View Single Post
Old 09-25-2008, 04:04 AM   #5
pepak
Guru
pepak has a spectacular aura aboutpepak has a spectacular aura aboutpepak has a spectacular aura aboutpepak has a spectacular aura aboutpepak has a spectacular aura aboutpepak has a spectacular aura aboutpepak has a spectacular aura aboutpepak has a spectacular aura aboutpepak has a spectacular aura aboutpepak has a spectacular aura aboutpepak has a spectacular aura about
 
Posts: 610
Karma: 4150
Join Date: Mar 2008
Device: Sony Reader PRS-T3, Kobo Libra H2O
Quote:
Originally Posted by kovidgoyal View Post
At some point I have to spend some time debugging this and figuring out exactly what the reader is getting its metadata from (the ebooks file itself or the media.xml database).
I can save you some time with this.

As far as I can tell, the Reader reads metadata only from media.xml. Even if the data is clearly incorrect or even not present in media.xml (e.g. you manually remove it), the Reader will not reread the metadata from the book itself.

But (you knew there would be a but, right?) there is a catch: That is only true as far as the XML record for the book is valid - that is, at the very least file path, file date (either creation date or modification date, depending on whether you use internal memory or a memory card) and file size match in XML and in the filesystem (and it is possible, but not certain, that the ID in XML must be sequential). If any of these values does not match, the old XML record is deleted and new one is created; at this point, metadata is reread from the book for the purpose of XML generation. But the actual display of the metadata is, once again, done from the XML itself - if you were to modify the XML file now, the Reader would display the modified metadata.
pepak is offline   Reply With Quote