My original mp3 file changes to unix executable. Why?

I have an imac running Catalina 10.15.7 I have the newest updated version of MBPicard. I am a newbie. Just installed this.
My Problem. I have a folder of random mp3’s on my desktop. I play them using Foobar 2000 for Mac. I NEVER USE ITUNES!
I take all the mp3 files in the desktop folder and run them thru the Cluster/Lookup/Scan aspects of MBP. It finds new tagging info on 95% of them. GREAT! I created another folder on my desktop called RENAMED FILES. I tell MBP to put all the retagged files there. I hit SAVE. Instantly my original mp3 disappears. I then go to the RENAMED FILE and here is a typical folder tagging that my old mp3 has been changed to:

Supertramp: >2014:>Crime Of The Century:>Supertramp - School:>

so all that is great… but the final file folder should contain my newly retagged mp3… but it doesn’t. It contains a little square icon of a black box with three tiny green dots in it. Next to that it says: _mp3 and the Kind (of file) has changed from .mp3 to now being listed as Unix Executable …which is not playable. My original mp3 is just gone!

Where is my newly retagged mp3?
Why did it change my file to something unplayable?
What am I doing wrong?

Any help you guys can give would be greatly appreciated!!

Thanks, Michael

Sounds like you can rename the _mp3 files and pop a .mp3 on the end to fix this. Something in your file naming script forgot to put .mp3 back on the end of the filename.

You know these are the real files as they are still the same size.

Lets see an actual filename. We’ll then help workout how to quickly rename them.

I tried to change it back back to .mp3 from _mp3 (a trick I’ve used before…) but it says:

You can’t use a name that begins with a dot because these names are reserved for the system.

Please post the file naming script that you are using in Picard. I suspect that you have an extra slash at the end that’s causing the problem.

1 Like

Please show an example of your filenames.

Don’t start the track name with a dot, it goes on the end.

MyTune.mp3 instead of MyTune_mp3

I am guessing you have lots of files with an “_” instead of a “.”

The warning about “reserved for system” is because you tried to put the dot at the front of the filename.

Sounds like you have a file naming script that generates an empty file name. As Picard ads the file extension that would lead to a file .mp3, and the leading dot than gets converted to an underscore to create a compatible file name.

Can you share the file naming script you are using?


Thanks guys for the help!

Here’s the file names… but they are exactly what I posted above:
Crime of the Century
Supertramp - School
and inside the last one is what i_mp3described above:
an icon and next to it:

As for the script: Under File naming script:
$if2(%artist%/%date%/%album%/%artist% - %title%)/

and Under Scripting:


1 Like

As @outsidecontext and I indicated, you have a slash at the end of your file naming script which is causing an empty filename. Try changing your renaming script to:

$if2(%albumartist%,%artist%)/%date%/%album%/%artist% - %title%

For more information about File Naming Scripts including a tutorial see the Picard User Guide.


That new script did the trick! Thanks.

Let me fool with this for a day or two to see if I have any follow-up problems/ questions… but thank you for your help guys.


If you still have the _mp3 files, try renaming them to something like 1.mp3, 2.mp3 and so on, and see if you can play them. If so, the files should contain the tagging information that was applied by Picard so you should be able to cluster and re-save them using the new file naming script.

1 Like

Ok… so I have a tangent problem.

I’ve loaded my giant file in (11,221) mp3’s into the program and run everything through cluster, lookup and finally scan. But none of them are loading album information. Not a one. It’s been hours.

As I said, I have a mac running Catalina with the most current update of MBP.

I’ve run it both with album art check and without.
Still: Album Art Loading

Any Ideas why?


Yeah, don’t load up 11,000 files. How would you check all of them for mistakes?

Look at the status bar at the bottom of the window. It is probably bogged down in getting artwork or similar.

The first time time you sort albums out, only do it in small batches. You will work quicker and can check better.

Also avoid scan if you have tags in your albums. Scan is audio fingerprinting mainly for unknown files. If you already have fairly good tags, just stick to lookup. And do a few albums at a time.

Ok I did 11 individual songs… and It’s still not finding any album information.

Maybe my expectations are to high for the speed of it being able to find the info. Should it take less than 5 minutes for 11 songs?

What is the average for 1 song?

Here’s a little more info: I went through my 40TB collection of music and picked out just the songs I liked for a playlist this came to be about 11,200 individual songs. I changed them all to mp3’s for space sake (The originals are all hi res flac files) and that’s what I’m plugging in to MBP to tag correctly.

Cluster + Lookup shouldn’t take too long per album (seconds), assuming:

  • Your files already have some metadata for lookup to work with (e.g. song name, album name)
  • The release is in the database

Scan is a last resort imo.
Is it particularly obscure music that may be missing from the DB?

Does Picard’s log show anything unusual? That might provide more information to work with. The section on Troubleshooting might provide some insights as well.

Also, as mentioned earlier, using the Lookup function is much preferred over Scan (which should be used pretty much as a last resort). There is more information available in the Retrieving Album Information section of the documentation.

1 Like

Should take a matter of seconds. FLAC or MP3. What state are your tags when you start? What do you have on your files before you do the lookup?

What other plugins have you selected? Artwork changes?

Only when I tick the box to grab loads of art do I see delays caused by fat PNGs or high quality art. Just picking a front image (or no art) then the lookups are fast.

Truthfully …this will take weeks to do if It’s taking roughly 2 minutes per mp3. Like I said…I have 11,000 individual mp3’s that I was just trying to get better tags on… but i think i’ll just live with whatever basic info i have on them which is mostly name of artist and title.

lol… I just did the math and i was wrong… it would take over 2 years!

I Think i’ll just pass unless someone wants to try and talk me out of it. lol.

I ran the cluster and the look up on one batch (The 01’s) in the file. Their are 788 of them. Cluster and look up only found about 300 easily. I would have to go track by track for the rest. Not worth the time for the eventual outcome.

But i appreciate the help in figuring out this will never work for my huge collection… before i’m dead. lol

Well, it’s up to you. If you check logs like rdswift suggested we can probably get to the bottom of why it’s going so slow all of a sudden

Since your files are not in albums at all, and lookup is giving minimal results, go ahead and use scan. To be honest, it sounds like it couldn’t make anything worse, that folder sounds like a nightmare :stuck_out_tongue_winking_eye:

But it should not take 2mins per MP3, so there has to be a different issue on that side. It should take seconds per album.

You are likely correct here. Only fix what is missing. If all your albums are already sorted out, all have artist\album\track details then you are probably in the best state already.

What was your plan to gain from MB data?

MB is mainly for addicts who want to get more details like performers, links into a database. This does take time as MB’s database is sorted with the idea you know exactly which version of a CD you ripped.

Your initial thump of throwing 11,000 tracks at Picard would basically have destroyed your old albums as Picard would have had to guess if each track was on the official album, the 25th anniversary release, or came from a NOW compilation. This is why manual checks are needed.

When time is spent on cataloguing your music it does lead to a mine of richer information, better art and many other goodies. But not everyone needs this. And Picard is such a huge multi-faceted tool it does take a little bit of time to get used to.

To sort out 11,000 tracks would be months, not years. But the more you got used to it, working a few albums a day, the faster you’d be in a short time.

Many of us have much larger collections than 11,000 catalogued in a much shorter time frame. :slight_smile:

Let us better know what you aim to achieve and we can help it happen. :slight_smile:

Thanks everybody for taking the time to reply!
I’ve decided to just live with it. Itunes gave up on me years ago lol.
I have thousands of rare, live and bootleg albums… no tagger knows what to do with it!
I’ll let the next generation figure it out LOL