[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Comment on SRFI-110 and Comparison to Genyris xyzzy

This page is part of the web mail archives of SRFI 110 from before July 7th, 2015. The new archives for SRFI 110 contain all messages, not just those from before July 7th, 2015.

Bill Birch <birchb1024@xxxxxxxxx> wrote:
> Re: http://srfi.schemers.org/srfi-110/srfi-110.html#genyris

Thanks so much for posting!

> I have made some syntactic decisions which I regard as restricting
> programmers to write in a better style. For example not allowing lists to
> wrap encourages more smaller functions.  That's OK for source code, however
> when loading data files one should not restrict the structure. One
> difficulty with a syntax that defaults to lists is that something special
> needs to be done for atoms.
> Example (a b c (d e f) xx) is problematic since xx is subordinate but is
> not a list. So in Genyris I was forced to add a leading 'continuation'
> character ~. Which gives me:
> > quote
> :   a b c
> :      d e f
> :      ~xx
> :
> (a b c (d e f) xx) # PairSource
> There is another (obvious) way to wrap lists in Genyris just place
> a = at the end of the line.  example:
> > list 1 2 3 4 5 =
> :    6 7 8
> :
> (1 2 3 4 5 6 7 8) # Pair
> In practice I don't often use line continuation in code.

Great points.   I plan to cut-and-paste the above text
(with credit) into the SRFI-110.  Although I've made different decisions
in sweet-expressions, I think it's helpful to see the contrast.

> One thing to remember is that indentation may force you to re-arrange
> functions to suit the constraints of indentation. For example I had a
> function 'tag' which took two arguments, an expression and a class. e.g.
>    tag (+ 2 3) Inches
> But when the expression begins to get complex it made more sense to
> re-order the function to give tag <class> <expression> simply so the
> expression could grow on the subordinate indents. example:
> So if you use indented syntax expect the language itself to change!

Fair point.  If it's just a matter of reordering arguments for
convenience, macros should be able to handle that nicely, but I
think that's a plausible general principle.  Lisp syntax has (in my opinion)
ossified.  Indentation adds a new dimension to syntax, and thus
creates new possibilities for ease-of-reading.

> Cheers,

Thanks!  And thanks very much for your comments.

--- David A. Wheeler