Puzzle. Exists a not any-type! datatype?
[1/3] from: rotenca:telvia:it at: 23-Apr-2003 16:49
The answer to this puzzle is "yes":
This is the proof, like a puzzle.
>From a console session (View 1.2.8):
object? o; == true
first o ;== [self a]
type? o/a ;== block!
length? o/a ;== 3 -- hu! it should be 2!
pick o/a 3 ;== -- mhhhh, unset?
unset? pick o/a 3; == false -- no? strange!
set/any 'a pick o/a 3
;** Script Error: set expected value argument of type: any-type
;** Near: set/any 'a pick o/a 3
; aaaarghhhhhhhhhhhhhhhhhh!!!!!!
length? append o/a 1 ;== 4 -- ok
pick o/a 4 ;== 1 -- ok again
set/any 'a pick o/a 3
;** Script Error: set expected value argument of type: any-type
;** Near: set/any 'a pick o/a 3
; aaaarghhhhhhhhhhhhhhhhhh!!!!!!
How to build the block o/a?
I think it is a Rebol bug, more than a feature.
---
Ciao
Romano
[2/3] from: lmecir:mbox:vol:cz at: 29-Apr-2003 9:36
Hi Romano,
I haven't been on line for about a week, but it may be interesting to know which version
of Rebol you used.
I would like to see the solution of the puzzle, although this looks like yet another
interpreter "leak", letting an internal value to be visible to the user, i.e. a bug.
Ciao
-Ladislav
[3/3] from: rotenca:telvia:it at: 29-Apr-2003 14:57
Hi Ladislav,
>I would like to see the solution of the puzzle, although this looks like yet
another interpreter "leak", letting an internal >value to be visible to the
user, i.e. a bug.
a bug for me (feedback sent)
o: context [a: 1 a: first self]
or
o: context [a: a: first self]
with every version i try (Core, View, standard and beta)
---
Ciao
Romano