PDF guides
PDF to RTF: How Table and Heading Detection Works
PDF to RTF reuses the exact same text-clustering, heading-inference, and table-detection engine as PDF to Word — the same underlying analysis, just written out as RTF instead of a DOCX table object. This guide explains how that detection works.
Ready to try the tool this guide describes?
How heading detection works
A line of text is treated as a heading when it stands alone (not part of a longer paragraph), is noticeably larger than the surrounding body text, and is reasonably short. Larger size ratios are inferred as more prominent headings.
How table detection works
The tool looks for runs of at least three consecutive lines of text that share recurring column positions beyond the page’s left margin, with most of those positions filled in across the rows. When a region matches that pattern closely enough, it is rebuilt as a real RTF table with `\trowd`/`\cellx`/`\intbl` markup — the same detection logic used by PDF to Word, just written to a different output format.
When detection doesn’t apply
Justified paragraphs that happen to share a left margin are deliberately not treated as tables — detection specifically requires a recurring interior column position, not just a shared starting edge. Irregular tables or tables with merged cells may not be detected and will come through as plain paragraph text instead.
PDF to RTF Table & Heading Detection FAQ
- Is this the same table detection as PDF to Word?
- Yes, exactly the same underlying engine — only the output format differs (real RTF table markup instead of a DOCX table object).
- Will every table convert correctly?
- No. This is a best-effort heuristic, the same one PDF to Word uses. Simple, regular tables convert reliably; irregular or merged-cell tables may come through as plain text.
- How are headings identified?
- By looking for standalone, short lines of text that are noticeably larger than the surrounding body text.
- Will justified paragraphs get mistaken for a table?
- No. Detection specifically requires a recurring interior column position, not just a shared left margin.
- What does the output table markup look like in the RTF file?
- A detected table is written as real RTF table markup (\trowd, \cellx, \intbl, \row control words), the same structure a word processor produces natively — not a plain-text imitation with tabs.
Related guides
- How to Convert PDF to RTFConvert a PDF into an editable Rich Text Format (.rtf) document in your browser. A best-effort conversion, similar in philosophy to PDF to Word.
- PDF to RTF Scanned PDF LimitationsPDF to RTF requires extractable text. Scanned or image-only PDFs are detected and rejected with a clear message rather than producing an empty document — no OCR is performed.
- PDF to Word Table ReconstructionHow PDF to Word detects simple, regularly-gridded tables and rebuilds them as real Word tables on a best-effort basis, and when a table comes through as plain text instead.