Need help to read value from web

---

Hello,

i´m trying to read the values of a forecast.

http://www.wetter.com/wetter_aktuell/wettervorhersage/3_tagesvorhersage/?id=DE0000416

I need the values from “rainy”

e.g.:
<span class=“umb-icon rainy”>
5,0 l/m² <<< This value.

I the summer i have to know if it will rain the next days so i can create a rule that the watering system in my garden won´t start one day before.

I know that i can achive it with:
curl -s http://www.wetter.com/wetter_aktuell/wettervorhersage/3_tagesvorhersage/?id=DE0000416 |grep rainy|
But don´t know the next parameters…

@hyperx
sorry, this won’t work that way, since you’ll get only the source code from the website and no data at all. you’ll have to use the API, but for the wetter.com API you’ll have to register first.
I’ve registered but do not get any activation email from them. Will give it a little more time.

@hyperx you can try that

curl -s http://www.wetter.com/wetter_aktuell/wettervorhersage/3_tagesvorhersage/?id=DE0000416 |grep 'l/m' | head -1 | sed 's#l/m².*##'

and

curl -s http://www.wetter.com/wetter_aktuell/wettervorhersage/3_tagesvorhersage/?id=DE0000416 |grep 'l/m' | head -2 | tail -1 | sed 's#l/m².*##'

for the day after

Perfect works ! but there are some empty spaces… hope that this has no effect for the variable

@hyperx said:

Perfect works ! but there are some empty spaces… hope that this has no effect for the variable

so you can do

curl -s http://www.wetter.com/wetter_aktuell/wettervorhersage/3_tagesvorhersage/?id=DE0000416 |grep 'l/m' | head -2 | tail -1 | sed 's#l/m².*##' | tr -d ' '

Perfekt, this is exactly what i need. …
btw. i have changed the url because it was for the 3 days forecast, so the value was wrong.

thanks yves!!
sorry for telling some crap above.
just learned some more

Hello,

i have to pick up this thread again with a new question:

I am still reading the values from this website, but something has changes, so i have to change my formular… but now i´m running in some new problems:

Still i want to pick the rain forecast from wetter.com, but look at the result for “today”

curl -s http://www.wetter.com/wetter_aktuell/wettervorhersage/7_tagesvorhersage/deutschland/au/DE0000416.html |grep 'l/m' | head -27 | tail -1 | sed 's#l/m².*##' | tr -d ' ' | tail -c 4

the result should be “2” but if the result arn´t numbers at the last 3 digits, the result is WRONG for pimatic.

The result for the forecast “tomorrow” is correct:

curl -s http://www.wetter.com/wetter_aktuell/wettervorhersage/7_tagesvorhersage/deutschland/au/DE0000416.html |grep 'l/m' | head -28 | tail -1 | sed 's#l/m².*##' | tr -d ' ' | tail -c 4

because here we have all numbers on the 3 last digits.

Any idea how to extract the correct values?
Thank you !