[trimbolim]

I've been working on importing large fakebook files with the publicly available CSV indexes. I'm finding that some of the indexes aren't as good as others and I ended up with a few hundred "bad" songs in my list (image attached). The bad song entries point to the wrong part of the pdf file, usually the first page. My thought was to simply remove the bad batch and re-import. The issue I'm having is that I don't see a way to identify just the "bad" songs. I don't have access to any metadata that I could use to identify the bad batch (i.e. import date, batch id, etc.) If I could do that, I could bulk delete a result set.

I know that I could delete my whole library and rebuild it, but I'm hoping there's a better way. I already did that, forgetting that I would lose my annotations. Lesson learned. I wrote a tiny bit of python to fix formatting and integrity in any CSVs that I'm trying to import and that looks like a good way forward, but I'm hoping to learn how to elegantly recover from a bad import should it happen again.

I'm using MSP on Windows.

Thanks,

Matt Trimboli

[Zubersoft]

The only things that come to mind are sorting by date created with the newest songs at the top so you can delete recently created songs, or you can go to Settings->Backup and Restore->Restore Automatic Database Backup and just revert the entire library to the point before you imported those bad songs. You could also go to Settings->Library Settings->Song Title Formatting, include page order as one of the fields, and then delete every song that points to the start of the PDF.

Mike

[Skip]

The only 'sure' way I've found is to do a backup before importing.

[Geoff Bacon]

When importing using CSV, I always include a collection name in the .csv file.
This allows one to filter the results by collection so that you can delete and reimport the songs.

Not much help now but might be useful for future imports

Geoff
PS the keyword in the .csv file is "collections"  (and not "collection")

[trimbolim]

The only things that come to mind are sorting by date created with the newest songs at the top so you can delete recently created songs, or you can go to Settings->Backup and Restore->Restore Automatic Database Backup and just revert the entire library to the point before you imported those bad songs. You could also go to Settings->Library Settings->Song Title Formatting, include page order as one of the fields, and then delete every song that points to the start of the PDF.

Mike

 That was exactly what I needed. When I sort by date added it exposed the timestamp which made it trivially easy to delete the bad batch. Thank you! - Matt

[itsme]

I had the same issue, see https://zubersoft.com/mobilesheets/forum...p?tid=8740
Scroll doen to see how I solved it

[lvr123]

Has there some improvement made here ?
Re-importing a CSV file should not duplicate the songs but only update them.
There is a new option in the Import dialog that says "Modify the songs after import" (*) which is NOT documented in the manual. I wish it was that option to update the songs instead of duplicating them. But it is not.

(*) freely translated from my french UI

[Zubersoft]

I haven't had time to work on this yet - it's on my list of things to do in one of the next big updates though.

Mike