Python Extension Module (bindings) for CCExtractor

Developed under Google Summer of Code, 2017 with CCExtractor Development By Diptanshu Jamgade

Introduction

CCExtractor is a software written in C language. The main aim of this project was to generate the python extension module for CCExtractor which would then help Python developers to extend the applications of CCExtractor in Python.

The main ideology of the module was to extract the captions using CCExtractor and parse the extracted captions to Python for processing them. The module was desired to be structured in such a way that the captions arrive in Python as soon as they are extracted by CCExtractor. Thus, the captions could be processed in real-time via Python rather than waiting for the extraction to finish. Also, the extension module is expected to be silent, i.e. not outputting any STDOUT information.

The module was needed to be silent in terms of all the output generated by CCExtractor. The output files were to be generated via Python and not via CCExtractor as the module's prime duty was to extract the captions in CCExtractor and parse the extracted captions to Python for processing.

Installation

For installing the Python extension module, the user needs to make sure that the user has compiled/installed all the dependencies for CCExtractor (mentioned here).

After the user has compiled/installed all the dependencies for CCExtractor, for installing Python extension module, the user needs to install two dependencies:

  • SWIG swig
  • Python-dev package ubuntu For other operating systems, the user has to install //python-dev package// as mentioned in the corresponding system's documentation.

After the user has compiled/installed the additional dependencies for Python Extension module, the user can install the extension module by

 sudo pip install ccextractor
             OR
 pip install ccextractor --user

(the above method has been successfully tested on unbuntu16.04 - Xenial)

In case of virtual environments, the user can install the Python Extension module by

pip install ccextractor

After the user has successfully installed the extension module, the user can test the module as per the test script which is used for processing a single sample. However, if the user desires to process all the samples in a particular directory then the user can use this test script.

NOTE: In case the installation fails, make sure you have updated setuptools, pip and also installed/compiled all the dependencies mentioned in this section.

Technical Documentation

Architectural Documentation

The overall documentation for the extension module as to how has the module architecture designed and how the further contributors can continue development has been done in this documentation.

Dependency Documentation

For the contributors who want to either just install the dependencies for the extension module documentation would help them understand the dependency compilation.

Contributor Documentation

For contributors who wish to understand how exactly is the CCExtractor Extension module compiled and installed on a system, this documentaion contains relevant information.

Contributions [PRs/Commits]

All the commits made by me to the CCExtractor repository could be found here.

All the pull requests by me to the CCExtractor repository could be found here.

Bugs Tracked

  • I have worked on solving issue #705 and submitted a PR(merged), thus solving the issue.
  • I have also worked on issue #304. I analyzed the issue's sample and found some errors in the way the sample information was encoded. A detailed discussion is present at our Slack under channel bug304.

GSOC 2017 Experience

The overall GSOC experience was really wonderful and the entire team here at CCExtractor was really helpful. The project was really challenging and helped me learn a great deal of basic as well as advanced concepts for software development.

The project included additions in CCExtractor source code as well as additions in Python extension module part to continue the processing of captions via Python. Thus, the project was really helpful in exploring concepts of C as well as Python. In addition to these, I also learnt a number of technologies such as SWIG, Python C-API, GCC- compiler basics as well as advanced topics, GDB, etc.

Moreover, this project helped me get a much deeper knowledge of Open Source Software Development.

About me

I am final year student studying at IIT Kharagpur, India. I would definitely be an active contributor not only for CCExtractor extension module but also the Sample-Platform as well as CCExtractor.

Email address : diptanshuj@gmail.com

Blog : https://diptanshujamgade.wordpress.com

GitHub profile: https://github.com/Diptanshu8

Slack : skrill

Thank you.