[TIP] Testify

[Rhett Garber]

On Fri, Oct 29, 2010 at 5:30 PM, C. Titus Brown <ctb at msu.edu> wrote:

> So we have a bunch of nearly identical test frameworks, all with
> slight syntactic differences, all serving nearly the same purpose within
> slightly different constraints, with various levels of internal mungery.
> People keep on telling me that their framework or approach is better,
> or somehow rawks, but since they mostly decline to document this
> simply enough for me to understand, or seem to regard arbitrary design
> decisions and limitations as an "improvement", I remain unconvinced.
> One more is hardly a catastrophe, I guess.

I'm going to argue just because I enjoy it, but keep in mind I'm not actually
trying to convert people to a new Testify-based religion. The fact
that it solves
the needs of Yelp's development team is sufficient reason for it's existence.

There are also lots of programming languages with slight syntactic differences,
all serving nearly the same purpose with slightly different
constraints. Lots of
python web frameworks, etc. etc. Most live in obscurity, and I think that's ok.
But the fact of the matter is, something like a testing library is
largely a matter of
taste. However, if you're developing a library for the core python
library, it's probably
more importantly a matter of constraints.

We (and I really can't take anything close to full credit for it) developed
Testify because unittest (the original) didn't fit our tastes AND
didn't provide some of the functionality we felt we needed to manage
our tests well.
Thankfully, we are only developing tools like this to serve our own
needs, so it
was really not too difficult to completely scrap the underlying
fundamentals of unittest
and start from scratch.