On Sun, Jul 26, 2009 at 04:36:31PM +0200, demerphq wrote:
> Hmm, no i dont think this is it. The changes im thinking of were to basic.t
>
> I think it was 1487aac67a72b9f87b24113f65b4d878401bee33, which appears
> to have stomped on the changes that were made in
> 4954abf77fcbac122d63699b1b3921b4d44d4570 that split basic.t into
> basic.pl and basic_finish.pl and then two separate basic.t's one that
> would be used by EUI, and would NOT be in the core, and one used by
> EUMM and that would be in the core.
>
> It looks like, probably due to me not following through properly, that
> this change didnt go upstream to EUMM proper, and thus Nicholas
> (probably accidentally) stomped on the change when he upgraded to EUMM
> 6.49_01. Which then meant that changes that might have gone in to
> basic_finish.pl from the EUI package wouldnt end up being executed, as
> basic.pl/basic_finish.pl are no longer used.
>
> Its a pity this happened, as the whole idea of the changes in 4954abf
> were meant to make it possible to have identical tests in EUMM/EUI and
> Core, but somehow it backfired and made things worse. :-(
So what implications are there for 5.10.1?
--
Overhead, without any fuss, the stars were going out.
-- Arthur C Clarke
