PG-2540 Fix buffer overread and simplify code for parsing comments

[jeltz]

The code was hard to read for no obvious gain so this attempts to make the code read more like one would expect. Also remove a redundant comment.

PG-0

[codecov]

Codecov Report

❌ Patch coverage is 36.84211% with 12 lines in your changes missing coverage. Please review.
✅ Project coverage is 87.68%. Comparing base (86edb37) to head (6fec9fd).

Files with missing lines	Patch %	Lines
src/pg_stat_monitor.c	36.84%	6 Missing and 6 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #752      +/-   ##
==========================================
- Coverage   88.33%   87.68%   -0.65%     
==========================================
  Files           3        3              
  Lines        1329     1332       +3     
  Branches      201      202       +1     
==========================================
- Hits         1174     1168       -6     
- Misses         72       79       +7     
- Partials       83       85       +2     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[dutow]

There's also a possible over-read in this code. While the pr only aims for simplification, should we also fix that too, if we are already changing this area of code?

SELECT 1 /*tail*/ or SELECT 1 /*oops

inner while loop consumes */ and points to the null. End of foor loop increases pointer again. *q_iter used by the for loop is now the character after null.

The unterminated comment is similar, the while loop already ends with q_iter pointing to the null byte.

[jeltz]

@dutow Does this fix look correct for the over-read?

[dutow]

Looks good now.

We should add more coverage to this (the cases I already mentioned for asan + maybe something like SELECT 1 /*a*//*b*/, but that can also be a separate PR.

[dutow]

Another question: since this now also includes a bugfix, it should have a proper ticket?

[jeltz]

Yeah, we probably should add a test case and a proper ticket now.

[jeltz]

Pushed the bugfix as a separate commit with tests and added a reference to a bug in Jira.