[jsword-devel] Performance problem

[Joe Walker]

This is the problem with o-o, by having interfaces that hide what is
really going on you no longer know what is going on!

I had a quick look and I think the problem is when it takes the
Gen-Rev passage and trims it down to the first 50 verses to fit in a
tab, it removes the remaining verses one by one and calls the Passage
changed listeners for each - quite a loop!

There must be some optimization to AbstractPassage.trimVerses() that
needs doing. Need to go out, back later.

Joe.

On Tue, 31 Aug 2004 13:46:10 -0400, DM Smith <dmsmith555 at yahoo.com> wrote:
> I did a Passage Lookup of Gen-Rev. Kind of dumb, but I wanted to
> robustly test a change I was making. Several places had a test of an
> element in a JDom tree being a String. It never happens, they are of
> type org.jdom.Text. So I put a break point in the code and
> 
> I had expected the following behavior:
> Convert the input into a VerseRange.
> Slice the verse range list into sub ranges of n verses where n is the
> verse per tab number.
> Display the first n verses in a tab.
> 
> I was expecting to have to click through many tabs.
> 
> I stopped it after 10 minutes, first pausing the debug session.
> It was in book 28 getting the text.
> So my guess is that it was going to get the entire passage and then
> re-get each tab's worth.
> And in the perf meter, it is pegging the system at 100% cpu and almost
> all the memory I have!
> 
> I have not looked into this any further. My test was successful!
> 
> _______________________________________________
> jsword-devel mailing list
> jsword-devel at crosswire.org
> http://www.crosswire.org/mailman/listinfo/jsword-devel
>