<div dir="ltr"><br><div class="gmail_extra"><br><br><div class="gmail_quote">On Fri, Jun 27, 2014 at 10:35 AM, Luca Delucchi <span dir="ltr"><<a href="mailto:lucadeluge@gmail.com" target="_blank">lucadeluge@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="">On 27 June 2014 15:46, Anna Petrášová <<a href="mailto:kratochanna@gmail.com">kratochanna@gmail.com</a>> wrote:<br>
><br>
><br>
><br>
> On Fri, Jun 27, 2014 at 3:03 AM, Luca Delucchi <<a href="mailto:lucadeluge@gmail.com">lucadeluge@gmail.com</a>> wrote:<br>
>><br>
>> On 26 June 2014 19:59, Markus Neteler <<a href="mailto:neteler@osgeo.org">neteler@osgeo.org</a>> wrote:<br>
>><br>
>> ><br>
>> > nice work, Luca (and thanks to Anna for code cleanup).<br>
>> ><br>
>><br>
>> Sorry Anna, but I don't understand because you remove all the 3D code,<br>
>> I would like to implement it later (sometime I'm careless but not this<br>
>> time :-P )<br>
>> Thanks for the improvement of coordinates field.<br>
><br>
><br>
> Sorry for that. I also fixed some other minor things in that commit. Do you<br>
> plan to support vector data as well?<br>
><br>
<br>
</div>I would like, first I need a vector temporal dataset to test :-)<br>
<div class=""><br>
> What I am not sure about is the usage of pygrass here. The problem is if we<br>
> would like to integrate it more in GUI. Currently when you click on the<br>
> g.gui.tplot in menu, a standard generated dialog opens and when you fill the<br>
> params and press ok, the actual plotting gui shows up, which is cumbersome.<br>
<br>
</div>yes I have your same feeling, I should create a function like for<br>
animation tool?<br>
(other tools have the same behavior of g.gui.tplot for example g.gui.timeline)<br>
<div class=""><br>
> The menu entry could bring directly the tplot, however then there is a<br>
> possibility of crashing entire GUI when something goes wrong (because of<br>
> using ctypes in pygrass). So one option is to call some t.* module but there<br>
> is no t.rast.what only t.vect.strds.observe which requires vector map.<br>
<br>
</div>I could use r.what instead pygrass but I would like to escape to call<br>
external modules<br>
<div class=""><br>
> Another option is to wrap the pygrass code and call it as a subprocess. I<br>
> don't have time to look at it more now, but we should be aware of this.<br>
><br>
<br>
</div>Just to know, we don't have the same problem using ctypes in the gui?<br>
In several modules we are using it....<br></blockquote><div><br></div><div>yes we have, digitizer and nviz, I just want to avoid another one when it would be relatively simple (comparing to those 2) make it safe. r.what could be a simple solution, although it would be probably slower than pygrass (not sure if significantly, maybe not).</div>
<div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="HOEnZb"><div class="h5"><br>
> Anna<br>
<br>
--<br>
ciao<br>
Luca<br>
<br>
<a href="http://gis.cri.fmach.it/delucchi/" target="_blank">http://gis.cri.fmach.it/delucchi/</a><br>
<a href="http://www.lucadelu.org" target="_blank">www.lucadelu.org</a><br>
</div></div></blockquote></div><br></div></div>