Closed kefo closed 5 years ago
The most recent version of EXIF is 2.31, which would mean that NLNZ's value in the above snippet is incorrect. It's probably not taking into account endianness, but, again, I digress.
Assuming that to be the case, is the best solution to rely on ExifTool's value?
Something like:
t.exif_version(path: 'exifVersion', attributes: { toolname: "Exiftool" })
See:
Resolved with https://github.com/samvera/hydra-works/pull/338
We have more than 200,000 pcdm:Files in our collection with multiple and unequal
exif:exifVersion
values.There are, of course, multiple values in the
fits
output:(I actually see 3 exifVersion values in our data sometimes and, for the life of me, I cannot figure out whence the third comes, but, again, I digress.)
Given the CONLICTing values above, I'm not sure what to do about this other than raise it as an issue.
In so far as this results in duplicate, conflicting values in the characterization metadata, it is related to https://github.com/samvera/hydra-works/issues/336.