How to extract text from a PDF (copy or save as .txt)

What a text extractor reads

A text-based PDF stores drawing instructions for glyphs and their positions, often with character mappings that let software recover words. An extractor interprets those objects and writes a linear text stream. Because PDF describes page appearance rather than a guaranteed reading sequence, columns, text boxes, headers, and unusual font encodings can produce unexpected order or missing characters.

Why extraction may be empty or incomplete

An image-only scan contains pixels rather than text objects, so ordinary extraction has little or nothing to read. OCR can recognize the image and produce text; some OCR workflows add a searchable layer to the PDF, while others return text separately. Extraction can also fail on nonstandard font mappings, text converted to outlines, permissions, damage, or pages that combine text and images.

Page markers provide useful context

Page markers help trace a passage back to its source and keep separate pages from running together. They do not reconstruct headings, paragraphs, columns, footnotes, or the document's semantic structure.

What plain text cannot preserve

Plain text does not preserve fonts, emphasis, positioning, tables, images, links, or form controls. Even a conversion to an editable document may only approximate the layout. Keep the PDF available whenever visual structure, citations, or exact page context matters.

Example: reviewing a long report

Suppose you need to review every occurrence of a clause in a 40-page report. A text export can be searched or processed with other tools, while page markers help you return to the original. Verify each important passage against the PDF because extraction can alter order, spacing, and characters.

Questions before extracting PDF text

Why is the result empty or short? The PDF may be image-only, use outlines or unusual character mappings, restrict access, contain damage, or mix extractable text with scanned pages. OCR addresses image text, but not every other cause.

Does extraction preserve formatting? No. Plain text discards visual formatting and cannot reliably represent tables, columns, images, or form fields.

Can I tell which page a piece of text came from? Yes, if the extraction marks page boundaries in the output. Look for page markers rather than one continuous block of text.

After extraction, scan the output for missing pages, scrambled columns, replacement characters, and broken words. Use OCR for image-only pages and return to the source PDF whenever accuracy or context is important.