simple question about loadStrings() and URLs

Programming Questions

NinjaChris

simple question about loadStrings() and URLs

in Programming Questions • 4 months ago

Hello, I'd like a widget to log my Xively (previously Cosm, Pachube) data to file. The data can be grabbed using their API. You enter a URL in your browser and you get CSV results. Eg: http://api.xively.com/v2/feeds/76854.csv gives me the six latest readings from the weather station I've got sending data to Xively in the CSV structure <datastream name>, <timestamp>, <data>.

When I use loadStrings() to try and grab this data I get a 401. Any other URL I enter gives me the text of the html. Eg.

http://processing.org/about/index.html

https://xively.com/dev/docs/api/data/read/

Can anyone tell me why this is the case?

Thanks,

EDIT

Here's some code you can comment/uncomment to test the different URLs. The first two are for my data and don't work. The other two are just web pages and work. One of them is from the Xively site itself so there's no problem connecting to that server in particular.

//String[] lines = loadStrings("http://api.xively.com/v2/feeds/76854/datastreams/BMPp.csv");

//String[] lines = loadStrings("http://api.xively.com/v2/feeds/76854.xml");

String[] lines = loadStrings("https://xively.com/dev/docs/api/data/read/");

//String[] lines = loadStrings("http://processing.org/about/index.html");

println(lines);

exit();

If you copy any of these URLs into your browser they all work as desired, just not in Processing.

Replies(5)

cageehv

Re: simple question about loadStrings() and URLs

4 months ago

Hey,

Maybe you should use the http://www.eeml.org/library/ library for retrieving the feeds.
I used it before and it worked fine for me.

peace,
Rolf

NinjaChris

Re: simple question about loadStrings() and URLs

4 months ago

I hadn't heard of that one before. I'll give it a try, thanks. I'd still like to understand what's going on with the URL request though.

cageehv

Re: simple question about loadStrings() and URLs

4 months ago

Hey NC,

While using your browser you are probably logged in to Xively... If you are using it from Processing you'll need your API key...

My two pence ;-)

peace,
Rolf

PhiLho

Re: simple question about loadStrings() and URLs

4 months ago

As said. If I go to the first URL you give, the browser is asking me credentials. loadStrings() isn't able to provide them. It won't provide the proper cookies. Check their API to see if you can provide them via the URL. Or perhaps you need to POST the requests, which is a bit more convoluted.

hex705

Re: simple question about loadStrings() and URLs

4 months ago

Just a note:

the EEML library uses API v1 -- which is deprecated.

You an use it with Xively but you must ensure that you change your request URL to indicate the old API number:

example -- a Xively API endpoint URL that is:

https://api.xively.com/v2/feeds/YOUR_FEED_NUMBER

would need to be changed to :

https://api.xively.com/v1/feeds/YOUR_FEED_NUMBER // <----- note v1 in URL

to be used in the EEML constructors.

Otherwise you will get lots of 404 errors.

Once you do this, also note that you will only be able to address channels in which the ID is an INT.

Xively (API v2) allows channel names to be strings -- but EEML can only handle IDs defined as INT.

see here for lots more detail:

https://xively.com/dev/docs/api/communicating/versions/

Top Reply