I know they use hash tables for symbols, but I've always assumed "lists" to 
be doubly-linked lists, because that nearly exactly mirrors the 
functionality they offer.  And if you correct my "bytes" into "bits" 
(sorry, it was getting late at night or something), then I think the total was:

4 bytes "Item" (e.g.: pointer to data)
4 bytes "Type" (e.g.: indicator int, string, etc.)
4 bytes ptr next
4 bytes ptr last
--
16 bytes total
If a property list, then another 8 bytes
Total: 24 bytes is my estimate for a property list, not including actual data.

Not much overhead, imo.

- Tab


At 10:09 AM 9/9/02, Mayuresh wrote:


>My First impression of a list was that they were implemented using linked 
>lists. A doubly linked list would certainly be robust which seems to be 
>the case since list implementation rarely breaks. That would be the more 
>straight forward way, though not the fastest as Robert pointed out.
>Hash table IMO would be when your indexing data and need to perform some 
>serious quick searches something which I have not really experienced with 
>lists.
>
>Though I can't think of a way to calculate how much each list or every 
>element from a list hogs, but I doubt if it's so much for a reasonably 
>sized list to seriously affect performance.
>
>Maybe MM can help out.
>
>Mayuresh

[To remove yourself from this list, or to change to digest mode, go to 
http://www.penworks.com/lingo-l.cgi  To post messages to the list, email 
[EMAIL PROTECTED]  (Problems, email [EMAIL PROTECTED]). Lingo-L is for 
learning and helping with programming Lingo.  Thanks!]

Reply via email to