PDF text cleaner
Rebuild paragraphs copied from PDFs by joining visual line wraps and repairing words split by hyphens.
How to use pdf text cleaner
Text copied out of a PDF arrives with a line break at every visual line ending and hyphens in the middle of split words. This tool reverses both, using the blank line between paragraphs as its signal for which breaks are real.
- 1Copy from the PDF and paste it here
Select the passage in your PDF reader, copy it, and paste it into the left editor unedited. The broken line endings and split words are exactly what the tool keys on, so pre-tidying only removes its evidence.
- 2Check the two repair rules
Join hyphenated words repairs splits like "docu-" / "ment", joining only where a hyphen sits at a line ending and the next line starts with a lowercase letter. Keep paragraph breaks treats every blank line as a paragraph boundary; switch it off to flow everything into a single block.
- 3Clean and verify the paragraphs
Select Clean text and check that the right panel reads as continuous prose. If two paragraphs merged, the PDF put no blank line between them — add one on the left and clean again.
When this tool is useful
How it tells a wrapped line from a real break
The core rule is simple: a blank line is a paragraph boundary, and every other line break is assumed to be a visual wrap. Within a paragraph, each break becomes a single space, which is what rebuilds flowing sentences out of the fixed-width lines a PDF reader hands the clipboard.
Hyphen repair is deliberately conservative. A word is joined only when a hyphen is immediately followed by a line ending and the next line opens with a lowercase letter — the signature a typesetter’s mid-word split leaves behind. A compound like "well-being" sitting mid-line is never touched. The one unresolvable case is a compound split across lines exactly at its own hyphen: in the file that is indistinguishable from a typesetting split, so the tool joins it, accepting the rarer of the two possible errors.
Multi-column PDFs are the hardest input. If the copy came out with the columns interleaved, the sentence order itself was scrambled during extraction, and no text-level repair can un-shuffle it. Copy one column at a time and the cleaner handles the rest.
PDF text cleaner questions
A PDF stores where each character is painted on the page, not which paragraph it belongs to. The reader reconstructs text by position and inserts a line break wherever the vertical position changes, so every visual line ending becomes a real newline on the clipboard.
It joins only a hyphen sitting directly at a line ending with a lowercase letter starting the next line — the pattern hyphenated typesetting produces. Hyphens anywhere else, including in compounds like "part-time", are never touched.
The tool recognises paragraphs by blank lines, and this PDF’s copy did not include any. Add a blank line where each paragraph should end and clean again, or accept the single block if the destination will reflow it anyway.
If each column copied as a contiguous block, yes. If the reader interleaved lines from both columns, the ordering damage happened during extraction and cannot be repaired from the text alone — copy one column at a time instead.
It depends on the scan’s OCR layer. If copying produces text at all, the same line-break and hyphen repair applies, though OCR misreads stay. A scan with no text layer has nothing to copy and nothing to repair.