Closed wblumberg closed 9 years ago
Going to give it a try right now!
First things first...
It looks like your code is using the old setup scripts. To be safe, call you do a pull on the upstream changes Patrick made regarding the setup.py script?
Hmm. I could have sworn I did that. Let me double check. Also, resetting the hodograph craps out. I know why.
It says everything is up to date...
I did git fetch upstream and git pull upstream master
Correction: It was bad on my end. You're fine.
Everything looks great! Just a few minor bugs I caught...
1) When you change the parcel by double-clicking the thermo window (say, remove EFF and add FCST) and the increment or decrement by a time step, the thermo window will reset it back to the default display. Ideally, if you change a displayed parcel, it should stick throughout time.
2) When selecting the USER parcel in the thermo window, if I set a new custom parcel (from lift this parcel in the SkewT window), it resets the parcels to the default display. Ideally this should not reset the parcels in the Thermo window.
If we can get those two bugs taken care of, I don't see anything else preventing this from being merged. It looks and feels outstanding! Great work!
Also: I stuck the IPython notebook in the notebook viewer - looks outstanding! I think this will be very helpful for users getting started. I've needed something like this on several occasions that I've tried to help some people use SHARPpy on their datasets.
Thanks! My contributes have been lurking in the shadows...
Dang. I thought I had fixed those bugs. Let me see what I can do about that.
Another thing I have noticed: upon creating the window, the thermo window defaults to selecting the effective parcel, but the mupcl is the one default displayed in the SkewT window. I would recommend making the default selected parcel in the thermo window be the mupcl.
Okay. i think those first two bugs are fixed. Now onto forcing the MU parcel to be a default.
Okay. It will default to the MU parcel if it's available and upon initialization of the thermodynamic inset. This way the parcel won't be defaulting to the MU parcel each time you increment through a profile. If the MU parcel isn't available, then it'll pick the first parcel in the list.
Sweet. When I get back from dinner, I'll test it and then we should be able to go ahead and merge the pull request.
Changes look good. One more thing (and this may not be a bug), but when you define a custom parcel and lift it for a given time, when you increment the time step, should the next time step lift a parcel from the same level, or be blank? Currently, if you choose a user defined parcel and advance the time step, the USER parcel becomes empty again. I just want to clarify how it is supposed to behave before merging.
Nope. Not a bug. There might be a way to propagate the parcel values over time, but I don't think it's a necessary function at this time.
Awesome. Merging.
Hurray!
This pull request includes a wealth of new things I've been working on for a while.
Scripting Tutorial
I've included a tutorial folder within the main SHARPpy directory for others to use. It includes a sample data file from the Pilger, NE 2014 tornado for the user to work with. The tutorial is within the IPython notebook that is included. It teaches the user about reading in the data, constructing the profile object, drawing parcel traces, computing convection indices, and plotting the data in Matplotlib.
New GUI Parcel Functions