[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Tests, Documentation
From: |
Tom Tromey |
Subject: |
Re: Tests, Documentation |
Date: |
25 Nov 2003 12:19:31 -0700 |
User-agent: |
Gnus/5.09 (Gnus v5.9.0) Emacs/21.3.50 |
>>>>> "graydon" == graydon hoare <address@hidden> writes:
This discussion should probably take place on mauve-discuss...
graydon> - mauve should be arranged such that:
graydon> [ ... great stuff ... ]
I think any improvements to Mauve are to be welcomed, and anybody
willing to put effort into improving the harness should go for it.
graydon> - it uses simple introspection conventions rather than scanning
graydon> text in a source tree and writing makefiles for itself
When I wrote the original mauve code, I set it up this way because
different free java implementations had different class coverage. So
if someone added a new test, there was no guarantee it could be
compiled against a given class library. Hence, source-level scanning
to eliminate tests you know you aren't interested in.
This need may be alleviated these days, now that everybody is moving
to Classpath. Things were a lot different 5 years ago.
graydon> these are not too much to ask; existing JUnit-based suites often
graydon> provide some or all these as extensions to the basic TestCase class,
graydon> and they make testing feel much more productive.
In the past people have also suggested converting Mauve to JUnit.
Tom
- Re: Tests, Documentation, (continued)
- Re: Tests, Documentation, Tom Tromey, 2003/11/23
- Re: Tests, Documentation, Brian Jones, 2003/11/24
- Re: Tests, Documentation, Mark Wielaard, 2003/11/25
Re: Tests, Documentation, Mark Wielaard, 2003/11/23
Re: Tests, Documentation, graydon hoare, 2003/11/25
Re: Tests, Documentation,
Tom Tromey <=