Authors split wrongly by Kobo

[gogu1904]

Hello,

I have epub books where the authors field is set via filename reading, as Surname, Name or Surname, Name & Surname, Name. All is good inside Calibre. But as I sideload the book in Kobo, Kobo creates two separate autors, namely Surname and Name. How can I teach Kobo that Surname, Name is a single author? Or how can I correct the problem? Thank you.

[gogu1904]

To clarify: the metadata is read from filename as:

Series [index] LN, FN & LN, FN = Title

This helps me have a clean archive of the original files. This is how I have the epub and mobi files when exported on hard drive. I noticed that Kobo separates by comma and sees two autors for each real autor (LN and FN seen as different authors). Therefore I need to convert on the fly the sequence LN, FN & LN, FN from the filename to FN LN & FN LN. And when exporting, as well, have the conversion back. The conversion after input works with Quality Check plugin. The conversion at output works with using author sort instead of authors. So all is good.

But I also use batch file conversions using epub-convert and epub-meta in cmd files (for all files in a folder etc). There all these do not work, unless I can use the quality check plugin in command line. This is where I would like to solve the problem.

Thank you!

[gogu1904]

It appears to be more complicated than that. When Calibre reads LN, FN & LN, FN and understands correctly two authors with those names, Kobo will have to read FN LN & FN LN. But now Kobo sees only the first author. I believe Kobo needs FN LN, FN LN as metadata in order to see all authors correctly.

[JSWolf]

Are you downloading ePub and removing the DRM if there is any? If so, let Calibre read the metadata from the ePub and only then if it's not correct should you correct it. That's what I do.

OH and you don't know the eBook is getting the author incorrect since you are not allowing Calibre to read the metadata. In most cases, the filename is not good to use to populate the metadata in Calibre. And if the metadata is incorrect, it's the fault of the publisher, not Kobo.

[DNSB]

Originally Posted by gogu1904

It appears to be more complicated than that. When Calibre reads LN, FN & LN, FN and understands correctly two authors with those names, Kobo will have to read FN LN & FN LN. But now Kobo sees only the first author. I believe Kobo needs FN LN, FN LN as metadata in order to see all authors correctly.

Correct. Kobo uses the comma as the separator between two or more authors with names in the FN LN format.

[davidfor]

@gogu1904: Yes, that is how the Kobo devices work. They expect the authors to be in the format "FN1 LN1, FN2 LN2". That is not actually wrong, just different than how you want it. Which is also different to what calibre uses.

There is a possible error in that Kobo only reads the first dc:creator tag in the OPF file. But, that is probably a lacking rather than an actual error.

This has been extensively discussed here and in the Kobo forum. The general recommendations are:

• Use "FN LN" for the authors in calibre.
• If you want to display "LN, FN", use the author sort for this
• For archiving the books elsewhere, use the author sort.
• For Kobo devices, use a metadata plugboard to set the author in a way the device will understand. Assuming you have done the first item above, the following template does this:
  

  program: raw_list('authors', ', ')

[chaley]

Originally Posted by davidfor

This has been extensively discussed here and in the Kobo forum. The general recommendations are:

• Use "FN LN" for the authors in calibre.
• If you want to display "LN, FN", use the author sort for this
• For archiving the books elsewhere, use the author sort.
• For Kobo devices, use a metadata plugboard to set the author in a way the device will understand. Assuming you have done the first item above, the following template does this:
  

  program: raw_list('authors', ', ')

You can keep "LN, FN & LN, FN" in calibre and convert it to "FN LN,FN LN" for Kobo using this template in the plugboard.

program:
	re(list_re_group(field('authors'), '&', '.', '(.*)', "program: swap_around_comma($)"), '&', ', ')