RE: [ROOT] RE: zoomed in TGraph bug

From: Colley, Tony (Tony.Colley@itt.com)
Date: Wed Feb 27 2002 - 15:08:05 MET


Hi Rene,

The message you replied to was sent prior to my receiving your email
notifying me of your fixes to TGraph. There seems to be a large delay in all
email I send from work... who knows what kind of trouble our IT department
has gotten our mail servers in.

Anyway, once I got your email, I updated ROOT from CVS, and the extra boxes
no longer appear! Occassionally, though, I still get the boxes apparently
wrapping around the edge of the graph (so that they are canvas-wide
rectangles rather than nice little squares). I haven't been able to come up
with a simple repeatable example of that, yet, though. If I do, I will pass
it along.

Thanks for the VERY quick resolution of my bug report!

Tony

-----Original Message-----
From: Rene Brun [mailto:Rene.Brun@cern.ch]
Sent: Wednesday, February 27, 2002 07:01
To: Colley, Tony
Cc: 'ROOT Talk'
Subject: Re: [ROOT] RE: zoomed in TGraph bug


Hi Tony,

Did you start from the CVS source with the tag 3-03-02 or from
the CVS head? I implemented this change only last night !

This morning, I have added additional protections in TGraph::ExecuteEvent
in case you zoom with very large factors.

Rene Brun

Colley, Tony wrote:
> 
> By the way, I just did a CVS update, remade & reinstalled ROOT, and now am
> running ROOT 3.03/02...
> 
> and the problem described below still exists.
> 
> -----Original Message-----
> From: Colley, Tony
> Sent: Tuesday, February 26, 2002 15:37
> To: 'ROOT Talk'
> Subject: zoomed in TGraph bug
> 
> The following demonstrates a "minor" bug in TGraph:
> 
> {
> Int_t i;
> Float_t x[1000];
> Float_t y[1000];
> 
> for (i=0; i<1000; i++)
> {
>    x[i] = 0.001*i;
>    y[i] = 6.2831853*x[i];
>    y[i] = sin(12*y[i]) + sin(17*y[i]) + sin(56*y[i]) + sin(74*y[i]);
> }
> 
> TGraph *graph = new TGraph( 1000, x, y );
> graph->Draw("AL");
> 
> printf(
> "\n\n===============================================================\n" );
> printf( "Right click on X axis and select SetRangeUser.\n" );
> printf( "Enter the range 0.387 to 0.389 and click OK.\n" );
> printf( "Move the cursor over one of the 3 points (at .387, .388,
.389).\n"
> );
> printf( "Press down the left mouse button.\n" );
> printf( "When I do it, there are 6 extra boxes drawn\n" );
> printf( " (in addition to the three that should be drawn)\n" );
> printf(
"===============================================================\n"
> );
> }
> 
> (I've attached the same thing as tgdemo.C, in case that's easier to use
than
> just cutting/pasting from this email)
> 
> The y function was chosen arbitrarily. The bug appears for many other
> functions I tried.
> 
> The bug is even more insidious when you significantly increase the number
of
> data points graphed. When you zoom in close enough to begin seeing
> individual points, in addition to having many extra boxes, some boxes are
> apparently drawn too close to the edges because there is a wraparound
effect
> (giving rectangles spanning the graph rather than little boxes around the
> points).
> 
> I also noticed in the example I provided, if you don't zoom in quite so
far,
> there are extra boxes drawn extending out of the left and right sides of
the
> graph. These boxes seem to correspond to data points outside the graph
> clipping area but within the canvas clipping area. I'm not sure if I would
> classify this behaviour as a bug or a feature... it could be useful and
may
> be intentional (i.e. a feature).
> 
> Tony Colley
> ITT Industries A/CD
> Fort Wayne, IN  USA
> 
> Using ROOT 3.03/00, last updated from CVS on 18 Feb 2002, on RedHat Linux
> 7.2
> 
> ************************************
> If this email is not intended for you, or you are not responsible for the
> delivery of this message to the addressee, please note that this message
may
> contain ITT Privileged/Proprietary Information.  In such a case, you may
not
> copy or deliver this message to anyone.  You should destroy this message
and
> kindly notify the sender by reply email.  Information contained in this
> message that does not relate to the business of ITT is neither endorsed by
> nor attributable to ITT.
> ************************************

************************************ 
If this email is not intended for you, or you are not responsible for the
delivery of this message to the addressee, please note that this message may
contain ITT Privileged/Proprietary Information.  In such a case, you may not
copy or deliver this message to anyone.  You should destroy this message and
kindly notify the sender by reply email.  Information contained in this
message that does not relate to the business of ITT is neither endorsed by
nor attributable to ITT. 
************************************ 



This archive was generated by hypermail 2b29 : Sat Jan 04 2003 - 23:50:43 MET