copy filename to epub metadata in batch mode

[parafux]

Hi all,

i am new here, and i have a serious problem, for what i did not find any solution with the common tools that are available for free. I am also willing to write skripts to do what i want.

here is my problem:

I have lots of ebooks (epub) that are all ordered in folders alphabetically.
The filename contains the author and the title in the following way:

lastname, surname - title.epub

If i read this files into my Tolino the books are not sorted like this. That is because the metadata do not match to the correct authors as they are stored in the filenames. So i want to correct the metadata automatically.

With calibre i have the problem, that it always generates duplicates of the file and the new filenames begin with the title not the author. Also it generates opf files, that i do not want. There must be a way to store the metadata inside the epub files without using opf files. Because otherwise i cannot explain how Tolino or Calibre know the already existing metadata without the opf, which i currently do not have.

With "epub metadata editor" i am not able to edit all the tags. Because if i read the modified files into Calibre, some tags are still missing or are still wrong. Most are working proberly. Epub metadata editor obviously modifies the metadata inside the epub files.

I am looking for a tool like I use for mp3 tag editing: Mp3tag
There i can easyly copy the filename into the several tags inside the mp3 file. And all can be done automatically in batch mode.

Does anyone have help.

cheers and thanks a lot.

[Turtle91]

You can use Calibre to read the metadata from the file name. Then you can use a specific template and "Save to disk". It will then use the format you gave it to save the epub to a folder.
I use the format: author’s last, first - series name series# - title.ePub
There are specific codes that I’ll need to copy from my computer when I get there, but it looks something like this:
“Herbert, Frank - Dune 04 - God Emperor of Dune.epub”

As for Calibre making duplicates, just create a different library for purposes of this metadata exercise.

Edit: if you don’t mind writing a program, and are familiar with python, one of the MR members is developing a module so you can programmatically acces the epubs. That may be more fun than learning the functions already in Calibre.

[Turtle91]

I am so sorry for the delayed update...had a run to the emergency room and been dealing with that and this post completely slipped my mind.

The filename template code can be entered into Calibre's preferences by going to Preferences/Saving books to disk/Save template and adding the codes to your preference. I used the following for the Herbert example I gave above:
{author_sort} - {series} {series_index:0>2s|| - } {title}

----

To read the metadata from the filename (I actually think it might be the default??) go into Prefences/Adding books and clear the checkbox Read metadata from file contents rather than filename then you can put a template on how your filenames are layed out. This will properly read the Herbert filename example I gave above:
(?P<author>[^_-]+) -?\s*(?P<series>[^_0-9-]*)(?P<series_index>[0-9]*)\s*-\s*(?P<title>[^_].+) ?

There is a nice little test function in there to help you build the regex and check matches against your particular filename method.