Why text copied from a PDF has broken line breaks

What actually goes wrong

You copy three paragraphs out of a PDF, paste them into a document, and get something unusable: every line ends where it ended on the page, words are split with hyphens in the middle, and the spacing between words is uneven. Retyping is often faster than fixing it by hand.

This is not a bug in Acrobat, Preview, or your browser. It is a direct consequence of what a PDF is, which is why every PDF reader has the same problem and why no setting turns it off.

A PDF does not contain paragraphs

A word processor document stores structure. It knows a paragraph is a paragraph, and the line breaks you see are calculated fresh whenever the window is resized.

A PDF stores the opposite: a description of where to paint each glyph on a fixed page. The format was designed so a document prints identically everywhere, and it achieves that by recording positions rather than meaning. A line of text in a PDF is a set of characters with coordinates, with nothing marking it as belonging to the same sentence as the line below.

So when you select text and copy, the reader has to reconstruct flowing text from painted positions. It walks the glyphs in order and inserts a line break wherever the vertical position changes. Those visual line endings — which existed only because the page is a certain width — become real newline characters in your clipboard.

This is why the problem scales with column width. Text from a narrow two-column academic paper breaks far more badly than text from a wide single-column report, because it has more line endings per paragraph.

Why words get split with hyphens

Typesetting hyphenates words at line endings to keep the right margin even. The word "environment" at the end of a line becomes "environ-" with "ment" starting the next line.

That hyphen is presentational. It exists to signal a continuation, and it should disappear when the text reflows. But the PDF has no way to mark it as different from a real hyphen in a word like "well-being", so a copy gives you:

This is a docu-
ment with hard line
breaks at the edge.

Fixing this correctly means joining a hyphen-plus-newline only when the next line starts lowercase — otherwise you would corrupt genuine hyphenated compounds that happen to fall at a line ending. That single rule is why find-and-replace often makes things worse.

Why the spacing looks wrong too

Justified text stretches the gaps between words so both margins line up. Those gaps are drawn as extra horizontal spacing, and when the text is extracted they frequently come through as two or three real space characters.

Some PDFs go further and adjust the space between individual letters. Extracted, that can produce spaces inside words, or invisible characters where the generator inserted positioning hints. Text that looks fine on screen can carry non-breaking spaces and zero-width characters that later break a form validation or a search.

Four ways to fix it

1. 1
   Export instead of copying

   If you have the whole document and permission to convert it, File → Export To → Word or plain text reconstructs paragraphs far better than a clipboard copy, because the converter can consider the whole page layout rather than just the selection.

2. 2
   Find and replace in Word

   Ctrl+H, search for ^p, replace with a single space. This works, but bluntly: it removes real paragraph breaks along with the unwanted ones, so you then have to put the paragraphs back by hand. Best for a single paragraph, painful for a page.

3. 3
   Paste into a plain text editor first

   This strips fonts and styling, which makes the remaining structural problems easier to see. It does not fix the line breaks themselves.

4. 4
   Use a tool built for the specific problem

   A cleaner that knows about this failure mode can rejoin wrapped lines while keeping genuine paragraph breaks, and repair hyphen splits without damaging real compound words.

What a good fix has to preserve

The reason this is not a one-line regular expression is that some line breaks are real. A blank line between paragraphs is meaningful. A line ending in a full stop followed by a capital letter is probably a real break. A line ending mid-clause almost certainly is not.

Anything that removes all line breaks indiscriminately will run your paragraphs together into one block. The useful behaviour is to treat a blank line as a paragraph boundary and join everything else within it.