feat(server): #21109 support leading wildcards in resource matching o…

[clifflau1120]

Overview

This PR closes #21109.

It adds support to leading wildcards while matching resources against events.

Checklist

• This pull request references any related issue by including "closes <link to issue>"
  • If no issue exists and your change is not a small fix, please create an issue first.
• If this pull request adds new functionality, it includes unit tests that cover the changes
• If this pull request removes docs files, it includes redirect settings in mint.json.
• If this pull request adds functions or classes, it includes helpful docstrings.

[codspeed-hq]

Merging this PR will not alter performance

✅ 2 untouched benchmarks

---

_{Comparing clifflau1120:21109-support-leading-wildcard-in-resource-matching-of-event-triggers (0dc2463) with main (447059a)}

[desertaxle]

Thanks for the contribution @clifflau1120 ! The in-memory matching change looks correct directionally, but I think there are a few things that need to be addressed before this is ready to merge:

1. Handle the double-wildcard case (*foo*)

As written, a pattern like *middle* will hit the startswith("*") branch and do value.endswith("middle*") — matching on a literal trailing *, which is almost certainly not what a user would intend. This needs to be handled, either as a contains check or explicitly rejected with a validation error:

if expected.startswith("*") and expected.endswith("*") and len(expected) > 1:
    match = expected[1:-1] in value
elif expected.startswith("*"):
    match = value.endswith(expected[1:])
elif expected.endswith("*"):
    match = value.startswith(expected[:-1])

2. LabelOperations in filters.py needs matching changes

This is the important one. The LabelOperations class translates wildcard patterns into SQL for database queries, and it currently only understands trailing wildcards (→ startswith/LIKE 'prefix%') or exact match. A leading wildcard like *.csv will fall into the simple (exact match) bucket and generate resource_id IN ('*.csv') instead of the correct LIKE '%.csv'.

This means triggers will evaluate events correctly in real-time (in-memory path), but historical event queries and the events API will silently return wrong results (DB path).

3. Minor things

• The docstring on matches() should be updated to mention leading wildcards
• The original test case ("any.old.*", "any.old.stuff") was removed rather than kept alongside the new case — would be good to keep it for trailing-wildcard coverage
• Worth adding test cases for the negation + leading wildcard combo ("!*.csv")

---