Order or reading meta data.

[krischik]

I am preparing subscription book for import into calibre and I use ebook-convert and ebook-meta to do so. Once the books are perfect I start the import.

However the result was not perfect after all. So I wonder: what takes priority when importing meta data: File-name extraction or embedded meta data. I begin to thing it might be file-name in which case I would which for a "do not use file-names for meta data" option.

Martin

[krischik]

I forgot: if you import multiple formats: which format is used to read the meta data?

[kovidgoyal]

file name extraction is not used unless set the preference option for it. metadata read from multiple formats will be a union of metadata from all formats. Which ones take priority are hard coded, look in the ebooks.metadata.meta module for details.

[krischik]

Originally Posted by kovidgoyal

file name extraction is not used unless set the preference option for it. metadata read from multiple formats will be a union of metadata from all formats. Which ones take priority are hard coded, look in the ebooks.metadata.meta module for details.

_METADATA_PRIORITIES = [
                       'html', 'htm', 'xhtml', 'xhtm',
                       'rtf', 'fb2', 'pdf', 'prc', 'odt',
                       'epub', 'lit', 'lrx', 'lrf', 'mobi',
                       'rb', 'imp', 'azw'
                      ]

If I understand the code right with my limited Python skills then mobi takes priority over epub. (Order is low priority -> high priority).

An opf file takes priority over everything - That is: if an opf file is successfully read then other files are skipped.

And file-names are used as fallback if reading meta-data fails for whatever reason.

That would explain my problems but also offers a solution: I just need to create an opf file from the epub .

Martin