|
eReader PDB's use PML internally. You will need a tool like what ficbot described to extract the actual contents. The "ereader2html" script actually converts the PML into HTML. This doesn't sound like what you want.
Do realize that there are two different types of eReader PDB files. An older and a newer one type. They are structured slightly differently. Basically anything created with makebook is using the older format and anything created with dropbook is using the newer.
Also, with the newer files some parts of the document actually compiled and otherwise not explicitly referenced in the text. I can't speak for ficbot's tool but calibre does handle these extra parts. However, PML does not map 1 to 1 to anything (EPUB, HTML, MOBI...) so some liberty is taken to make it look correct.
Have you tried using the HTMLZ output when converting? That should give you the least amount of "special processing" and give you plain HTML to work on correcting.
|