Hi, Alex Shinn <alexsh...@gmail.com> writes:
> On Mon, Sep 6, 2010 at 9:12 PM, Ludovic Courtès <l...@gnu.org> wrote: >> >> Well, since there are only 9 of them, they could probably be implemented >> as special cases, with an augmented ‘match-gen-ellipses’, which would be >> told the minimum number of elements expected? > > Oh, the Wright syntax limited you to 9 forms, so "..10" is illegal? >From Wright’s 1995 paper & doc I thought k > 9 wasn’t supported but apparently it is: --8<---------------cut here---------------start------------->8--- guile> (use-modules (ice-9 match)) ;; <- Wright’s code guile> (match (iota 30) ((a ..10) a)) $1 = (0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29) guile> (match (iota 30) ((a ..22) a)) $2 = (0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29) guile> (version) $3 = "1.8.7" --8<---------------cut here---------------end--------------->8--- :-( > "..1" is actually useful - it's the analog of "+" in regular > expressions, and allows simplifying many syntax-rules > patterns you see written (elt0 elt1 ...) as (elt ..1). If > the elements are more complex patterns this is a big > win. Agreed. So perhaps one solution would be to: 1. Have a special case for ‘..1’, since it’s quite handy. 2. Have a new syntax form, like ‘.. k’, as you suggested. What do you think? Thanks, Ludo’.
