Currently, the parser is very context-sensitive. This means various errors will appear as parsing errors. For instance, a(1) results in
Syntax error detected: Unexpected "T_BRACE" at character 2.
even though it is intuitively "parsable", and could give a better error message, saying that the function a doesn't exist.
Therefore, the parser should be more lenient and defers various checks to later phases (at compile-time)
| Subject | Repo | Branch | Lines +/- | |
|---|---|---|---|---|
| Add a static analyzer for the filter language | mediawiki/extensions/AbuseFilter | master | +1 K -171 |
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | Nullzero | T260903 Create a static analyzer for the filter language | |||
| Resolved | Nullzero | T238709 The parser should be more lenient |
Change 551966 had a related patch set uploaded (by Nullzero; owner: Nullzero):
[mediawiki/extensions/AbuseFilter@master] [DO NOT MERGE] Add a static analyzer for the filter language
Change 551966 merged by jenkins-bot:
[mediawiki/extensions/AbuseFilter@master] Add a static analyzer for the filter language