Re: The beauty of LATEX
Joel de Guzman <[email protected]>
| Newsgroups | gmane.comp.lib.boost.documentation |
|---|---|
| Message-ID | <[email protected]> |
On 10/26/2011 11:02 PM, Dave Abrahams wrote: > > on Wed Oct 26 2011, Joel de Guzman <joel-AT-boost-consulting.com> wrote: > >> On 10/26/2011 4:40 PM, Daniel James wrote: >>> On Tuesday, 25 October 2011, Joel de Guzman wrote: >> >>> >>> On 10/25/2011 6:01 PM, Daniel James wrote: >>> > >>> > Yes, that works. Although now when you want to pass the function as a >>> > parameter, you have to define another lambda function. Which could get >>> > very verbose - especially if the function takes a few arguments. It >>> >>> That's how it's done in lisp/scheme, it's foolish to go invent >>> something else. >>> >>> > also requires you to always know how many arguments a function takes. >>> >>> What's wrong with that? Ah, you want it like C function pointers >>> where you pass only f and be done with it? That's too limiting. >>> You can't even curry with it. >>> >>> >>> You can do that in Scheme: >>> >>> http://mitpress.mit.edu/sicp/full-text/book/book-Z-H-12.html#%_sec_1.3.1 >> >> Oh right, I forgot about that. Yes, we can have both. > > No offense, but this is starting to sound a little silly... Could > someone explain to me why we don't use a real scheme (or some other) > interpreter? Actually, we have a real scheme interpreter. Behind the scenes, I'll probably use that. The template processor is a bit different though in that it is purely a text generation engine. Also, I'm trying to get as close as possible to the current Qbk syntax and there are big differences. For example, notice that [* This Is Bold] does not mean the same thing in scheme which will try to multiply three symbols, This, Is and Bold. It's a totally different beast. Don't be misled by the similarity in syntax. What we have is a closer to an FP version of Terence Parr's StringTemplate (http://www.stringtemplate.org/). Regards, -- Joel de Guzman http://www.boostpro.com http://boost-spirit.com