Remove folder from Database

[mopperle]

I have a folder added to the database and assigned keywords and categories to the pictures in this folder. Unfortunately I ran a metadata template which overwrites exiting keywords. So I removed the folder from the database, expecting that also the keywords/categories asigned are also removed.
But when I add the folder again to the database, former keywords and categories are automatically assigned again.
How to avoid this.

[Mario]

This sounds rather unplausible.
Have you written back the changed metadata before removing the folder?
If not, where could the keywords come from...?

When IMatch removes a file from the database, all attributes and associated metadata is deleted and the file is also removed from all categories to which it is assigned.
Categories exist only in the database, with the exception of data-driven categories which are dynamically calculated based on the metadata and @Keywords categories of course.

[thrinn]

If you remove a folder from the database, category assignments are indeed removed from the database, as far as I know. I am talking about "regular" IMatch categories here, not the special @Keyword categories. For keywords, the situation is different, because keywords are stored in the files themselfes (or corresponding XMP files). Now, it depends if you performed a write-back after you run you metadata template. If you did, the keywords are now part of the files (so to speak). Adding the folder again to the database will read / import the keywords from the files, so whatever is there will be in you database. If you did not perform a write-back, the keywords are whatever was in your files before.

[mopperle]

Yes Mario, you are right. Nonsesne what I wrote. Of course have I written back the data, so no wonder that they are there now bach again. So the problem is now a bit more complicate:
During my scan process keyword are written to IPTC-IIM in the tiff file (picture 2)
after Processing in Lightroom, it ends up in XMP (picture 1)
So how to get those kewords into iMatch

[Mario]

IMatch automatically imports keywords from legacy IPTC and XMP (both the flat keywords in dc:Subject and the hierarchical keywords in lr:hierarchcialSubject).

Make sure you write metadata back in Lightroom (menu command). By default, Lr keeps metadata only in it's catalog, unreachable for other applications.
Please see the Lightroom documentation to learn more about how Lightroom works with metadata and also the Using Adobe Lightroom® and IMatch together article in the IMatch knowledge-base.

[mopperle]

Well, when you look at the first picture, LR has written the 3 keywords in the section "lr", but i only get the first keyword. Is there a way to make imatch read the metadata again? Rescan did not solve it.

[Mario]

There is only one keyword: Schweiz.
Not sure what metadata weightedFlatSubject is, but it's not a standard keyword. I could not find anything useful about this proprietary Lightroom tag.
All I could find out is that this seems to be some newly introduced tag where  Lightroom stores weighted (ordered keywords) of sorts? In addition to normal keywords it stores?

If you consider these keywords, please input them as keywords, and IMatch will pick them up.
IMatch does not use proprietary tags written by cameras or applications as input for keyword mapping.

Which version of Lightroom did you use and what did you do so only one of your keywords ended up in standard keywords, and two others in the proprietary tag?

[mopperle]

I'm using LRC 11, but that doesnt matter. I think I know what the problem caused:
When LrC writes the jpeg, everything is OK (screenshot 1), but after adding it to iMatch, I used a template writing a certain part of the Filename to the keywords and this obviously overwrites the existing data and you get what you see in screenshot 2.
This can be avoided by setting the checkbox "merge with existing comtent".

[Mario]

This can be avoided by setting the checkbox "merge with existing comtent".

Yes. This is what this option is for. To add more data to existing data instead of replacing. See Repeatable Tags

[sinus]

mopperle,

OT: Just curious: are these scans from slides? Because I see Silverfast, hence it could be.

I ask, because in the next time I will also scan some slides (well, in fact about 3'000) and I do check the possibility not to scan them with a scanner, but "scan" (take photos) with a 1:1 Micro Nikkor.

[mopperle]

Yes, scanned all my slides (~11.000) with a Nikon LS4000ED using Silverfast Studio 9; slides where taken with Kodachrome, Ektachrome, Agfa, Fuji....
next project will be scanning Negatives.

[sinus]

Yes, scanned all my slides (~11.000) with a Nikon LS4000ED using Silverfast Studio 9; slides where taken with Kodachrome, Ektachrome, Agfa, Fuji....
next project will be scanning Negatives.

Thanks, mopperle
Boah, you already scanned quite a lot of slides, very good, congratulations.

For me the simplicity and speed is very important, hence I will evaluate the best (fastest) way.
Because, well, finally I have about 200'000 color-slides and over 2 millions of negative (black and white).

But of course I will not scan (or let scan) all of them, to scan e.g. all variations does not make sense.

[mopperle]

The 11.000 were left out of around 50.000. If you need more details contact me via PM, just started scanning the Negatives.

[sinus]

Thanks, mopperle!