The code here is a general driver code for the Matplotlib package, similar in intent to the xmgrace code that the author has used for a long time (the xmgrace web site is at https://plasma-gate.weizmann.ac.il/Grace/, but since it depends on X11 and Motif it does not seem to be a viable option for the future).
The code is a package that can be installed and imported in the usual way. It is also possible to run the code from the command line if ones $PATH environment variable is set properly (the author does not use Microsoft Windows and so does not know what the equivalent is in such computers). The code uses matplotlib and tkinter as well as other standard Python packages.
** Note: MacOS 10.14 Mojave Issue With Python version 3.7
The code here uses the Tkinter package that is part of the standard Python distribution. In the MacOS version 10.14 (Mojave) there is an issue that causes the use of the Tkinter package functions with Python 3.7 to crash the user's session. This issue does not exist in Python versions 3.6 or 3.8 with this version of MacOS, and it did not exist in the previous MacOS version as far as the author is aware. There is no solution for this issue within Python. Hence, do not use this code with that particular combination of MacOS 10.14 and Python 3.7.
** Installation and Set-Up
To install the code one needs to copy the repository to some directory, which will be called /your/path/to/the/files/ in the documentation. Substitute the actual path on your system for this dummy path. It is best to put the package in a new directory.
When the code is unpacked in the target directory you should see the following files in the directory:
LICENSE
README.md
matplotlib_driver_code_documentation.docx
matplotlib_driver_code_documentation.pdf
setup.py
src
tests
The tests directory is a placeholder and is currently empty. The code itself is in the src directory.
To install the package run the normal command
python setup.py install
in the target directory /your/path/to/the/files. This will add build and dist directories. Do not issue this command in the src directory, it must be done in the upper level directory.
If one wishes to be able to start the interface from the command line one needs to add the target directory to the $PATH environment variable. This is done on Unix-type systems via the command
setenv PATH /your/path/to/the/files:${PATH}
in tcsh/csh or
export PATH='/your/path/to/the/files:$PATH'
in bash. This assumes that the $PATH variable is already defined.
** Starting the code
Assuming that the installation worked as expected, one can then import the plot interface and start it via Python commands such as
import matplotlib_user_interface as mui
root, myplot = mui.startup()
This will cause a new window to open with the interface controls. See the main documentation in the matplotlib_driver_code_documentation.pdf or matplotlib_driver_code_documentation.docx files for more information about using the interface and adding data sets from the Python interpreter. The "myplot" variable is an object that runs the interface. The "root" variable is a Tkinter variable for the plot window.
** Use of the code
The code assumes that one will want to read in data values, possibly with associated uncertainties, from an ascii file. One can have one or more "sets" of data values in one or more plots in the display area. The code also has some capability to create a data set via a formula or in a text area window created by the code. One can add sets from the Python interpreter as well if the code was started that way. The latter process is to do the following, once root and myplot are created as noted above. First create the x and y values in Python as one-dimensional numpy arrays, for example
import numpy
xvalues = numpy.arange(0, 100, 1)
yvalues = numpy.sqrt(xvalues)
Once the x and y values are defined in the interpreter, one can load them into the plot object
myplot.add_set(xvalues, yvalues)
and then tell the code to replot the plot(s) associated with the myplot object
mui.make_plot.make_plot(myplot)
One can add multiple sets to the interface by making additional calls to the add_set method of the object.
The code also has the machinery to add vectors/boxes/lines/ellipses/text to a given plot.
Once a plot has been made, one can save the state of the code variables to an ascii file for later re-loading. This is analogous to the save/open plot function in xmgrace.
One has control of the symbols/lines/legend entries/error bars for each set, and control of the axis limits/labels/tick marks for each axis of each plot in the display area.
The code is written in Python 3 and will not function in Python 2. The code has been tested on a MacOS system and on a Linux system, but not on a Microsoft Windows system.