[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kfm-devel
Subject:    Re: khtml
From:       Lars Knoll <Lars.Knoll () mpi-hd ! mpg ! de>
Date:       1999-06-07 15:13:31
[Download RAW message or body]

On Mon, 7 Jun 1999, Waldo Bastian wrote:

> Hi Lars,
> 
> Since the new KHTML seems to be doing quite well, I guess the unicode
> branch can be merged back with the HEAD branch.

I was also thinking about it. I'll do it tonight.

> It might still need a bit more tuning, I have a test-page which seems to
> spend most of its time in the parser. I don't get profiling to work 
> correctly, but I suspect the QString -> char * conversion done before
> we enter the hash functions. We can probably build a 'char *' right 
> away, since tag-, entity- and attribute-names are ascii anyway.
Don't think so. If it spends time in the parser it can't be the hash
functions, since they're all called from the tokenizer. Anyway, for each
token/attribute/entity, the conversion to char * is only done once. I
would really wonder if this is the problem.

Perhaps you could send me the test-page you have. I could have a look at
it too.

Cheers,
Lars

[prev in list] [next in list] [prev in thread] [next in thread]