fix bad for range loop

[neheb]

This loop is supposed to break when it encounters a match.

Signed-off-by: Rosen Penev [email protected]

I tried converting this to std::find_if. Didn't work.

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (main@091fd77). Click here to learn what that means.
The diff coverage is n/a.

❗ Current head c458e70 differs from pull request most recent head 0970053. Consider uploading reports for the commit 0970053 to get more accurate results

@@           Coverage Diff           @@
##             main    #1635   +/-   ##
=======================================
  Coverage        ?   66.91%           
=======================================
  Files           ?      151           
  Lines           ?    20805           
  Branches        ?        0           
=======================================
  Hits            ?    13921           
  Misses          ?     6884           
  Partials        ?        0           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 091fd77...0970053. Read the comment docs.

[neheb]

Maybe the find is better.

[neheb]

find causes address sanitizer issues. Oh well.

[piponazo]

Great! It looks you have fixed the tests that were failing on Windows from time to time. It is interesting to see that those tests were not failing on Linux nor on Mac.

I can backport this change to the 0.27-maintenance.

Thanks!

[piponazo]

Ups ... weird, in the PR view all the checks look green, but when I go to the github actions page, I see that some of these flaky tests (on BMFF code) on windows are still failing:

https://github.com/Exiv2/exiv2/runs/2573006142?check_suite_focus=true

Just for clarification. Were you trying to fix some defect in Exiv2 that was already reported?

[piponazo]

I was not looking to the right github actions link ... that is the one were all the windows jobs are passing:
https://github.com/Exiv2/exiv2/actions/runs/837867978

What I do not understand is that the failing tests seemed to be on BMFF while your change is in the epsimage.cpp file.

[piponazo]

I tried to cherry-pick the change to include it in 0.27-maintenance, but we do not need it there

[neheb]

No this was a bad transformation on my part.