Sample work · Data & Documents
Accessible PDF: Operating manual with structure, alternative texts and language
An operating manual was generated from a semantic source as a tagged PDF and post-processed: Headings, lists, table with header cells, images with alternative text, headers and footers as artifacts. veraPDF checked it against PDF/UA.
Sample work (demo), not a client job. Built as a trial task with invented example data; the check figures below come from the automatic test run and are not polished.
What was built
- Structure adopted from the source: Headings, lists, table, footnote, link
- Alternative texts and language come from the source and are never invented
- Page numbers, lines and decorative images marked as artifact
Download bedienungsanleitung-barrierefrei.pdf
The automated check does not replace a test with a screen reader. Form fields and multi-column layouts are not proven in this sample.
How it was checked
Automatic test run on 2026-10-31, taken from the log (test data invented, figures measured):
- Manual source (HTML, lang=de, 9 headings H1-H3, 2 lists, 2 images with alt + 1 decorative, table with 6 TH scope, footnote, link) -> Chromium tagged PDF (2 pages, 98 structure elements) -> pikepdf post-processing (artifacts: 1 image group, 8 footer text objects, 1 line
Limits
What this sample does not cover, so nobody expects more:
- Automated testing (custom structure check + veraPDF) does not replace manual control with a screen reader (NVDA/JAWS) and PAC; comprehensibility of alt texts and sensible reading order in the ...
- Proved is the new creation from a semantic source (HTML/Word). Post-processing foreign, untagged PDFs (scans, legacy inventory from prepress) is significantly harder: Structure must be done by hand ...
- No form fields (AcroForm/Widget with TU labeling and tab order) in the sample - accessible forms are therefore NOT proven
- Alt texts and language are adopted from the source, never invented: if they are missing there, the PDF fails (counter-checks without_alt/without_lang)
- Chromium as generator: no THead/TBody groups, no Note tags for footnotes, paragraphs as NonStruct children; sufficient for PDF/UA-1, PDF/UA-2 / WCAG PDF techniques not tested in detail
- Artifact marking here is blanket (everything Chromium did not tag); only secured because the check limits free text to the page number - for foreign PDFs, per object ...
- Single-column layout only, one language; language switches in text (Lang per element), multi-column layouts, MathML and complex tables (colspan/rowspan, Headers IDs across multiple levels) not proven
Something like this for your business?
Describe briefly what you need. You get a fixed price before I start; nothing happens without your yes.