PDF Contact Details Extraction

I have a collection of PDFs that vary widely in layout and often run over several pages. From every file I only need three things captured—mobile numbers, email addresses, and website URLs—and delivered in a single CSV. Each type of data should sit in its own column so the final file is ready for quick filtering and analysis.

Because there’s no consistent template inside the PDFs, the script or program you build must be able to scan the full text, detect the three patterns reliably, and ignore any surrounding noise. Handling dozens of pages without missing buried contact info is essential.

Here’s how I will consider the job complete: I hand you a small sample set of PDFs, you return one CSV where every mobile number, email ID, and website appears in the correct column with no misplaced characters or duplicates. Once verified, you run the same process on the remaining files and provide the final CSV plus the executable script or notebook so I can reuse it on future documents.

If your preferred tools are Python with libraries like PyPDF2, pdfplumber, or similar, that’s perfect, but I’m open to other approaches as long as the output is accurate and repeatable.