R: Create Python wrapper for windrose method in Climatol

[isedwards]

The rosevent (windrose) method in Climatol has the following documentation (see reference manual at CRAN):

---

Description
Plot of a wind-rose. Wind frequencies may be supplied by speed intervals, and can be absolute or relative.

Usage
rosavent(frec, fnum=4, fint=5, flab=2, ang=3*pi/16, col=rainbow(10,0.5,0.92,start=0.33,end=0.2), margen=c(0,0,4,0), key=TRUE,
uni="m/s", ...)

Arguments
Data parameters

• frec Data frame containing the wind frequencies.

Other parameters

• fnum Number of reference circles to plot.
• fint Frequency steps (in %) between reference circles.
• flab Parameter indicating which circles must be labelled:
  1: Label outer circle only,
  2: Label all circles,
• Other value: Do not label any circle.
• ang Angle along which circles will be labelled.
• col Colors to fill the frequency polygons.
• margen Margins vector for the plot (to be passed to par)
• key Set to FALSE if you do not want a legend of the wind-rose, that will otherwise be plotted if frequencies are supplied by speed intervals.
• uni Speed units for the legend header.
  ... Other graphic parameters

Details
The number of direction classes is indifferent: As many as elements of the frequency vector (or
columns of the matrix) will be considered, but the first element (column) must correspond to North.
Frequencies will be converted to percentages, no matter how they were supplied.

Value
No value is returned by this function.

See Also
plot, par

Examples
data(windfr)
rosavent(windfr, 4, 4, ang=-3*pi/16, main="Annual windrose")

---

• Create an (initially empty) Python function/method for Climatol's rosevent and add this to the process library created in R: Create Python wrapper for windrose method used by R-Instat #4
• Add an initial Python docstring to the new method
• Contribute to the automated test suite, also created in R: Create Python wrapper for windrose method used by R-Instat #4, with an (initially failing) image test for rosevent
• If possible, check characteristics of the plot numerically with an initially failing test