Yet Another Dayfile Reader (PHP)

[Dennisdg]

beteljuice

Thanks, I must have mixed some file versions.
Now OK.
Looks great.

[beteljuice]

Weather By You (meteo) experimenters
Has the beteljuice ever mentioned he hates css ?

Try this replacement css file please: Edit: Attachment removed

I've got an in-line Vs stylesheet precedence conflict (sounds good anyway)

... off to the PUB

[joewith]

Hi beteljuice,

Has the beteljuice ever mentioned he hates css ?

Yes, I think I've seen you mention that on a couple of ocassions.

I managed to download your replacement code (before you removed it) and tried it out. The replacement code at least got the troublesome areas inside the box.
I tried tweaking some of the numbers you changed, but only managed to make things worse so I currently am using the replacement code as you sent it out.

I've got an in-line Vs stylesheet precidence conflict (sounds good anyway)

Way above my pay grade!

... off to the PUB

I wish I could go as well.

Thanks for all your efforts on this whole project, I'm sure its appreciated by quite a few others as well!

[westonweather]

Thanks for all your efforts on this whole project, I'm sure its appreciated by quite a few others as well!

Amen!

[KarlS]

Once the Beteljuice has had some (well-earned) rest, I would like to make some suggestions for the upcoming ver3.0 of this script?

• Tdiff: Does it make sense to treat a temperature difference of 0 as a missing value (like you do for the windrun)? Please look at my Tdiff data for 2013 <here>
• Tdiff: Is it possible to have different colours for the temperature difference if the original min/max values were below 0? At the moment it shows a balmy orange-red for a min value of -35 and a max of -10 … anything but balmy. Examples: February 2014
• Rainfall: We tend to have <100 rainy days/year, so over 2/3 of all cells contain the value 0. Also, for 5-6 months every year the precipitation in our part of the world falls in the form of snow, so a value of 0 is actually misleading. Is it possible to just leave these cells empty?
• Windrun: The windrun is only displayed as kilometres, a conversion to miles is not offered.

I know that these changes are mostly only cosmetic and I could possibly do them myself. However, maybe some other users of this great script might benefit from these suggestions too.

[beteljuice]

Weather By You (meteo) Experiment

Can somebody please try this replacement main file AND css file.
The beteljuice has pretty much lost the plot trying css tweaks !!!!
EDIT: link removed - no longer relevent
This should correctly keep space for the title and 'pickers' but ...
they may be 10 px low of middle in the border - in which case try modding ccs:

[beteljuice]

[*]Tdiff: Does it make sense to treat a temperature difference of 0 as a missing value (like you do for the windrun)? Please look at my Tdiff data for 2013

30th Nov ?
It is assumed that your weather station at least records temperature !
Further - zero for temperature is a valid value. - What more can I do ???

[*]Tdiff: Is it possible to have different colours for the temperature difference if the original min/max values were below 0? At the moment it shows a balmy orange-red for a min value of -35 and a max of -10 … anything but balmy. Examples: February 2014

Err ... T diff is not a temperature it's err ... T diff and the colour reflects the magnitude of the DIFFERENCE.

[*]Rainfall: We tend to have <100 rainy days/year, so over 2/3 of all cells contain the value 0. Also, for 5-6 months every year the precipitation in our part of the world falls in the form of snow, so a value of 0 is actually misleading. Is it possible to just leave these cells empty?

You could add it to $zero_as_null - but what would be the point ? - averages etc. would be total rubbish.

[*]Windrun: The windrun is only displayed as kilometres, a conversion to miles is not offered.[/list]

Windrun is displayed as miles, kilometres and nautical miles - guess what, it makes the decision on what windspeed you've asked for

I know that these changes are mostly only cosmetic .....

Did you not read the pages of problems of dealing with zeros and nulls ?

[KarlS]

Sorry Beteljuice, I don't want to get into a heated discussion about this, especially since I highly appreciate the work you put into this. However:

It is assumed that your weather station at least records temperature !

Why is that? Like any other component of a weather station a temperature sensor can fail, hence missing data.

Further - zero for temperature is a valid value. - What more can I do ???

That is not the point. I’m talking about temperature differences, where a value of 0 is HIGHLY unlikely.

Err ... T diff is not a temperature it's err ... T diff and the colour reflects the magnitude of the DIFFERENCE.

Absolutely right: it’s not a temperature. So why use the temperature colour scheme?

You could add it to $zero_as_null - but what would be the point ? - averages etc. would be total rubbish.

Why should I introduce an error when all I want to do is having a better display of the data. You could argue how meaningful averages for rainfall are anyway. That’s why you included the (much more useful) Totals! And that still does not address the fact that a value of 0 is misleading when you had 10cm of snow at the same time.

Windrun is displayed as miles, kilometres and nautical miles - guess what, it makes the decision on what windspeed you've asked for

This is obvious to everybody who reads the source code, but not to the visitors of your web site. For them it’s just an error/omission. At the very least, decisions that the program makes for the user should be transparent to the user. And why shouldn't I be able to change my mind about km vs. mi vs. nm at ANY time?

[beteljuice]

All getting a bit esoteric

Some weather stations and software (including some stuff I've coded) use the outdoor temp as proof of valid data, else that entire log entry is junked.

Tdiff:
The code is already 'validating' all over the place, and as long as the code believes there is a Hi temp AND a low temp it carries on regardless.
O.K. yet another check for the unlikely result of zero could be introduced, but by the reasons above there really shouldn't be a temp 'null' as that would would also mean several other chunks of data are incorrect.

If you don't want temp gradient for Tdiff because it's a relative temperature, why do you want to somehow give a different display if the difference happens to have occured in the 'sub-zero' temperatures ?

Rain Vs Snow - entirely different ball game, all the code can deal with is bucket preciptation !

At the very least, decisions that the program makes for the user should be transparent to the user.

Just how many decisions at how many levels do you suppose are taking place, it would take a few pages and a glossary to cover them.

...And why shouldn't I be able to change my mind about km vs. mi vs. nm at ANY time?

... because that's the way I wrote it ... that and because the reverse process (setting windspeed by windrun) can't be done.

[KarlS]

... because that's the way I wrote it ...

Fair enough. But if you put your code into the public domain, you should expect others to make suggestions to improve the usability of the program. And a few suggestions were all I made ...

[joewith]

Hi beteljuice,

I tried the recent Version you made available that becomes Version 2.1 and as far as I can see every thing appears to be OK!

This should correctly keep space for the title and 'pickers' but ...
they may be 10 px low of middle in the border - in which case try modding ccs:

The title and 'pickers' all appear within the borders without the extra modding of the css unless you see something differenty.

http://www.tigerdenweb.com/tigerdenwx/d ... ary900.php

[beteljuice]

Still not right - must be some interaction somewhere ...

Try adding
height: 22px;

to the #table_name .table_title {

[joewith]

Hi beteljuice,

I tried "height: 22px;", but only saw a little change so I started experimenting and it looks like "height: 32px;" seems to do the trick...I think.

http://www.tigerdenweb.com/tigerdenwx/d ... ary900.php

Thanks,

[beteljuice]

Yep - That's it - Thank You

Typo by me - the beteljuice had become a litle distracted.

There will be another 'release' later.

[beteljuice]

V3.0 Release

Download on first post ..

23 Feb 2014 - v3.0

• ALL css files updated because
  betel_readDayfile modified - DO NOT MIX OLD AND NEW !
  *new* Weather By You (meteo) css
  readDayfile modded - fix to language working in saratogaWX environment
  language files - Czech added, thank you milos.jirik
  "How To .." file updated