tballison commented on PR #2917:
URL: https://github.com/apache/tika/pull/2917#issuecomment-4842640631
From claude's review:
```
1. The comments are mutation-tooling exhaust and will rot. Lines like //
kills the surviving mutants on lines 70-71 (EQUAL_ELSE + removed call...), //
InlineConstant (1.0 -> 2.0) ... on lines 81-82, // line 94 (NO_COVERAGE)
describe why PIT generated the test and hard-code production line numbers. The
moment
StandardHtmlEncodingDetector.java shifts by a line, those comments are
wrong. I'd ask the contributor to rewrite them to state the behavior under test
(e.g.
"charset comes from Content-Encoding when Content-Type is absent") and
drop the mutant/line-number references entirely.
2. customMarkLimit's comment is slightly off. "the meta tag beyond 100
bytes won't be found" — the tag actually starts at byte 80 (inside the limit);
it's the
charset value that gets truncated at byte 100. The test logic is correct;
only the wording misleads.
3. Minor/stylistic: assertEquals(1.0f, getConfidence()) has no delta. It
compiles (JUnit 5 has the (float,float) overload) and passes because production
hard-codes 1.0f, so it's fine — a purist might add a delta.
```
These make sense to me.
Thank you for opening this and improving our unit tests.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
