DocParser

A local web app that converts PDF, DOCX, DOC, XLSX, XLS, and CSV files into Markdown or JSON — no cloud, no file size limits, no accounts.

Engines

Format	Engine
PDF	opendataloader-pdf
DOCX / DOC	python-docx
XLSX / XLS / CSV	openpyxl

Requirements

• Python 3.10+
• Java 11+ (required by opendataloader-pdf — install from Adoptium)

Setup

# 1. Install dependencies
pip install -r requirements.txt

# 2. Run the server
python app.py

# 3. Open in your browser
# http://localhost:5000

Optional: Hybrid mode for complex PDFs

For scanned PDFs, borderless tables, formulas, or charts, enable hybrid mode:

# Terminal 1 — start the AI backend
pip install "opendataloader-pdf[hybrid]"
opendataloader-pdf-hybrid --port 5002

# Terminal 2 — run the app as normal
python app.py

Then in app.py, add hybrid="docling-fast" to the opendataloader_pdf.convert() call.

Features

• No file size limit — convert one file or a hundred at once
• Batch ZIP download — download all converted files in one click
• Per-file preview — see the first 2,000 characters inline
• Per-file download — save individual converted files
• Runs 100% locally — your documents never leave your machine

Output formats

• Markdown (.md) — clean text with headings, tables, and lists
• JSON (.json) — structured data with element types (heading, paragraph, table…)