Database-Converter using the existing MetaData instead of running exiftool

[MyMatch]

I tried to convert some v-4 databases to v-5 ...
Most of them never seemed to finish and i needed to shut down the computer meanwhile.

I was suspecting that it may depend on the fact that i did not convert the offline cache directory ...

Today, i tried again and started the resourcemonitor - and then i noticed that the convert runs exiftool and each and any file in the database!
Hundreds of thousand times!

This data is already contained in the v-4 database!
Why the data cannot just be converted!

It is a Very Bad Idea to re-read such many files, IMHO!
Even more serious, i have many "removed" volumes!
What will happen with those offline files?

I think, the database converted is not very usable in it´s current way to convert the DB.
It should just use the DB as input and not all those files!

Any production database will have 200k, 500k or much images!
This is not handable with the converter ...
:-(

[sinus]

I tried to convert some v-4 databases to v-5 ...

v-4 database?

What is this?

[MyMatch]

I quote from the Database Converter:

"The IMatch 5 Database Converter allows you to convert existing IMatch 3 databases (with the .IMD4 file extension) into the new IMatch 5 database format."

Results in a .IMD5 file ...

So, from v-4 to v-5.

[Mario]

Today, i tried again and started the resourcemonitor - and then i noticed that the convert runs exiftool and each and any file in the database!
Hundreds of thousand times!

Please see the database converter help topic in the IMatch help (click on Help or Press F1) for all the details and background information.


IMatch 3 uses only a subset of XMP and comes not even close to the level and richness of metadata produced by ExifTool for IMatch 5. This is the reason why the database converter imports all metadata using the IMatch 5 routines and options after the conversion.  And IMatch 5 caches all metadata, including IPTC, EXIF, GPS, PDF, QuickTime, FlashPix, Office, ID3, ... so it has to read the metadata once from your files anyway. IMatch 3 did only cache a subset of XMP data.

I think, the database converted is not very usable in it´s current way to convert the DB.
It should just use the DB as input and not all those files!

Several thousand databases have been converted during the Beta test and in the past two weeks since IMatch 5 is officially released. It works quite well for the majority of users.

[MyMatch]

I would be happy with the XMP part from the existing database!

Maybe an option to reduce the XMP to the v-4 data?

What happens with those people who have tons of removed volumes - offline images?

Will they get no XMP at all?
Or just the v-4 parts?
Or are you forced to re-enter all those volumes and images for the conversion?

Maybe this is addressed in the documentation, you mentioned.
Will read it ... :-)

Thank you!

[Mario]

Maybe an option to reduce the XMP to the v-4 data?

Will not work. IMatch 5 / ExifTool use a totally different XMP naming convention and adressing system. And I'm really happy to get rid of the flawed XMP data produced by various incarnations of the Adobe XMP toolkit.

What happens with those people who have tons of removed volumes - offline images?

As explained in the manual, you need to bring the files on-line once and then rescan the metadata. Then you can take them off-line again.

Maybe this is addressed in the documentation, you mentioned.

Yes. In great detail.

Will read it ... :-)

Very good. Actually you are supposed to read the database converter help and to perform the prerequisites explained there before you convert your database.