Well, at least it looks pretty same with WU forecasts now. And, yes, there are still some cosmetic corrections, such as the different views on the same html file, when nothing else is changed but the orientation from horizontal to vertical.
Today I detected another (cosmetic) misbehavior in multiple_tables. When toggling hours in the first day (see the appended html file), the alignment is not correct. It looks like, if there is no 07:00 line and the moon rise should be displayed within this line, in the 19:00 line the table cells <td class='col-moon'></td> and <td class='col-moonphase'></td> are empty and thus the following table cells shift to the left. This might be the same problem as with the alignment of <div class='legend'> and the following <div class='summary'> table, where there is col width defined in the css being not appropriate for the data. So, the cell width depends on the content. This may be not so obvious, when the set/rise data are displayed without seconds. But I don't know, where the format is defined. In my case, showing the format hh:mm:ss needs more space than provided with the min col width defined in the css file. So, at what point should this cosmetic problem be taylored? [email protected] schrieb am Mittwoch, 26. August 2020 um 22:13:27 UTC+2: > I’ll continue to look for what that number should be. There does not > appear to be an equivalent for NWS. Either a rain line appears or not in > the NWS forecast downloaded. > > Having said that, I’m OK with changing 0 to 20, as you have done. I’ll > push that. > > > On Aug 26, 2020, at 11:38 AM, Vetti52 <[email protected]> wrote: > > > > And, I just modified forecast.py at line 2563, and changed "0" to "20": > > if precip_chance > 20 and (precip_type == 'rain' or precip_type == > 'snow'): > > Now, only rain icons are shown, when precipitation chance is above 20%. It > looks like the example at sailing.mit.edu, as linked above. I am happy > with that result for forecast_table.inc now. We could let Matthews work on > the rest, for my opinion. And there is more time for hard real time jobs... > And I have learned a little bit about python. Just a tiny little bit.... > Vetti52 schrieb am Mittwoch, 26. August 2020 um 16:31:23 UTC+2: > >> Thanks for your copy of the skin. >> I visited Matthew's wiki today, because I was wondering, that the link to >> an image there did not work anymore. Indeed, most of the images there are >> broken. Bad news, because I found the table presentation very impressive. >> But then I followed his link to >> http://sailing.mit.edu/weather/forecast.html as an example. And there >> you can see an example of the forecast_table, when choosing NWS, which >> currently seems to be the only one, that is working. There are currently >> icons for rain showers and thunderstorm, but only at lines with more than >> 30% POP. The legend tells weewx 3.9.0b2, so I guess, this is an early >> forecast version as well. This design looks pretty nice. So, I doubt, that >> Matthew would change this kind of data evaluation, if you would reproduce >> it the same way. Whenever you can spend some time.... >> >> [email protected] schrieb am Mittwoch, 26. August 2020 um 16:08:18 >> UTC+2: >> >>> I did make a change to better align the header icons on the single >>> strip. That’s something I did a long time ago for my copy when I >>> integrated it into my skin. You’ll have to download it by choosing Code -> >>> Download ZIP from the main page as I didn’t do a release for it. >>> >>> For your requests, such as not seeing a row in the summer months, or >>> not seeing a rain icon if POP is under 20%, it’s best if you copied these >>> examples and customized in a way that’s best for you. >>> >>> The request to not see a row for wind gust when the forecast has no wind >>> gust information seems reasonable. I might make that change sometime when >>> I am not so busy, but I do worry that all of these changes are going to >>> disappear when Matthew Wall updates his forecast extension. Remember, my >>> fork is planned to go out of existence once Matthew updates his code for >>> WeeWX 4. >>> >>> On Aug 26, 2020, at 3:37 AM, Vetti52 <[email protected]> wrote: >>> >>> >>> >>> When looking at single-strip.html from the sampler, there are only the >>> cloud icons, but no precipitation icons. So it is not completely comparable >>> with forecast_table. >>> >>> Besides, an there is an empty line below the wind (km/h) line, which I >>> guess to be for gust data. But WU does not provide gust. How to disable >>> this line for WU? I don't get into the concept of selecting the data >>> appriopriate to the forecaster. Although, there should be an overall >>> directive of which data are provided, e.g. driven by #set >>> $forecast_source = 'WU' >>> >>> Your question about, what is wrong in single-strip.html? >>> The date line is missing.The table starts with hours. Next, there are >>> three temp lines. The three temp icons have no descriptions. I interpret >>> them as temp-max, temp, and temp-min. The data from day values are shown in >>> temp-max and temp, those from night values in temp and temp-min. Is this >>> the intention? >>> After the empty gust line, I see wind direction icons followed by the >>> cloud icons. That's perfect. >>> Then, the table ends with three text lines: POP %, precipitation in mm >>> and snow in mm. Actually, the last line should only show, when needed, at >>> least not now, here in August. >>> Curiously, in the vertical strip the date is present. But the cloud >>> icons are double in size, so that the table gets huge. The rest of the >>> content is identical to the horizontal strip. Both forecasts looks the same >>> with Firefox, Safari, and Edge. I have no idea, why the date line is not >>> shown in horizontal mode, but in vertical mode, and the cloud icon size >>> differs, although both are absolutely identical, except of the orientation >>> setting. >>> >>> Concerning the rain icons, I think, that there should be no rain icon, >>> when POP is <=20% and/or the amount of precipitation is (close to) zero. >>> This would be logically, or not? >>> [email protected] schrieb am Dienstag, 25. August 2020 um 21:40:03 >>> UTC+2: >>> >>>> I had a look at single-strip.html and it looks good when I plug in youR >>>> latlong. >>>> >>>> Please concentrate on single-strip.html. And look at the logic in >>>> forecast_table.inc. >>>> >>>> Do you see anything that is wrong? I’m looking for exactly one thing >>>> to start with. Don’t say something like, I see the same icon for 10% >>>> chance of rain vs. 100% chance of rain unless you think the logic in >>>> forecast_table.inc is differentiating them and presenting different icons. >>>> >>>> My goal is not to change the logic and the icons, it is for the >>>> forecast to be working as intended. >>>> >>>> On Aug 25, 2020, at 12:03 PM, John Kline <[email protected]> wrote: >>>> >>>> >>>> >>>> I see your latlong, let me run it. >>>> >>>> On Aug 25, 2020, at 12:01 PM, John Kline <[email protected]> wrote: >>>> >>>> >>>> Please point me to a exactly one specific thing you think is not >>>> correct. I’m not suggesting you are wrong, but I’m very busy (I have a >>>> real job) and that will help me quickly find the issue. Also, please send >>>> your lat long. >>>> >>>> On Aug 25, 2020, at 11:57 AM, Vetti52 <[email protected]> wrote: >>>> >>>> >>>> Well, I am not pretty sure about the syncronization of json and html >>>> file. I append the most recent data as a ZIP file again. The latest WU >>>> forecast seems to be from 25-Aug-2020 19:05. And the json file is >>>> retrieved >>>> from just now, which should be in the same 3h period. >>>> May be, there are differences in time, but the forecast seems to be the >>>> same. >>>> And, there is some rain outside.... >>>> >>>> [email protected] schrieb am Dienstag, 25. August 2020 um 17:52:08 >>>> UTC+2: >>>> >>>>> Can you state with certainty that you waited for a new WU forecast to >>>>> be downloaded? >>>>> >>>>> On Aug 25, 2020, at 8:32 AM, Vetti52 <[email protected]> wrote: >>>>> >>>>> >>>>> >>>>> Ok, I understand (hopefully). Meanwhile I have an update to v3.4.0b8, >>>>> where I could see, that precipitation is in a separate column next to >>>>> POP. >>>>> So, you divide the WU icon set into two separate sets, one for cloud >>>>> cover, >>>>> and one for precipitation. That is a little bit more complex and needs >>>>> more >>>>> intelligence than just reading the iconCode and display, whatever it >>>>> means, >>>>> in one single row (or column). So you can gain simultaneous information >>>>> from different json entries and display the results separately, right? >>>>> >>>>> Actually, the next forecasts are full of "rain". But it would be >>>>> nicer, if the icons would differ, when the precipitation chance is <20% >>>>> or >>>>> 80%. Actually, all of the week forecast looks the same for rain, although >>>>> there are days with POP 100% and <20%. >>>>> Forecast for next Sunday is displaying the SC icon, presumably due to >>>>> WU cloudCover=42, WU iconCode would show icon 45, which is scattered >>>>> showers. I think, that different icons, as WU provides here, would reduce >>>>> irritation. >>>>> Just to let you reproduce, what I am trying to explain, I attach >>>>> another ZIP with the actual index.html and json files. >>>>> >>>>> And, again, I will give v3.4.0b9a spin. >>>>> [email protected] schrieb am Dienstag, 25. August 2020 um 16:35:16 >>>>> UTC+2: >>>>> >>>>>> >>>>>> WU_SKY_DICT is for cloud cover. What you are proposing is simply not >>>>>> a good fit for this extension. I won’t be creating these new icons. >>>>>> >>>>>> Having said that, there is definitely information to be gathered from >>>>>> the iconCode. Using that code, I can get many more precipitation types >>>>>> which can be combined with precipitation chance to derive which icon to >>>>>> add. >>>>>> >>>>>> As such, I’ve pushed v3.4.0b9. It adds support for the following >>>>>> icons in the WU forecast: tstms, hail, dirzzle, sleet, frzngdrzl, >>>>>> frzngrain, flurries. >>>>>> >>>>>> On Aug 25, 2020, at 4:03 AM, Vetti52 <[email protected]> wrote: >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> ok, I was thinking of keep it simple. So the correlation of iconCode >>>>>> to one icon each looked straight forward to me. Probably you can see, >>>>>> what >>>>>> I mean in the attached ZIP. There should be no need to have separate >>>>>> icons >>>>>> for each forecaster, but the idea was, that there is a simple one to one >>>>>> relation of WU iconCode to PNG file. So, there should be a complete set >>>>>> for >>>>>> all of the 47 iconCodes. The icon files could nevertheless be replaced, >>>>>> matching a common schema. >>>>>> >>>>>> I will give b8 a spin tonight. You may know, that I replaced the >>>>>> Zambretti narratives with German translations. So, I need some extra >>>>>> seconds for spinning. >>>>>> [email protected] schrieb am Montag, 24. August 2020 um 22:18:43 >>>>>> UTC+2: >>>>>> >>>>>>> That’s good news. >>>>>>> >>>>>>> The forecast extension relies on coercing the various forecast to a >>>>>>> standard representations that is stored in a common schema. >>>>>>> >>>>>>> As such, supporting arbitrary icons would not fit in with the scheme. >>>>>>> >>>>>>> WU does provide precipType and precipChance. precipType can be >>>>>>> rain, snow or precip. >>>>>>> >>>>>>> For cases where precipType is rain, or snow, I have now added the >>>>>>> appropriate icon (based on precipChance). >>>>>>> >>>>>>> Please give 3.4.0b8 a spin; and do wait for a new forecast to be >>>>>>> downloaded. That will be no sooner then 3 hours after the previous >>>>>>> forecast. >>>>>>> >>>>>>> On Aug 24, 2020, at 9:25 AM, Vetti52 <[email protected]> wrote: >>>>>>> >>>>>>> You are right! I assumed, that the forecast would be downloaded >>>>>>> immediately after installing and restarting. Now, three hours later, >>>>>>> the >>>>>>> cloud column is completely filled and all is correct. Great job! >>>>>>> >>>>>>> >>>>>>> >>>>>>> So, what about the idea to use the complete set of icons, including >>>>>>> day/night versions? Well, it is pure luxury. But to me mapping iconCode >>>>>>> versus PNG icons seems the most simple adoption. >>>>>>> >>>>>>> >>>>>>> I have already written down the WU codelist with descriptions >>>>>>> provided from the weather-watch forum, just for better understanding >>>>>>> and >>>>>>> may be for future use. And, if I can help drawing the png icons, I >>>>>>> could >>>>>>> support. >>>>>>> [email protected] schrieb am Montag, 24. August 2020 um 17:45:41 >>>>>>> UTC+2: >>>>>>> >>>>>>>> Actually, the most likely problem is that you didn’t wait for a new >>>>>>>> WU forecast to be downloaded after installing and restarting. That >>>>>>>> won’t >>>>>>>> happen until at least three hours after the last forecast. Before you >>>>>>>> do >>>>>>>> what I asked below, please check these reports again. Do they not >>>>>>>> look >>>>>>>> correct? >>>>>>>> >>>>>>>> On Aug 24, 2020, at 7:08 AM, John Kline <[email protected]> >>>>>>>> wrote: >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> Send the contents of forecast/single-table.html and >>>>>>>> forecast/forecast-periods.html that are generated. Also send the >>>>>>>> contents >>>>>>>> of a contemporaneous WU API call. Don’t send excerpts of either. >>>>>>>> >>>>>>>> Are you sure you installed and restarted weewx? I ask because I >>>>>>>> get icons for every day in single-table.html and descriptions for >>>>>>>> every >>>>>>>> period in forecast-periods.html. Please verify that you see 3.4.0b7 >>>>>>>> in the >>>>>>>> log. >>>>>>>> >>>>>>>> As for the extra icons you would like to see on the right, I would >>>>>>>> need a way to generate them. Probably, over time I could do it from >>>>>>>> the >>>>>>>> English descriptions. I could also do it by looking at the icons they >>>>>>>> provide and map them (as was done by the author of the post you >>>>>>>> provided). >>>>>>>> I note that these extra icons were not generated for the prior WU >>>>>>>> daily >>>>>>>> forecast. Let me think about whether or not I want to add that >>>>>>>> functionality. >>>>>>>> >>>>>>>> On Aug 24, 2020, at 3:36 AM, Vetti52 <[email protected]> wrote: >>>>>>>> >>>>>>>> >>>>>>>> I hate the new google forum surface! >>>>>>>> >>>>>>>> The link should read: >>>>>>>> >>>>>>>> https://www.weather-watch.com/smf/index.php/topic,67656.msg543684.html#msg543684 >>>>>>>> >>>>>>>> >>>>>>>> and for clarification of the missing entries, it is much better to >>>>>>>> look at the Forecast Sampler: Periods >>>>>>>> >>>>>>>> *Periods from Selected Forecast Sources* >>>>>>>> >>>>>>>> Display specific portions of a forecast, from various forecast >>>>>>>> sources. This illustrates the use of the forecast search list >>>>>>>> extension. >>>>>>>> *Weather Underground* >>>>>>>> >>>>>>>> 24-Aug-2020 19:00 11,7°C >>>>>>>> 25-Aug-2020 07:00 20,6°C >>>>>>>> 25-Aug-2020 19:00 14,4°C >>>>>>>> 26-Aug-2020 07:00 17,8°C >>>>>>>> 26-Aug-2020 19:00 12,8°C >>>>>>>> 27-Aug-2020 07:00 18,9°C Considerable Cloudiness >>>>>>>> >>>>>>>> Issued at 24-Aug-2020 09:20 for geocode=54.09,10.26 >>>>>>>> To my intension, every line should end with a description, such as >>>>>>>> 'Mostly cloudy'. Well, actually I miss 'Clear', but this is, due to >>>>>>>> the >>>>>>>> current weather situtation, not an option. >>>>>>>> >>>>>>>> Vetti52 schrieb am Montag, 24. August 2020 um 11:45:12 UTC+2: >>>>>>>> >>>>>>>>> I found this link for assigning the iconCode entries from the WU >>>>>>>>> json file to the meaning, along with some icons: >>>>>>>>> >>>>>>>>> https://www.weather-watch.com/smf/index.php/topic,67656.msg543684.html?PHPSESSID=9cntj0deu62o4c9q6jasslpqph#msg543684 >>>>>>>>> >>>>>>>>> After update to version b7, there is still only one icon visible, >>>>>>>>> which is on thursday. I guess, this corresponds to number 6 >>>>>>>>> (Considerable >>>>>>>>> Cloudiness). No matter, which of the forecast sampler files I look at. >>>>>>>>> Here my actual request: >>>>>>>>> cloudCover >>>>>>>>> 0 85 >>>>>>>>> 1 64 >>>>>>>>> 2 76 >>>>>>>>> 3 84 >>>>>>>>> 4 96 >>>>>>>>> 5 64 >>>>>>>>> 6 60 >>>>>>>>> 7 76 >>>>>>>>> 8 84 >>>>>>>>> 9 65 >>>>>>>>> 10 72 >>>>>>>>> 11 77 >>>>>>>>> iconCode >>>>>>>>> 0 11 >>>>>>>>> 1 29 >>>>>>>>> 2 11 >>>>>>>>> 3 12 >>>>>>>>> 4 12 >>>>>>>>> 5 11 >>>>>>>>> 6 30 >>>>>>>>> 7 45 >>>>>>>>> 8 11 >>>>>>>>> 9 11 >>>>>>>>> 10 11 >>>>>>>>> 11 11 >>>>>>>>> wxPhraseShort >>>>>>>>> 0 "Showers" >>>>>>>>> 1 "P Cloudy" >>>>>>>>> 2 "PM Lgt Rain" >>>>>>>>> 3 "Rain" >>>>>>>>> 4 "Rain" >>>>>>>>> 5 "Rain Early" >>>>>>>>> 6 "P Cloudy" >>>>>>>>> 7 "Shwrs Late" >>>>>>>>> 8 "Showers" >>>>>>>>> 9 "Showers" >>>>>>>>> 10 "Showers" >>>>>>>>> 11 "Showers" >>>>>>>>> >>>>>>>>> Am Montag, 24. August 2020 10:18:12 UTC+2 schrieb Vetti52: >>>>>>>>>> >>>>>>>>>> I was thinking about a graphic demonstration like that in the >>>>>>>>>> wiki: >>>>>>>>>> >>>>>>>>>> https://camo.githubusercontent.com/8b5474131edc812687002ab075ec51e1ddb8bd82/687474703a2f2f6c616e6365742e6d69742e6564752f6d77616c6c2f70726f6a656374732f776561746865722f77656577782d666f7265636173742d7461626c652e706e67 >>>>>>>>>> where all states are shown. Thus, there is no empty field in the >>>>>>>>>> cloud column at all. >>>>>>>>>> Does "pull head" mean just to update to the latest version, >>>>>>>>>> 3.4.0b7? Sorry, but I am dumb.... >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> -- >>>>>>>> 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 [email protected]. >>>>>>>> To view this discussion on the web visit >>>>>>>> https://groups.google.com/d/msgid/weewx-user/5e5f50cb-19fe-42e7-a558-9c12da34ea85n%40googlegroups.com >>>>>>>> >>>>>>>> <https://groups.google.com/d/msgid/weewx-user/5e5f50cb-19fe-42e7-a558-9c12da34ea85n%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>>>>> . >>>>>>>> >>>>>>>> -- >>>>>>> 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 [email protected]. >>>>>>> >>>>>>> To view this discussion on the web visit >>>>>>> https://groups.google.com/d/msgid/weewx-user/e80b093b-2357-47b2-9e87-b0846febe3can%40googlegroups.com >>>>>>> >>>>>>> <https://groups.google.com/d/msgid/weewx-user/e80b093b-2357-47b2-9e87-b0846febe3can%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>>>> . >>>>>>> >>>>>>> -- >>>>>> 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 [email protected]. >>>>>> >>>>>> To view this discussion on the web visit >>>>>> https://groups.google.com/d/msgid/weewx-user/a5fbeeac-0357-4303-852f-1eba383709b2n%40googlegroups.com >>>>>> >>>>>> <https://groups.google.com/d/msgid/weewx-user/a5fbeeac-0357-4303-852f-1eba383709b2n%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>>> . >>>>>> <WU_Code.zip> >>>>>> >>>>>> -- >>>>> 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 [email protected]. >>>>> >>>>> To view this discussion on the web visit >>>>> https://groups.google.com/d/msgid/weewx-user/6e2d2064-655c-4ba2-b944-8d4b76ef9845n%40googlegroups.com >>>>> >>>>> <https://groups.google.com/d/msgid/weewx-user/6e2d2064-655c-4ba2-b944-8d4b76ef9845n%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>> . >>>>> <5day.zip> >>>>> >>>>> -- >>>> 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 [email protected]. >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/weewx-user/6b7e5e2c-d1c0-41a6-8e84-1de1fe3695ecn%40googlegroups.com >>>> >>>> <https://groups.google.com/d/msgid/weewx-user/6b7e5e2c-d1c0-41a6-8e84-1de1fe3695ecn%40googlegroups.com?utm_medium=email&utm_source=footer> >>>> . >>>> <5day.zip> >>>> >>>> -- >>> 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 [email protected]. >>> >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/weewx-user/65bc4520-5b6d-4391-bf24-28fb3231b154n%40googlegroups.com >>> >>> <https://groups.google.com/d/msgid/weewx-user/65bc4520-5b6d-4391-bf24-28fb3231b154n%40googlegroups.com?utm_medium=email&utm_source=footer> >>> . >>> >>> -- > 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 [email protected]. > > To view this discussion on the web visit > https://groups.google.com/d/msgid/weewx-user/cc859425-625c-4ee4-a8c8-e322f1fd20b7n%40googlegroups.com > > <https://groups.google.com/d/msgid/weewx-user/cc859425-625c-4ee4-a8c8-e322f1fd20b7n%40googlegroups.com?utm_medium=email&utm_source=footer> > . > > -- 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 [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/37f613df-4d79-4497-a24e-a0739936c51an%40googlegroups.com.Title: Forecast Sampler: Multiple Tables
Forecast Table
Display forecast data in multiple tables, one table for each forecast source.
Weather Underground
°C |
°C |
km/h |
mm |
|
Do 27 Aug |
|
11 11 |
N/A N/A |
10 |
06:18:17 20:21:12 |
17:17:11 00:32:33 |
71% |
50%
0,51 |
|
|
Fr 28 Aug |
|
20 12 |
N/A N/A |
10-16 |
06:20:05 20:18:49 |
18:18:50 00:32:33 |
76% |
80%
5,08 |
|
|
Sa 29 Aug |
|
21 11 |
N/A N/A |
11-19 |
06:21:52 20:16:25 |
19:06:34 01:27:06 |
84% |
20%
|
|
So 30 Aug |
|
20 11 |
N/A N/A |
13-14 |
06:23:40 20:14:01 |
19:41:58 02:32:21 |
91% |
40%
1,02 |
|
|
Mo 31 Aug |
|
18 8 |
N/A N/A |
11-16 |
06:25:27 20:11:36 |
20:08:11 03:44:21 |
96% |
20%
|
|
Di 01 Sep |
|
19 11 |
N/A N/A |
10-13 |
06:27:14 20:09:10 |
20:28:13 04:58:47 |
99% |
30%
0,76 |
|
WU forecast
for geocode=54.09,10.26
issued on 27-Aug-2020 14:05
report generated in 0.187 s
NWS
No data for forecast NWS
NWS forecast
report generated in 0.008 s
Aeris
No data for forecast Aeris
Aeris forecast
report generated in 0.008 s
