fix: Regenerate language_pack #502
Conversation
|
@howlger What do you think? |
|
In my personal extension repo I have a CI job that generates PRs automatically, maybe something like this could be implemented https://github.com/sebthom/extra-syntax-highlighting-eclipse-plugin/blob/main/.github/workflows/update-syntaxes.yml |
|
+1, (based on the code review without testing): +1 also for @sebthom's idea Independent from this, would it be better to have separate plugins and features, instead of a single plugin/feature containing everything? |
This was done by running `mvn package` in `_update`. The autogenerated files have one small error that was previously fixed in 242dcf1 and this change therefore preserves that fix by excluding this diff from the commit: ```diff diff --git a/org.eclipse.tm4e.language_pack/javascript/Regular Expressions (JavaScript).tmLanguage b/org.eclipse.tm4e.language_pack/javascript/Regular Expressions (JavaScript).tmLanguage index 54c7651c..1dda7806 100644 --- a/org.eclipse.tm4e.language_pack/javascript/Regular Expressions (JavaScript).tmLanguage +++ b/org.eclipse.tm4e.language_pack/javascript/Regular Expressions (JavaScript).tmLanguage @@ -230,7 +230,7 @@ </dict> </dict> <key>scopeName</key> - <string>lngpck.source.js.regexp</string> + <string>source.js.regexp</string> <key>uuid</key> <string>AC8679DE-3AC7-4056-84F9-69A7ADC29DDD</string> </dict> ``` And bumping version numbers due to updated language packs Fixes eclipse-tm4e#501
jonahgraham
force-pushed
the
update_lang_packs
branch
from
9edb497
to
24c5e09
Compare
I have updated the commit and the description of the PR. An updated look over it would be appreciated.
+1 From me too.
If you mean having as many plug-ins as there are languages in the language pack, I'm not keen on that idea. |
Looks good to me. It would be good to test this by opening a simple example for each language/format so that the grammar is loaded. If I remember correctly, not all regular expressions work; lookbehind of an unknown number of characters is not supported and causes an error when loading the grammar. But that might have changed in the meantime. @sebthom, please correct me if this is not true. If you'd like me to test it rather than just roughly look it over, I can do that the week after next. |
|
@howlger I'd appreciate if you could test the changes. it would actually also be nice to have an automated test that tries to load/parse the grammars. |
|
TM4E is switching to |
sebthom
changed the title
|
+1 Sorry for the delay. I tested this pull request manually with sample files for the different formats. I will try to fix the generation to avoid overwriting of the manual fix 242dcf1, and if I am successful, I will submit it as a separate pull request. After that, I also plan to look at how to automatically test it, to at least find the regular expressions that cause errors at runtime. But I can't promise anything and would be happy if someone else would do that. |
|
Superseded by #550 |
This was done by running
mvn packagein_update.The autogenerated files have one small error that was
previously fixed in jonahgraham@242dcf1
and this change therefore preserves that fix by excluding this
diff from the commit:
And bumping version numbers due to updated language packs
Fixes #501