We Keep Coding

WebClient failing on physical device and emulator

UPDATE:
Maybe it's just me not understanding how oAuth works? I tried running
the query manually on http://www.apikitchen.com and I get a 400 error
there too! Just to be sure, am I constructing the URL correctly here?
POST URL:
https://api.bufferapp.com/1/oauth2/token.json?client_id=[hidden]&client_secret=[hidden]&redirect_uri=http://apps.joel-murphy.com/buffer/code/success.php&code=[the
access code I get from buffer starting with
1/]&grant_type=authorization_code
Original post:
I'm building a Windows Phone application which requires the use of data from a website. The website uses oAuth to authenticate users.
I used the built in web browser control to make a GET request to authenticate users. The official documentation requires the URL structure to be like this:
GET https://bufferapp.com/oauth2/authorize?
client_id=...&
redirect_uri=...&
response_type=code
This part of my app works. Although when it comes to exchanging the Authorization token for an access token from the server, I am facing problems. The official documentation requires the URL structure to be like this:
POST https://api.bufferapp.com/1/oauth2/token.json?
client_id=...&
client_secret=...&
redirect_uri=...&
code=...&
grant_type=authorization_code
Correct me if I'm wrong, but from what I know there is no way to make a POST request from a browser, unless submitting a form. For this reason, I have decided to use the WebClient class to submit data to the server. However, no matter if I run my code on an actual device or on the Visual studio emulator I always receive the following error:
The remote server returned an error: NotFound
Does anyone have any idea what's wrong with the following code? I've spent over 5 hours across 2 days trying to solve this error, but nothing seems to be working.
The code I'm using:
WebClient wc = new WebClient();
wc.UploadStringCompleted += new UploadStringCompletedEventHandler(wc_UploadStringCompleted);
wc.UploadStringAsync(new Uri(access_token_url), "POST", GetPostParameters());
void wc_UploadStringCompleted(object sender, UploadStringCompletedEventArgs e)
{
try
{
MessageBox.Show(e.Result);
}
catch (Exception ex)
{
MessageBox.Show(ex.Message);
}
}
string GetPostParameters()
{
string data = "";
data += "client_id="+client_id + "&";
data += "client_secret=" +client_secret + "&";
data += "redirect_uri=" + redirect_uri+ "&";
data += "code=" + App.AccessToken + "&";
data += "grant_type=authorization_code";
return data;
}
Does anyone have any idea what's wrong? this is driving me crazy and it's a real shame that oauth has to be so complicated when it's such a used technology nowadays.
Thanks in advance.

Can you try URL-encoding the redirect_uri variable?

How to read the query string when it contains unencoded data?

I have the below asp.net page which accepts a "url" query string key whose value can be an un-encoded url:
http://localhost:4104/WebSiteForTest/TinyUrl.aspx?url=http://www.google.co.uk/#hl=en&q=life&oq=life&aq=f&aqi=g-s1g9&aql=&gs_sm=3&gs_upl=2803373l2803701l2l2803826l4l4l0l0l0l0l188l453l0.3l3l0&bav=on.2,or.r_gc.r_pw.r_cp.,cf.osb&fp=94681dc4659502d1&biw=1680&bih=883
Now from this page, how would that be possible to read the text after ".aspx?"?
I checked the Request.Url.AbsoluteUri property and it only showed
"http://localhost:4104/WebSiteForTest/TinyUrl.aspx?url=http://www.google.co.uk/"
I also checked with the Request.QueryString with the below code:
private void getQueryString()
{
var sb = new StringBuilder();
var queryStringCount = Request.QueryString.Keys.Count;
for (int keyIndex = 0; keyIndex < queryStringCount; keyIndex++)
{
sb.Append(Request.QueryString.Keys[keyIndex]).Append("=").Append(Request.QueryString[keyIndex]);
if (keyIndex != (queryStringCount - 1))
{
sb.Append("&");
}
}
}
However, the code after "#" doesn't appear in any query string.
how would that be possible to read the text after ".aspx?"?
if you say it's not possible, how Google uses "#" in their url then when you search for something?!
http://www.google.co.uk/#hl=en&site=&q=life&oq=life&aq=f&aqi=g-s1g9&aql=&gs_sm=3&gs_upl=3317l3630l0l3755l4l4l0l0l0l0l125l391l3.1l4l0&bav=on.2,or.r_gc.r_pw.r_cp.,cf.osb&fp=94681dc4659502d1&biw=1680&bih=849
Thanks,

It's not possible to get value after anchor on server side, you can check this with fiddler or something similar, you should deal with this on client. Browser simply strips all after anchor.
Retrieving Anchor Link In URL for ASP.Net
c# get complete URL with "#"
Update:
I don't know how google exactly do this, but if you look with fiddler after initial request there goes another without #, here is a fidller log for request from your question :
so my advice is look with fiddler how google do this, or maybe ask another question

Use Request.QueryString
http://localhost:4104/WebSiteForTest/TinyUrl.aspx?url=http://www.google.co.uk/#hl=en&q=life&oq=life&aq=f&aqi=g-s1g9&aql=&gs_sm=3&gs_upl=2803373l2803701l2l2803826l4l4l0l0l0l0l188l453l0.3l3l0&bav=on.2,or.r_gc.r_pw.r_cp.,cf.osb&fp=94681dc4659502d1&biw=1680&bih=883
<%=Request.QueryString("url")%> will get the ?url parameter

I assume you're using C# to do this. You can easily get the parameters and their values by iterating through the request object. Or in this case, since you know the name of the parameter, simply do this:
String url = Request.QueryString["url"];
More information on iterating through your request parameters can be found here.

The Uri Type works as well.
String yourHttpUri ="....";
Uri yourURI = new Uri(yourHttpUri);
yourURI.query // "?url=http://www.google.co.uk/"
yourURI.fragment // "#hl=en&q=life&oq=life&aq=f&aqi=g-s1g9&aql=&gs_sm=3&gs_upl=2803373l2803701l2l2803826l4l4l0l0l0l0l188l453l0.3l3l0&bav=on.2,or.r_gc.r_pw.r_cp.,cf.osb&fp=94681dc4659502d1&biw=1680&bih=883"
Edit:
Have you tried Request.Url.ToString(); (And create a new Uri from the result)

How to read returned xml value from google maps

I am trying to call google maps geocode and am following the example on their webpage to try and apply it to mine
http://code.google.com/apis/maps/documentation/geocoding/index.html
in this example, the Geocoding API requests an xml response for the
identical query shown above for "1600 Amphitheatre Parkway, Mountain
View, CA":
http://maps.googleapis.com/maps/api/geocode/xml?address=1600+Amphitheatre+Parkway,+Mountain+View,+CA&sensor=true_or_false
The XML returned by this request is shown below.
Now i am trying to run that url like this in my c# winforms application
string url = "http://maps.googleapis.com/maps/api/geocode/xml?address=1600+Amphitheatre+Parkway,+Mountain+View,+CA&sensor=true_or_false";
WebRequest req = HttpWebRequest.Create(url);
WebResponse res = req.GetResponse();
StreamReader sr = new StreamReader(res.GetResponseStream());
try
{
Match coord = Regex.Match(sr.ReadToEnd(), "<coordinates>.*</coordinates>");
var b = coord.Value.Substring(13, coord.Length - 27);
}
finally
{
sr.Close();
}
However it doesnt seem to be returning anything and as such my var b line gives an index out of bounds error. Can anyone point me in the right direction for at least getting the example to work so i can apply the logic to my own application?
Thanks

If you visit your link "http://maps.googleapis.com/maps/api/geocode/xml?address=1600+Amphitheatre+Parkway,+Mountain+View,+CA&sensor=true_or_false" directly in a browser you can see what it's returning. It's giving me a REQUEST DENIED error.
The problem is caused by the sensor=true_or_false parameter. You have to choose if you want it to be true or false. Google put it this way in their example so that you have to explicitly decide for yourself. This setting indicates if your application is using a location sensor or not. In your case, I'm guessing not, so set it to false.
If you change the link you're using to http://maps.googleapis.com/maps/api/geocode/xml?address=1600%20Amphitheatre%20Parkway,%20Mountain%20View,%20CA&sensor=false, I think you'll get the results you were expecting.

Add incoming querystring to all outgoing links/traffic c#

If i have a page called www.testsite.com and I link to that page using a querystring in the url, is it possible to attach that same querystring to all the outgoing links/traffic somehow?
For instance, say I link to that page like this www.testsite.com?parameter1=somevalue
And then, on that page I click a link to for instance google, and I would like the parameter to be passed so the url would be www.google.com?parameter1=somevalue.
I know how to do this manually, but is there someway it can be appended to the url automatically? Like catch the outgoing request before happening and add the parameter?
Thanks!
Eric

you could do this pretty simply with javascript or javascript using jQuery
$("a").click(function(e){
e.preventDefault();
window.location = $(this).attr("href") + window.location.search;
});
that should get you started
if you wanted a non-jQuery solution you can look into binding events using javascript only
http://www.digimantra.com/tutorials/change-onclick-function-anchor-tag-javascript/
http://forums.techarena.in/software-development/1308647.htm

You can implement an intercepting filter through an httpmodule which modifies the html you emit back to the client. For more details see: http://msdn.microsoft.com/en-us/library/ff649096.aspx
And a few more examples:
Http Response Filter
SO: Get HTML content for modification

Unless you have a central location where you output the outgoing links, you will have to do this manually (possibly the visual studio search/replace can help here, but I would be careful as you may end up changing links you didn't want to).

You can use following to use Query String parameters
You can loop through all your query string parameters like this
int loop1, loop2;
// Load NameValueCollection object.
NameValueCollection coll = Request.QueryString;
// Get names of all keys into a string array. String[] arr1 = coll.AllKeys;
string parameters = "";
for (loop1 = 0; loop1 < arr1.Length; loop1++)
{
parameters += Server.HtmlEncode(arr1[loop1]) + "=" + coll.GetValues(arr1[loop1]);
for (loop2 = 0; loop2 < arr2.Length; loop2++)
{
parameters += Server.HtmlEncode(arr2[loop2]);
}
}
its just basic idea. But you can loop through parameters without knowing the names.

C# HTTP programming

i want to build a piece of software that will process some html forms, the software will be a kind of bot that will process some forms on my website automatically.
Is there anyone who can give me some basic steps how to do this job...Any tutorials, samples, books or whatever can help me.
Can some of you post an working code with POST method ?

Check out How to: Send Data Using the WebRequest Class. It gives an example of how create a page that posts to another page using the HttpWebRequest class.

To fill out the form...
Find all of the INPUT or TEXTAREA elements that you want to fill out.
Build the data string that you are going to send back to the server. The string is formatted like "name1=value1&name2=value2" (just like in the querystring). Each value will need to be URL encoded.
If the form's "method" attribute is "GET", then take the URL in the "action" attribute, add a "?" and the data string, then make a "GET" web request to the URL.
If the form's "method" is "POST", then the data is submitted in a different area of the web request. Take a look at this page for the C# code.

To expand on David and JP's answers':
Assuming you're working with forms whose contents you're not familiar with, you can probably...
pull the page with the form via an HttpWebRequest.
load it into an XmlDocument
Use XPath to traverse/select the form elements.
Build your query string/post data based on the elements.
Send the data with HttWebRequest
If the form's structure is known in advance, you can really just start at #4.
(untested) example (my XPath is not great so the syntax is almost certainly not quite right):
HttpWebRequest request;
HttpWebResponse response;
XmlDocument xml = new XmlDocument();
string form_url = "http://...."; // you supply this
string form_submit_url;
XmlNodeList element_nodes;
XmlElement form_element;
StringBuilder query_string = new StringBuilder();
// #1
request = (HttpWebRequest)WebRequest.Create(form_url));
response = (HttpWebResponse)request.GetResponse();
// #2
xml.Load(response.GetResponseStream());
// #3a
form_element = xml.selectSingleNode("form[#name='formname']");
form_submit_url = form_element.GetAttribute("action");
// #3b
element_nodes = form_element.SelectNodes("input,select,textarea", nsmgr)
// #4
foreach (XmlNode input_element in element_nodes) {
if (query_string.length > 0) { query_string.Append("&"); }
// MyFormElementValue() is a function/value you need to provide/define.
query_string.Append(input_element.GetAttribute("name") + "=" + MyFormElementValue(input_element.GetAttribute("name"));
}
// #5
// This is a GET request, you can figure out POST as needed, and deduce the submission type via the <form> element's attribute.
request = (HttpWebRequest)WebRequest.Create(form_submit_url + "?" + query_string.ToString()));
References:
Link
http://www.developerfusion.com/forum/thread/26371/
http://msdn.microsoft.com/en-us/library/system.xml.xmlelement.getattribute.aspx
http://msdn.microsoft.com/en-us/library/system.xml.xmlelement.selectnodes.aspx

If you don't want to go the HttpWebRequest route, I would suggest WatiN. Makes it very easy to automate IE or Firefox and not worry about the internals of the HTTP requests.