eDocAPI

A lightweight Python framework for building document-processing and document-automation APIs.

Simple API for the developer. Powerful document processing underneath.

Built on Starlette. Version 0.1.0.

---

Installation

Core (PDF, DOCX text/HTML, images, Markdown text/HTML, uploads, CLI):

pip install edocapi

With enhanced HTML/Markdown/DOCX → PDF layout (pulls in WeasyPrint):

pip install edocapi[html]

WeasyPrint may need system libraries for enhanced layout. On Debian/Ubuntu:

sudo apt-get install -y libcairo2 libpango-1.0-0 libpangocairo-1.0-0 \
  libgdk-pixbuf-2.0-0 libffi-dev shared-mime-info

With PDF → images (requires poppler):

pip install edocapi[pdf-images]
# system: sudo apt-get install -y poppler-utils

Everything:

pip install edocapi[all]

Development (from source):

pip install -e ".[dev,html]"

Build and publish

From the repository root, build and check the package before uploading:

py -m pip install --upgrade build twine
py -m build
py -m twine check dist/*
py -m twine upload dist/*

Use __token__ as the Twine username and your PyPI API token as the password when prompted. Do not commit the token. See PUBLISH.md for the full checklist.

---

Quick Start

Create main.py:

from edocapi import App, Document

app = App()

@app.post("/convert")
def convert(file):
    return Document(file).to_pdf()

@app.post("/compress")
def compress(file):
    return Document(file).compress()

@app.post("/extract")
def extract(file):
    return {"text": Document(file).extract_text()}

@app.post("/merge")
def merge(files):
    return Document.merge(files)

@app.get("/")
def home():
    return {"message": f"eDocAPI v{__version__}", "supported": Document.supported_types()}

Run the development server:

edocapi run
# or with auto-reload:
edocapi run --reload

Open http://127.0.0.1:8000 in your browser to see the eDocAPI developer dashboard for your app. It shows your registered routes, supported document types, upload limit, and available Document operations, with a Try it explorer for testing endpoints. The dashboard is enabled by default when you have not registered your own GET / route. Set App(dashboard=False) to disable it. The separate eDocAPI company site lives in webapp/; see webapp/README.md to run it.

Upload a document to POST /convert and receive a PDF.

---

Core Concepts

Application

from edocapi import App

app = App(
    debug=True,
    max_file_size="25MB",
)

Document

from edocapi import Document

# From uploaded file (Path)
doc = Document(file)

# From HTML string
doc = Document.html("<h1>Invoice</h1><p>Total: ₦50,000</p>")

# From Markdown
doc = Document.markdown("# Hello World")

Conversions

Document(file).to_pdf()          # → Document (PDF)
Document(file).to_text()         # → str
Document(file).extract_text()    # → str
Document(file).to_html()         # → str
Document(file).to_images()       # → list[Document]

PDF Operations

Document(file).compress(level="medium")
Document(file).split()                    # list of single-page Documents
Document(file).extract_pages(1, 5)        # 1-based inclusive
Document(file).rotate(90, pages=[1, 2])
Document.merge([file1, file2])
Document(file).info()                     # dict of metadata

Document(file).compress(target_size=20 * 1024, level="medium") optionally targets a maximum output size in bytes. PDFs are optimized and their output size is checked; text-based PDFs can be rebuilt with simplified layout and without original graphics if needed. Image files with JPG/JPEG, PNG, or WebP extensions are recompressed and retain their image format. Image compression can reduce color detail. Some target sizes cannot be reached. Without a target, the compressor returns the original if recompression would make it larger. Other document types are not accepted by compress().

---

Supported Formats

Input	to_pdf	to_text	to_html	to_images	Notes
PDF	✓	✓	✓*	✓**	*text-based HTML
DOCX	✓	✓	✓	–
HTML	✓	✓	✓	–
Markdown	✓	✓	✓	–
JPG/PNG/WEBP	✓	–	–	✓
TXT	–	✓	–	–	via Document constructor

* PDF→HTML is a simple text extraction wrapper, not a visual rendering.
** PDF→Images requires optional pdf2image + system poppler.

---

Deploying

The package exposes an ASGI application. In your application's main.py, create app = App() and start it with an ASGI server. For example:

uvicorn main:app.asgi --host 0.0.0.0 --port ${PORT:-8000} --workers 2

Keep debug=False in hosted environments. HTML, Markdown, TXT, and DOCX PDF conversion have a Pillow text-rendering fallback; install WeasyPrint's operating-system libraries for richer layout. Set the reverse proxy's request-body limit to match max_file_size; for larger workloads, use a worker queue and monitor temporary disk usage.

Configuration

app = App(
    debug=False,           # more detailed errors when True
    max_file_size="10MB",  # "10MB", "25MB", 10485760, etc.
    temp_dir=None,         # custom temp directory (optional)
)

---

Error Handling

eDocAPI raises specific exceptions that become consistent JSON responses:

{
  "error": "UnsupportedFileType",
  "message": "The uploaded file type is not supported. (extension: .xlsx)"
}

Available exceptions: EdocAPIError, UnsupportedFileType, InvalidDocument, FileTooLarge, FileNotFound, ConversionError, ProcessingError, ValidationError.

---

CLI

edocapi --version
edocapi run                  # main:app on http://0.0.0.0:8000
edocapi run --reload
edocapi run myapp:app --port 8080

---

Architecture

eDocAPI
   │
   Document abstraction
   │
   +───────────+───────────+
   │           │           │
  PDF        DOCX        Images …
   │           │           │
 pypdf    python-docx   Pillow
              │
         weasyprint (HTML/PDF)

Processors are registered internally and can be extended in future versions.

---

Development

pip install -e ".[dev]"
pytest

---

Release status

eDocAPI is alpha software. Review conversion behavior, optional system dependencies, and workload limits before exposing it to untrusted public traffic. See PUBLISH.md for the release checklist.

Roadmap

• v0.1 – OpenAPI generation, more formats, plugin system
• v0.2 – OCR, background jobs, cloud storage adapters
• v0.3+ – Document signing, AI integrations, advanced workflows

---

Author

EMMANUEL EMMANUEL ETIM
Email: emmanuel224etim089@gmail.com
GitHub: https://github.com/emmanuelemmanueletim/edocApi

---

License

MIT — Copyright (c) 2026 EMMANUEL EMMANUEL ETIM