How to write test cases for the below scenario???
I had argument with my friend for the below given scenario and argument was for what is good practise and number of test cases.
Though there was no conclusion, I am trying to get help from the forum.
Below Scenario is repeated for 15 different of data files:
1. Download the data file (Every type has a file to be downloaded)
2. Trigger a job which will populate the data in DB (Job is different for each of teh type)
3. Validate the data against the downloaded data for data correctness. (SQL query used for validation differs slighly for each of the 15 type)
Argument was where to write 15 test cases for 15 types so that there would be clarity
Or to write 1 test case and pass 15 test data.
I personally supported the first, 15 test cases so that I can pin point which type failed.
Onthe other hand there is repeatativeness.
Whats the best approach in the above mentioned scenario. Please help
I am making some assumptions for this scenario
The data in all 15 file types will be same and the 15 file types just signify the different formats in which the file can be downloaded e.g. .xls, .pdf, .doc, .csv etc
If this assumption is correct, then it would be better to write a single test case with 15 different sets of data as the backend program to select data and download file will remain same. Another program will be rendering the file format for different types of files and we need to just check the format for all the different types.
Thanks for reply.. Let me clarify the assumptions, Data might/will change from file to file and file type do not signify different formats..
I meant I have 1.csv, 2.csv,3.csv .... 15.csv (15 different files)
also there is no relationship between these files.
DB compare is different for each of the file
Last edited by shailesh07us; 09-23-2013 at 02:44 AM.
In this case, it is best to have 15 different test cases....
Thanks for reply...
I had debate with my friend on the number of manual test cases. I personally preferred 15 manual test cases as it simplifies things and easy for me to log defect against type of file. and also I can certify certain file type as passed or failed,Which I cant do when I have only 1 manual test case. More over I feel that Management will get clear picture of the things/status of the project.
Whats your thought on this????
Your point is valid and also as these 15 files will have different data and DB compare will also be different for these, so we need to have different test cases for each type of file.
there is a solution i found. try a trail demo given on testgleam. Navigate to Test Toolkit-> Test data generator