Skip to content
Use case

Converting a three-column Federal Register document to EPUB

Updated:

The Federal Register prints three narrow columns per page with rotated text in the margin, which defeats converters that read line by line. On this public-domain document the EPUBLab engine keeps the column order and removes the running headers.

Original and result

Three columns, rotated production text in the margin, running headers and footnotes.

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
Federal Register document 2026-22443
Licence
Public domain (work of the U.S. Government)
Pages
45
Traits
text PDF, three columns, repeated headers, footnotes

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)0.859
EPUBLab engine (plain text)0.853
pdf_oxide · extract_text0.334
pdf_oxide · to_plain_text (XY-cut)0.035
pdf_oxide · Markdown0.339
PDFium · text().all()0.859
OpenDataLoader PDF (local mode)0.801
pdftotext (Poppler)0.835

Limitations we observed

  • The rotated margin text is dropped, which is right here but would lose real rotated text elsewhere.

Convert a similar PDF