Forum Comments

Describe all the phases of SDLC with a real project. Please try to explain each phases with 3-4 sentences.
In Thread Discussions
Data-driven framework ideas
In Thread Discussions
cs1
May 05, 2019
We are using csv files for our homegrown TC framework. Test maintainence is really just annoying in general. For acceptance level testing, there's just so many scenarios and cases you can test. I think the key to find ways of modularizing your tests. That will reduce the amount of data you have to maintain. For example.. let's compare the following scenario. MyTest Step 1 (3 possible inputs) Step 2 (1 possible input) Step 3 (2 possible input) Test #1 Data Data11, Data12, Data13 Data21, Data22, Data23 Data31, Data32, Data33 Data41, Data42, Data43 Data51, Data52, Data53 Data61, Data62, Data63 In this scenario we have you large tests, but the data is all in one place and is harder to maintain. If you have your tests in this sort of scenario, then it's better to use a database as you can do queries and SQL updates to maintain you data. But notice since the data is stored in a flat database the amount of data goes up expoentially with each input combination. But I prefer the next example.... MyTest Step 1 (3 possible inputs) Step 2 (1 possible input) Step 3 (2 possible inputs) Step1Data Data1 Data2 Data3 Step2Data Data1 Step3Data Data1 Data2 Here your data is seperated by where it's applied. It's easier to update and manage in spreadsheets. If one step goes away, you can simple delete the entire sheet. Or if you add a step, add a new sheet. Here you have the benefit of keeping smaller sets of data. You have more sheets to maintain, but you don't need to store each possible dataset for each test combination.
1
0

cs1

More actions

IT Consulting, Training and Job Placement services