Python + Syncfusion Document SDK Samples

This repository contains two complete, runnable samples demonstrating how to generate Word (.docx) documents and convert them to PDF using the Syncfusion .NET Document SDK (DocIO and DocIORenderer) from Python 3.9+.

Microsoft Office is not required. All document generation and PDF rendering run locally.

Two Approaches

Feature	Approach 1: Local .NET Worker	Approach 2: Python.NET Wrapper
Location	local-dotnet-worker/	pythonnet-wrapper/
How it connects	Separate process spawned via standard subprocess	In-process via Python.NET (pythonnet CoreCLR host)
Python requirements	Python 3.9+, zero pip packages	Python 3.12 (or 3.9+ with compatible CPython), pythonnet>=3.1.0
.NET requirements	.NET 8 SDK (build), .NET 8 Runtime	.NET 8 SDK (build), .NET 8 Runtime
IPC mechanism	JSON over standard input (stdin)	Direct managed memory function calls
Best for	Strict Python environments, batch jobs, CLI tools, maximum isolation	High-frequency calls, latency-sensitive workflows

---

Approach Documentation

The two approaches are independent self-contained samples. Pick one and read its README.md; the table below links to each:

Approach	Quickstart README	Python client
Local .NET Worker	local-dotnet-worker/README.md	local-dotnet-worker/document_sdk.py
Python.NET Wrapper	pythonnet-wrapper/README.md	pythonnet-wrapper/document_sdk.py

Both clients expose the same public surface (load_service() returning a DocumentService with create_docx, create_pdf, excel_to_pdf, powerpoint_to_pdf, watermark_pdf, and two sample-input helpers), so the same caller code works with either approach. For the C# side, see the DocumentBridge/ folder inside each approach.

---

Samples

Every approach ships with a small samples/ folder of runnable scripts that build on document_sdk.load_service():

Sample	Local Worker	Python.NET
01 — Hello World (DOCX + PDF)	link	link
02 — Batch generation loop	link	link
03 — Custom output paths & bundle	link	link
04 — Excel → PDF	link	link
05 — PowerPoint → PDF	link	link
06 — PDF watermark	link	link
Per-approach index	local-dotnet-worker/samples/README.md	pythonnet-wrapper/samples/README.md

Run a sample (example, local worker):

cd local-dotnet-worker
python -m samples.01_hello_world

Samples are run as modules (python -m samples.XX) so the parent folder containing document_sdk.py is on sys.path. Running them as plain scripts (python samples/XX.py) fails with ModuleNotFoundError: No module named 'document_sdk'.

---

Quick Tour

Approach 1: Local .NET Worker (local-dotnet-worker/)

cd local-dotnet-worker

# 1. Publish the .NET worker (Windows x64 example)
dotnet publish DocumentBridge/DocumentBridge.csproj -c Release -r win-x64 --self-contained false -o artifacts

# 2. Run with plain Python (creates output/Sample.docx)
python document_sdk.py create-docx

# 3. Render directly to PDF without an intermediate DOCX
python document_sdk.py create-pdf

# 4. Convert an existing XLSX/PPTX file to PDF, or watermark a PDF
python document_sdk.py excel-to-pdf      --input documents/input.xlsx --output output/workbook.pdf
python document_sdk.py powerpoint-to-pdf --input documents/input.pptx --output output/slides.pdf
python document_sdk.py watermark-pdf     --input output/workbook.pdf  --output output/watermarked.pdf --label CONFIDENTIAL

Approach 2: Python.NET Wrapper (pythonnet-wrapper/)

cd pythonnet-wrapper

# 1. Publish the .NET class library
dotnet publish DocumentBridge/DocumentBridge.csproj -c Release -r win-x64 --self-contained false -o artifacts

# 2. Install pythonnet in a virtual environment
python -m venv .venv
.venv\Scripts\activate      # On Linux/macOS: source .venv/bin/activate
pip install -r requirements.txt

# 3. Run in-process
python document_sdk.py create-docx
python document_sdk.py create-pdf

# 4. Convert an existing XLSX/PPTX file to PDF, or watermark a PDF
python document_sdk.py excel-to-pdf      --input documents/input.xlsx --output output/workbook.pdf
python document_sdk.py powerpoint-to-pdf --input documents/input.pptx --output output/slides.pdf
python document_sdk.py watermark-pdf     --input output/workbook.pdf  --output output/watermarked.pdf --label CONFIDENTIAL

---

Licensing & Evaluation

Both samples run in Syncfusion's trial mode by default — no license flag is required, and no Python method takes one. Evaluation watermarks are added to the output documents in this mode.

For licensed production use, set the SYNCFUSION_LICENSE_KEY environment variable before running the command. The .NET code reads the variable via Environment.GetEnvironmentVariable("SYNCFUSION_LICENSE_KEY") and registers it via SyncfusionLicenseProvider.RegisterLicense(key). Keys are never passed on the command line.

# Windows PowerShell
$env:SYNCFUSION_LICENSE_KEY="YOUR_KEY_HERE"

# Linux / macOS
export SYNCFUSION_LICENSE_KEY="YOUR_KEY_HERE"

# Then run any sample as usual — no extra flag is needed.
python document_sdk.py create-pdf

---

Tested Configurations

• .NET SDK: .NET 8.0 LTS
• Syncfusion Packages: Syncfusion.DocIORenderer.Net.Core, Syncfusion.XlsIORenderer.Net.Core, Syncfusion.PresentationRenderer.Net.Core, Syncfusion.Pdf.Net.Core (all 34.2.5)
• Python: 3.9.6, 3.12.x
• Platforms: Windows 10/11 x64 (win-x64), macOS Apple Silicon (osx-arm64), Linux Ubuntu x64 (linux-x64)

For the C# implementation, see local-dotnet-worker/DocumentBridge/ and pythonnet-wrapper/DocumentBridge/.