Bronek Kozicki <brok@rubikon.pl> writes:
I still feel there is urgent need to simplify regression tests in order to make future releases simpler.
I'm not sure that testing fewer compilers simplifies regression tests in any important way. It might make the turnaround time at release shorter, but then, that could be accomplished other ways (e.g. by dedicating more machines to testing)
I'd also like to give choice to developers if they want to support old compilers (like MSVC6, GCC 2.95 or BCC32) or not.
They already have that choice.
Here's another idea: two sets of regresion tests.
More complicated already. <snip>... I hate to be a wet blanket, but making policy changes that require new docs, procedures, etc., had better yield simplifications that balance out the added complexity. -- Dave Abrahams Boost Consulting www.boost-consulting.com