Yep, better with nice menu an “import wizard” and so on. However, at least “we” have an ability to use and modify (add records) to the PL database - what i think is nice. May you already see (previously in this thread), i create a SQL script for simple Keyword import and Keyword + Photo relation. May running SQL script is not so difficult in the real life - of course its easy to say as someone like me work with some SQL in the past. But i still think its not so difficult to run (use) it, no worries.
That said, I have a 40,000 line item structured keyword list that I have built and use in LRC that I wish to move across to PL9 (9.7.0) could I use the database tool you indicated in you post to help do this.
SQL script can do it in the matter of seconds (if its well written, etc). Its can also show issues/conflicts (even before importing)
My personal notes about “importing” and PL stuffs on that area:
I think PL not so far for more better “Keyword search” lists, groups (metadata filtering) - “inside” everything is ready → may nice UX/Gui needed. May its less a priority for them to improve this area.
Folder / Photo (and not Keyword/Ratings, etc.) import (to PL database) via SQL script anyhow need PL re-index → So, i think PL folder and photo indexing is necessary anyhow.
Keyword import seems easy (to PL database) via SQL script.
Keyword + Photo relation import seems easy (to PL database) via SQL script, but it has some logical issues → see a bit later about that.
PL Project import (like from Lrc) seems more-or-less easy (to PL database) via SQL script.
Export any kind of metadata from PL database seems easy via SQL script → see a bit later about that.
I think running SQL scripts is more easy in the real world than usually expected.
But see some issues:
Imported data structure can arrive in various forms (file, like: csv, xml, etc.) → import need to be flexible / configurable. What’s not so trivial to do. My keyword import example works fine with 2 column CSV file → But this things can arrive in various files, various data field names, various logic, etc → So translation logic needed → what need an ability for user configurable.
Need to handle uniqueness / duplicity cases → and not just handle, but list/show it, example: 40000 keyword in the source (import) data, but this 4532 is duplicated, this 2654 is orphaned, etc. → what you like to do with them? Some various issues can also come up, example: import data folder not exist in PL database, etc.
And i wonder: even the import of metadata (keyword, stars/color ratings) question come up sometimes, seems no one really interesting on SQL scripts and similar. A few day back i post about: " Lr → PL metadata conversion (rating, keywords) script - Someone interested?" not a reply at all - zero, nuku, nada, null. So, seems quite often the “need” to import come up, but in the reality no one interested for SQL solution. I wonder, if i write some plain GUI for import (what practically does noting, just run SQL scripts step-by-step) and say: “Hello everyone, here one the app what does the magic!” → Is anyone really interested in the end of the day? May the same not. And if even someone interested on that, i’m quite sure the source (import) data (file) and field mappings issue come up instantly, example: i write sample import script, what use 2 column CSV file - 1st column for filename, 2nd column for keyword. But high chance the following come up: “But my source file is XML, and i have no idea what field is what”.
So, seems not so trivial the import of metadata - at least my standpoint of view.
Of course xmp can fine (as others describe previously) as its more-or-less universal interchange file/data structure (with PL index/re-indexing all folders)