PDF/A and PDF/UA

PDF/A is the ISO standard for documents that must stay readable for decades. PDF/UA is the standard for accessible PDFs. Convert existing PDFs here, or ask for either while converting.

POST https://api.pdfdispatch.com/v1/pdf/pdfa

Form fields

Field	Type	Default	What it does
`files` required	file(s)	–	The PDFs to convert.
`pdfa`	string	–	`PDF/A-1b`, `PDF/A-2b` or `PDF/A-3b`. Other levels are refused (400).
`pdfua`	boolean	`false`	Make the file PDF/UA (universal accessibility).

At least one of pdfa and pdfua is required.

Which level

Level	Based on	Use it when
`PDF/A-1b`	PDF 1.4	a recipient insists on PDF/A-1. No transparency, no attachments.
`PDF/A-2b`	PDF 1.7	archiving in general. Keeps transparency and layers. A safe default.
`PDF/A-3b`	PDF 1.7	the file must carry attachments, such as the XML of an e-invoice.

Examples

An existing PDF:

curl --fail-with-body \
  --header "Authorization: Bearer $PDD_KEY" \
  --form files=@invoice.pdf \
  --form pdfa=PDF/A-2b \
  --output invoice-archive.pdf \
  https://api.pdfdispatch.com/v1/pdf/pdfa

Better: ask for it while converting, which saves a request and keeps more of the document’s structure:

curl --fail-with-body \
  --header "Authorization: Bearer $PDD_KEY" \
  --form files=@index.html \
  --form pdfa=PDF/A-2b --form pdfua=true --form generateTaggedPdf=true \
  --output invoice.pdf \
  https://api.pdfdispatch.com/v1/convert/html

Good to know

PDF/A documents cannot be encrypted; the API refuses the combination.
Fonts are embedded and colours fixed to a profile; the file usually grows a little.
A tool can only make a PDF accessible when the source has the structure: real headings, lists, table headers, alt text on images, a lang on <html>. For HTML, write semantic markup and add generateTaggedPdf=true.