Add Ghostscript-based PDF compressor and update docs (fixes #129)

- Add pdf_compressor_ghostscript.py using open-source Ghostscript - Update README.md with both legacy and recommended methods - Update requirements.txt to note system dependencies - Fixes issue #129: PDFTron/PDFNet is now commercial and requires license - Provides free alternative with same functionality and API

Author: x4nth055

handling-pdf-files/pdf-compressor/README.md

@@ -1,8 +1,48 @@
 # [How to Compress PDF Files in Python](https://www.thepythoncode.com/article/compress-pdf-files-in-python)
-To run this:
-- `pip3 install -r requirements.txt`
-- To compress `bert-paper.pdf` file:
- ```
- $ python pdf_compressor.py bert-paper.pdf bert-paper-min.pdf
- ```
- This will spawn a new compressed PDF file under the name `bert-paper-min.pdf`.
+
+This directory contains two approaches:
+
+- Legacy (commercial): `pdf_compressor.py` uses PDFTron/PDFNet. PDFNet now requires a license key and the old pip package is not freely available, so this may not work without a license.
+- Recommended (open source): `pdf_compressor_ghostscript.py` uses Ghostscript to compress PDFs.
+
+## Ghostscript method (recommended)
+
+Prerequisite: Install Ghostscript
+
+- macOS (Homebrew):
+ - `brew install ghostscript`
+- Ubuntu/Debian:
+ - `sudo apt-get update && sudo apt-get install -y ghostscript`
+- Windows:
+ - Download and install from https://ghostscript.com/releases/
+ - Ensure `gswin64c.exe` (or `gswin32c.exe`) is in your PATH.
+
+No Python packages are required for this method, only Ghostscript.
+
+### Usage
+
+To compress `bert-paper.pdf` into `bert-paper-min.pdf` with default quality (`power=2`):
+
+```
+python pdf_compressor_ghostscript.py bert-paper.pdf bert-paper-min.pdf
+```
+
+Optional quality level `[power]` controls compression/quality tradeoff (maps to Ghostscript `-dPDFSETTINGS`):
+
+- 0 = `/screen` (smallest, lowest quality)
+- 1 = `/ebook` (good quality)
+- 2 = `/printer` (high quality) [default]
+- 3 = `/prepress` (very high quality)
+- 4 = `/default` (Ghostscript default)
+
+Example:
+
+```
+python pdf_compressor_ghostscript.py bert-paper.pdf bert-paper-min.pdf 1
+```
+
+In testing, `bert-paper.pdf` (~757 KB) compressed to ~407 KB with `power=1`.
+
+## Legacy PDFNet method (requires license)
+
+If you have a valid license and the PDFNet SDK installed, you can use the original `pdf_compressor.py` script. Note that the previously referenced `PDFNetPython3` pip package is not freely available and may not install via pip. Refer to the vendor's documentation for installation and licensing.

handling-pdf-files/pdf-compressor/pdf_compressor_ghostscript.py

@@ -0,0 +1,103 @@
+import os
+import sys
+import subprocess
+import shutil
+
+
+def get_size_format(b, factor=1024, suffix="B"):
+ for unit in ["", "K", "M", "G", "T", "P", "E", "Z"]:
+ if b < factor:
+ return f"{b:.2f}{unit}{suffix}"
+ b /= factor
+ return f"{b:.2f}Y{suffix}"
+
+
+def find_ghostscript_executable():
+ candidates = [
+ shutil.which('gs'),
+ shutil.which('gswin64c'),
+ shutil.which('gswin32c'),
+ ]
+ for c in candidates:
+ if c:
+ return c
+ return None
+
+
+def compress_file(input_file: str, output_file: str, power: int = 2):
+ """Compress PDF using Ghostscript.
+
+ power:
+ 0 -> /screen (lowest quality, highest compression)
+ 1 -> /ebook (good quality)
+ 2 -> /printer (high quality) [default]
+ 3 -> /prepress (very high quality)
+ 4 -> /default (Ghostscript default)
+ """
+ if not os.path.exists(input_file):
+ raise FileNotFoundError(f"Input file not found: {input_file}")
+ if not output_file:
+ output_file = input_file
+
+ initial_size = os.path.getsize(input_file)
+
+ gs = find_ghostscript_executable()
+ if not gs:
+ raise RuntimeError(
+ "Ghostscript not found. Install it and ensure 'gs' (Linux/macOS) "
+ "or 'gswin64c'/'gswin32c' (Windows) is in PATH."
+ )
+
+ settings_map = {
+ 0: '/screen',
+ 1: '/ebook',
+ 2: '/printer',
+ 3: '/prepress',
+ 4: '/default',
+ }
+ pdfsettings = settings_map.get(power, '/printer')
+
+ cmd = [
+ gs,
+ '-sDEVICE=pdfwrite',
+ '-dCompatibilityLevel=1.4',
+ f'-dPDFSETTINGS={pdfsettings}',
+ '-dNOPAUSE',
+ '-dBATCH',
+ '-dQUIET',
+ f'-sOutputFile={output_file}',
+ input_file,
+ ]
+
+ try:
+ subprocess.run(cmd, check=True)
+ except subprocess.CalledProcessError as e:
+ print(f"Ghostscript failed: {e}")
+ return False
+
+ compressed_size = os.path.getsize(output_file)
+ ratio = 1 - (compressed_size / initial_size)
+ summary = {
+ "Input File": input_file,
+ "Initial Size": get_size_format(initial_size),
+ "Output File": output_file,
+ "Compressed Size": get_size_format(compressed_size),
+ "Compression Ratio": f"{ratio:.3%}",
+ }
+
+ print("## Summary ########################################################")
+ for k, v in summary.items():
+ print(f"{k}: {v}")
+ print("###################################################################")
+ return True
+
+
+if __name__ == '__main__':
+ if len(sys.argv) < 3:
+ print("Usage: python pdf_compressor_ghostscript.py <input.pdf> <output.pdf> [power 0-4]")
+ sys.exit(1)
+ input_file = sys.argv[1]
+ output_file = sys.argv[2]
+ power = int(sys.argv[3]) if len(sys.argv) > 3 else 2
+ ok = compress_file(input_file, output_file, power)
+ sys.exit(0 if ok else 2)

handling-pdf-files/pdf-compressor/requirements.txt

@@ -1 +1,7 @@
-PDFNetPython3==8.1.0
+# No Python dependencies required for Ghostscript-based compressor.
+# System dependency: Ghostscript
+# - macOS: brew install ghostscript
+# - Debian: sudo apt-get install -y ghostscript
+# - Windows: https://ghostscript.com/releases/
+#
+# The legacy script (pdf_compressor.py) depends on PDFNet (commercial) and a license key.