$twcweather.alerts Observations

weather
#243

I will add all the new data later on today…

Those who know me well, know that I am blessed/cursed with the ability to see multiple (opposing) perspectives simultaneously…

… So even though I love the concept and streamlining of @Pantheon’s modified piston, it also dawns on me that this section of the Wiki will never get alternative data.

For example, if that piston was in use 6 months ago, we would have missed out on 38% (5 out of 13) of the following dataPoints:

pic

Essentially, the modified piston will help us to quickly get new & unique data into the Wiki, but at the cost of never showing us any alternative dataPoints.

#244

All 3 of the latest posts have been added to the Wiki now.

New Additions (and contributors):

Thanks guys!!

I also added a new section identifying patterns in phenomena.

#245

Any suggestion as to what changes I can make to address this?

#246

Not really… I mean, we could add another couple of variables up top to also track previous data from phenomena and/or productIdentifier… but even this is not fool-proof… There are tons of data points still not confirmed on the Wiki that can show up in practically any alert.
(IE: responseType = Evacuate… or severity = Unknown)

But even if the piston was massive, and we tracked those also, we’d still miss on alternative dataPoints. (IE: some dataPoints appear in more than one eventDescription)

Without writing 5 thousand lines of code, (with extensive cross-checking), I don’t think there is a way to only show new information without also missing out on tons of alternative responses.

Or as my Grandma used to say, “We can’t have our cake, and eat it too.” … LOL

#247

So we drop back to the original piston so that we don’t miss further data?

#248

I’m on the fence with mixed feelings, really…

  • On the one hand, if we get too many texts, we’ll stop paying attention to the details.
  • On the other hand, too few texts means data may remain missing.

I do not know the best answer, besides running both pistons, LOL

#249

Hello! What is the “AccuWeather” device shown in the example?

#250

Never mind… found it!!

#251

I’m going to start playing around with this.

Lately, it seems like I keep getting the same ‘generic’ alerts from $twc. I seem to only get “Special Weather Statement” instead of the actual alert. So, instead of “Lake Effect Snow Warning”, I get “Special Weather Statement” from $twc. Instead of “High Wind Warning”, I get “Special Weather Statement”. So I’m forced to go look up what the ACTUAL weather alert is.

I’ll see what happens with $twc now that we are going into spring, but I want to see what’s available with AccuWeather.

#252

For others finding this thread later, here’s the link:

AccuWeather Connect

#253

I guess I am confused. Before, if I got any of these alerts, since there were already in the database, I would just ignore them and not send you the data. Was I missing things doing that?

#254

If the first person posted:
SV, SVR, 0106, Severe Thunderstorm Warning
Then we would have missed:
SV, SVS, 0333, Severe Thunderstorm Warning

Likewise, if the first person posted:
SV, WCN, 0547, Severe Thunderstorm Watch
Then we would have missed:
thunderstorm, ECA, 0000, Severe Thunderstorm Watch

In other words, two alerts, with identical eventDescriptions, may each have unique data…

#255

Well then, I am probably guilty of missing some things. Because if I see Severe Thunderstorm Watch, and I know that it is already in the database, I just ignore it. The same with all of the other “known” warnings.

#256

Yeah… It is bound to happen. Maybe we just got lucky with the extra data above because someone did not check the Wiki first, LOL

#257

So maybe I am just stupid…but how do I know from the $twcweather.alerts data if something is new, especially if the alert Severe Thunderstorm Watch is in the database?

#258

Perhaps look at the entire alert, and compare other dataPoints, besides just the eventDescriptions…

Yes, it’s a bit more work for you guys, but keep in mind that I have to do that for every single post here on this thread…

I typically spend 15-20 minutes with a single alert (consisting of 49-63 dataPoints):

  • I do a quick visual scan looking for uniques,
  • Tidy up the JSON, (so it’s easier to read)
  • Add it to a spreadsheet, (so previous data is accessible)
  • Cross reference looking for new data,
  • Update the Wiki,
  • and then update the alternative sections in the Wiki…

… for every alert posted here!

#259

Got a new one:

$twcweather.alerts[0].eventDescription
“River Flood Warning”

$twcweather.alerts[0].headlineText
“River Flood Warning until FRI 8:00 PM EDT”

$twcweather.alerts
[[adminDistrict:Ohio, adminDistrictCode:OH, areaId:OHC093, areaName:Lorain County, areaTypeCode:C, categories:[[category:Met, categoryCode:2]], certainty:Unknown, certaintyCode:5, countryCode:US, countryName:UNITED STATES OF AMERICA, detailKey:9af5a1d2-6e62-3fd1-92d6-b1e7281d968d, disclaimer:null, effectiveTimeLocal:null, effectiveTimeLocalTimeZone:null, eventDescription:River Flood Warning, eventTrackingNumber:0016, expireTimeLocal:2020-03-20T20:00:00-04:00, expireTimeLocalTimeZone:EDT, expireTimeUTC:1584748800, flood:[floodCrestTimeLocal:2020-03-19T20:00:00-04:00, floodCrestTimeLocalTimeZone:EDT, floodEndTimeLocal:2020-03-20T14:00:00-04:00, floodEndTimeLocalTimeZone:EDT, floodImmediateCause:Excessive Rainfall, floodImmediateCauseCode:ER, floodLocationId:ELRO1, floodLocationName:Black River at Elyria, floodRecordStatus:A record flood is not expected, floodRecordStatusCode:NO, floodSeverity:Minor, floodSeverityCode:1, floodStartTimeLocal:2020-03-19T09:18:00-04:00, floodStartTimeLocalTimeZone:EDT], headlineText:River Flood Warning until FRI 8:00 PM EDT, ianaTimeZone:America/New_York, identifier:759478a2c4743e86152971c85acdd472, issueTimeLocal:2020-03-19T15:31:00-04:00, issueTimeLocalTimeZone:EDT, latitude:0, longitude:0, messageType:Update, messageTypeCode:2, officeAdminDistrict:Ohio, officeAdminDistrictCode:OH, officeCode:KCLE, officeCountryCode:US, officeName:Cleveland, onsetTimeLocal:null, onsetTimeLocalTimeZone:null, phenomena:FL, processTimeUTC:1584646287, productIdentifier:FLS, responseTypes:[[responseType:Avoid, responseTypeCode:5]], severity:Severe, severityCode:2, significance:W, source:National Weather Service, urgency:Unknown, urgencyCode:5]]

#260

That one is a gold mine… Thanks @dejavux2!!

In the near future, I will add:

  • eventDescription:River Flood Warning
  • floodSeverity:Minor
  • floodSeverityCode:1
  • floodRecordStatus:A record flood is not expected
  • floodRecordStatusCode:NO

plus at least two other dataPoints…

#261

The latest data has been added to the Wiki now.

#262

Hey @Pantheon, I just remembered a previous conversation that I had with @dejavux2… Would you please consider removing “Special Weather Statement” from {Alert_Data} in your earlier post?

The reason I am asking is because I have a hunch that there may be unique data yet unseen linked to phenomena and/or productIdentifier… but without that eventDescription slipping past your censor, I’ll never know for sure…

IE: “Special Weather Statement” sounds generic, so it may apply to many different possibilities.