Text only | Text with Images

newLISP Fan Club

Forum => newLISP newS => Topic started by: cormullion on November 28, 2008, 10:18:54 AM

Title: removing items from hash table contexts...
Post by: cormullion on November 28, 2008, 10:18:54 AM
You can use contexts as hash tables and see tham as assoc lists (manual: "The contents of the namespace can be shown as an association list:") But it appears you can't use all assoc-list techniques on them:


Code Select
newLISP v.9.9.95 on OSX IPv4 UTF-8, execute 'newlisp -h' for more info.

> (define Doyle:Doyle)
nil
> (Doyle "baker" 221)
221
> (Doyle "street" 26)
26
>
> (Doyle)
(("baker" 221) ("street" 26))
>
> (pop-assoc "baker" (Doyle))
("baker" 221)
> (Doyle)
(("baker" 221) ("street" 26))


although you can use others:


Code Select
> (find-all '(? 26) (Doyle) (println $0))
("street" 26)

What's the difference?
Title:
Post by: Lutz on November 28, 2008, 10:51:25 AM
The list returned from (Doyle) is generated from  the namespace Doyle on the fly every-time you execute (Doyle).



So when you change that list you don't change the namespace, and then calling (Doyle) creates the complete list again.

But of course you could do:


Code Select
(set 'lst (Doyle))

(pop-assoc "baker" lst)

lst => (("street" 26))


Perhaps the manual should say instead of: "hash tables can be seen as association lists" to: "association lists can be created from hash tables".
Title:
Post by: cormullion on November 29, 2008, 02:56:46 AM
Thanks - I confused myself with my own writing, it seems... :)



BTW: How about a table for the manual showing which functions work on nested lists, and which on flat lists? I might have a go, but you might know of a shortcut... !
Text only | Text with Images