Another 2.5.1 request - this for reports with charts

[Geert Janssens]

On Wednesday 15 May 2013 07:22:20 David Carlson wrote:


    
On 5/15/2013 6:18 AM, Geert Janssens      wrote:        
                  
On Tuesday 14 May 2013 14:43:03 David Carlson        wrote:      
>       
> Since I was in my download directory, I        re-installed 2.5.0 and found      
> that the chart reports are blank in tat        release too.      
>       
> David C      
      
If fact this problem turned out to be easy to        repeat and (more importantly) easy to fix.      
      
The charts weren't displaying in my Windows        test setup either. The cause turned out to 
be that path names on        Windows start with C: or something similar. Webbrowsers take        
this as an unknown network protocol and due to this will fail to        load the necessary 
javascript and css files on that platform if        not properly prefixed. This also explains why 
this problem        didn't happen on linux or OS X.      
      
I have just committed a fix in r22978. You        can try to download tomorrow's nightly build 
or wait for 2.5.2        in a couple of weeks to verify this works.      
      
The memory issues you were having are a        different problem. Do you still run into those ? 
If so, that        will require a separate evaluation.      
      
Geert            I knew it wasn't my imagination, and I suspected that it was Windows    related 
because nobody else believed me.  Thank You!        All I can say about memory usage is that 
the Windows Resource    monitor reports much higher RAM usage when I am running release    
2.5.1 than when I am running 2.4.13.  Release 2.5.0 and release    2.5.1 also run a lot slower 
than release 2.4.13.  I believe Robert    expected some of that, but this seems to be 
debilitating.  I am not    running the new releases enough to see a good history, though.  The    
other clue that I have is that there is no RAM issue on this    computer when release 2.5.1 is 
not installed and running        David C.  
David,

For me it was never a matter of not believing you. I asked my questions in an attempt to get 
more details about the exact cause. Your experience is real, but what causes it is the first 
thing to find out. In this case it turned out to be a bug in the software which I probably could 
have determined early on if I had tried to replicate it in my Windows testbox. It's just become 
a habit to first ask for more details before trying myself.

In other cases it might be an issue on the user's PC in which case it makes sense to ask lots 
of questions - I am not in front of that PC to see what is going on...

Anyway. Your remark about memory consumption is well received. It looks like we better run 
some memory/performance testing and optimizations before we release 2.6.

For me personally, it will have to wait a bit. I have more pressing matters to attend to, but 
perhaps another dev is curious (and experienced!) enough to give it a shot.

Geert