On Jul 12, 2014, at 6:17 PM, Asumu Takikawa wrote: > On 2014-07-12 16:18:55 -0400, Brian Adkins wrote: >> 1) It's odd to me to specify the l argument, and then never refer to it. >> 2) The syntax of the former seems less "noisy". > > I agree that the syntax can be noisy for simple cases. > > The design rationale for having the header is to allow optional, > keyword, and rest arguments. Also it lets you avoid using an `and` > pattern to refer to the whole argument. > > Just wanted to note that in case you or anyone else was wondering why > the seemingly redundant header is there. > > Cheers, > Asumu
In hindsight, I was probably being overzealous in trying to get rid of everything I didn't need for my simple example. Being able to reference the original argument will certainly be needed at times, and a couple extra [ ] chars is really not that big of a deal. Plus the advantage of someone else being able to look at my code and immediately understand define/match vs. some ad-hoc macro I whipped up is significant. As I mentioned in another reply, I quickly ran into a limitation with my macro that I wouldn't have if I had used define/match. I'll wait until I have enough code to see what the real pain points might be. ____________________ Racket Users list: http://lists.racket-lang.org/users
