Convert messy PDF-extracted text into clean Markdown files
Project description
pdfcleanerx
Convert messy PDF-extracted text into clean Markdown files.
- Fully offline – no AI APIs, no cloud services
- Modular pipeline – swap or extend any layer
- CLI + Python SDK
- Production-ready – structured logging, custom exceptions, typed
Installation
pip install pdfcleaner
From source
git https://github.com/namanbhola1888/pdfcleaner
cd pdfcleanerx
pip install -e ".[dev]"
Quick Start
CLI
# Single file → writes output/report.md
pdfcleaner convert report.pdf
# Named output
pdfcleaner convert report.pdf --output clean.md
# Print to stdout
pdfcleaner convert report.pdf --stdout
# Batch (glob)
pdfcleaner convert docs/*.pdf --output-dir ./markdown
# Verbose / debug
pdfcleaner convert report.pdf --verbose
Python SDK
from pdfcleanerx import Converter
# Default: full pipeline → Markdown string
converter = Converter()
markdown = converter.convert("report.pdf")
print(markdown)
# Write directly to file
written_path = converter.convert_to_file("report.pdf", "report.md")
Dependency injection (custom pipeline)
from pdfcleanerx import Converter
from pdfcleanerx.cleaners import CleanerPipeline, WhitespaceCleaner, PageNumberCleaner
# Only run two cleaners
pipeline = CleanerPipeline([PageNumberCleaner(), WhitespaceCleaner()])
converter = Converter(pipeline=pipeline)
markdown = converter.convert("report.pdf")
Configuration
Copy .env.example to .env and edit:
cp .env.example .env
Key settings:
| Variable | Default | Description |
|---|---|---|
LOG_LEVEL |
INFO |
DEBUG / INFO / WARNING / ERROR |
OUTPUT_DIR |
./output |
Default output directory |
HEADING_FONT_SIZE_THRESHOLD |
1.2 |
Font size multiplier for heading detection |
CLEANER_PAGE_NUMBERS |
true |
Enable/disable page-number removal |
CLEANER_LINE_WRAP |
true |
Enable/disable line-wrap merging |
CLEANER_WHITESPACE |
true |
Enable/disable whitespace normalisation |
CLEANER_HEADINGS |
true |
Enable/disable heading detection |
FOOTER_REPEAT_THRESHOLD |
3 |
Pages a footer must repeat on to be stripped |
Architecture
CLI (Typer)
│
▼
Converter ← SDK entry point (dependency injection)
│
├── BaseExtractor ← pdfplumber (swappable)
│ └── Document (pages → blocks + font metadata)
│
├── CleanerPipeline ← ordered chain
│ ├── PageNumberCleaner
│ ├── WhitespaceCleaner
│ ├── HeadingDetector
│ └── LineWrapCleaner
│
└── BaseFormatter ← MarkdownFormatter (swappable)
Adding a custom cleaner
from pdfcleanerx.cleaners.base import BaseCleaner
from pdfcleanerx.models import Document
class MyCustomCleaner(BaseCleaner):
def clean(self, document: Document) -> Document:
for page in document.pages:
for block in page.blocks:
block.text = block.text.replace("©", "")
return document
# Inject into Converter
from pdfcleanerx.cleaners import CleanerPipeline, PageNumberCleaner, WhitespaceCleaner
from pdfcleanerx import Converter
pipeline = CleanerPipeline([PageNumberCleaner(), WhitespaceCleaner(), MyCustomCleaner()])
converter = Converter(pipeline=pipeline)
Adding a custom formatter (e.g. HTML)
from pdfcleanerx.formatter.base import BaseFormatter
from pdfcleanerx.models import Document
class HtmlFormatter(BaseFormatter):
def format(self, document: Document) -> str:
parts = ["<html><body>"]
for page in document.pages:
for block in page.blocks:
level = getattr(block, "_heading_level", 0)
if level:
parts.append(f"<h{level}>{block.text}</h{level}>")
else:
parts.append(f"<p>{block.text}</p>")
parts.append("</body></html>")
return "\n".join(parts)
converter = Converter(formatter=HtmlFormatter())
Development
# Install dev dependencies
pip install -e ".[dev]"
# Run tests
pytest
# Run with coverage
pytest --cov=pdfcleanerx --cov-report=html
# Lint
ruff check src/ tests/
# Type check
mypy src/
Build & Publish to PyPI
Prerequisites
pip install build twine
Build
python -m build
# Creates dist/pdfcleanerx-0.1.0.tar.gz and dist/pdfcleanerx-0.1.0-py3-none-any.whl
Test on TestPyPI first (recommended)
twine upload --repository testpypi dist/*
pip install --index-url https://test.pypi.org/simple/ pdfcleanerx
Publish to PyPI
twine upload dist/*
You will be prompted for your PyPI API token. Store it in ~/.pypirc or set:
export TWINE_USERNAME=__token__
export TWINE_PASSWORD=pypi-<your-token>
Bump version
Edit version in pyproject.toml and src/pdfcleanerx/__init__.py, then rebuild.
License
MIT
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
pdfcleaner-0.1.0.tar.gz
(27.3 kB
view details)
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file pdfcleaner-0.1.0.tar.gz.
File metadata
- Download URL: pdfcleaner-0.1.0.tar.gz
- Upload date:
- Size: 27.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
3abf7ee58980ef8909fb4b96ce543928a8fe2636fd732ed8dd7ee8f816c8daaa
|
|
| MD5 |
fc0f2f1dd2480af663260beeb7829ec7
|
|
| BLAKE2b-256 |
96a8cc7992236ba9704345b4cc5da73a546bb117d6fabdbab0545789002e0ce1
|
File details
Details for the file pdfcleaner-0.1.0-py3-none-any.whl.
File metadata
- Download URL: pdfcleaner-0.1.0-py3-none-any.whl
- Upload date:
- Size: 23.2 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
51ed5071ad9f55067eda0369a446191e10ef557701342f945d473dfb13e74a24
|
|
| MD5 |
29d928f746a565cde17f9be4187d5043
|
|
| BLAKE2b-256 |
5b77f04dfb109480b5ade537b5765f3d91dfa63f6385dce13375faa61bd21ffc
|
