[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

updated testing.scm implementation

This page is part of the web mail archives of SRFI 64 from before July 7th, 2015. The new archives for SRFI 64 contain all messages, not just those from before July 7th, 2015.

I've made some changes at:
I'm still working on this (so no urgency updating
the version on schemers.org), but anyone who wants to
look at the code should probably use this version.

Among the major changes:
- Support for mzscheme (requires an an implementation of
cond-expand, at http://per.bothner.com/mztesting.ss).
- A reasonable implementation of test-error.  (The test
is now skipped on "default" implementations that don't
handle an error-catching mechanism defined.)
- A new test-read-eval-string function that is useful for
testing syntax, including erroneous syntax.

The main thing that still needs nailing down is "test specifiers"
and how they should be used.  E.g. should we specify a mechanism
for running just a specified set of tests with a specified
or default test-runner?  The use case is that you're working on
a specific test case (that is presumably failing), and you want
to run just that test, not the entire test-suite.  What is the
best way to handle that case?  Is it acceptable to just
comment out the other tests?

There are also some aspects in the implementation I'm not very
happy about, including some code duplication.
	--Per Bothner
per@xxxxxxxxxxx   http://per.bothner.com/