[musichawk]

MSP is set to Let MobileSheets Manage My Files, which are stored here, same location since MSP installed some time ago:
/storage/extSdCard/Android/data/com.zubersoft.mobilesheetspro/files

I use flat storage, no subdirectories for songs.

The problem, possibly existing for months or longer, is that some songs I add to tablet library via MSPC do not actually get added. I've had "I thought I had that song" moments but figured I was forgetting. I can add individual songs, no problem. But now, I think there's a Batch Import problem, because that is how I've added the vast majority of songs. I have a mix of missing songs and missing files, and can't seem to correct this via Batch Import or any other MSP action.

Using the latest updates of MSP and MSPC, with Batch Import always set to Avoid Duplicate Songs, the problem is that the two forms of mismatch do not get corrected. When a duplicate file is found, selecting Use Original File, or Overwrite, or Skip File doesn't seem to matter, at least not in my test records/files. (If I'm incorrect about this, please advise what tests to run.)

I find instances of two types of database-file mismatches, and they persist, can't find a way to have Batch Import "fix" them:

Source of files is always the same, a single folder on PC that has all the PDFs I want on the tablet. Every file is precisely named, and MSPC is set to use the Filename as the Song name. For Batch Import I simply select this folder, and MSPC seems to process all the files in the folder.

When...
-Song already in database
-File not on tablet
Result=The missing file does not get added to the bad record.
Instead, the missing file gets added as a new record, so the song name exists twice in the database. In one test, the prior file-less database record was then found to link to the newly-added PDF file. Presumably this was because the db record had originally pointed at the file, which disappeared somehow but the pointer was still there.
Note that the song name that gets duplicated is identical to what was already in the library, because it is the Filename. There seems to be no checking for an existing db record of the same name.
The problem is that duplicate records appear and then must be manually spotted and deleted. The risk is if the two (or more) records do not all point at the file, probably they do, but the only verification is to open each song.

When...
-Song not in database
-File already on tablet
Result=Nothing happens with Batch Import, the song does not get added to the database, unless....
The orphan file is identified during import via a Duplicate filename dialog. If I select Overwrite, the song gets added to the database and presumably the file gets replaced. However, with more than 4000 files the only practical method is selecting Duplicate file Overwrite to do this with all files by checking the box, which results in almost the entire library being re-copied from PC to tablet, just to get any "lost" songs added back to the database. This would take a while, but is it safe? Any other side effects?
The orphan PDF file does not show up via Find Missing Files, not sure what it actually does, though in a previous test it found orphan mp3 files.

I'm trying another Batch Import, specifying Overwrite for all files. Hoping this does not result in the topic that started this post, double records for almost every song, yet some songs skipped. Here goes...

NOPE, FAILED. Importing with Overwrite of all existing files, the library on tablet grew by ONE file. The Library still is missing more than 500 files in the PC's source folder total of 4658 files. The tablet folder that stores MSP songs has about 4940 files, apparently are on the tablet but not in the database therefore unidentified by MSP.

This is a mystery. I have never deleted a song without also deleting its file (except for a few recent tests), so how did these files get on the tablet and/or their database records get deleted? Does MSP copy new files from PC to tablet, THEN update the database, which perhaps failed for some reason? It will be difficult to delete them because they look just like every other MSP library file. This is where the solution could be for MSP having a mode where it scans all files for matchng database records.

The Proper outcome would be for ALL files from the source get loaded onto the tablet, no more or less. I see no cause for these large, persistent anomalies. Or am I thinking about this wrong?