from:"themattharris"

Hey Dave,

Looking through your signature base string just a couple of things
which jump out and would be worth checking:
* There are some spaces in the nonce and timestamp where there
shouldn't be. In all honesty this is most likely email formatting
problems but I wanted to point it out just in case.
* Your text is encoded three times which isn't good and could be the
source of the problem depending on what else you are doing when you
send your request. I want to rule it out as a symptom so can you take
a look at that, and why it is encoded one extra time. When I run tests
I see my text string similar to text%3Djust%2520a%2520simple
%2520message%2520test. yours, in it's current encoding, would read as
just%252520a%252520simple%252520message%252520test.

Matt

On Jun 18, 9:54 am, Acme Dave acmedav...@gmail.com wrote:
 Thanks Matt,

 I am still having problems. Here is the output of my test app.
 I am able to update status. So I know my POST logic is working correctly.

 Info sigBase: POSThttp%3A%2F%2Fapi.twitter.com
 %2F1%2Fdirect_messages%2Fnew.xmloauth_consumer_key%3DCKEY%26oauth_nonce%3D 
 5855976111049200858%26oauth_signature_method%3DHMAC-SHA1%26oauth_timestamp% 
 3D1276879684%26oauth_token%3DATOKEN%26oauth_version%3D1.0%26text%3Dyes%2525 
 20very%252520good%252520message%26user%3DTOUSER

 Info url:http://api.twitter.com/1/direct_messages/new.xml

 ERROR during token access exchange: java.io.IOException: Server returned
 HTTP response code: 401 for 
 URL:http://api.twitter.com/1/direct_messages/new.xml

 Extended error response: ?xml version=1.0 encoding=UTF-8?hash
 request/1/direct_messages/new.xml/request  errorIncorrect
 signature/error/hash

 On Fri, Jun 18, 2010 at 9:21 AM, themattharris 
 thematthar...@twitter.comwrote:



  Hi Dave,

  I seemed to have missed your message originally. Are you still
  experiencing problems with this?

  In answer to your question there isn't anything special required for
  signing requests to send direct messages. If you are still having
  problems can you provide us with your signature base string (without
  any tokens/keys).

  Matt

  On Jun 11, 9:32 am, ds acmedav...@gmail.com wrote:
   I am having trouble sending a direct message with:
 http://api.twitter.com/1/direct_messages/new.xml

   I have been able to successfully post status updates with my code
   using oAuth.
   I have a java based custom application that is working fine for these
   updates.

   When trying to post a direct message I get 402 errors with incorrect
   signature.

   Basically I am using the same signature generation as update except I
   replace
   status with text and add user at the bottom of signature generation.
   Just like
   the oAuth example page on twitter.

   Then I pass the two parameters in the request body.

   I also tried moving everything to url parameters but got the same
   error.

   Is there anything special about parameters for direct message when
   using
   oAuth?

   Does it sound like I am doing this correctly with my post/body method?

   Thanks,
   ...Dave

[twitter-dev] Re: Oauth Echo and Drupal

Hi Tim,

That error you are getting is often thrown when curl tries to verify
the remote host. You can bypass this check in PHP using:
  curl_setopt($link, CURLOPT_SSL_VERIFYPEER, FALSE);
BUT this isn't ideal as it only hides the problem on your server.

Instead, what you need to do is check the permissions on your server
in the certificates folder. Reading the error message you get it
implies your certificate folder is:
  /etc/ssl/certs/ca-certificates.crt

We can't help specifically on how to do this for your environment but
your hosting provider should be able to help you.

Hope that helps,
Matt

On Jun 21, 11:59 am, Taylor Singletary taylorsinglet...@twitter.com
wrote:
 Hi Tim,

 That sounds like your machine might be having some issues connecting via
 SSL. Are you able to use Curl for any other SSL-based sites?

 Taylor

 On Mon, Jun 21, 2010 at 11:57 AM, Tim Millwood 
 t...@millwoodonline.co.ukwrote:



  Tried using Curl instead of drupal_http_request and got the following
  error.

  error setting certificate verify locations:\n  CAfile: \/etc\/ssl\/
  certs\/ca-certificates.crt\n  CApath: none\n

  Not sure what that means

  On Jun 21, 4:40 pm, Tim Millwood t...@millwoodonline.co.uk wrote:
   I have a test air app that posts to Twitpic perfectly fine using oAuth
   Echo, if I change the URL to my web app I get 401.

   Here are the headers I am passing to twitter to verify credentials. Am
   I missing some?

   [HTTP_X_AUTH_SERVICE_PROVIDER] =
 https://api.twitter.com/1/account/verify_credentials.json

   [Authorization] = OAuth

  realm=twitter,oauth_consumer_key=t94eBtc4Pz2zqo4KhABseQ,oauth_token=62
  66632-

  e0NRaGReqpzR84Floyg565BMJbBH4lYMxsJD9LNZY,oauth_version=1.0,oauth_timest
  amp=1276986312,oauth_nonce=KhWw0N,oauth_signature_method=HMAC-
   SHA1,oauth_signature=ubKN3OQy8xC5Sdkn%2BD%2Bcq9c1ywY%3D

   Tim

   On Jun 21, 4:28 pm, Taylor Singletary taylorsinglet...@twitter.com
   wrote:

Awesome. There's much untapped potential in OAuth Echo beyond just the
TwitPic, yFrog, etc. use cases.

This is an area where you're going to have to be very exacting. Have
  you
confirmed that the request you are building would actually execute
  against
Twitter correctly before you've sent it through your process? As in,
  have
you verified that the HTTP Authorization header you've created will
  work
against the end point before you've stuffed it into some other header,
processed the request, etc.

Can you share the authorization header you are using and how you've
  defined
the OAuth Echo headers? Your signature base string for the same?

Taylor

On Mon, Jun 21, 2010 at 8:06 AM, Tim Millwood 
  t...@millwoodonline.co.ukwrote:

 I am trying to do something else.

 I am not trying post to twitpic, I am trying to post to my own web
  app
 (similar to twitpic).

 I am getting the HTTP_X_AUTH_SERVICE_PROVIDER and
 HTTP_X_VERIFY_CREDENTIALS_AUTHORIZATION headers from the test app,
 then my web app is renaming the
 HTTP_X_VERIFY_CREDENTIALS_AUTHORIZATION header to Authorization and
 POSTing both to the URL in HTTP_X_AUTH_SERVICE_PROVIDER.
 This returns the 401 error.

 On Jun 21, 3:11 pm, Taylor Singletary taylorsinglet...@twitter.com
 wrote:
  Hi Tim,

  The call specified in your HTTP_X_* headers is for the OAuth Echo
 provider
  to execute against the API. Since they execute the call, it
  invalidates
 the
  oauth_nonce you provided. Really, it's a different API call that
  your
  application should be executing following an OAuth Echo
  transaction..

  1. You've got something to post with TwitPic
  2. You setup a mock request to Twitter to verify credentials so
  that
 TwitPic
  can identify your user with Twitter
  3. You send that mock request in HTTP_X_* headers to Twitpic, along
  with
  your API request to Twitpic with the image
  4. TwitPic executes the Twitter API call specified in the HTTP_X_*
 headers,
  verifying the user
  5. On success, TwitPic sends you in its response information about
  the
 media
  you just uploaded on behalf of your user
  6. You take that response and append it to a tweet, or whatever
  other API
  operation you're doing, and send a brand new request to Twitter

  Is this the flow you're following or are you trying to do something
  else?

  Taylor

  On Sat, Jun 19, 2010 at 12:40 PM, Tim Millwood 
 t...@millwoodonline.co.ukwrote:

   So... I now have a test app which is sending oAuth Echo request
   successfully to Twitpic.

   If I change the URL to my web app I get a 401 error back from
  Twitter,
   so there is something I am doing wrong.

   I am getting the HTTP_X_AUTH_SERVICE_PROVIDER and
   HTTP_X_VERIFY_CREDENTIALS_AUTHORIZATION headers from the test
  app,
   then my web app is renaming the
   HTTP_X_VERIFY_CREDENTIALS_AUTHORIZATION header

[twitter-dev] Re: Which IETF standard has the year appearing after the time?

The time format is a little weird and as far as I know, doesn't match
any RFC. Instead it matches the ruby default and is represented in
tokens by:
  %a %b %d %H:%M:%S %Z %Y

The format has been like this since the API was first released which
means, for backwards compatibility with other applications, we can't
easily change it with this version of the API.

I hope that explains the why it is still in the format it is.
Hopefully you can use the token string above to parse the date using
the time parsing functions of your chosen language.

Matt

On Jun 21, 12:40 pm, Peter Cross zootl...@gmail.com wrote:
 This date is from a call 
 tohttp://api.twitter.com/1/statuses/user_timeline.xml:

 created_atMon Jun 21 19:06:21 + 2010/created_at

 begin rant

 I've never seen the year come after the time... in any standard date
 format.  It's as if someone thought Hmmm... how can we make this date
 format more difficult to work with?.  Why, why why?  Now I have to
 write a special handler for this one exception.  It's sloppy.

 /end rant

 This isn't an XML standard date format either.

 -ZPC

[twitter-dev] Re: oAuth and Direct Message

Thanks for the update and sharing the solution Dave. I'm glad it's all
working.

Regarding nonce values. I recommend something like the MD5 of the
concatenation of some kind of time value with a random number/string
to ensure a unique nonce every time.

Hope that helps
Matt

On Jun 21, 1:11 pm, ds acmedav...@gmail.com wrote:
 I solved the problem.

 I did not have an  between my two parameters in the POST body.

 Adding  between the text and user parameter fixed it.

 Thanks Matt.

 On Jun 21, 1:10 pm, themattharris thematthar...@twitter.com wrote:



  Hey Dave,

  Looking through your signature base string just a couple of things
  which jump out and would be worth checking:
  * There are some spaces in the nonce and timestamp where there
  shouldn't be. In all honesty this is most likely email formatting
  problems but I wanted to point it out just in case.
  * Your text is encoded three times which isn't good and could be the
  source of the problem depending on what else you are doing when you
  send your request. I want to rule it out as a symptom so can you take
  a look at that, and why it is encoded one extra time. When I run tests
  I see my text string similar to text%3Djust%2520a%2520simple
  %2520message%2520test. yours, in it's current encoding, would read as
  just%252520a%252520simple%252520message%252520test.

  Matt

  On Jun 18, 9:54 am, Acme Dave acmedav...@gmail.com wrote:

   Thanks Matt,

   I am still having problems. Here is the output of my test app.
   I am able to update status. So I know my POST logic is working correctly.

   Info sigBase: POSThttp%3A%2F%2Fapi.twitter.com
   %2F1%2Fdirect_messages%2Fnew.xmloauth_consumer_key%3DCKEY%26oauth_nonce%3D

   5855976111049200858%26oauth_signature_method%3DHMAC-SHA1%26oauth_timestamp%

   3D1276879684%26oauth_token%3DATOKEN%26oauth_version%3D1.0%26text%3Dyes%2525
20very%252520good%252520message%26user%3DTOUSER

   Info url:http://api.twitter.com/1/direct_messages/new.xml

   ERROR during token access exchange: java.io.IOException: Server returned
   HTTP response code: 401 for 
   URL:http://api.twitter.com/1/direct_messages/new.xml

   Extended error response: ?xml version=1.0 encoding=UTF-8?hash
   request/1/direct_messages/new.xml/request  errorIncorrect
   signature/error/hash

   On Fri, Jun 18, 2010 at 9:21 AM, themattharris 
   thematthar...@twitter.comwrote:

Hi Dave,

I seemed to have missed your message originally. Are you still
experiencing problems with this?

In answer to your question there isn't anything special required for
signing requests to send direct messages. If you are still having
problems can you provide us with your signature base string (without
any tokens/keys).

Matt

On Jun 11, 9:32 am, ds acmedav...@gmail.com wrote:
 I am having trouble sending a direct message with:
   http://api.twitter.com/1/direct_messages/new.xml

 I have been able to successfully post status updates with my code
 using oAuth.
 I have a java based custom application that is working fine for these
 updates.

 When trying to post a direct message I get 402 errors with incorrect
 signature.

 Basically I am using the same signature generation as update except I
 replace
 status with text and add user at the bottom of signature generation.
 Just like
 the oAuth example page on twitter.

 Then I pass the two parameters in the request body.

 I also tried moving everything to url parameters but got the same
 error.

 Is there anything special about parameters for direct message when
 using
 oAuth?

 Does it sound like I am doing this correctly with my post/body method?

 Thanks,
 ...Dave

[twitter-dev] Re: oAuth and Direct Message

2010-06-18 Thread themattharris

Hi Dave,

I seemed to have missed your message originally. Are you still
experiencing problems with this?

In answer to your question there isn't anything special required for
signing requests to send direct messages. If you are still having
problems can you provide us with your signature base string (without
any tokens/keys).

Matt

On Jun 11, 9:32 am, ds acmedav...@gmail.com wrote:
 I am having trouble sending a direct message 
 with:http://api.twitter.com/1/direct_messages/new.xml

 I have been able to successfully post status updates with my code
 using oAuth.
 I have a java based custom application that is working fine for these
 updates.

 When trying to post a direct message I get 402 errors with incorrect
 signature.

 Basically I am using the same signature generation as update except I
 replace
 status with text and add user at the bottom of signature generation.
 Just like
 the oAuth example page on twitter.

 Then I pass the two parameters in the request body.

 I also tried moving everything to url parameters but got the same
 error.

 Is there anything special about parameters for direct message when
 using
 oAuth?

 Does it sound like I am doing this correctly with my post/body method?

 Thanks,
 ...Dave

[twitter-dev] Re: Repeat Tweets

2010-06-16 Thread themattharris

Thanks for letting us know about these Andy. As Taylor was saying the
instabilities the other day are likely to have been the cause for
these getting through. It shouldn't happen now but anything that did
get through will still be there.

Matt

On Jun 15, 6:39 pm, Andy Matsubara andymatsub...@gmail.com wrote:
Taylor,

There's a sample of six consecutive
tweets.http://twitter.com/AppleWalker/status/16204554355http://twitter.com/AppleWalker/status/16204551280http://twitter.com/AppleWalker/status/16204548302http://twitter.com/AppleWalker/status/16204545368http://twitter.com/AppleWalker/status/16204542279http://twitter.com/AppleWalker/status/16204539134

Andy Matsubara

On Wed, Jun 16, 2010 at 4:21 AM, methnen meth...@gmail.com wrote:
Can you provide some links to the repeated tweets? While things were at
their worse last night, it was very easy for duplicate updates to get
processed but that shouldn't be the case now.

We manually deleted them and it now appears to have stopped.

Are you sure that the client you were using isn't trying to re-post tweets
on your behalf, thinking they haven't been posted yet?

At first I thought it might be on our end. As we use the API to make
tweets via a custom script in our CMS. However, the script records
successful posts and then stops posting and the repeating tweet was
recorded as a successful post and our system never sent it to Twitter
again.

Anyway it appears to have stopped so I guess all is good. :)

[twitter-dev] Re: oauth status update returning error 401 invalid / used nonce

2010-06-15 Thread themattharris

Hey Craig,

Could you let us see what the authorization header you are sending
looks like (obfuscating your oauth keys and secrets)?

What you're aiming for is a header similar to this:

POST /1/statuses/update.xml HTTP/1.1
Accept: */*
Connection: close
User-Agent: OAuth gem v0.4.0
Content-Type: application/x-www-form-urlencoded
Authorization: OAuth
  oauth_consumer_key=\\,
  oauth_nonce=\\,
  oauth_signature=\\,
  oauth_signature_method=\HMAC-SHA1\,
  oauth_timestamp=\1276634179\,
  oauth_token=\xx\,
  oauth_version=\1.0\
Content-Length: 27
Host: api.twitter.com

With your post body being the status you are sending. Similar to:

status=something%20to%20say

Matt

On Jun 13, 8:32 pm, Craig chanson9...@gmail.com wrote:
 Ok, so I believe my signature is correct because if I take out the
 oauth_token parameter from the signature base I get an incorrect
 signature error instead of the invalid/used nonce error.  I am
 definitely posting to the correct url: https://api.twitter.com/1/
 statuses/update.xml because if I change it I get an invalid url
 response.  The nonce I'm generating is unique for every request I'm
 sending.  The timestamp is accurate to the number of seconds that have
 passed since unix epoch.  Obviously the invalid/used nonce error is
 garbage.

 Here's the formatting of my URL request in obj c:

 // Create the request
 NSString* requestString= [NSString
 stringWithFormat:@%@,TWITTER_STATUS_UPDATE_URL];
 NSMutableURLRequest* request = [[NSMutableURLRequest alloc]
                                 initWithURL:[NSURL 
 URLWithString:requestString]
                                 
 cachePolicy:NSURLRequestReloadIgnoringCacheData
                                     timeoutInterval:30.0];
 [request setHTTPMethod:@POST];
 [request addValue:authHeader forHTTPHeaderField:@Authorization];
 [request setValue:@application/x-www-form-urlencoded
 forHTTPHeaderField:@Content-Type];
 [request setValue:postBodyLength forHTTPHeaderField:@Content-
 Length];
 [request setHTTPBody:[postBody
 dataUsingEncoding:NSUTF8StringEncoding]];

 Any other ideas?  I'm at a total loss as to why this might be
 happening.

 Thanks,
 Craig

 On Jun 13, 3:42 pm, Craig chanson9...@gmail.com wrote:



  No dice.  I just tried creating a new app and using the new keys/token
  to tweet.  I'll go over my steps once more and see if I can track down
  an issue somewhere...

  On Jun 11, 8:53 pm, Taylor Singletary taylorsinglet...@twitter.com
  wrote:

   xAuth in this case, I think, is unrelated to the issue. If you can use
   a different key and use the my token feature to get your access
   token, then try to tweet using that token, it will sufficiently
   express the problem I think.

   Taylor

   On Jun 11, 2010, at 2:48 PM, Craig chanson9...@gmail.com wrote:

Hello,

I have an iphone app that is using xauth.  I am able to obtain the
access token with no problem at all.  When I go to post a status
update, I receive an invalid / used nonce error.  It can't actually be
a used nonce since I have checked this multiple times.  Here are the
details of the post:

url:
   http://api.twitter.com/1/statuses/update.xml

signature base string:
POSThttp%3A%2F%2Fapi.twitter.com%2F1%2Fstatuses
%2Fupdate.xmloauth_consumer_key%3Dmyconsumerkey%26oauth_nonce
%3D397vi5Ug1YHC3UAVUAoB%26oauth_signature_method%3DHMAC-
SHA1%26oauth_timestamp%3D1276292596%26oauth_token%3Dmytoken
%26oauth_version%3D1.0%26status%3Dmy%2520tweet

Authorization header:
OAuth oauth_nonce=397vi5Ug1YHC3UAVUAoB, oauth_signature_method=HMAC-
SHA1, oauth_timestamp=1276292596,
oauth_consumer_key=myconsumerkey, oauth_token=mytoken,
oauth_signature=yOh2zQPGDBlVEP5cDWhjddQWTLc%3D, oauth_version=1.0

Content-Type:
[request setValue:@application/x-www-form-urlencoded
forHTTPHeaderField:@Content-Type];

I can see no reason why this shouldn't work.  Any help would be
greatly appreciated!

Thanks,
Craig

[twitter-dev] Re: Where to post documentation bugs?

2010-06-09 Thread themattharris

Hi Andrew,

Thanks for your question. You can file any issues with the docs on
dev.twitter.com on the api-issues list [1]. The list has a component-
docs which you can mark the issue with.

We're working on a number of issues with the docs at the moment and
hope to have them and the rest of the docs from the apiwiki updated
and correct in the next few weeks. Thanks for your patience.

Matt

1. http://code.google.com/p/twitter-api/issues/list

On Jun 5, 5:46 pm, Andrew andrew.don...@gmail.com wrote:
 Folks,

 I've just successfully implemented xAuth for my iPad app. During this
 process, I went down a few ratholes due to what I now consider
 documentation bugs. I list them below but I would like to know where I
 send these things in the future?

 All of these bugs are on: http://dev.twitter.com/pages/xauth.

 1) HMAC method string, oauth_signature_method, is wrong. It is:
 HMAC_SHA1. It should be: HMAC-SHA1.

 2) The below alludes to a POST body in the xAuth /oauth/access_token
 method.

 Your HTTP POST body would look like this:
 x_auth_mode=client_authx_auth_password=xyz12242134x_auth_username=episod

 In fact, the above is your parameter/query string. The method requires
 that the POST place these parameters on the URL.

 BTW, it would be nice to just see the whole URL. There is little
 reason to save bytes here. It would look like this:

 https://api.twitter.com/oauth/access_token?
 x_auth_mode=client_authx_auth_password=xyz12242134x_auth_username=
 episod

 Anon,
 Andrew

[twitter-dev] Re: Sign out of Twitter through API

2010-06-09 Thread themattharris

Hey GHengeveld,

There was a conversation about this back in April which might be help
[1]. In it Taylor explains that OAuth is stateless and that the logged
in state of a user is based on your system rather than ours.

Your application would be interacting with Twitter using the OAuth
tokens for the user as you have identified them. If the user isn't
recognised I would expect your widget would detect that and ask the
user to authenticate or authorize again. That being said, the method
you use to identify the user depends on what you want to do. Also, as
the user isn't logged into Twitter through the API the account/
end_session method will have no affect.

I'm wondering, have you looked at @anywhere. It may be suitable for
what you want to do.

Matt

1.
http://groups.google.com/group/twitter-development-talk/browse_thread/thread/02e44b27d7ba3661

On Jun 8, 7:25 am, GHengeveld korad...@gmail.com wrote:
I'm developing an application designed to run on a public computer,
where many users will sign in with Twitter to register for our app.

I've been looking for a way to sign the user out of Twitter when they
exit the application (though a sign out button). Since we cannot
destroy the cookies set by twitter.com, all it does right now is
destroy our own session and cookies and show a button which links to
the Twitter sign out page (in a popup).

We would prefer not to use the popup, but instead sign the user out of
Twitter automatically when they sign out of our application. This will
greatly reduce the likeliness of people forgetting to logout (and thus
allowing access to their own account by the next user). We're already
using force_login=true to force a new login when someone connects to
our application, but then it could already be too late.

Is there any way to call a sign out through the API?

Account/end_session seems to be what I'm looking for, but I can't get
it to work. Another approach would be to scrape twitter.com for the
authenticity_token and call the logout from through cURL, but this
isn't the nicest way to do things and my first attemps have failed so
far.

[twitter-dev] Re: Search API: searching for Don and finding don't instead

2010-06-07 Thread themattharris

Hi Jeffrey,

Thanks for bumping this to our attention. Some of the threads fall off
our radar so a prompt is always welcome.

Search treats separate words as an AND search meaning a search for:
  Don SomeLastName
will translate to:
  Don AND SomeLastName.

For a complete phrase search you would instead want to search for:
  Don SomeLastName.

The problem you are experiencing with Don matching Don't is, as
you suggested, managed by appending -don't to the query. You don't
need to escape the apostrophe and the quotes are not necessary, making
your search query:
  Don SomeLastName -don't

You can read more about the supported advanced search operators on the
search site [1].

Hope that helps,

Matt Harris
Developer Advocate, Twitter
http://twitter.com/themattharris

1. http://search.twitter.com/operators

On Jun 7, 9:09 am, Jeffrey Greenberg jeffreygreenb...@gmail.com
wrote:
 Hello Twitter,
 Anyone home?
 j

 On Jun 2, 11:28 pm, Jeffrey Greenberg jeffreygreenb...@gmail.com
 wrote:



  We have a user that is causing us to create a search of the form:
     Don SomeLastName
  which is returning tweets containing don't and SomeLastName.

  Thats a no good!

  Is there a decent workaround for this by modifying the search? e.g.
      Don SomeLastName -don't
  but how do you escape the single quote?  Like this?
      Don SomeLastName -don't

[twitter-dev] Re: Trends not currently supporting the callback parameter?

Thanks for letting us know about this. You can follow the progress of
this on the issues log [1].
We'll get onto the problem and have it resolved soon.

Matt
1. http://code.google.com/p/twitter-api/issues/detail?id=1668

On May 27, 9:37 am, Steven steven.pul...@gmail.com wrote:
 I have noticed a change with regards to the trends resource.

 for reference please check the following URLs

 1)http://api.twitter.com/1/trends/current.json
 2)http://api.twitter.com/1/trends/current.json?callback=foo
 3)http://search.twitter.com/trends.json?callback=foo

 Currently, for me api.twitter.com trends/current.json is returning the
 same pure JSON payload regardless of the callback parameter.  This is
 new behavior that has changed since I last used this resource.

 What I would expect, and is not the case, is that URL #2 with the
 callback parameter specified would return JSON wrapped in a function
 call, similar to how URL #3 behaves.

 Has anyone else noticed this?  I have confirmed it myself with more
 than one location.

 Is this on purpose?  Was a bug rolled out to production in a release?
 Am I missing the obvious. Please help.

 Thank you very much.

[twitter-dev] Re: Thoughts on annotations

Zak,

Thanks for coming along and sharing your thoughts and feedback with
everyone. We're all impressed with what you and other developers came
up with this weekend and for finding those bugs and gotchas. We're
taking that information on board and making the necessary changes to
protect the API and improve the documentation.

For those who are interested in knowing about the other projects
please visit the wiki page [1] for a list of them.

Matt

1. http://wiki.github.com/themattharris/annotations-hackfest/

On May 31, 12:56 pm, zbowl...@gmail.com zbowl...@gmail.com wrote:
 I was still waking up and recovering from yesterday when I wrote this
 so forgive the typos, grammar, and mixing up attributes as
 annotations.

 I also wanted to mention that I'm glad that twitter didn't confuse
 annotations by using the term namespace or providing some kind of
 mechanism that goes that direction. I originally was wondering about
 conflicting but this method is better. Namespace gives developers a
 sense of ownership with the data stored in that annotation type. You
 can easily still easily namespace your types if you want (possibly in
 the reverse DNS format used in Java) but you can't prevent other
 developers from using them and that needs to be conveyed. If you need
 to strongly store annotation data, you can use signatures, hashes,
 version markers, or whatever in the annotation that you want to
 provide that, but it's up to the developer to what they want to fit
 their needs.

 Also wanted to playfully rant that parking overnight on the weekend
 near Twitter HQ sucks, especially when there is a baseball game. The
 closest garages all close at night on the weekends, so I ended up
 moving my car 3 times. Next time I'm taking BART and riding my
 bike. :-)

 Also

 On May 31, 11:39 am, Zac Bowling zbowl...@gmail.com wrote:



  This weekend's hackfest was at Twitter HQ was fun. About a couple dozen of 
  us stayed awake for about 30 hours and still had enough to energy to 
  present. Some pretty amazing things created and we helped identified a 
  bunch of bugs.

  Now that I've had a chance to go home and catch up on some sleep, here is a 
  brain dump of my thoughts.

  * One of the documented recommended types is place/location, but this data 
  is similar to what we store in the geo fields. I'm not sure what issues we 
  may run into privacy using it rather then storing the Geo fields (users can 
  enable/disable geo and remove geo data from all previous status updates).

  * We will always have twitter clients that will not understand or look even 
  look at our attributes. This means that we can't can't have annotations 
  that change the meaning of a tweet or make the meaning of the tweet 
  useless. This is basically graceful degradation, and not progressive 
  enhancement. We joked that want to see tweets that say: This tweet can 
  only be read in clients that support X annotations. Please upgrade your 
  twitter client or try X client..

  * You have to treat annotations as potentially hostile attack vectors.  As 
  was proved with some awesome cornfied and flashing unicorn injections this 
  weekend, any raw data can be store in annotations. Just because you stored 
  it there, anyone can do store any raw data and anyone can post tweets that 
  copy your annotation format. Twitter may sanitize javascript injections, 
  but it doesn't stop other types of injections from occurring if you don't 
  check. It's extremely important to validate, html encode, or whatever you 
  need to with the data stored in the annotations.

  As I did with my twitter remote shell execution example, I added my own 
  signature and noance of my own into the twitter annotation to validate the 
  sender had my secret. It may be one solution.

  * Attributes work at the time of creation because status updates are 
  immutable. This may be obvious to most, but its a limitation that hits you 
  a few times as you develop. Because of that we need to make sure that we 
  can get most of the clients, including Twitter.com, support the most 
  popular annotation formats. We can't fix update status updates after the 
  fact so we have to get it right.  (Adding annotations to new style retweets 
  is in theory possible)

  * Can't remind people enough to switch from twitter.com to api.twitter.com. 
   A bunch of little differences between the two that give you headaches. Our 
  board of wasted time at the hackfest summed it up pretty well.

  * A good number of us spent a good deal of time on just getting past OAuth 
  this weekend. We had a lot of people that understood the OAuth spec fairly 
  well thankfully and @jmhodges was there to help (although not his area he 
  deals with in the code). Since you update twitter with POST, it's optional 
  to store the authentication data in the postdata instead of the 
  authentication header according to the spec, and some our libraries were 
  doing just that, but twitter only works

[twitter-dev] Re: Bug with source parameter in Twitter search ?

Hi Ram,

Thanks for bumping this to our attention. We're aware a number of uses
of source are being missed out and we are tracking this as ticket 1632
on our issues log [1]. You'll be able to track progress on that
thread.

Matt

1. http://code.google.com/p/twitter-api/issues/detail?id=1632

On Jun 1, 11:40 am, Ram group...@cascadesoft.net wrote:
 ping



  Over the past several weeks, I've seen an intermittent issue with
  Twitter search.

  A search ofhttp://search.twitter.com/search?q=%23ceconomyshowsall
  recent tweets with this hashtag. Multiple people have used this
  hashtag.

  However, asource-based search 
  ofhttp://search.twitter.com/search?q=%23cEconomy+source%3AiPhoneEconomyApp
  returns zero results (even though almost all the tweet-results from
  the previous search are from thissource)

  The samesource-based search works fine on some days (and it returns
  all tweets made over the past 5-6 days with thissourceand hashtag).
  On days like today, it returns zero results. On other days, it returns
  partial results.

  I'm assuming that adding asourceparameter should not suppress any
  results for tweets (made from the specifiedsource) that show up in
  searches made without thesourceparameter.

  Can someone from Twitter confirm that this is a bug (or does anyone
  else have any thoughts on this issue) ?

[twitter-dev] Announcing Snowflake

Hey everyone,

In March we sent a message to you about upcoming changes to the way
status IDs are sequenced (http://bit.ly/upcoming-status-id-changes).
Today we announced Snowflake - the service we will be using to
generate those new IDs.

It isn't going live yet! We just know a lot of you are concerned about
how the Tweet IDs are going to affect your applications and wanted you
to have the chance to familiarize yourself with how Snowflake works.

You can find the a link to the code and read more about Snowflake on
the Twitter Engineering blog: http://bit.ly/announcing-snowflake

Best,
Matt

[twitter-dev] Re: users/lookup seems not to be working

2010-05-29 Thread themattharris

Hey Abacadabrados,

You need to give the user lookup request some valid user ids or screen
names. The error you are receiving is because you are sending a
user_id of 0 which is not a valid account. Maybe try with your screen
name first.

You can find more information on how the method works on
dev.twitter.com [1].

Hope that helps,
Matt

1. http://dev.twitter.com/doc/get/users/lookup

On May 29, 5:17 am, Abacadabrados rene.rikk...@gmail.com wrote:
 users/lookup seems not be working in the API console.

 For instance: I signed in, and selected my application in the console
 Headers and methods: GET / 1 / users/lookup
 Parameters and values: user_id / 0
 Response body: ... No user matches for specified terms ...

 The same for parameters and values: screen_name / twitterapi

 Interesting is maybe that I get a 404-error when using users/lookup in
 my own application.
 For your information: I do succesfully use OAuth, and other queries
 like friends/ids are doing fine.

[twitter-dev] Re: Issue with posting to twitter: http://twitter.com/home?status=doesnotworkifyouarenotloggedin

Hi Priyanka,

Thanks. We're aware of the problem and it is being tracked as ticket
1650 [1] on the twitter-api issues log.

1. http://code.google.com/p/twitter-api/issues/detail?id=1650

Matt

On May 27, 8:24 pm, newtothisworld priyankalut...@gmail.com wrote:
 Hi,

 It appears that if you are not logged in, and hit the following 
 URL:http://twitter.com/home?status=doesnotworkifyouarenotloggedin, you
 lose the status you were trying to post.

 I'm very certain this used to work until a few days ago.

 Has something changed?

 Thanks for your help!
 Priyanka

[twitter-dev] Re: users.lookup() pulls by friendship date

Hey cballou,

What is the actual API call you are making?
I ask because the users/lookup.json method requires you pass it user
ids for it to work. If you don't pass it any user ids or screen names
you will told that no user matches the specified terms.
The data is then returned in an undefined order. As nischalshetty
said, this method can be used to request information about any user so
the sort order is arbitrary.

Are you getting results when you pass no IDs?

Matt



On May 27, 7:43 pm, nischalshetty nischalshett...@gmail.com wrote:
 So it's the same one that I was talking about. They haven't specified
 any sorting rules in the doc, are you sure about it?

 Before all that, I hope you know that the lookup API can be used to
 retrieve info about ANY twitter user. So it does not matter if that
 user is your friend. How will the sorting be applied then?

 -Nischal

 On May 27, 5:18 am, cballou ball...@gmail.com wrote:



  The link is:

 http://dev.twitter.com/doc/get/users/lookup

  It actually returns the full dataset for up to 100 users. The returned
  data is sorted by your newest friendship in descending order.  This
  functionality is quite minimal and could definitely be expanded upon
  like I suggested above.  I was just wondering if there were any
  possibly hidden parameters I could pass in to change the count, cursor
  position, etc.

  On May 26, 1:22 am, nischalshetty nischalshett...@gmail.com wrote:

   Are you talking about this 
   -http://api.twitter.com/version/users/lookup.format

   The above API returns whatever ids you have passed. Am I missing
   something?

   -Nischal

   On May 26, 4:38 am, cballou ball...@gmail.com wrote:

Nobody?

On May 25, 12:14 pm, cballou ball...@gmail.com wrote:

 I really don't like the fact that calling users.lookup() returns the
 last 100 users I have friended.

 Is there a way to retrieve users in a more random fashion or with some
 kind of ordering (ascending/descending)?  I'm looking for more
 optional parameters.

 Suppose, for instance, that I have 500 friends.  I would not want to
 pull my last 100 friends when making this API call.  I might want to,
 however, pull a random sampling of 100 of those friends.  I may also
 want to pull a particular number of friends (i.e. imposing a limit on
 the request).

 I would propose that there be additional filtering parameters for this
 request:

 sort: asc/desc/random (default desc)
 limit: 1-100 (default 100)

 Can anybody clarify and expand upon this for me?

[twitter-dev] Re: Search spam??

I'm wondering whether the search is getting popular tweets mixed in
(the default behavior) and that those are what you are seeing.
Can you give an example of one or two of the searches that are doing
this?

Thanks,
Matt

On May 27, 2:19 pm, M. Edward (Ed) Borasky zn...@borasky-
research.net wrote:
 I just noticed this today - may have been going on for a while,  
 though. I'm logged in on twitter.com. I have a few saved searches for  
 some friends - about four of them with different groups of friends.  
 What's in the search term is, for example, screen name 1 OR screen  
 name 2. The screen names appear without @ signs.

 One of these searches is returning tweets that don't match either of  
 the two screen names! They are returning tweets from the two screen  
 names *plus* some tweets that appear to be from people trying to get  
 me to click on links. These tweets do *not* have the characteristics  
 of a Promoted Tweet. They aren't showing up at the top of the search  
 - they're showing up in time sequence order.

 Has someone figured out how to game the search? Is Twitter testing  
 something and not telling us? When I search for sn1 OR sn2 I do  
 *not* want to receive tweets like this!

 http://twitter.com/rx8mall/status/14859299274

[twitter-dev] Re: Social Graph Methods Page Size change?

Hi Miles,

Which API method are you using. The preferred method for retrieving
friends and followers is friends/ids [1] and followers/ids [2]. In
both cases it is recommended that you add cursor=-1 to the end.

Matt

1. http://dev.twitter.com/doc/get/friends/ids
2. http://dev.twitter.com/doc/get/followers/ids

On May 27, 1:37 pm, Miles  Parker milespar...@gmail.com wrote:
 I'm noticing now that page for friends and followers (for arbitrary
 user ids) are returning 100 users, not 5,000 as I'm seeing in the API
 docs. Is it my imagination that this has changed just in the last
 couple of days?

[twitter-dev] Re: Widget Search problem

Hi cfalar,

This is a known bug and is being tracked as ticket 1404 [1].

Matt

1. http://code.google.com/p/twitter-api/issues/detail?id=1401

On May 27, 12:55 pm, cfalar carolfalard...@hotmail.com wrote:
 Hi,
 I have a problem with the Search widget.
 If i go on this pagehttp://twitter.com/goodies/widget_searchand type
 #habs OR attaquea5 on the search input, i got no result.
 If i take a look at the get parameter the q value is #habs+OR
 attaquea5.
 Look like something add the + sign.
 It always do that and its return no result.

[twitter-dev] Re: Important Notice on Incorrect API Endpoints for Search, REST, OAuth (+ some general tips!)