Re: Having trouble getting number-columns-spanned to work

[Glen Mazza]

Richard Mixon (qwest) schrieb:
Glen Mazza wrote:
 

Andreas L. Delmelle wrote:
   

-----Original Message-----
From: Richard Mixon (qwest) [mailto:[EMAIL PROTECTED]
       

<SNIP>
 

Anyway, I'm not sure why that functionality was removed. Maybe
someone else can fill you in here? 


     

Sure, Driver.reset() was removed in 1.0 because Driver itself was
removed, and the 1.0 API equivalent, the Fop class, has no need for
it.  
0.20.5 and 1.0 are completely different products, and 1.0 has many
modernizations/optimizations already built-in. 

Glen
   

Glen,
Thanks for the explanation. So, what I hear is that in most situations,
the Fop class is light weight enough that there is not need to worry
about creating a new one for each document request we receive. If I
notice any bottlenecks in the future, I can always profile and report
back.
Thanks so much to both Andreas and you for the help.
- Richard
 

Yes--it's hard to speed up Driver when it doesn't exist, and the Fop 
class is all of 180 LOC.  The .reset() class was basically a placebo in 
1.0 anyway (it may have done more in 0.20.5, however, I don't know)--it 
just set your output stream to null, which you can do anyway.  (But if 
it makes you feel better, we *can* add an empty .reset() method, as well 
as empty .runFaster() and .makeReallyNicePDF() methods...  ;-)  The XSLT 
process, as well as layout and document building, will remain 98%+ of 
the story.  However, please profile and report should it turn out I/we 
were in error with our assumptions--quite possible, as I'm hardly the 
brightest light on this team.

Thanks,
Glen
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]