Archives of the TeradataForum

Message Posted: Mon, 09 May 2011 @ 14:18:01 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

Hi,

How do you do your teradata testing, especially bteq code testing?

With testing becoming more and more important in the software developing world, more and more environments are providing default unit testing solutions. The best ones that I know of are Java and Perl. But for teradata, forgive me ignorance, I don t know any.

Thus, I d like to know what is your strategy and steps in unit and QA testing of teradata code.

Intuitively, in betq, I would insert some testing data into my testing tables, carry out my ETL job (optionally compare the results), then insert more testing data and continue the above steps until all cases are covered.

But I found it is hard to do testing data insert in teradata. All exiting loading/importing tools seems to be an overkill for such small testing jobs to me. Another alternative is to do the insert SQL code in bteq, but unlike pg_dump in psql, I didn t know any easy way to convert exiting data into insert SQL statements, with both data and corresponding fields names, which is important for testing that, in further there might be columns reordering and/or more columns added, etc.

All in all, I d like to hear your testing considerations and solutions.

BTW, I know even shell scripts have unit testing solutions now...

Thanks a lot!

FM

	<Prev	Next>		<<First	<Prev	Next>	Last>>