suzanne.soy/racket
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update redex enumeration docs

Browse Source
This commit is contained in:
Max New 2014-02-10 06:21:57 -06:00
parent 1b1dc9cfb8
commit 09f05678a5
2 changed files with 16 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
pkgs/redex-pkgs/redex-doc/redex/scribblings/ref.scrbl
View File

@ -1869,9 +1869,9 @@ Generates terms in a number of different ways:
(for/list ([i (in-range 20)])
(generate-term L any #:i-th i))]
In addition, all other pattern types are supported except for mismatch @racket[_!_]
patterns, mismatch repeat @racket[..._!_] patterns and @racket[side-condition]
patterns.
In addition, all other pattern types are supported
except for mismatch repeat @racket[..._!_] patterns
and @racket[side-condition] patterns.
The enumerators do not repeat terms unless the given pattern is ambiguous.
Roughly speaking, the enumerator generates all possible ways that a pattern
@ -1898,7 +1898,17 @@ Generates terms in a number of different ways:
(overlap ::= natural integer))
(for/list ([i (in-range 10)])
(generate-term L overlap #:i-th i))]
For similar reasons, enumerations for mismatch
patterns @racket[_!_] are unsound in the presence of
ambiguity, but work as expected for unambiguous
grammars.
@examples[#:eval
redex-eval
(define-language Bad
(ambig ::= (x ... x ...)))
(generate-term Bad (ambig_!_1 ambig_!_1 #:i-th 4))]
}
@item{@racket[from-judgment-form]: Randomly picks a term that satisfies

4
pkgs/redex-pkgs/redex-test/redex/tests/enum-test.rkt
View File

@ -105,13 +105,13 @@
(x number)
;; Example of poorly behaved mismatch
(ambig (x ... x ...)))
(ambig (y ... y ...)))
(try-it 100 M m)
(try-it 100 M n)
(try-it 100 M p)
;; Ambiguity kills us here
;; (try-it 5 M (ambig_!_1 ambig_!_1))
;; (try-it 2 M (ambig_!_1 ambig_!_1))
;; test variable filtering
(define-language Vars