XOM API for Processing XML with Java

[Wolfgang Hoschek <whoschek AT lbl.gov>]

On Nov 14, 2004, at 3:54 AM, Elliotte Harold wrote:

> Wolfgang Hoschek wrote:
>
>
> > #2 UTF conversion of large strings is fast, but for small strings 
> > (the common case) it is the opposite (setup costs), and hence Text is 
> > a hotspot as well.
>
> Could you define small? Is there an inflection point somewhere?

I'd say small is < 10K or perhaps < 1K, but that's just merely an 
educated guess. If I remember it correctly, 2 MB strings can be 
converted at 50MB/s or somewhere in that ballpark.

> Possibly I could look at optimizing this better for small strings, or 
> even use String for small strings and UTF-8 for large ones, though 
> that probably would cost more than it would save. I'd have to profile 
> and see.

It's not clear to me how one could improve the situation. Doesn't look 
like a "one Text class fits all" scenario. Might be better to just have 
two Text classes, each addressing a separate specific need.

Wolfgang.