Hello everyone! As I stated in my last mail, as part of Google Summer of Code, '13, I'm re-writing the Test Suite for Wget in Python. I have recently pushed a commit to the Release branch [1] that has a functioning HTTP Server and a Test driver script. The basic structure of the Test Suite is as follows:
All tests are written as XML files. The XML Filenames are passed to the the driver script (runTest.py) as command line arguments. The TestEnv module contains the various internal details about how the environment is implemented the checks implemented to determine whether a Test is passed or not. The HTTPServer module contains the HTTP Server that is spawned and used for the Test. A currently undocumented feature is the use of the NO_CLEANUP environment variable which will prevent the temporary files created during the test from being deleted. I do kindly request everyone here to checkout the Release branch and try running the tests. It would be extremely kind of you'll to try and write a few new tests and give me feedback on the process. My aim is to make it as easy as possible to write new tests. Also, if there is a certain feature / check that I have missed, please do let me know and I will implement it as soon as I can. Important notice: The use of this Test Environment requires Python 3.3 and Automake >= 1.12. Are these versioning requirements fine? Does anyone have a problem with bumping up the required version of Automake to 1.12? This is required since the parallel-tests driver being used was changed between v1.11 and v1.12. If you use Automake 1.12, you may need to run automake --add-missing manually to get the required automake scripts. I once again request everyone to checkout the codebase and try and write a few more tests. Let's ensure that Wget stays as regression free as possible. [1]: https://github.com/darnir/wget-gsoc/tree/test-env-release -- Thanking You, Darshit Shah
