Well, I tried using Calibre, turning on the heuristics.
It converted better, with a few anomolies:
1. Triple copies of the cover
2. Numbering is off. A few pages show 1 of 4, then many show 2 of 4. There are over 200 pages though.
3. Margins are too large - too much white space around each page.
4. It converted all the text pages, which I could highlight in Adobe Reader, into png. Only 2 pages, the front and back are graphical.
Trying to correct the above. I tried the "tweak book" tool. I opened content.opf. I don't know that much html, which I think will help. Am I on the right track here?
|