On Monday, January 2, 2017 at 3:04:48 PM UTC-5, Dan'l B wrote:
>
> In oder to use something other than Zambretti in the ticker-tape/scroller 
> of the steel series gauges I understand from a post by Matthew that you 
> need to "hijack" the $fc variable here:
>
> #set $fc = $forecast.label('Zambretti', $forecast.zambretti.code)
>
> What is the correct replacement to use a WU or NWS current forecast there? 
> I can't seem to locate an example in the docs.
>

dan'l,

i'm afraid there is not yet a consistent way to do this.

many of the forecast sources provide a textual description.  unfortunately, 
they do not do it consistently.  there is a field 'desc' in the forecast 
database that is *supposed* to hold the textual description.  it is not yet 
populated.

NWS - no textual description in the point forecast matrix
WU - the 'condition' field is text, but it is same as cloud cover
OWM - weather.description is same as the WU condition field
UKMO - no textual description in the 3hourly forecast
Aeris - no textual description
WWO - weatherDesc is same as the WU condition field

so what do we do?

many of these forecast sources provide additional forecast data.  for 
example, nws has marine forecasts and non-marine textual forecasts that do 
not have any of the point forecast numerics.  to properly address the 
textual descriptions, i'll probably have to make the forecast extension 
recognize those textual sources, then provide a generic way to access them 
using the search list extension.

pull requests are welcome :/

despite my best efforts, this is fairly complicated code - not because it 
is difficult maths or physics, but because there are so many different 
permutations and variations between the forecast sources and their outputs.

m

-- 
You received this message because you are subscribed to the Google Groups 
"weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to weewx-user+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to