Search Text in Multiple PDFs

When the answer is "somewhere in one of these forty PDFs", opening them one by one is the slow way. Select all the files, type a word, phrase or pattern, and this tool searches every page of every PDF, listing each match with the file name, page number and the text around it. Use it for contracts, invoices, papers or exported statements — the files are searched in your browser and never uploaded.

How to use this tool

1. Select all the PDFs you want to search.
2. Type the word or phrase. Spaces also match line breaks in the PDF.
3. Tick Match case, Whole words or Regular expression if you need them.
4. Read the matches with their context, or download the CSV.

Formula

Each page's text layer is read in order, pieces on the same line are joined, whitespace is collapsed to single spaces and the text is Unicode-normalised (NFKC), so ligatures like "fi" and full-width letters match normal ones. The search then runs over the whole page, so a phrase split across two lines is still found.

Whole words means the match is not touching another letter, digit or underscore, in any script, so accented and Korean words are handled too. In Korean a particle written straight after a word (계약서는) counts as part of that word, so leave Whole words off when searching Korean nouns.

Example

Searching 25 supplier invoices, each printing its number once, for the regular expression INV-\d{4} returns 25 matches in 25 files, each with its page and about 60 characters either side, e.g. "…Invoice no. »INV-2048« issued 3 March…". The CSV has columns File, Page, Match, Before and After.