I was working on a patch that visualises and sonifies weather data posted online in an HTML document. I was using [httpget] to grab the document and using [listsplit] and [symbol2list] to parse the necessary data. However, using this method, I will often get the error "error: unpack: type mismatch". I was wondering if someone could look at the patch and advise me of a cleaner way of doing this or showing
me what I might be doing wrong? It works 70% the time. Initially, I was just increasing the frequency of the [httpget] to ensure that the data changes are captured but this method isn't very clean and would appreciate any corrections or feedback.

This Weather:
[httpget]: http://puredata.info/docs/tutorials/SimpleWebclient

http://www.pdpatchrepo.info/hurleur/york.pd

I've managed to do this more efficiently using Ruby.
http://philipcunningham.org/2011/03/05/university-york-weather-parser/

However, I'd still like a Pd-only solution.

bump.

i looked at your patch and i have no idea where this type missmatches are coming from.
i tried to parse the list charakter by charakter, but this makes a new problem, as the length of the values in digits may change. very interesting patch though...sorry that i cant help.

It seems that, for whatever reason, [httpget] isn't always getting the full html page. Try connecting [httpget] to a [prepend set] -> [messagebox( to see what I mean. Whenever you get the type mismatch errors, it's because you're missing the first few lines of code.

You could do a test to make sure you received the whole page. I noticed that it works every time the list length is 228. So you could just test that it's the correct length first. If so, parse it; if not, retrigger [httpget].

This is very interesting.

i'm looking to extract the values for

• Arg of perigee:

-Inclination:

-Mean motion:

Using the york.pd httpget patch on this url:

http://spaceflight.nasa.gov/realdata/sightings/SSapplications/Post/JavaSSOP/orbit/ISS/SVPOST.html

Any suggestions on how I could do it?

Thanks in advance