https://bugs.kde.org/show_bug.cgi?id=386959
mau <b-m...@gmx.ch> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |b-m...@gmx.ch
--- Comment #15 from mau <b-m...@gmx.ch> ---
Hmm, for me it looks like we have two different problems here:
First one is the mapping:
The Metadata Working Group (MWG) [1] recommends
for Tag Name "CreateDate"
Composite:SubSecCreateDate
EXIF:CreateDate
IPTC:DigitalCreationDate
IPTC:DigitalCreationTime
XMP-xmp:CreateDate
CurrentIPTCDigest
IPTCDigest
which is the "creation date of the digital representation" (MWG),
and for Tag Name "DateTimeOriginal"
Composite:SubSecDateTimeOriginal
EXIF:DateTimeOriginal
IPTC:DateCreated
IPTC:TimeCreated
XMP-photoshop:DateCreated
CurrentIPTCDigest
IPTCDigest
which is the "creation date of the intellectual content being shown" (MWG).
So apparently a photoshop-named tag should be used for DateTime information -
at least as a fallback if no other field contains the DateTime information.
Second problem is: Resolution (time information included or not).
Since there seem to be cases when Xmp.photoshop.DateCreated does only contain
the date and not the time (and that could at least theoretically be the case
for all DateTime fields), I'd propose to prefer fields wich contain both date
and time over fields which contain only the date part. So I'd use two priority
rankings:
First go through all DateTime fields in the defined order, if a valid value for
Date and Time is found -> use that and exit.
If no entry is found, go through all DateTime fields again, in the same order,
and accept the first date value (withouth time) -> use that and exit.
Another way of handling all the different fields would be to group them (just
as MWG does) and check for and report any inconsistencies found (like different
dates / times or different resolution (missing time)). That would allow to
correct them, but the resulting question would be which cases of inconsistent
information could be corrected silently and which cases would require human /
user intervention.
[1] https://sno.phy.queensu.ca/~phil/exiftool/TagNames/MWG.html
--
You are receiving this mail because:
You are watching all bug changes.
