Re: polymarkers

Keith,

I'm afraid there's no way to tell the gsn_poly* code to use the right
Y axis for its data space. What you are doing to transform the
coordinates yourself beforehand is the way to go.

--Mary

On Thu, 13 Sep 2007, Keith Oleson wrote:

> Excellent! That works fine (and Mary's suggestion worked also).
>
> A related question:
>
> I'm using the following to create curves of model data on an x-y plot with different left
> and right y-axis:
>
> plot(0) = gsn_csm_xy2(wks,hour,tg_west_wall_model,fira_west_wall_model,resL,resR)
>
> The model data are being plotted correctly.
>
> Then I'm using the following to create polymarkers representing observed data for the left
> and right x-axis:
>
> dum1 = gsn_add_polymarker(wks,plot(0),hour,tg_west_wall_obs,polyres1)
> dum2 = gsn_add_polymarker(wks,plot(0),hour,fira_west_wall_obs,polyres2)
>
> But I have to transform the second set of data "fira_west_wall_obs" to get the data plotted
> correctly with respect to the right y-axis. Is there any way to tell dum2 = gsn_add_polymarker...
> to use the right y-axis coordinates?
>
> I don't mind transforming the data and actually I don't have to transform the data if I use
> Mary's method, but I'm just curious if there's a way to accomplish this.
>
> Thanks,
>
> Keith
>
>
> ____________________________________________________________________________
> Keith Oleson oleson_at_ucar.edu
> Terrestrial Sciences Section Work Phone: 303-497-1332
> Climate and Global Dynamics Division Fax: 303-497-1348
> National Center for Atmospheric Research Express mail: NCAR
> P.O. Box 3000 1850 Table Mesa Dr.
> Boulder, CO 80307 Boulder, CO 80305
>
> On Thu, 13 Sep 2007, David Brown wrote:
>
>> Hi Keith,
>>
>> Actually although it is not documented as working, it turns out that
>> you can control the
>> clipping of polymarkers and other poly primitives using the resource
>> 'gsClipOn'.
>> Just set it to False to allow primitives to be drawn unclipped.
>>
>> Exceptions: when using data coordinates you cannot draw outside the
>> defined data space:
>> the data space is undefined for map and 'irregular' transformations
>> outside the viewport, as
>> well as for log plots with data values of 0 or less.
>>
>> Contrary to what the documentation says, it has been implemented ever
>> since Version 4.2.0.a010,
>> which was released on 31 July 2000. I will update the docs and I
>> have attached a script that
>> illustrates the use of gsClipOn with gsn_polymarker.
>> -dave
>>
>> On Tue, 11 Sep 2007, Keith Oleson wrote:
>>
>>> I have a feeling I've asked this question before but I can't find
>>> any record of it
>>> (on the website or in my mind).
>>>
>>> When I generate an x-y plot with lines and markers, the markers at
>>> the left and rightmost
>>> x-axis values are getting cut in half by the y-axis. I've attached
>>> an example plot
>>> (the filled circles and open circles at x=0 and x=10 are cutoff).
>>> Any way to fix this?
>>> I don't want to expand the x-axis to -1 to 11.
>>> I'm using (e.g.):
>>> plot(0) = gsn_csm_xy2(wks,hour,tg_west_wall,fira_west_wall,resL,resR)
>>> dum1 = gsn_add_polymarker(wks,plot(0),hour,tg_west_wall_obs,polyres1)
>>> dum2 = gsn_add_polymarker(wks,plot
>>> (0),hour,fira_west_wall_obs,polyres2)
>>>
>>> with:
>>>
>>> polyres1 = True
>>> polyres1_at_gsMarkerIndex = 16
>>> polyres1_at_gsMarkerSizeF = 0.0045
>>> polyres1_at_gsMarkerColor = "black"
>>> polyres2 = True
>>> polyres2_at_gsMarkerIndex = 4
>>> polyres2_at_gsMarkerSizeF = 0.0045
>>> polyres2_at_gsMarkerColor = "black"
>>>
>>> and then paneling plots.
>>>
>>> Thanks,
>>>
>>> Keith
>>>
>>> ______________________________________________________________________
>>> ______
>>> Keith Oleson oleson_at_ucar.edu
>>> Terrestrial Sciences Section Work Phone: 303-497-1332
>>> Climate and Global Dynamics Division Fax: 303-497-1348
>>> National Center for Atmospheric Research Express mail: NCAR
>>> P.O. Box 3000 1850 Table
>>> Mesa Dr.
>>> Boulder, CO 80307 Boulder,
>>> CO 80305
>>
>>
>>
>
_______________________________________________
ncl-talk mailing list
ncl-talk_at_ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Mon Sep 17 2007 - 10:14:19 MDT