GoLicense-Classifier

---

A Python package to find license expressions and copyright statements in a codebase.

Based on Google LicenseClassifer V2, GoLicense-Classifier (or glc for short) focuses on performance without compromising with accuracy.

Installation

Note: Currently, this package only supports Linux Platform. Work is in progress for Windows and Mac.

Installing GoLicense-Classifier is as simple as

pip install golicense-classifier

Or, you can build the package from source as

git clone https://github.com/AvishrantsSh/GoLicense-Classifier.git
make dev
make package

Usage

To get started, import LicenseClassifier class from the module as

from LicenseClassifier.classifier import LicenseClassifier

Note: Work on Copyright Statement is still in beta phase. Expect some issues, mostly with binary files

The class comes bundled with some handy functions, each suited for a different task.

1. scan_directory

   This method is used to recursively walk through a directory and find license expressions and copyright statements. It returns a dictionary object with keys header and files.

   Usage

   ---

   classifier = LicenseClassifier()
   res = classifier.scan_directory('PATH_TO_DIR')

   Optional Parameters

   ---

   • max_size

     Maximum size of file in MB. Default is set to 10MB. Set max_size < 0 to ignore size constraints

   • use_buffer

     (Experimental) Set to True to use buffered file scanning. max_size will be used as buffer size.

   • use_scancode_mapping

     Set to True if you want to use Scancode license key mappings. Default is set to True.

2. scan_file

   This method is used to find license expressions and copyright statements in a single file.

   Usage

   ---

   classifier = LicenseClassifier()
   res = classifier.scan_file('PATH_TO_FILE')

   Optional Parameters

   ---

   • max_size

     Maximum size of file in MB. Default is set to 10MB. Set max_size < 0 to ignore size constraints

   • use_buffer

     (Experimental) Set to True to use buffered file scanning. max_size will be used as buffer size.

   • use_scancode_mapping

     Set to True if you want to use Scancode license key mappings. Default is set to True.

Further Customization

You can set custom threshold for scanning purpose that best suits your need. Simply change the parameter threshold during object creation as

classifier = LicenseClassifier(threshold = 0.9)

Contributing

Contributions are what makes the open source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

To get started, read the Contributing Guide.

References

1. Google LicenseClassfifer V2 https://github.com/google/licenseclassifier/
2. Ctypes Shared Library Code https://github.com/AvishrantsSh/Ctypes_LicenseClassifier
3. Ctypes https://docs.python.org/3/library/ctypes.html