TAS: A Data-driven Test Framework 一个数据驱动测试的自动化框架

Data-driven test has two major parts, one is a number of sets of data, and the other,  a test logic. One set of data here is called a test case, and all the test cases form. a test suite.

Data-Driven test are considered one kind of the most suitable test to be automated. This is because Data-driven test have two special characteristics, which make it able to achieve a high mark in terms of input-output ratio, relative to none-data-driven test.

The first one is that all the test cases can be automated once the test logic has been implemented and at least one test case is able to run with it, because one test logic serves for all test cases.

The second one is, the execution of one test case is absolutely independent to another. That is to say, theoretically, if running a test case multiple times, the test result of a test case is always the same no matter in what a order it is executed..

As nearly all datat-driven testcaes shares those characteristics, to implement a framework to deal with the common things is meaningful. This is exactly why TAS(TestAutomationSystem) is invented. 

====================================================
数据驱动测试有两个主要部分构成，一个是一套数据集，另外就是测试逻辑。一套数据在这里叫做一个testcase，而所有的数据一起构成一个testsuite。

数据驱动测试被认为是最适合自动化的测试之一，这是因为数据驱动测试有两个重要特性，使得它能具有较高的 投入产出比。

第一个特点是，一旦成功自动化一个testcae，那么所有的testcae都能被自动化起来。因为所有的testcae都使用同一个测试测试逻辑。

第二个特点是，每一个testcase的执行是和另外的testcae完全独立的。理论上，在多次运行同一个testcaes，无论它是以什么样的顺序运行，结果都应该是一样的。

因为几乎所有数据驱动的测试都具有这样的性质，实现一个自动化框架来处理这些公共部分是有意义的。这也就是TAS产生的原因。

========================================
tas 1.1 released.

1. simplify create_everything.py
2. more robust in executing testtask, adding exception handling
3. other improvements