Your Format string should look like this: %artist%;%title%;%peak_level%
Please note that a semicolon inside your artist name or title cause difficulties for the import.
The separator should be as unique as possbile and should not exist in the data itself.
Thanks for the quick answer, but it doesn't work. I tried with this syntax and went through the Convert/Clipboard -> Tag menu. But if I try to insert the values in a custom field, it doesn't work. On the other hand, it's ok if I use a stock field, for example:
%artist%,%title%,%comment%
I don't know why it doesn't work with a custom field.
Is this menu point available on Mac?
I don't find it on my Windows Mp3tag v3.28c.
If I import the CSV line Name of artist;Title of song;-0.14 dB
with the menu Convert -> Text file - Tag
and the Format string %artist%;%title%;%peak_level%
Please show us some content of your real CSV file.
In the initial question you write about fields separated by a semicolon.
In your second answer you seem to use a comma a separator in the Format string.
The separator in the Format string has to be exactly the same as the one your CSV file.
Okay, I think I've got it. It looks like it's the underscores that are the problem in Puddletag. Why? I don't know. I renamed the field to peaklevel and it works (with import function or clipboard).
I hope that you can find some function in Puddletag that can copy your content from PEAKLEVEL to PEAK_LEVEL and then deletes the no more needed field name without underscore.
It's not as easy as I thought. I'm French. I have many pieces of music that contain “special” characters.
Example: Étienne Daho
In my spreadsheet (LO Calc), when I sort the file name alphabetically, É comes after E. But in the Puddletag sorting, it comes after Z. And it's the same for accented lower-case letters within words (Cé comes after Cu instead of Ce).
So I have to be very careful and classify my text file in the same way as Puddletag. I had naively thought that the function system would automatically look for the line corresponding to the file (or to the artist/title) to enter the values.
Please understand that this is the Mp3tag Community. If you're looking for support regarding puddletag, please refer to their respective support channels.
The alphabetical arrangement of files under Linux is correct. It's the same as under Windows; it's in the Puddletag (and MP3Tags) interface that something goes wrong (and there are no setting options in either of these programs).
In any case, we can consider the issue resolved. But rather than inserting values into fields with an exact correspondence of lines, wouldn't it be possible, by specifying the file name, to have the values inserted according to that name, thus, without having to worry about the order?
You can define the property "Sort by" for each column in which you e.g. replace accented characters with the plain ASCII one.
Also it may be worthwhile to check the options Ctrl-O > General and the settings for sorting.