Wind computations seldom give the right direction

[MacFlorent]

As pointed by Skynight on the VEAF Discord: https://discord.com/channels/471061487662792715/1333160268829823078

The weathermark method used by the veafWeather module doesn't work properly when computing wind direction. It calculates the wind azimuth based on the vector (x, z) given by DCS by doing math.atan(z, x). It's the right principle, but unfortunately, DCS apparently uses Lua version 5.1, and in this version, math.atan takes only one parameter (it should have been math.atan(z/x)). As a result, the angle is often wrong, more or less.
In Lua 5.3 or higher it would have been okay, but here we need to either pass the correct parameter or use the atan2 function.

Remediation will be to rewrite the wind calculation function.
In addition it's the only thing still being used in weathermark, so we would not be to far from dumping the library altogether.

[davidp57]

On utilise pas mal de fonctions de weatherMark dans veaf.weatherReport

[MacFlorent]

C'est pour ça que j'avais indiqué "not too far". Mais veaf.weatherReport devrait être remplacé par veafWeather.

[davidp57]

pourquoi pas ; à creuser, on peut rouvrir pour étendre