Copy the text out of a PDF in four steps
- Open PDF to Text and choose the PDF, or drop it on the page.
- Leave the Pages box empty for the whole document, or type pages like
1-3, 5, 8-. - Choose As in the PDF or Paragraphs for the line breaks, and turn Page markers on to see where each page starts.
- Press Copy to put the text on the clipboard, or Download to save a .txt file named after the PDF.
If the tool says the PDF looks like scanned pictures, there is no text in the file to copy. ClipLime does not read text from pictures (no OCR). The next section shows how to recognize a scan, and a later one covers what to do about it.
Why you can't select anything
A PDF can hold a page in two very different ways. A PDF made from a document, a web page or a spreadsheet contains real text: every piece has its characters and its position. That text layer is what you select in a viewer. A PDF made by scanning paper holds a picture of each page, with no characters in the file to copy.
A ten second test: drag across a word, then press Ctrl+F (Cmd+F on a Mac) and search for a word you can see. If nothing highlights and the search finds nothing, the page is a picture.
| What happens | What it means | What to do |
|---|---|---|
| You can select words, but they paste in the wrong order | Real text, with a layout your viewer reads badly, such as columns or tables | PDF to Text rebuilds the order from where the text sits |
| Nothing highlights and search finds nothing | A scan or a photo: a picture with no text | You need OCR software. ClipLime does not read text from pictures |
| Selecting works but pasting gives squares or odd symbols | The fonts do not say which letters they draw | You need a version of the PDF with a proper text layer |
| The viewer says copying is not allowed, or asks for a password | The PDF has a password or permission settings | ClipLime cannot open these. If you are allowed to, print or export a new unprotected PDF from a viewer |
One twist: some scanned PDFs carry an invisible text layer added by a scanner or an OCR program. That text selects normally and is read by the tool, mistakes included.
What PDF to Text does
It reads the text layer page by page and gives you plain text: no fonts, bold, colors or images. Text typed into form fields (with the name of the field) and the text of comments are added at the end of each page, and you can switch that off. Right to left text and Chinese, Japanese and Korean text come out in reading order.
A PDF does not store paragraphs. It stores pieces of text, each with a position, in the order the program drew them, which is not always the order you read them in. The tool rebuilds reading order from the positions: pieces on the same baseline become a line, lines stack from top to bottom, and a strip of empty space running down the page marks a column.
Columns, tables and odd layouts
- Columns are read one after the other. A two column article comes out as the left column, then the right, and a title across both stays above them.
- Tables are not split into columns. When short cells line up in rows, each row is read across with tabs between the cells, so the result pastes into a spreadsheet. A table with long cells that wrap onto several lines can come out in an odd order.
- Headers, footers and page numbers are text on the page, so they appear with each page: a running head first, a footer or page number last.
- Overlapping text boxes, text wrapped around pictures and unusual designs can come out in a different order. A vertical label in a margin is added after the main text.
Line breaks, paragraphs and hyphens
As in the PDF gives one line of text for every line of the page, with a blank line where reading jumps to another column or block. It is the safest choice for tables, addresses, code and poems.
Paragraphs joins the lines of a paragraph into one line and puts a blank line between paragraphs, while short lines such as a list stay lines. It is a careful guess, so check the result if the layout matters. A paragraph that runs onto the next page is not joined across the page break.
In Paragraphs mode, a word split at the end of a line is joined again only when the evidence is clear, for example when the hyphen is a soft hyphen or the word appears whole elsewhere in the text. Otherwise the hyphen stays, so a real compound such as long-term is usually left alone, and a split word may keep its hyphen. As in the PDF leaves split words as drawn.
A scanned PDF: what to do instead
A scan is a picture of a page. ClipLime does not offer OCR, the step that turns pictures of letters into text, so for a PDF made only of scans this tool gives you nothing and says so. Your honest options:
- Use OCR software or an OCR feature you trust. It can make mistakes, so read the result. An online OCR service is an upload, so read is it safe to upload a PDF to a free website first.
- Prepare the pages. Save them as pictures with PDF to JPG. 300 dpi is the usual choice for programs that read text from pictures, but stay near the scan's own resolution, because extra pixels add no detail. Turn sideways pages upright first, as in rotating a PDF and saving it.
- Only some pages are scans? The pages with text are read and the pages without are named in the result. Pull out just those pages with Split PDF into pages and send only them to OCR.
- For a paragraph or two, retyping from the picture is often faster than any tool.
The guide to converting a PDF to JPG covers resolution and quality in more detail, and the guide to splitting a PDF into pages shows how to separate single pages.
Text from several PDFs at once
To get the text of several PDFs in one go, join them first with Merge PDF, then read the merged file once with Page markers on, and use the page count of each file to find where each source begins. Merging copies the pages as they are, so selectable pages stay selectable and scanned pages stay pictures.
After you have the text
Count or tidy it with Word Counter or Text Cleaner. If you extracted two versions of a contract, the guide to comparing two texts and seeing every difference shows what changed.