃
2017-06-29 11:29:55 +01:00
48eb6f251b Update comment 2017-06-28 13:44:44 +01:00
30f80b57f2 Instead of making this part of the build process, let the devs generate stripped-emoji.json
With the expectation that when the file needs to be regenerated to include other keys, the developer will run this script.
2017-06-28 13:28:48 +01:00
2b8da85726 Strip the emoji meta-data for the data we need
This is done at build time by parsing emojione/emoji.json, stripping it and then writing to ./lib/stripped-emoji.json.
2017-06-28 11:36:19 +01:00
4eb7419335 Don't put in spurious newline 2017-06-13 11:54:28 +01:00
89c7477bd6 Add script to copy translations between files
So we can fill in missing translations from different dialects

Use it to fill in missing strings in pt from pt_BR
2017-06-13 11:39:37 +01:00
74e4260556 fix fullstops 2017-06-08 15:14:46 +01:00
9bda212cb1 Make travis/jenkins check for lint on files which are clean to start with 2017-06-08 14:03:30 +01:00
fe487232ad sync fullstops everywhere 2017-06-07 18:15:56 +01:00
650d45466c fix up missing strings caused by punctuation changes 2017-06-01 22:06:02 +01:00
b4284cf000 fix up Can't connect to HS i18n with _tJsx 2017-06-01 01:17:39 +01:00
85f830db71 warn for duplicate translation keys 2017-05-28 21:28:11 +01:00
dba35f4901 extend script to check for broken variables 2017-05-27 20:14:52 +01:00
796158422c list the strings remaining to be translated 2017-05-27 19:03:26 +01:00
b697c0472f fix fullstop 2017-05-27 18:41:43 +01:00
bc433bc2a2 ignore country codes 2017-05-27 18:38:37 +01:00
8d79ab6ce2 track entries that exist in en_EN but not in the src... 2017-05-27 18:35:34 +01:00
3302fc690f fix all missing translations picked up by check-i18n.pl 2017-05-27 18:20:44 +01:00
1305fd619e fix quoting 2017-05-27 17:42:58 +01:00
f442a665c8 fix more i18n punctuation fails 2017-05-27 17:36:02 +01:00
730258bd3c fix punctuation on e2e warning 2017-05-27 17:30:32 +01:00
e684cd5d28 track broken i18n files 2017-05-27 17:27:58 +01:00
44f8ee44e6 check for missing src strings too 2017-05-27 17:13:17 +01:00
c8dab3ae65 script for checking health of i18n 2017-05-27 16:47:15 +01:00
c34edbec3e include fullstop fixup 2017-05-27 16:02:51 +01:00
0cacf5b1a0 script to make punctuation consistent in i18n strings 2017-05-27 15:02:05 +01:00
4df51a2b0b Reskindex to a temp file, to confuse the babel watcher less (#908)
I'm seeing a lot of instances where the babel watcher picks up a half-written src/component-index,
and generates an empty lib file - which it then doesn't update when src/component-index is updated.

Empirically, this seems to make it better.
2017-05-19 11:44:04 +01:00
15201d86aa Prevent reskindex -w from running when file names have not changed 2017-05-15 10:16:47 +01:00
9af9603373 Only watch indexed files 2017-05-08 16:29:53 +01:00
2eaaa97451 Give reskindex.js a watch mode (-w)
`scripts/reskindex.js -w` will run reskindex in watch mode whereby FS events will cause a reskindex to occur.

This depends on `chokidar`
2017-05-08 15:07:05 +01:00