Skip to content
Use case

Building chapters and a table of contents from a PDF textbook

Updated:

Chapters come from the PDF's own outline when it has one, or from the heading level that repeats across pages. On this Project Gutenberg textbook typeset in LaTeX, every engine scores low on text because of the formulas, ours included.

Original and result

Chapters, running headers, footnotes and many mathematical formulas.

Original PDF
Page of the original PDF
Result (EPUB, as a reader shows it)
The same passage in the converted EPUB

Download source PDF Download the sample EPUB

Source and licence

Document
Calculus textbook typeset in LaTeX (Project Gutenberg #33283)
Licence
Public domain in the U.S. (Project Gutenberg; its trademark only under its licence)
Pages
292
Traits
text PDF, book, chapters, repeated headers, footnotes, formulas

Settings used

  • Engine: epublab-core, build of 2026-11-03 (the one measured in the benchmark).
  • Pages: the whole document.
  • Options: none. The engine has no per-document settings yet; the same defaults apply to every file.

Validation and results

EPUBCheck EPUBCheck on our EPUB: 0 errors, 0 warnings.

8-gram recall · Like 4-gram recall with 8 words: drops when reading order or line joins break sentences (continuity). Methodology
Engine8-gram recall
EPUBLab engine (EPUB output)0.487
EPUBLab engine (plain text)0.487
pdf_oxide · extract_text0.479
pdf_oxide · to_plain_text (XY-cut)0.496
pdf_oxide · Markdown0.481
PDFium · text().all()0.494
OpenDataLoader PDF (local mode)0.533
pdftotext (Poppler)0.511

Limitations we observed

  • Formulas are written as plain text; they are not converted to MathML.
  • OpenDataLoader has a higher 8-gram recall than our engine on this document.

Convert a similar PDF