the relevant PRs to ImageMagick and FileIO are now merged. this PR should work (i.e. pass tests) after they are both tagged.

Thanks! Can you (or someone) post here when the tags are in metadata? I will then retrigger CI and merge

Oh, speaking of tags, the minimum version of FileIO would then need a bumb in REQUIRE. But I could also take care of that little detail after this PR is merged

please retrigger CI as FileIO and ImageMagick have been tagged:

JuliaLang/METADATA.jl#15632 (comment)
JuliaLang/METADATA.jl#15257

thanks!

Codecov Report

Merging #714 (f3e49ec) into master (a80031e) will not change coverage.
The diff coverage is n/a.

❗ Current head f3e49ec differs from pull request most recent head d170ee3. Consider uploading reports for the commit d170ee3 to get more accurate results

@@           Coverage Diff           @@
##           master     #714   +/-   ##
=======================================
  Coverage   85.96%   85.96%           
=======================================
  Files           9        9           
  Lines        1026     1026           
=======================================
  Hits          882      882           
  Misses        144      144           

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 a80031e...d170ee3. Read the comment docs.

running CI again

tests pass on julia 0.6 for windows and linux. quartzimageio.jl just needs metadata support for this PR to work on os x...

Is there any reason this has to be in Images? I have a few issues:

1. metadata is too useful of a name to grab of this. I use it all the time for more elaborate information about experimental conditions. Something like imgsize and imgtype may be better? I wold also suggest not exporting these names.

2. I don't think that QuartzImageIO can be refactored to separately grab image info, then the image data, without repeating a lot of work. It would be like peeling an onion to get to that layer you want (with the metadata), reassembling it, then peeling it again to a deeper layer to get the data. I don't look forward to that 😄

I think I'm with @rsrock on this one. As a non-exported method of ImageMagick it's fine, but as a general method it would need a more specific name.

i've removed the export of metadata and modified the tests accordingly. can think about an alternate name if desired at another time.

Oops... I made a mistake by accidentally dropped the commits when resolving the conflicts locally. @bjarthur do you happen to have a local copy of it?

Edit:

Found it back using git reflog and git reset --hard command (from https://stackoverflow.com/questions/134882/undoing-a-git-rebase)