On 8 Jun 2007, James Youngman said: > On 6/8/07, Nix <[EMAIL PROTECTED]> wrote: >> I'd say this behaviour violates the principle of least astonishment, at >> least. Mind you, avoiding it does seem like it could be expensive: [...] > > Maybe. For the issue-diagnostic-message use case, performance is not > such an issue. But I'm sure there are valid use cases where ultimate > performance is really vital. Use-cases vary a lot.
I wonder if what we need is some sort of floating-point-number validation function that we could call if we thought the number we were about to print may be suspect. (But even then, *any* double may be rendered suspect by bad RAM, and that's a problem that will only increase as RAM densities rise.) -- `... in the sense that dragons logically follow evolution so they would be able to wield metal.' --- Kenneth Eng's colourless green ideas sleep furiously
