The TAP protocol is both pretty simple and time tested. I'd suggest to adopt it.
On Thu, Aug 20, 2015 at 12:44 AM, Alex Guo <chessnut@xxxxxxxxxxx> wrote:
well, you can't cover everything with PASS or FAIL unit tests. Also consider
performance tests. Whether LuaJIT's performance regresses on any significant
Subject: RE: LuaJIT project governance
Date: Wed, 19 Aug 2015 18:40:56 +0000
Or at least a simple agreement on "returns 0 == success, returns '1' ==
failure", or something?
Some guidelines such as:
A single unit test should be fully independent and live in a single file
A unit test is responsible for setting up its environment and tearing it
- Shaping clay is easier than digging it out of the ground.
Date: Wed, 19 Aug 2015 10:30:47 -0700
Subject: Re: LuaJIT project governance
I think a good start would be the list of things tested for if the test
code can not be shared. Just knowing the questions to ask and possibly
the answers can go a long way.
On Wed, Aug 19, 2015 at 1:54 AM, Lesley De Cruz
On 19 August 2015 at 00:31, Geoff Leyland
"vegorov@xxxxxxxxxx<mailto:vegorov@xxxxxxxxxx>" for DMARC)On 18/08/2015, at 9:02 pm, Vyacheslav Egorov (Redacted sender
On Tue, Aug 18, 2015 at 9:59 AM John Graham-Cumming
1. The test suite
I am happy to help transitioning the tests.
Likewise. I’d like to help where I can.
Same goes for me. How should we proceed, given that Mike's test suite
cannot be made public as-is due to license issues? Start collecting