How to Clean Up Text Copied From a PDF

You select a paragraph in a PDF, copy it, paste it into your document, and it looks like it went through a shredder. Every line ends in the wrong place, words are split by hyphens, and there are double spaces where nothing should be. It’s one of the most common small annoyances in office work, and it has a simple cause and a simple fix.

Why it happens

A PDF isn’t a document in the way a word-processor file is. It’s closer to a printed page: the file records where each piece of text sits on the page, line by line. There’s no stored idea of a “paragraph” that flows from one line to the next. When you copy from it, your computer has to guess, and the easiest guess is to treat the end of every visual line as the end of a line of text.

That gives you the usual symptoms:

• A line break at the end of every line, so sentences are cut in half.
• Hyphenated words split across lines, like “informa- tion”, left behind from the original justification.
• Double spaces and odd gaps, from justified text or tabs.
• Page numbers, headers and footers dropped into the middle of your paragraph.
• Bullets and numbers pasted as strange symbols.

A five-step routine

You can fix most of it in a few minutes with the same sequence each time.

1. Remove the line breaks. Paste the text into the remove line breaks tool. The default option turns single line breaks into spaces and keeps blank-line paragraph breaks, which rebuilds proper paragraphs. Leave “Collapse repeated spaces” and “Trim spaces” on to clear the double spaces at the same time.

2. Repair split words. Words that were hyphenated at the end of a line now appear as a hyphen followed by a space, such as “informa- tion”. Use find and replace to join them: find a hyphen followed by a space and replace with nothing. A word of caution: this will also join genuine hyphens followed by a space, for example in a list or a dash used as punctuation, so scan the result.

3. Delete headers, footers and page numbers. These are best removed by hand, or with a pattern such as a line that is only digits. In find and replace with regular expressions on, a pattern like \n\d+\n matches a number alone on a line, though check the matches against your document, since some numbers are real.

4. Fix special characters. PDFs often use “smart” quotes, ligatures such as the combined fi and fl characters, and special bullets. Replace any that look wrong, using find and replace.

5. Check the paragraphs. Read through once. Headings, captions and lists may have been merged into body text, because they looked like ordinary line breaks to the tool. Add the breaks back where they belong.

Tips for fewer problems

Copy less at a time. A single page or section is easier to check than a whole document.

Try the original source. If the PDF came from a Word file or a web page, the original is cleaner than anything you can copy out of the PDF. Ask for it if you can.

Use the PDF reader’s text export. Some readers have an option to save as plain text, which can handle columns and paragraphs better than copy and paste, especially for multi-column layouts.

Watch for columns. Text laid out in columns may paste with lines from different columns interleaved. Copy each column separately.

Scanned PDFs are different. A scan is a picture of text, so there’s nothing to copy until it goes through optical character recognition, which tends to introduce its own errors: wrong characters, extra spaces and stray symbols. Proofread carefully.

Cleaning lists and tables

Copying a table from a PDF often produces one cell per line, which is hard to use. For a single column you can paste the lines straight into a spreadsheet. For several columns, you may need to rebuild the structure with find and replace. A list of items can be tidied with remove duplicate lines and put in order with sort lines, and add text to each line can turn it into a comma-separated list.

Count what you’ve cleaned

Stray breaks and extra spaces can skew counts, so clean before you count. The word counter gives you words, sentences and paragraphs afterwards, and it helps you check that nothing was lost in the clean-up: if the word count dropped sharply, something got deleted.

The short version

The mess comes from PDFs storing lines, not paragraphs. Remove single line breaks, collapse spaces, repair hyphenated words, delete headers and footers, and read the result once. For documents you’ll process again and again, a saved sequence of find-and-replace patterns will save you a lot of time.