All good suggestions John. Problem is, I've already done all that including the earthboundlight web site. He got about as far as I did on reverse engineering the structure.
The Lightroom Catalog is just a SQLite database which explains the upper limit on the catalog size. So it is browsable by any SQL browser - which I have several including the one I have written.
Since I am sure the Adobe Engineers are highly sophisticated and capable when it comes to database design, I am sure it is my limitations at work here. Because SQLite does have a size limit and because catalog integrity is paramount, I can only believe that some kind of very high level normalization is at work here. There is no one key that you can use to access all the information in all the tables related to any given image. You take the key from one table and access the next table with another field from the first table. I am now in the process of figuring out the pattern of keys. Let's just say "It ain't easy"
Of course to make matters worse, I have added custom metadata fields to LR to identify the records to the application I am working on outside of Lightroom.
I first envisioned that I would write a plug-in to read the metadata in an image record and write it out in some form that I can import into my app. Problem is, the SDK has very limited functions that can be programmed (Metadata, Export, Web Gallery) and the Lua language that you have to use for plug-ins can't write to external files.
I know I am way out on an edge here but if anyone can offer any suggestions I would greatly appreciate it....
William (Bill) Plunkett
NANPA Board Member
Last edited by plunkettphoto; 06-03-2010 at 06:13 PM.