How the text is rebuilt
Inside a PDF, text isn’t stored as sentences. It’s stored as small pieces placed at exact spots on the page, often a word or a few letters at a time and not always in reading order. The tool reads where every piece sits and puts them back together: pieces at the same height become a line, read left to right; lines are read top to bottom; and a bigger gap between lines starts a new paragraph. Spaces are added where there’s a visible gap between words.
For most documents, such as reports, letters, articles and e-books, this gives clean text you can paste into an email, a document or a translation app. Pages with two columns, sidebars or tables come out line by line across the page, so columns may be interleaved. For tables, PDF to Excel keeps the rows and columns apart.
Scanned PDFs
If a PDF was made from photos or a scanner, each page is a picture, and there’s no text inside to extract. The tool tells you when it finds no text, or which pages had none. Getting words out of pictures needs OCR (optical character recognition), which this tool doesn’t do. Many phone scanner apps and Google Drive offer OCR.
Languages and special characters
Text is taken out exactly as stored, so accented letters, Urdu, Hindi, Arabic, Chinese and other scripts come through when the PDF stores them properly. Right-to-left text may appear in visual order in some files, depending on how the PDF was made.
Copy or download
“Copy text” puts everything on your clipboard. “Download .txt” saves a plain UTF-8 text file named after the PDF, which opens in any text editor and in Word. Your file is opened and processed in your browser, so it is never uploaded to a server. For a password-protected PDF, unlock it first with Remove PDF Password.