Okay, I have the weather integration working. I’ll spare you the process I went through to figure it out, but the short version is that I am a worse developer than I originally suspected. The remoteFile capability works by replacing the three compliment arrays (morning, afternoon & evening). The weather integration works (sensibly, IMHO) by adding weather-specific compliments to the existing arrays.
So in order to get them both working, you need to specify both the original arrays and the weather-specific arrays. You can delete any weather-specific arrays you wish, but you must have the morning, afternoon & evening arrays. Copy & paste this compliments.json file to your mirror and then modify accordingly.
{
"morning": ["Morning compliment"],
"afternoon": ["Afternoon compliment"],
"evening": ["Evening compliment"],
"day_sunny" : [
"Day Sunny Compliment 1",
"Day Sunny Compliment 2"
],
"day_cloudy" : [
"Day Cloudy Compliment 1",
"Day Cloudy Compliment 2"
],
"cloudy" : ["Cloud 1"],
"cloudy_windy" : ["Cloud windy"],
"showers" : ["Showers"],
"rain" : ["Rain"],
"thunderstorm" : ["Thunderstorm"],
"snow" : ["Snow"],
"fog" : ["Fog"],
"night_clear" : ["Night clear"],
"night_cloudy" : ["Night Cloudy"],
"night_showers" : ["Night showers"],
"night_rain" : ["Night Rain"],
"night_thunderstorm" : ["Night thunderstorm"],
"night_snow" : ["Night Snow"],
"night_alt_cloudy_wind": ["Night Cloudy Winds"]
}
I’m going to work on this a bit and hopefully the next release of the MirrorMirror software will be a bit more friendly. As an aside, I’ve noticed that the weather integration technique continually adds to the arrays. So the memory usage will grow and grow at whatever frequency you set the compliment refresh. This isn’t terrible, as straight text doesn’t take a whole lot of memory. But it may have unintended effects if you leave your mirror running for months at a time or if you are expecting a true randomization of the compliments you enter.