Custom graph anomaly

[Blicj11]

I have a custom graph that displays a weird blip I cannot explain. The graph displays accumulated wind run. At the beginning of most, but not all, hours, it displays an inaccurate increase in wind run, followed by a decrease back to the accurate amount. See the attached screenshot.

Any ideas?

[elagache]

Dear Blick and WeatherCat troubleshooters,

I have a custom graph that displays a weird blip I cannot explain. The graph displays accumulated wind run. At the beginning of most, but not all, hours, it displays an inaccurate increase in wind run, followed by a decrease back to the accurate amount. See the attached screenshot.

I have not observed exactly the phenomenon you are seeing, but I graph the accumulated wind run weekly and instead see a puzzling "blip" at the very start of the week.  Here is an example:



Here is the AppleScript I use for the synthetic channel:

Code: [Select]

-- Channel: Weekly Wind Run

set Query to "STAT$WINDRUN:TOTAL:THISWEEK$"
set result to QueryResult
I don't see anything that would explain the blip in the AppleScript code unless it has something to do with the STAT$ query itself.

In my case I was wondering if WeatherCat has an issue with the transition from one week to the next.  I'll have to see where the wind run ends up to see if that has something to do with the blip.

Could you share with us the exact AppleScript in your synthetic channel so that we can attempt to reproduce it?

Cheers, Edouard

[Blicj11]

Here is my Apple script.

Code: [Select]

-- This Query tracks daily wind run and resets each day
set Query to "STAT$WINDRUN:TOTAL:TODAY$"
set result to QueryResult

[elagache]

Dear Blick and WeatherCat troubleshooters,

Here is my Apple script.

Okay, I've replicated your synthetic channel and graph.  Let me run it for a day to see if it repeats what you are seeing.

Cheers, Edouard

[Blicj11]

Thank you my friend. Hope you are feeling better each better.

[elagache]

Dear Blick and WeatherCat troubleshooters,

Okay, I've replicated your synthetic channel and graph.  Let me run it for a day to see if it repeats what you are seeing.

24 hours have passed and the answer is . . . . . . . .



So it isn't your imagination.  Alas, I can reproduce it as well.  We will have to wait for Stu to take a look at this and see if he can understand what is going on.

Cheers, Edouard

[xairbusdriver]

Looks like something is triggering a 1mph signal on nearly every hour. Is there a wind event that occurs during the previous hour but that is computed for Wind Run at the end of that hour? I’m not sure how/when a Wind Run value is created/computed. I use that only in a custom gauge.

[Blicj11]

So it isn't your imagination.  Alas, I can reproduce it as well.  We will have to wait for Stu to take a look at this and see if he can understand what is going on.

Edouard, a sincere thank you for running this script. The hourly blip is not a new thing; this has been happening for several years. I just never paid any attention to it.

[elagache]

Dear X-Air, Blick, and WeatherCat troubleshooters,

I had to reboot my Mac and had some troubles with this graph.  It "lost" the interval to graph the data.  When I straightened all that out, the hourly anomaly disappeared:



You might try simply restarting WeatherCat during the day sometime and see if the anomaly persists.  That might be a clue for Stu to untangle what the issue is.

Cheers, Edouard

[Blicj11]

Thanks for the suggestion, but restarting WC  did not clear out my not quite every hourly anomaly.

[elagache]

Dear Blick and WeatherCat troubleshooters,

Thanks for the suggestion, but restarting WC  did not clear out my not quite every hourly anomaly.

Strangely enough, for the second day I am no longer seeing the blips:



The other thing I did was to fuss with the window settings.  In particular I changed the period of the anchored data back to a day.  You might try changing the period and changing it back to a day and see what that does.   Even if it isn't a cure, it might give Stu some clues about where the problem lies in the code.

Cheers, Edouard

[Blicj11]

That's weird. I see 'em every day, but thanks for checking up on my script. I've seen it on two different iMacs over the past few years.

[elagache]

Dear Blick and WeatherCat troubleshooters,

That's weird. I see 'em every day, but thanks for checking up on my script. I've seen it on two different iMacs over the past few years.

Alas, after a while they returned and my attempt to make the go away has failed.  Switching to scrolling data at 3 day interval makes them go away.  However, even after a WeatherCat restart, returning to anchored data, and a 1 day interval brings back the dreaded tick marks.

So I suppose we will have to wait for Stu to scrape up the time to look into this.

Cheers, Edouard

[Blicj11]

Thank you so much for taking the time to play with this and see what you could learn. You have gone beyond the call of duty and I am grateful for your efforts.
Hope you are feeling better too!