> This breaks the API, which is not good at all. Does anybody use the > vector method the old API provided (i.e. supplying a vector as t1 and > get the output over the whole vector?). What do you propose? Leave the > old API? I could rename the eval to e.g. eval_single.
True, the API breaks-- but the API was both confusing (for my part I thought it was a bug I'd found a workaround for) and undocumented (the example in the docs didn't work, which is always discouraging when you're trying to sort out what's going on). My vote is that the proposed API change is one to the "correct" behaviour (i.e. most like raw GSL), and qualifies as a two-for-one bugfix. Unless someone's very, very unlucky, it's also likely to break code immediately and be noisy about it. Doug -- Queen Mary College, University of London "Still creating worlds.. Mathematical Sciences, Astronomy Unit .. but now with an accent!" ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ pygsl-discuss mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/pygsl-discuss
