Using the Google Reader API – Part 3

24 Responses to "Using the Google Reader API – Part 3"

• By Martin Doms : Using the Google Reader API – Part 2 January 20, 2010 - 11:10 am

  [...] Part 3 – Editing API [...]

• By Martin Doms : Using the Google Reader API – Part 1 January 20, 2010 - 11:10 am

  [...] Part 3 – Editing API [...]

• By Rosebush January 24, 2010 - 10:02 pm

  Does this url “http://www.google.com/reader/api/0/subscription/edit?client=[your client]” can use now?

  This is my code, it works normally until 2010/01/, but after that date, it always return error 401.The SID and token are all right.

  who can help me? Thank you first!

  *********************************************************

  public int addnewfeed(String feedurl,String foldername) {
  int res = -1;
  String posturl = “http://www.google.com/reader/api/0/subscription/edit”;

  List nvps = new ArrayList();
  if (GoogleLoginSID == null || GoogleLoginSID.compareTo(“”)==0)
  {
  return res;
  }
  else
  {
  nvps.add(new BasicNameValuePair(“Cookie”,”SID=”+mySID));
  }
  nvps.add(new BasicNameValuePair(“client”,”-”));
  nvps.add(new BasicNameValuePair(“s”,”feed/” + feedurl));
  nvps.add(new BasicNameValuePair(“ac”,”subscribe”));
  nvps.add(new BasicNameValuePair(“T”,getToken()));
  if (foldername != null)
  {
  nvps.add(new BasicNameValuePair(“a”,”user/-/label/” + foldername));
  }

  nvps.add(new BasicNameValuePair(HTTP.CONTENT_TYPE, “application/x-www-form-urlencoded”));
  // HttpResponse response = gPostMethod(posturl, nvps, HTTP.DEFAULT_CONTENT_CHARSET);
  HttpResponse response = gPostMethod(posturl, nvps, HTTP.UTF_8);
  if (response == null)
  {
  return -1;
  }
  return response.getStatusLine().getStatusCode();
  }

• By Martin Doms January 25, 2010 - 3:29 pm

  Hmm, your code looks OK to me. Are you properly URL encoding the data? I can’t seem to reproduce the 401 error. 401 usually indicates an authentication issue, are your SID and token valid?

• By AJ February 19, 2010 - 6:54 am

  Hi Martin,

  Great article. However, I can’t seem to get Item Editing EDIT-TAG to work. I always get a 401. I am using C#. Do you have any code that is working?

  Thanks.
  AJ

• By Edwin Lee February 21, 2010 - 7:57 am

  Thanks! Great Stuff! Got a question though…

  Using the web-based Google Reader interface, i can either open a feed for *real* reading, or just “mark it as read” (using the “mark all as read” function).

  The difference it seems between these 2, is how it is refkected in the Trends. Items that are actually opened for reading are counted as “Items Read”, while items that are just “marked as read” are not.

  In terms of the API, how are these 2 actions differentiated? (Because, what i understand is that to mark an item as read is to add a “read” state to the item, or is there a different way?)

  Thanks!

• By ann April 9, 2010 - 10:19 pm

  inyteresting

• By Daniel Baulig April 21, 2010 - 12:56 am

  Hi, nice work so far! I am currently working on a google reader api implementation myself. I’m just wondering if you found anything out about the lifetime of the token. For how long will it remain valid usually? Keep it up!

• By Martin Doms April 28, 2010 - 7:41 pm

  Regarding token lifetime I have never discovered a good answer to this. If anyone knows please let me know. I’ve never personally experienced an expired token but I’ve also never set out to explicitly test it either.

• By cosina1985 June 1, 2010 - 1:26 am

  hi. good job!
  but i found ‘ot’ value it’s not just firstitemmsec/1000.
  do you have any new idea about this paramter?

• By Andy June 24, 2010 - 4:15 am

  I was using this flawlessly for just under 2 months, and then yesterday, it suddenly stopped working and errors with a 401 Permission Denied response.

  The initial Auth, via getSid works fine and it retrieves an SID just as one would expect.

  When the SID is passed to the “http://www.google.com/reader/api/0/token” URL to obtain the token the google responds with the 401 permission denied.

  My initial thought were they don’t like us accessing the reader api but they aren’t basing the 401 denial on IP as if I use the same server to login to Reader via chrome/IE it’s fine.

  I fired up fiddler and had a look at what the reader is current doing, it appears to use a number of different URLs, I wonder if they have changed the api slightly.

  I’m going to try and modify your great example to work around the 401 error, will let you know if I get anywhere.

  Thanks for the get article.

• By Andy June 24, 2010 - 4:23 am

  Update : Reading the google group http://groups.google.com/group/fougrapi/ it appears that a bunch of people are getting the same error.

  But what’s even better, there’s now OAuth support for Reader. Whoopie!

• By Martin Doms June 24, 2010 - 3:17 pm

  Hey thanks for letting me know Andy! I will update my articles with OAuth information as soon as I have time (in the midst of exams right now!). Again, thanks so much for letting me know.

• By Andy June 26, 2010 - 3:35 am

  Update ::

  Okay I’ve managed to get it to work.

  Although I’ve implemented OAuth and have had been able to get some of the features to work it appears that it not quite perfect and I’m waiting on a response from the Unofficial Group on Google, in the mean while I followed through some other post on there and have seen there’s another way, w/o using OAuth.

  Much the same as you have Martin, only you’re sending an AccessToken in the Header on the Request, and there’s no need for the cookie that’s passed.

  The getSid method is much the same, with the exception that the body now contains an additional “Auth=” parameter, this is what you use in the header over the cookie’s SID.

  Here’s a quick hack where I’m modified your original getSID code:

  private void getSid()
  {
  string requestUrl = string.Format(“https://www.google.com/accounts/ClientLogin?accountType=GOOGLE&service=reader&Email={0}&Passwd={1}”, _username, _password);
  HttpWebRequest req = (HttpWebRequest)WebRequest.Create(requestUrl);
  req.Method = “GET”;

  HttpWebResponse response = (HttpWebResponse)req.GetResponse();
  using (var stream = response.GetResponseStream())
  {
  StreamReader r = new StreamReader(stream);

  /* TODO : Use propper SubString & IndofOf over Split & Replace */
  string cSID = “”;
  string cLSID = “”;
  string cAUTH = “”;

  foreach (string cLine in r.ReadToEnd().Split(‘\n’))
  {
  if (cLine.StartsWith(“SID=”)) { cSID = cLine.Replace(“SID=”, “”); }
  if (cLine.StartsWith(“LSID=”)) { cLSID = cLine.Replace(“LSID=”, “”); }
  if (cLine.StartsWith(“Auth=”)) { cAUTH = cLine.Replace(“Auth=”, “”); }
  }

  _sid = cSID;
  _lsid = cLSID;
  _auth = cAUTH;

  }
  }

  There’s no longer a need for the SID or LSID either.

  Initially I thought there wasn’t a need for the getToken, but my initial tests’ still return 401 if I don’t obtain a token, even if I don’t actually pass it. So in order to make getToken work replace the cookie with the following request Header & Content Type :

  ……

  string url = “http://www.google.com/reader/api/0/token”;

  HttpWebRequest req = (HttpWebRequest)WebRequest.Create(url);
  req.Method = “GET”;
  req.ContentType = “application/x-www-form-urlencoded”;
  req.Headers.Add(“Authorization”, “GoogleLogin auth=” + _auth + “”);

  HttpWebResponse response = (HttpWebResponse)req.GetResponse();

  ……

  Then the common httpGet / httpPost methods that are used for the other interaction require the same modifications.

  Here’s httpGet, as that’s all I’ve had a chance to test so far :

  private HttpWebResponse httpGet(string requestUrl, string getArgs)
  {
  string url = string.Format(“{0}?{1}”, requestUrl, getArgs);

  HttpWebRequest request = (HttpWebRequest)WebRequest.Create(url);
  request.Method = “GET”;
  request.ContentType = “application/x-www-form-urlencoded”;
  request.Headers.Add(“Authorization”, “GoogleLogin auth=” + _auth + “”);
  return (HttpWebResponse)request.GetResponse();
  }

  Hope that helps anyone else who wants to make use of the API.

  Once I get time / If I get any feedback on the OAuth mechs, I’ll update you with my findings, personally I’d rather use OAuth as I’m not particularly happy with keeping Username & Passwords, much nicer to use the OAuth system.

  Cya

• By Eric Mann June 29, 2010 - 1:06 pm

  Andy, thanks for such a quick and easy-to-understand solution for this. I guess this is what we get for building on top of an “unofficial” undocumented API. Though I find it incredibly frustrating that they would change the authentication process on the fly like that.

  In any case, I look forward to seeing what you put together with oAuth. Hopefully that will prove more stable in the longrun. I’ve rewritten much of this tutorial in PHP on my own site (for use in plug-ins for WordPress) and my own tutorial could benefit from an overview of using oAuth with the Google API.

• By Andy June 30, 2010 - 8:13 pm

  Eric

  The Google Group (http://groups.google.com/group/fougrapi/) does have some Googlers on it and there was a post on there sometime ago that announced the change ahead of time.

  If you do intend to use the API then I recommend you join the group to be sure of receiving any future announcements / changes.

• By Fred. July 10, 2010 - 8:31 am

  A C# wrapper class to get unread count on a Google Reader Account…

  http://www.chapleau.info/cs/blogs/fchapleau/archive/2010/07/09/google-reader-api-to-check-for-unread-article.aspx

• By sam October 11, 2010 - 11:32 pm

  Very useful,thanks!

  Is there an API to post comment or notes for feed item??

• By rajani April 26, 2011 - 10:35 pm

  how to post a note to google reader from android app.Is there an API to post comment or notes ?

• By rajani April 26, 2011 - 10:36 pm

  how to post notes to google reader from android application.Is there an API to post comment or notes ?

• By rajani April 26, 2011 - 10:37 pm

  how to post notes to google reader from android application.Is there an API to post comment or notes ?

  ….Rajani

• By rajani April 26, 2011 - 10:37 pm

  how to post notes to google reader from android application?

• By aze August 5, 2011 - 6:56 pm

  Very useful,thanks!

• By thaison November 19, 2011 - 3:39 pm

  I have th 400 BAD Request when add a new feed?Does anybody help me?
  StringBuilder buf = new StringBuilder();
  HttpsURLConnection request = null;
  OutputStreamWriter post = null;
  try {
  URL url = new URL(StaticValue.URL_ADDFEED);
  //URL url = new URL(“https://www.google.com/reader/api/0/subscription/quickadd”);
  request = (HttpsURLConnection) url.openConnection();
  request.setDoOutput(true);
  request.setDoInput(true);
  request.setUseCaches(false);

  buf.append(“s”).append(“=”).append(“feed/http://vnexpress.net/RSS/GL/trang-chu.rss”);
  buf.append(“&ac”).append(“=”).append(“subscribe”);
  //buf.append(“&a”).append(“=”).append(“user/18030984262532758053/label/tab4″);
  buf.append(“&t”).append(“=”).append(“TrangChu”);
  buf.append(“&T”).append(“=”).append(getGoogleToken(auth, sid));

  /*buf.append(“quickadd”).append(“=”).append(URLEncoder.encode(“http://vnexpress.net/RSS/GL/trang-chu.rss”));
  //buf.append(“&ac”).append(“=”).append(“subscribe”);
  buf.append(“&T”).append(“=”).append(getGoogleToken(auth, sid));*/

  request.setRequestMethod(“POST”);
  request.setRequestProperty(“Content-Type”, “application/x-www-form-urlencoded”);
  request.setRequestProperty(“Content-Length”, buf.toString().getBytes().length+”");
  request.setRequestProperty(“Authorization”, “GoogleLogin auth=” + auth);
  request.setRequestProperty(“Cookie”, “SID=” + sid);
  post = new OutputStreamWriter(request.getOutputStream());
  post.write(buf.toString());
  post.flush();

  /* BufferedReader in = new BufferedReader(new InputStreamReader(request.getInputStream()));
  buf = new StringBuilder();
  String inputLine;
  while ((inputLine = in.readLine()) != null) {
  buf.append(inputLine);
  }*/
  post.close();
  //in.close();
  Log.e(“Content-Length:” , buf.toString().getBytes().length+”");
  Log.e(“response from server:” , buf.toString());

  int code = request.getResponseCode();
  Log.e(“response code: “, String.valueOf(code));
  Log.e(“response message: “, request.getResponseMessage());

  }
  catch (ProtocolException e) {
  // TODO: handle exception
  Log.e(“e0″, e.getMessage());
  }catch (MalformedURLException e) {
  // TODO: handle exception
  Log.e(“e1″, e.getMessage());
  }
  catch (FileNotFoundException e) {
  // TODO: handle exception
  e.printStackTrace();
  Log.e(“e2″, e.getMessage());
  }
  catch (IOException e) {
  // TODO: handle exception
  e.printStackTrace();
  Log.e(“e3″, e.getMessage());
  }
  catch (Exception e) {
  Log.e(“exception”, e.getMessage());
  //TODO: do something about it
  }

Leave a reply

Name

Email(will not be published)

Website

CAPTCHA Code *