Morning Christian,
I did a little digging, and here are my conclusions:
tag() was introduced primarily to deal with situations where there is differing data in different tag blocks and it allows us to override MC's default order of importing data from the various tag blocks contained within a .jpg file; different dates being the classic example. When the function was first introduced, I had trouble getting it to work and Matt offered the following explanation:
I cannot get this to work with anything other than MJMD tags. I've tried as many combinations as I can think of, including capitalisations and spaces:
tag(exif: date)
tag([exif: date])
tag(exif: [date])
tag(exif: date taken)
tag(exif: date/time)
If I try them in the [date] field of the tag window, "Unknown Date" is returned and the file's date is changed to 0000/01/01 (used with a leading = sign)
If I try them in a standard field, the field is left unchanged.
If I try them as thumbnail text, (3rd line), the entire thumbnail text is shown as "Expression Error (Unknown)"
-marko
Capitalization is ignored for the prefix and function, but you need correct capitalization for the field, so use "Date". Here are complete examples that will work:
=Tag(Exif: Date)
=tag(exif: Date)
=tAG(EXIF: Date)
Note that "Date" must be correctly cased. This is because you're talking directly to an input / tagging plugin. It compares what you pass it against a defined list of fields. There's never been a need to consider case here since previously only the program itself called into these, where the case would always be consistent.
I didn't pick up on it at the time, but I think the key here is the "defined list of fields" part that I've bolded above.
The following quote details the current iptc > MC mappings:
As was mentioned elsewhere, this is very early in development. All these comments are very welcome and helpful as we develop this further. In addition to reading IPTC, we are looking into reading other TAG formats--is everyone using IPTC these days, or do people use competing or superseeding standards? We are also planning to implement writing out the tags.
The way the IPTC tags are read currently, as I understand it, is any matched field that is not aready being used is read from IPTC. This means once read, the IPTC tags won't be read again from that photo, unless the tags are cleared in MC. The IPTC and existing tags are not merged. Any of this, of course is subject to change.
Here are the current IPTC tags being read, and MC tags the are mapped to (also subject to change):
BYLINE -> Artist
CREDITS -> Notes
DATECREATED ->[changed] Date
CITY STATE, COUNTRY ->(combined) Places
COPYRIGHTNOTICE ->Copyright
CAPTION ->(same: no longer Comment)
CATEGORY->Genre
KEYWORDS->(same)
Suggestions to mapping as well as all other suggestions, are welcome.
JimN
And in my quick tests, the only data I could get MC to import using tag() was the stuff that is in that field mapping list. Nothing else would import.
Someone from JRiver will need to confirm whether or not my theory is sound, but as far as I can see, you are using the correct syntax, and I am getting the same results as you are.
Sorry I can't be more helpful than that at this time,
-marko.