Exiv2 v0.27.6

[nehaljwani]

It is time to cut a release 🙂

xref: #2406

[codecov]

Codecov Report

Merging #2463 (d2ed6f4) into 0.27-maintenance (eae83b4) will increase coverage by 1.02%.
The diff coverage is 100.00%.

@@                 Coverage Diff                  @@
##           0.27-maintenance    #2463      +/-   ##
====================================================
+ Coverage             58.89%   59.92%   +1.02%     
====================================================
  Files                   149      149              
  Lines                 23203    23921     +718     
  Branches              12711    12664      -47     
====================================================
+ Hits                  13666    14334     +668     
+ Misses                 6700     6692       -8     
- Partials               2837     2895      +58     

Impacted Files	Coverage Δ
src/preview.cpp	51.30% <100.00%> (+2.32%)	⬆️
include/exiv2/pgfimage.hpp	0.00% <0.00%> (-100.00%)	⬇️
unitTests/test_helper_functions.cpp	66.66% <0.00%> (-13.34%)	⬇️
unitTests/test_futils.cpp	63.85% <0.00%> (-5.59%)	⬇️
include/exiv2/slice.hpp	87.28% <0.00%> (-4.24%)	⬇️
samples/Jzon.h	32.55% <0.00%> (-2.33%)	⬇️
samples/iptcprint.cpp	85.71% <0.00%> (-1.79%)	⬇️
unitTests/test_FileIo.cpp	64.28% <0.00%> (-1.43%)	⬇️
include/exiv2/error.hpp	54.54% <0.00%> (-1.02%)	⬇️
unitTests/test_jp2image.cpp	82.53% <0.00%> (-0.48%)	⬇️
... and 90 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[nehaljwani]

The failures in the macOS release build in CI are consistent and I haven't been able to reproduce them locally.

Combo macOS | x64 | Release:

build/bin/exiv2 -f -pp test/data/exiv2-pre-in-xmp.xmp
Preview 1: image/jpeg, 150x91 pixels, 4602 bytes

Combo macOS | x64 | Debug:

build/bin/exiv2 -f -pp test/data/exiv2-pre-in-xmp.xmp
Preview 1: image/jpeg, 150x91 pixels, 4745 bytes

Anyone know what's going on here? cc @postscript-dev, @kmilos, @piponazo, @neheb, @kevinbackhouse

[piponazo]

First time that I see this Mac CI errors. I could try to reproduce locally too, now that I have a Mac at home. I will let you know if I find the time to do it.

[nehaljwani]

I was able to reproduce the problem with Apple Clang 14.0.0
More details at: https://stackoverflow.com/q/75153729/1005215

[neheb]

I remember converting this to std::array. vector seems better here because of the constructor.

[piponazo]

LGTM! ❤️

[kmilos]

Finally, I know this is a worn out subject, but should we perhaps make ENABLE_BMFF=ON the default for this release and on this branch as well?

[nehaljwani]

For a patch release, I'd rather not. But I'm open to cutting a minor release with that flip.

[kevinbackhouse]

@nehaljwani: nice work tracking down that macOS bug! Amazing that it's actually caused by a compiler bug!

[postscript-dev]

@nehaljwani
Thanks for your work on the new release.

One minor thing, would it be possible to change my contributor name from Peter to Peter S. This is what Robin used in previous releases. I have updated my GitHub profile to help with any future automation.