>>>>> On Sun, 18 Feb 2001 16:08:49 -0500, [EMAIL PROTECTED] (Michael G. Schwern) said:
> On Sun, Feb 18, 2001 at 07:20:55PM +0000, [EMAIL PROTECTED] wrote:
>> <[EMAIL PROTECTED]> writes:
>> >Why is t/TEST anything more than a thin wrapper around Test::Harness?
>>
>> Because t/TEST pre-dates Test::Harness by years, and no one has got
>> round to changing 'make test' call the new style.
> Is that it? Historical reasons?
No. For me the reason was this reminder in TEST:
# This is written in a peculiar style, since we're trying to avoid
# most of the constructs we'll be testing for.
> Would anyone scream if I gave a shot at gutting TEST?
No, but I do believe, we should keep something like TEST for at least
base/, cmd, and op/ as a fallback. When some construct within
Test::Harness breaks, the poor pumking should get a quicker
alternative than running single tests manually.
--
andreas
