Re: [q-lang-users] Matching with special constructors
Brought to you by:
agraef
From: John C. <co...@cc...> - 2006-05-12 16:07:04
|
Albert Graef scripsit: > Specifically, the pattern matcher, when trying to match > a pattern inside a suspended argument, should evaluate that argument > recursively before proceeding. AFAICT, that's also the way that this > situation would be handled in other languages which provide pattern > matching on lazy data structures. Indeed, in Haskell Core (the output of the front end of the GHC compiler) pattern matching is the *only* way in which evaluation happens. > Of course this changes the way that special forms work, so it will not > be backward-compatible. I still think that I should do this change. What > do you think? Any other solution approaches? Any other opinions on what > would be "the right thing" here? I wonder whether it would be useful to have a way of declaring precisely which special constructors (or special forms in general?) are affected by this. Maybe not. -- What has four pairs of pants, lives John Cowan in Philadelphia, and it never rains http://www.ccil.org/~cowan but it pours? co...@cc... --Rufus T. Firefly |