Re: Progressively worse performance from a table

[scott <scott+lua@...>]

On Sat, Jan 10, 2004 at 07:28:00PM -0200, Leandro Candido wrote:
> Hello Edgar/scott,
> 
>     I tested the 2 codes (the one from scott and the one from Edgar), and
> they aren't the same. The code from Edgar doesn't allow to add duplicated
> words (lines, in code examples).
> 
> before
> 
> print ("-mark- beginning delete")
> 
> in the 2 codes, add:
> 
> totalItemsInTable = 0;
> table.foreach(words,function() totalItemsInTable = totalItemsInTable + 1
> end);
> print("Number of items in table words:",totalItemsInTable);
> 
> The result will be different for scott's and for Edgar's code.
> This is because in the Edgar's code, if the item is already in the table, it
> isn't added twice. This is a good thing, but the idea of Edgar was to mimics
> the behavior from scott's code, and in this case, it isn't equal.
> 
>                                                             The God's Peace,
> 
> 
> Leandro.
> 

Good call!  The original code was intended to store lists of words, which
of course needs to be able to handle duplicates.  My example data
('/usr/share/dict/') and even my unit tests have no duplicates so I wasn't
even considering this situation.

For anyone interesed in what I was doing, take a look at
http://www.pragprog.com/pragdave/Practices/Kata/KataTwentyOne.rdoc  I'm
using the programming exercises suggested by Dave Thomas as a way to
experiment with Lua.

scott

-- 
------------------------------------------------------------------------
scott jacobs                                   scott+lua@escherichia.net
------------------------------------------------------------------------