"Newbie" questions - "unique" sorting ?

Kim Petersen kp at kyborg.dk
Tue Jun 24 11:16:47 CEST 2003


Cousin Stanley wrote:
> | How about the simple approach?
> | ...
> 
> Kim ...
> 
> The approach I used is fairly simple and similar
> to the one you posted, basically just
> stuffing words from split lines into
> a dictionary ...

yeah it is - a reminder to me not to do any scanning of code before my 
morning coffee ;-)

> Perhaps skipping the dictionary word count update
> in the following line might speed things up ...

totally skipping the word count should speed it up - and i believe that 
the approach of dict_words[this_word]=dict_words.get(this_word,0) btw. 
is a bit faster than doing has_key() or the least of doing:

    if this_word not in dict_words.keys() :

which aught to be extremely slow on a large dictionary (creating and 
dropping lists of thousands + doing a O(n) search over it). And that may 
very well be the culprit of the slow run you see....

> 
>       else :
> 
>             dict_words[ this_word ] += 1
> 


-- 
Med Venlig Hilsen / Regards

Kim Petersen - Kyborg A/S (Udvikling)
IT - Innovationshuset
Havneparken 2
7100 Vejle
Tlf. +4576408183 || Fax. +4576408188





More information about the Python-list mailing list