[racket-dev] [racket] tests/eli-tester feedback (Was: Racket unit testing)

From: Neil Van Dyke (neil at neilvandyke.org)
Date: Tue Feb 15 09:09:02 EST 2011

Eli Barzilay wrote at 02/15/2011 07:32 AM:
> * It uses TR -- but it has to be very minimal.  (*Extremely* minimal, since it's a candidate for inclusion in the `racket' language, which is why I was thinking of not much more than agreed structs and parameters.)

A related consideration is that, the more dependencies the test 
mechanism has, in theory, the more potential points of failure that 
could corrupt the test mechanism itself.

I doubt that TR, for example, is going to break in a way that causes 
tests that should fail to pass, but I think that the consideration 
should be mentioned.


Posted on the dev mailing list.