Initial support for BPLIST in Apple Makernotes

[nosnhojbob]

This change introduces BPLIST (binary property list) support in Apple makernotes. The Apple makernote tag that comes with the most format documentation is 0x0003, RUN_TIME. I provided a new directory and descriptor for this field, supporting the raw properties and human-readable descriptions for the CMTime fields represented by this tag. I left some @see Javadoc pointers to the data structures and descriptions of the CMTime fields. Essentially, the CMTime structure provides a value and scale, where (value / scale) = seconds.

Two of the test "data" objects were taken from the existing image library samples of Apple/IOS images (read as Metadata, extracted as byte array) for the 0x0003 (RUN_TIME) and 0x0002 (unknown) fields. On a Mac, you can right-click on the files and it conveniently provides a "Get Info" preview function, which shows the BPLIST as an XML document. I used this output in my test assertions.

Limitations:

• 0x0003 (RUN_TIME) was relatively easy to implement since the data lengths and pointer sizes were all 1. More complex objects serialized as BPLIST would require a "fuller" parsing implementation.
• 0x0002 (unknown); I spent some time researching, but only found references to this tag's BPLIST contents being a byte array. I did not implement any iterable tag sub-directory - will wait for reviewer feedback/comments before proceeding on that part.
• XML serialization of the BPLIST is included in the BinaryPropertyListUtil class, but proper XML escaping/encoding is not included. For the CMTime structure, string concatenation works.

[nosnhojbob]

A quick note to the reviewer(s). I included a definition of 0x0015 (TAG_IMAGE_UNIQUE_ID), per the ExifTool's referenced page. However, after running against all of my personal iPhone photos, I could not find any that contained 0x0015. Based on the data in the fields, I suspect that 0x0016 is actually the "image unique identifier". I did some more looking on the ExifTool forums and found:

http://u88.n24.queensu.ca/exiftool/forum/index.php/topic,8750.msg44954.html#msg44954

I can't find why I called 0x0015 ImageUniqueID. In fact, right now I can't find that tag in any of my samples at all.

Shall I change the tag value to 0x0016 here and make a note of the discrepancy with http://owl.phy.queensu.ca/~phil/exiftool/TagNames/Apple.html?

[drewnoakes]

@BobJohnsonGitHub thank you for this. I am currently in the process of moving country and don't have a lot of bandwidth. I encourage others to provide feedback here, and I'll provide a review as time permits.

[scrappydog]

@BobJohnsonGitHub iPhone 8 Plus potential Unique ID's

Apple Makernote - Unknown tag (0x0020) = 57E29167-7C4B-44BF-BC68-4280C6951F7F Apple Makernote - Unknown tag (0x002b) = 054B8D8A-9250-4F2E-986C-B91A91F0A8C0

Do these tags vary based on Camera model?

[scrappydog]

@BobJohnsonGitHub

iPhone 6 - This smells like UUID (no guids found in meta data)

Apple Makernote - Unknown tag (0x0018) = AQCrunJxXx4X6kaWWCP6eL9bKpi1

[drewnoakes]

@scrappydog thank you for drawing my attention to this PR again.

@BobJohnsonGitHub apologies it took over a year to get to this. I've been doing a bit of work on the library again during the quarantine. Your changes look good and I've merged them with a few minor tweaks. Thanks very much.

[drewnoakes]

@scrappydog if you're able to extract more info from the Apple makernote, a PR would be great.

[nosnhojbob]

Thanks for the merge, @drewnoakes

[nosnhojbob]

@drewnoakes - given @scrappydog comments on the ImageUniqueId, should we open a separate issue to track? From my initial research, it looked like 0x0015 was just invalid (per the ExifTool forum) and 0x0016 was the right tag. But it appears that there might be more candidate tags within Apple metadata. (see the earlier comments on this PR)

[drewnoakes]

If new tags are identified we can add support for them. Creating a new issue or pr is the best way. To answer an earlier question, yes these tags often vary by camera model.