Text Copied From a PDF Is a Mess: How to Fix Line Breaks, Hyphens and Odd Characters

A PDF describes where to draw text on a page, not what the text means. There is no paragraph object, only lines of glyphs at coordinates. When you copy, your software reconstructs reading order from positions, and it records a line break at the end of every visual line because that is the only break it can see. Headers, footers and page numbers sit in the same stream as the body, and in multi column layouts the reader may jump between columns.

• ▸Hard line breaks inside paragraphs. Replace single line breaks with spaces but keep paragraph breaks. The Remove Line Breaks tool does this in one pass and can keep blank lines as paragraph boundaries.
• ▸Hyphenated words split at the end of a line, such as informa- and tion on the next line. A rule that removes a hyphen followed by a line break before a lowercase letter rejoins most of them. Leave hyphens that belong to compounds such as well-known, which normally appear mid line.
• ▸Ligatures and special characters. Typesetting software often stores fi, fl and ffi as single ligature characters, and your paste shows them as odd symbols or breaks your search. Unicode compatibility normalization turns them back into plain letters. The Remove Text Formatting tool has an option for this.
• ▸Repeated headers, footers and page numbers. Search for the repeated phrase and delete it, or use Find and Replace with a pattern for lines that contain only digits.

The order matters because early steps change what later steps can see.

• ▸Paste into a plain text box so no styles come along.
• ▸Normalize characters first: ligatures, curly quotes if you need straight ones, and non-breaking spaces.
• ▸Remove repeated headers, footers and page numbers while they are still on their own lines.
• ▸Rejoin hyphenated words, then join lines inside paragraphs.
• ▸Collapse repeated spaces and trim line ends.
• ▸Read through once, scanning for numbered lists and bullets, which the earlier steps may have merged into running text.

Some PDFs paste without any blank lines between paragraphs, so a join step would merge the whole document into one block. In that case, add paragraph breaks where a line ends with a full stop and the next line starts with a capital letter and is clearly the start of a new thought, or work page by page. Another option is to skip copy and paste and use the PDF to Markdown Converter, which reads positions and font sizes to find headings and paragraph gaps before it produces text.

• ▸Two column articles: select and copy one column at a time. Copying across both columns interleaves their lines.
• ▸Tables: pasted cell text appears line by line. Rebuild the table in a spreadsheet, or select a single column at a time and recombine.
• ▸Scanned PDFs: there is no text layer, only pictures of pages. Nothing can be copied. Use image to text recognition on the page images instead, and expect to proofread the result.

If you are the author, export PDFs from tagged sources such as Word with accessibility options enabled, and keep a plain text or Markdown source. Tagged PDFs carry paragraph and heading structure that copy tools and screen readers can use, which makes both copying and accessibility much better.