PDF to Text Extractor

Extract the text of PDF files, page by page or as one document. Give it direct links to PDFs (or upload one file) and get clean text in a dataset you can download as JSON, CSV or Excel, or read through the Apify API.

What it does

• Downloads only the PDFs you list (up to 500 per run, up to 100 MB each). It does not crawl websites.
• Reads the text layer of each page with pdf.js (through unpdf).
• Returns one row per page, or one row per PDF with the pages joined and marked --- Page N ---.
• Tells you which pages had no text, and which files failed and why.

What it does not do

• No OCR. A scanned page is an image: it has no text layer, so it comes back empty with hasText: false. Pages without text are never charged. If you need OCR, use an OCR Actor for those pages.
• No password-protected PDFs (reported as an error, not charged).
• Layout is simplified: text comes in reading order, line by line; tables become plain lines of text.

Pricing (pay per event)

Event	Price	When
Page with text	$0.003 ($3 per 1,000 pages)	One page whose text was extracted

Free: pages without text (scans, blank pages), files that fail to download, files that are not PDFs, broken or password-protected PDFs, and pages beyond your "Max pages per PDF". Your maximum cost per run is respected: when it is reached the run stops and the pages left out are not charged.

Example: a 6-page text PDF plus a 1-page scan costs 6 × $0.003 = $0.018.

Input

{
    "pdfUrls": [
        "https://www.irs.gov/pub/irs-pdf/fw9.pdf"
    ],
    "outputFormat": "perPage",
    "maxPagesPerPdf": 1000
}

• pdfUrls: direct links to PDF files.
• pdfFile (optional): upload one PDF from your computer in the Apify Console.
• outputFormat: perPage (one row per page, default) or perDocument (one row per PDF).
• maxPagesPerPdf: read only the first N pages of each PDF.

Output

One row per page (perPage):

{
    "url": "https://www.irs.gov/pub/irs-pdf/fw9.pdf",
    "fileName": "fw9.pdf",
    "pageNumber": 1,
    "totalPages": 6,
    "hasText": true,
    "charCount": 5658,
    "text": "Form W-9(Rev. March 2024)\nRequest for Taxpayer\nIdentification Number and Certification\n..."
}

A file that fails gives one free row with error set, for example "download failed: HTTP 404" or "the file is not a PDF".

The OUTPUT record of the run's key-value store has a summary: PDFs read, pages with and without text, pages charged and the errors.

Use cases

• Feed PDF reports, papers, manuals or forms into a search index, a spreadsheet or an LLM pipeline.
• Check which PDFs in a list are scans (no text layer) before sending only those to OCR.

Notes

• Only process files you have the right to use. The Actor reads the files you give it and nothing else.
• Questions or a PDF that comes out wrong? Open an issue on this Actor's page.