Problem with pan/ZOOM on semilog plot

Hi,

I've just discover a problem with the pan/zoom tool.
When using the pan/zoom tool (in the toolbar) on a semilogy plot the
zoom does not work correctly.
To visualize the problem launch the small script in attachment press the
pan/zoom button and use the zoom on the x axis direction -> you will see
the button part of the curve directly jumping at the bottom of the graph
  whereas you would expected it not to move...
I evidence this problem using GTKAgg but also with the OSX backend (did
not try other backend).
Matplotlib version is the latest svn on a MacOS X platform.

Thanks in advance,

David

logscale_zoom_problem.py (275 Bytes)

I don't think I see this -- if you hold down the 'x' while you are
panning, the zoom will be constrained in the x direction and I see no
unusual movement. If I release the 'x' but still zoom horizontally,
there is some motion in the vertical direction, but that is because I
cannot move precisely horizontally. Could you be more precise about
the artifact you are seeing. For me, the bottom part of the curve is
at (0, 10^0) and and I do not see any unusual behavior there

···

On Sat, Jan 10, 2009 at 12:28 PM, David Trem <david.trem@...149...> wrote:

Hi,

I've just discover a problem with the pan/zoom tool.
When using the pan/zoom tool (in the toolbar) on a semilogy plot the
zoom does not work correctly.
To visualize the problem launch the small script in attachment press the
pan/zoom button and use the zoom on the x axis direction -> you will see
the button part of the curve directly jumping at the bottom of the graph
whereas you would expected it not to move...
I evidence this problem using GTKAgg but also with the OSX backend (did
not try other backend).
Matplotlib version is the latest svn on a MacOS X platform.

John Hunter wrote:

Hi,

I've just discover a problem with the pan/zoom tool.
When using the pan/zoom tool (in the toolbar) on a semilogy plot the
zoom does not work correctly.
To visualize the problem launch the small script in attachment press the
pan/zoom button and use the zoom on the x axis direction -> you will see
the button part of the curve directly jumping at the bottom of the graph
whereas you would expected it not to move...
I evidence this problem using GTKAgg but also with the OSX backend (did
not try other backend).
Matplotlib version is the latest svn on a MacOS X platform.

I don't think I see this -- if you hold down the 'x' while you are
panning, the zoom will be constrained in the x direction and I see no
unusual movement. If I release the 'x' but still zoom horizontally,
there is some motion in the vertical direction, but that is because I
cannot move precisely horizontally. Could you be more precise about
the artifact you are seeing. For me, the bottom part of the curve is
at (0, 10^0) and and I do not see any unusual behavior there

John,

I reproduced the problem (jumpy boundary with log scale and zoom) and fixed it in my last two commits (6773 to maint and 6775 merging to trunk).

Eric

···

On Sat, Jan 10, 2009 at 12:28 PM, David Trem <david.trem@...149...> wrote:

------------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
_______________________________________________
Matplotlib-devel mailing list
Matplotlib-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-devel

It works perfectly now using svn 6776

Thanks very much to be so efficient in killing bugs !
Less than three hours after my small bug report... I'm really impressed!

David

Eric Firing a �crit :

···

John Hunter wrote:

On Sat, Jan 10, 2009 at 12:28 PM, David Trem <david.trem@...149...> >> wrote:

Hi,

I've just discover a problem with the pan/zoom tool.
When using the pan/zoom tool (in the toolbar) on a semilogy plot the
zoom does not work correctly.
To visualize the problem launch the small script in attachment press the
pan/zoom button and use the zoom on the x axis direction -> you will see
the button part of the curve directly jumping at the bottom of the graph
whereas you would expected it not to move...
I evidence this problem using GTKAgg but also with the OSX backend (did
not try other backend).
Matplotlib version is the latest svn on a MacOS X platform.

I don't think I see this -- if you hold down the 'x' while you are
panning, the zoom will be constrained in the x direction and I see no
unusual movement. If I release the 'x' but still zoom horizontally,
there is some motion in the vertical direction, but that is because I
cannot move precisely horizontally. Could you be more precise about
the artifact you are seeing. For me, the bottom part of the curve is
at (0, 10^0) and and I do not see any unusual behavior there

John,

I reproduced the problem (jumpy boundary with log scale and zoom) and
fixed it in my last two commits (6773 to maint and 6775 merging to trunk).

Eric

------------------------------------------------------------------------------

Check out the new SourceForge.net Marketplace.
It is the best place to buy or sell services for
just about anything Open Source.
http://p.sf.net/sfu/Xq1LFB
_______________________________________________
Matplotlib-devel mailing list
Matplotlib-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-devel