Disclaimer: thats my interpretation of things, if I'm wrong please
someone correct me

On Thu, Apr 16, 2009 at 1:47 PM, Bruno Deferrari <[email protected]> wrote:
> On Thu, Apr 16, 2009 at 1:09 PM, Stefan Schmiedl <[email protected]> wrote:
>> On Thu, 16 Apr 2009 11:16:12 -0300
>> Bruno Deferrari <[email protected]> wrote:
>>
>>> And now that we are at it, integers as sequences can be confusing too
>>> (like in your example, 0 matches because it is the first element of
>>> the CHAR: a sequence, matches as the key and 1 being the second
>>> matches as the value), is that necessary now that factor has ranges?
>>
>> Well, I'm thoroughly confused already by my string example. I have
>> not yet really thought about the sequenciability of integers...
>>
>> On the one hand, "abc" seems to be interpreted as an association with
>> three keys: "abc" keys    -> { 0 0 0 }
>> On the other hand, requesting a value for one such key yields something
>> obviously not contained in the association.
>
> 1. Strings and Integers are Sequences (Integers are sequences from 0
> to n-1 and empty for 0)
> 2. Sequences are instances of assoc (to make alists work I guess)
> 3. "abc" is being treated as an alist (alist exampe: { { "key1"
> "value1" } { "key2" "value2" } } )
> 4. being 'CHAR: a' one of the elements of the sequence, and being a
> character an integer, and being integers sequences too, then CHAR: a
> is like { 0 1 2 3 .. 96 } ('CHAR: a' is 97)
> 5. On an alist, the key is the first element of the inner sequence (0
> here, or "key1" on the alist example)
> 6. On an alist, the value is the second element of the inner sequence
> (for 'CHAR: a' or any integer greater than 1 it is 1), thats why you
> are getting 1
>
>> On the third hand: Does any of the available assoc implementations actually
>> support multiple usages of the same key?
>>
>> What I'd have expected of an assoc-conforming string, would be
>> access to its characters: 2 "abc" at   -> either b or c
>>
>> s.
>>
>>>
>>> On Thu, Apr 16, 2009 at 11:11 AM, Bruno Deferrari <[email protected]>
>>> wrote:
>>> > I may be wrong, but I think that sequence is an instance of assoc to
>>> > make alists work (look at the definition of at*). But yeah, I don't
>>> > like how interacts with sequences in general.
>>> >
>>> > On Thu, Apr 16, 2009 at 10:35 AM, Stefan Schmiedl <[email protected]> wrote:
>>> >> Recently on #concatenative
>>> >>
>>> >> [14:40]  <swsch> What am I doing wrong:  H{ { "a" 1 } } "a" at .
>>> >> ---> f [14:41]  <tizoc> swsch: "a" goes before H{ ... }
>>> >> [14:42]  <tizoc> at's stack effect is ( key assoc  -- value/f )
>>> >> [14:48]  <swsch> so ... I'm being stupid, that's ok
>>> >> [14:48]  <swsch> why does "a" implement assoc-protocol?
>>> >> [14:57]  <swsch> ah ... INSTANCE: sequence assoc  does it
>>> >> [14:58]  <swsch> but a string is a somewhat strange assoc ...
>>> >> "abc" keys   --> { 0 0 0 } [15:00]  <swsch> and even funnier   0
>>> >> "abc" at  -> 1 ... what's the idea behind that?
>>
>> ------------------------------------------------------------------------------
>> Stay on top of everything new and different, both inside and
>> around Java (TM) technology - register by April 22, and save
>> $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
>> 300 plus technical and hands-on sessions. Register today.
>> Use priority code J9JMT32. http://p.sf.net/sfu/p
>> _______________________________________________
>> Factor-talk mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/factor-talk
>>
>

------------------------------------------------------------------------------
Stay on top of everything new and different, both inside and 
around Java (TM) technology - register by April 22, and save
$200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
300 plus technical and hands-on sessions. Register today. 
Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________
Factor-talk mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/factor-talk

Reply via email to