[freetds] The ideal tester

Lowden, James K LowdenJK at bernstein.com
Fri Apr 23 11:54:18 EDT 2004


> From: Jeff Fisher [mailto:JFisher at Emery-Waterhouse.com]
> Sent: April 23, 2004 8:47 AM
> 
> > One of the growing impediments is a lack of facilities and 
> testers.  Take for example "Oracle heterogeneous 
> 
> We have Oracle on both Solaris and Linux.  We would like to 
> get this working
> and so we would be available to help test.  What would be involved?

Hi Jeff, 

Thanks for the offer!  

The goal of the programmer is to reduce the problem to something the machine can understand.   The goal of the tester is to reduce the problem to something the programmer can understand.  This is especially true when the programmer cannot fill in for the tester because he has no access to the testing environment.  

Let me describe the ideal tester as I see it.  Perhaps others would like to add their thoughts.  

You must be willing to be a light spot between two dark ones.  You don't have access to Oracle's or Sybase's code; you can only observe their behavior through the TDSDUMP logs.  You have to like reasoning by inference.  

You should be ready to install and reinstall new versions of FreeTDS, and to triangulate problems by creating different tests to try to isolate the problems.  It would be good if your C skills aren't too rusty, because the short circuit in the problem-identification loop is adding logging statements at strategic points in the code.  

You need to know or be willing to learn the capabilities of your host.  Does it support transactions, remote procedure calls, computed rows, 8-byte integers?  Your SQL skill should be good enough to exercise the various parts of the supported functionality.  

You also need discipline and patience.  You'll want to create tests for all datatypes at the boundary of their representable values.  You'll pass the information between the servers and check the results against your expectations.  It's good you have both endians available; if there are data conversion issues, we want to flush them out.  Just as important, and more time consuming, you want to raise error conditions on the remote side and make sure they propogate back correctly.  

I think at this point the ODBC driver is pretty good at logging requests it doesn't support.  Trickier, especially when you can't read the caller's logic, is identifying things it does that are out of spec.  

>From all this, you maintain a list:

1.  Functions/options not implemented, and what features they support.
2.  Bugs, by function/option, with logs and user-level behavior. 
3.  Broken user-level behavior, cause undetermined, with associated logs.  

Your list serves as guideposts and reference points for anyone seeking to implement a missing feature or convert a #3 problem into a #1 or #2 problem.  

There are many users and few programmers.  The more you can to do make accurate black-and-white this-should-do-that statements, the more straightforward is the effort to making this do that.  

I hope this doesn't discourage you.  A group working together would be able to supply all the skills even if one person couldn't.  

Regards, 

--jkl

-----------------------------------------
The information contained in this transmission may contain privileged and confidential information and is intended only for the use of the person(s) named above. If you are not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, any review, dissemination, distribution or duplication of this communication is strictly prohibited. If you are not the intended recipient, please contact the sender immediately by reply e-mail and destroy all copies of the original message. Please note that we do not accept account orders and/or instructions by e-mail, and therefore will not be responsible for carrying out such orders and/or instructions.
If you, as the intended recipient of this message, the purpose of which is to inform and update our clients, prospects and consultants of developments relating to our services and products, would not like to receive further e-mail correspondence from the sender, please "reply" to the sender indicating your wishes.  In the U.S.: 1345 Avenue of the Americas, New York, NY 10105.





More information about the FreeTDS mailing list