[ncl-talk] dummy wks for get_isollines()- best/better practice?
Alan Brammer
abrammer at albany.edu
Tue Jan 12 11:49:34 MST 2016
Thanks. Yeah, I’m not seeing any files created anywhere. The ps/pdf info is interesting, I didn’t try with those formats but that behaviour was what I wanted to avoid.
The calls to create the dummy wks and plot don’t seem to take any appreciable amount of time so seems good for now.
> On 12 Jan 2016, at 13:33, Rick Brownrigg <brownrig at ucar.edu> wrote:
>
> Hi Alan,
>
> I'm not expert at this, but to the best of my knowledge, get_isolines is the only way (currently) to get at the contour lines, and it is fundamentally based upon the contour plot object. So your approach is pretty much what is required.
>
> There should be no hidden files being created however -- at most, there may be an empty file with the name you gave to the gsn_open_wks() call. For PNG output however, the file is not created until the workstation is "cleared", which is invoked by a "frame()", so you shouldn't be seeing any plot output files from your script. (is this true?). This is not the case for ps/pdf/svg output, where the plotfile is created upon workstation open.
>
> Hope that answers your questions....
> Rick
>
>
> On Mon, Jan 11, 2016 at 9:51 AM, Alan Brammer <abrammer at albany.edu <mailto:abrammer at albany.edu>> wrote:
> Hi all,
>
> I’m writing a function that utilises the get_isollines function. I don’t want a plot however, so am currently generating a png workstation then creating a plot but never drawing or framing. This works fine and seems to be fast enough for now, just wondering if there is a cleaner way?
>
> Am I unnecessarily creating a hidden png somewhere on my filesystem or does that not happen until draw and frame are called?
> If there are system calls occurring in the background is there any difference (in terms of computation time) about the type of “fake” workstation I create?
> Is it possible to use get_isollines without the dummy wks and plot calls .e.g Is there a background marching squares algorithm already setup for wrapping into NCL?
>
> Thanks,
>
> Alan.
> _______________________________________________
> ncl-talk mailing list
> ncl-talk at ucar.edu <mailto:ncl-talk at ucar.edu>
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk <http://mailman.ucar.edu/mailman/listinfo/ncl-talk>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20160112/d996aa28/attachment.html
More information about the ncl-talk
mailing list