EpubCCThe quality warnings in your conversion report
EpubCC's quality report shows a warning whenever the engine detects a problem it could not fix, instead of hiding it. There are six: scanned pages, characters the PDF doesn't let us recover, a long book without chapters, tables to check, simplified tables, and pages that could not be read.
Updated:
Scanned pages
Shown when a page has no selectable text and an image covers the whole page. Such pages are scanned images; without OCR there is no text to convert, so they are left out of the EPUB and listed by number.
What to do: if only a few pages are affected (a cover, a scanned insert), the rest of the book converts normally. A fully scanned book needs OCR first, which EpubCC does not offer yet.
Characters without Unicode
Shown when a page has many characters drawn with a font that has no ToUnicode map: the PDF shows the letters but does not say which letters they are. They appear in the EPUB as the replacement character �, and copying text from the PDF shows the same problem.
What to do: export the PDF again from its source with fonts embedded properly, if you can. Otherwise those passages need OCR or retyping.
No chapters detected
Shown when a book of more than 30 pages ends up as a single section: the PDF has no outline and no heading level the engine could rely on. The text is complete; only chapter navigation is missing.
Tables to check
Shown when the engine rebuilt a table as a real EPUB table (rows, columns, header cells and caption) on the pages listed. Tables are one of the hardest things to read from a PDF, so the report asks you to check them in the sample before converting the whole book.
What to do: open the sample and compare each table with the original page in the preview.
Simplified tables
Shown when a table's grid could not be rebuilt reliably, for example with cells that span several rows or headers on several levels. Each row is kept as one line with its cells separated by a dot, so no text is lost, but it no longer reads as a grid.
What to do: if those tables matter, keep the PDF next to the EPUB for them, or check whether the source document exists as HTML or a spreadsheet.
Unreadable pages
Shown when some pages could not be read at all, usually because they are damaged. Those pages are missing from the EPUB and listed by number.
What to do: open the PDF and check those pages; exporting or printing the PDF to a new file often repairs them.
FAQ
Will the EPUB be perfect?
No converter can promise that, and we don't. When the engine detects problems (unreadable glyphs, doubtful reading order) it shows a warning instead of hiding it, and the sample shows you the hard parts before you pay.
Does it work with scanned PDFs?
Not yet. EpubCC starts with PDFs that have selectable text. Scanned books need OCR, which comes later as a separate option and will never be sold as perfectly accurate.