How to use matplotlib(metpy) to display data created in the background

[vbdyron]

Context

---

Hi there, I am new to python coding (and Github) and have been studying it for the last month and a half. I have been trying to implement what I learned that is directly in relation to my field of work/expertise (weather). I have created a script that can do some data retrieval. The input for this data retrieval goes through appJar. Internally I can manipulate the data to create a figure which can be viewed in the console output. However, I cant seem to figure out how to display it in the appJar interface using matplotlib. I did read on the website that the matplotlib widget is intended to create some simple graphs. But I was hoping that I could just create it in the background and let the widget just do the display work in the SKEW T tab. Is it possible to do what I want in appJar or should I wait for a future release? I have attached a PDF and RST(just to make copy pasting easier) that shows the libraries that I am using, the code used to build the appJar interface and the ouput.

Sample code, demonstrating the issue

---

Sounding Retrieval Utility(pre skew output).pdf

Sounding Retrieval Utility(pre skew output).zip

Version Information

---

appJar 0.93.0 matplotlib 2.2.2 py36_1 conda-forge metpy 0.8.0 py36_0 conda-forge numpy 1.15.0 py36_blas_openblash8d851b4_200 [blas_openblas] conda-forge numpy-base 1.14.3 py36h5c71026_0

[jarvisteach]

There are a lot of dependencies there, that I don't have installed, and a lot of code to try to work my way through...

I think what you're asking, is: "is there a way to run a process in the background, which generates data, and then updates a plot in appJar"

If so, then the answer is yes.

Have a look here: http://appjar.info/pythonThreads/

You can write a function that generates all the data you want, run it in a thread, and have it use the update queue to modify your GUI..

[jarvisteach]

Here's an example:

from appJar import gui 
from numpy import sin, pi, arange
import random, time

def getXY():
    x = arange(0.0, 3.0, 0.01)
    y = sin(random.randint(1,app.slider('yVal') +1) * pi * x)
    return x,y 

def generate():
    while True:
        app.queueFunction(app.updatePlot, 'p1', *getXY())
        time.sleep(1)

with gui() as app:
    app.label('MatPlotLib Demo')
    app.slider('yVal', show=True, pos=[2,0])
    app.addPlot("p1", *getXY(), row=1)

    app.thread(generate)

[vbdyron]

Thank You Jarvis, This is interesting, you got general idea of what I wanted to do, I'll try this out and give you the feedback on my attempt. The way I managed to solve it on my end (while waiting for your input), was to generate the PNG image file and then display it with Canvas. It is a bit limiting solution, but it did get the job done.

[JaredMGoldman]

I was wondering if you could include any more advanced labeling techniques in a graph displayed on a gui (i.e. multiple plots with different colors and a legend). These features are available on a matplotlib.pyplot graph and I was wondering if there would be any way to directly instantiate a pyplot graph into the gui. My initial ideas are by creating a new widget type that displays the image of a pyplot graph, as seen from the show() method in pyplot or display a .png as mentioned above.

The latter solution is not ideal and I would like to use this ability to display a live graph within the gui that updates as data is received from an outside source. I could use your built-in software, but I may need to display up to 9 graphs, so I would prefer to compress the data into two graphs (one with up to 8 subplots) which I could easily accomplish with a pyplot graph.

Any help is greatly appreciated!