On 3/19/07, Chris Hostetter <[EMAIL PROTECTED]> wrote:
Thanks for comitting Mike, sorry i never got around to reviewing the patch. Just out of curiousity though, can anyone explain why ConvertedLegacyTest was functioning properly before this change? I'm just wondering if we're breaking a case people might have been relying on because it functioned (even if it wasn't intended to).
That's a good point--the change didn't affect the the point of the test (which was sorting-related), but there was likely some reason that the original test used id_i. Ryan, what problem did you encounter that prompted the change? cheers, -Mike