Member Avatar for divyakrishnan

Hi...

I need a batch script to convert a PDF file to a text file on window xp.Any idea ?please help me...

  • 2 Contributors
  • 1 Reply
  • 196 Views
  • 1 Day Discussion Span
  • Latest Post Latest Post by gerbil

Dani AI

Generated

Short answer: when the PDF contains selectable text, the simplest, scriptable tool is the command‑line pdftotext (part of Poppler / Xpdf). If the PDF is a scanned image, OCR is required — Tesseract is a mature, free CLI OCR engine. As pointed out, confirm whether text is selectable first (select/copy in a reader). pdftotext man page. (manpages.debian.org) Tesseract downloads/docs (Windows binaries info). (tesseract-ocr.github.io)

Example batch (process all PDFs in the current folder with pdftotext — drops .txt beside each PDF):

@echo off
rem requires pdftotext.exe in PATH (Poppler/Xpdf)
for %%F in (*.pdf) do (
  pdftotext -layout -enc UTF-8 "%%~fF" "%%~dpF%%~nF.txt"
  for %%S in ("%%~dpF%%~nF.txt") do if %%~zS equ 0 (
    echo Empty output for "%%~nxF" - likely scanned images
  )
)

Windows builds of Poppler (pdftotext, pdftoppm, etc.) are available as prebuilt packages; place the poppler binaries in PATH or the script folder. Poppler for Windows (prebuilt packages). (github.com)

When pdftotext yields an empty file (image-only PDF) use OCR: convert pages to images with pdftoppm then run tesseract, for example:

pdftoppm -png -r 300 "input.pdf" "input_page"
tesseract "input_page-1.png" stdout -l eng >> input.txt

Repeat/loop for all page images, or use a small script to append each page’s OCR output. See Tesseract docs for installers and language models. (tesseract-ocr.github.io)

Troubleshooting notes: try pdftotext -layout to preserve columns; use -enc UTF-8 for proper character encoding; if Poppler/Tesseract executables fail on very old Windows XP installs, prefer 32‑bit/older builds or run the tools on a newer machine/VM. For PDFs with embedded fonts but garbled output, OCR may still be the only reliable option. (manpages.debian.org)

Member Avatar for gerbil

PDFs and text within. Have you ever viewed the coding of sample pdfs? The text content is in binary. PDFs with text can also in fact be images of documents. My advice would be to invoke a pdf reader such as Foxit, and the text view within, select all and to copy. Batch that. Of course that won't work with document images.
Else use something like PDF Text Extractor; there are also command-line versions.

Edited by gerbil because: n/a
Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.