Re: [twitter-dev] Re: Tweet button and iphone
On Jul 27, 2011, at 3:43 AM, Victor Ruiz wrote: On Android, working with intents and after sharing successfully, the window closes automatically so it returns to the previous page. The window that automatically closes should have set up a timeout because it takes some seconds to be closed. If for example client browser doesn't support Javascript, there is a cancel button that also works perfectly. On iPhone, is different than in Android, because after sharing successfully, you don't have possibility to close and return back. I guess Twitter detects the user agent and show different pages in Android and iPhone, isn't it? iPhone and Android both have the same Web Intents UI (as do all devices, actually.) The behaviour you describe for Android, where the Intent page automatically closes and the browser takes you back to the original page is the correct behaviour. You should get the same experience on Android and iPhone. The behaviour you're seeing on iPhone there is odd, but could occur (on any device) if the Tweet Intent is invoked in a different way (either without JavaScript, or without a window.opener being set.) Can you link to the specific case where the iPhone UI isn't behaving as expected? I'll take a look to debug it. Thanks, Ben On 21 jul, 07:21, anaj...@ibs.com.jo anajjar %ibs.com...@gtempaccount.com wrote: It's working perfectly,Thanks Ben Note: This issue is also noticed in another url,please check my other thread here: https://groups.google.com/group/twitter-development-talk/browse_threa... -- Have you visited the Developer Discussions feature on https://dev.twitter.com/discussions yet? Twitter developer links: Documentation and resources: https://dev.twitter.com/docs API updates via Twitter: https://twitter.com/twitterapi Unsubscribe or change your group membership settings: http://groups.google.com/group/twitter-development-talk/subscribe -- Have you visited the Developer Discussions feature on https://dev.twitter.com/discussions yet? Twitter developer links: Documentation and resources: https://dev.twitter.com/docs API updates via Twitter: https://twitter.com/twitterapi Unsubscribe or change your group membership settings: http://groups.google.com/group/twitter-development-talk/subscribe
[twitter-dev] Can you do one query and leave it running to get subsequent results?
-- Have you visited the Developer Discussions feature on https://dev.twitter.com/discussions yet? Twitter developer links: Documentation and resources: https://dev.twitter.com/docs API updates via Twitter: https://twitter.com/twitterapi Unsubscribe or change your group membership settings: http://groups.google.com/group/twitter-development-talk/subscribe
Re: [twitter-dev] Re: Tweet button and iphone
Hi Anajjar, On Jul 17, 2011, at 7:35 AM, anaj...@ibs.com.jo wrote: Thank you very much,can you please inform me when it's going to be deployed this week? or at least update me on this thread once it's deployed? Thanks again This fix is live now. You should now find that cancel buttons link to the original_referer where appropriate. Ben -- Have you visited the Developer Discussions feature on https://dev.twitter.com/discussions yet? Twitter developer links: Documentation and resources: https://dev.twitter.com/docs API updates via Twitter: https://twitter.com/twitterapi Unsubscribe or change your group membership settings: http://groups.google.com/group/twitter-development-talk/subscribe
Re: [twitter-dev] Re: Tweet button and iphone
Hi Anajjar, The fix for the cancel button links is done, and should be deployed next week. Ben On Jul 13, 2011, at 11:05 PM, anaj...@ibs.com.jo wrote: Can someone please respond on my issue? Why the support doesn't follow up with my issue to the end?!! :\ ,Is there any other place to get better support than here ? -- Have you visited the Developer Discussions feature on https://dev.twitter.com/discussions yet? Twitter developer links: Documentation and resources: https://dev.twitter.com/docs API updates via Twitter: https://twitter.com/twitterapi Unsubscribe or change your group membership settings: http://groups.google.com/group/twitter-development-talk/subscribe -- Have you visited the Developer Discussions feature on https://dev.twitter.com/discussions yet? Twitter developer links: Documentation and resources: https://dev.twitter.com/docs API updates via Twitter: https://twitter.com/twitterapi Unsubscribe or change your group membership settings: http://groups.google.com/group/twitter-development-talk/subscribe
Re: [twitter-dev] What is the best strategy to get this volume of data?
The intention was to do lots of the same query, perhaps 80 unique queries, but every 30 seconds or less -- Twitter developer documentation and resources: https://dev.twitter.com/doc API updates via Twitter: https://twitter.com/twitterapi Issues/Enhancements Tracker: https://code.google.com/p/twitter-api/issues/list Change your membership to this group: https://groups.google.com/forum/#!forum/twitter-development-talk
Re: [twitter-dev] Re: Tweet button and iphone
Hi Anajjar, On Jul 7, 2011, at 6:28 AM, anaj...@ibs.com.jo wrote: the login page will appear and you can see the cancel button,click on cancel button,now it should return the user back to that link and as you can see the link has a querystring original_referer with the original url so the cancel button SHOULD return back to that url,but you will notice that it doesn't return the user back to that page and instead it redirects to twitter homepage Ah, you've found a bug. Thanks! I'll fix that. The cancel button there should behave the same way as cancel buttons elsewhere (e.g. in the after-Tweet screen) and redirect to the original referrer. On Jul 2, 2011, anaj...@ibs.com.jo wrote: In Safari I don't have any issues,but I do have issues with tweet button within the container,once the login window of twitter appears and user clicks on cancel it doesn't return the user back to my page,instead it redirects the user to twitter homepage,Is there any solution for this? For what it's worth, this is the full cascade of behaviours that Cancel buttons have within Web Intents: 1. If the intent is a pop-up window (e.g. if `window.opener` is available) then clicking a cancel button will close the window. On desktop, this closes the window as you'd expect. On mobile platforms such as the iPhone, this closes the page and flips the user back to the page that invoked the Intent. 2. If `window.opener` isn't available, then the page is assumed to have been navigated to conventionally. In this case, the destination of cancel links is (or should be) set with the following priority: 2a. To the original_referrer, if available and valid. 2b. If original_referrer is not set, or if the original_referrer is an invalid URL, or that URL falls foul of our known-malware-URL filter, then the cancel buttons will simply point to twitter.com. Finally, note that if you “pop up” an intent within a WebView in a native app, that does not set `window.opener`, so the page will believe it's been invoked directly in a browser (because it has.) If you want to handle the close event, you'll need to watch the source URLs. If it's possible to create a WebView on your platform of choice, and set a default window.opener *and* override window.close() with a call to native ‘close the WebView’ code, then you may be able invoke the Intent in a WebView and have all the autoclose behaviour from the web work. I've never tried this though. Ben -- @benward Twitter, Web Intents developer -- Twitter developer documentation and resources: https://dev.twitter.com/doc API updates via Twitter: https://twitter.com/twitterapi Issues/Enhancements Tracker: https://code.google.com/p/twitter-api/issues/list Change your membership to this group: https://groups.google.com/forum/#!forum/twitter-development-talk
[twitter-dev] What is the best strategy to get this volume of data?
I want to send 100 queries to Twitter every minute via my server application and would like to know if there is a method that will allow me to do this? -- Twitter developer documentation and resources: https://dev.twitter.com/doc API updates via Twitter: https://twitter.com/twitterapi Issues/Enhancements Tracker: https://code.google.com/p/twitter-api/issues/list Change your membership to this group: https://groups.google.com/forum/#!forum/twitter-development-talk
Re: [twitter-dev] Unwanted T.CO shortening
On Jun 10, 2011, at 1:21 PM, Kosso wrote: The massive trouble I have with all this is that I like to know what the hell I'm clicking on before clicking a link. It's kind of my right as a citizen of the web. I personally can't stand it when, for example a link fires up iTunes or goes to some site I don't want to waste (possibly mobile and limited) bandwidth on. I like to choose WHO I give MY visit/traffic to. To be clear, the API returns all the information for all clients to display the original short URL, and navigate via t.co. We also look up the full destination URL and return that too, allowing even clearer navigation of where you as a user will end up when following a link. You can see this implemented on twitter.com today: https://twitter.com/joshtpm/status/79283124747501568 * The URL destination points to t.co * The displayed text of the URL is a cropped and shortened version of the real URL * The title (tooltip) of the URL displays the full address of the destination. I've further illustrated it with a screenshot here: https://skitch.com/benward/frff8/ The documentation for the URL entities that provide all of this information in the API response is here: http://dev.twitter.com/pages/tweet_entities Ben -- Platform Developer, Twitter -- Twitter developer documentation and resources: https://dev.twitter.com/doc API updates via Twitter: https://twitter.com/twitterapi Issues/Enhancements Tracker: https://code.google.com/p/twitter-api/issues/list Change your membership to this group: https://groups.google.com/forum/#!forum/twitter-development-talk
Re: A Problem with the web view. [was: Re: [twitter-dev] Please confirm this OAuth flow ...]
Hi Andrew,
On May 27, 2011, at 7:02 AM, Andrew W. Donoho wrote:
My question: Am I in error? Does Twitter return a different callback value
for the No, Thanks case. If not, I humbly request that Twitter add this
important, for a better user experience, notification.
We actually do have a simple callback format for when the user denies access.
It's not an automatic redirect, but we've always provided a link for users back
into the app.
It works like this:
When a user doesn't click authorize they go to a screen informing them either
that the application does not have access to their account, or that the
application *does* still have access to the account (if they choose ‘Cancel’ on
the `/authenticate` flow for an already authorized, for example.) This callback
is the same URL as for the success case, but returns without an
`oauth_verifier` parameter (since there isn't one) and with an additional
`denied` parameter, in which we pass the value of the `oauth_token` that the
request was originally made with.
Now, on the cancellation pages as your currently see them there's a link back
to the app that is neither big enough or obvious enough for users to respond
to. Therefore, we already have a design tweak implemented for next week that
makes it a much clearer “Return to {{Application}}” button on the denied and
cancel screens. Nothing is changing about the callback structure itself, which
you can already access for testing.
This should keep both ends of the flow hooked into all kinds of apps. As ever,
let us know if there's ways we can improve it for you; I'm particularly
interested in suggestions of extra data to send through in that callback to
improve the feedback in apps.
Thanks,
Ben
--
@benward
Platform Developer, Twitter
--
Twitter developer documentation and resources: https://dev.twitter.com/doc
API updates via Twitter: https://twitter.com/twitterapi
Issues/Enhancements Tracker: https://code.google.com/p/twitter-api/issues/list
Change your membership to this group:
https://groups.google.com/forum/#!forum/twitter-development-talk
Re: [twitter-dev] Re: Visual refresh of the OAuth screens
Clive, Congratulations on the new release—it works great! And thank you so much for writing up your implementation in this detail. I'm glad it's come together and I'm glad you've been able to use the callbacks mechanism to improve the flow. Hopefully we're at a point in mobile platform development now where OOBs become a real rarity for users, and we can use callbacks for better app integration in all sorts of ways. Also, thanks for your patience and enthusiastic working with me to get the problem diagnosed. It was incredibly helpful to get your feedback and assistance, and took a bit of stress out of that weekend! I really appreciate it. On May 25, 2011, at 3:53 AM, bitrace wrote: The main reason we opted for the embedded UIWebView was because our TweetIgnite App allows the user to add multiple accounts, and we found that the cookies twitter sets after adding the first account makes it difficult to add a second, as the first account is still logged in and there is currently no option on the new screens to log in as different user. For what it's worth, the `force_login` parameter which we'll release soon on the authorize flow is a solution to this problem. (As mentioned in some of the other threads.) Regards, Ben -- @benward Platform Developer, Twitter Thankfully, we have now fixed the issue that caused the new OAuth templates to break our iPhone App and an update is now live on the App store. I thought I'd share our findings as they may prove useful to anyone else who has experienced problems or who is currently migrating from xAuth to oAuth in order to meet the new Direct Message access requirement. The problem we experienced following the rollout of the new OAuth screens occurred when a user tried to add their twitter account using the in-App UIWebView. This rendered blank instead of prompting the user to enter their username and password. We traced the problem to how our OAuth helper was interacting with the UIWebView. In our initial implementation the OAuth helper code was both creating and issuing the OAuth requests, and passing html responses back to the UIWebView, as text to be rendered. Although this worked fine with the old style screens, changes to twitter's anti- click jacking implementation in the new templates resulted in the UIWebView being rendered blank. The solution was simply to get the OAuthHelper code to create the OAuth signed requests, and pass the NSURLRequest directly to the UIWebview for loading directly. Our initial OAuth implementation made use of a PIN based out-of-band flow, however, as part of our redesign we found that we were able to move over to an OAuth callback implementation, which provides a much slicker user experience. To do this: 1. Login in to twitter's developer page and change your applications settings to browser and register a nominal callback URL, say your twitter page (this callback URL is not actually used). 2. Back in your App, define a protocol specific callback url e.g. yourappname://oauth/callback ours was tweetignite://oauth/callback 3. Define the oauth_callback as your protocol specific url in the RequestToken request. [Take care not to over URL encode the callback URL in the header - a big time waster! 8-( ] e.g. OAuth oauth_callback=tweetignite%3A%2F%2Foauth%2Fcallback, oauth_consumer_key=x, oauth_nonce=C8612F28- D307-446B-88E6-4804282CCB4F, oauth_signature=xOKzHrQk4HwhOzdGEdWalVPxF6E%3D, oauth_signature_method=HMAC-SHA1, oauth_timestamp=1306318989, oauth_version=1.0 4. Append the protocol specific url as an oauth_callback parameter in the Authorize URL e.g. https://api.twitter.com/oauth/authorize?oauth_callback=tweetignite%3A%2F%2Foauth%2Fcallbackoauth_token=xxx Once you have implemented this in your OAuth code there are two choices for implementing the webview. You can either spawn the request externally in mobile safari, and add your protocol specific callback URL to your Info.plist, which will cause your app to re- launch when it gets the callback. To get the tokens add - (BOOL)application:(UIApplication *)application handleOpenURL:(NSURL *)url to your AppDelegate and grab the tokens from the oauth_verifier to complete the flow. Or alternatively, as in our case continue to use the embedded UIWebview, which actually receives the callback URL directly in -(BOOL)webView:(UIWebView *)webView shouldStartLoadWithRequest: (NSURLRequest *)request navigationType: (UIWebViewNavigationType)navigationType here you can grab the tokens from the oauth_verifier to complete the flow. With the embedded UIWebView approach you need to programmatically clear any cookies twitter sets, before launching a new UIWebView so a new account can be added. Thanks to Ben at twitter for his help with this, and hopefully this will be useful to others migrating from xAuth to a full OAuth flow. Best regards
Re: [twitter-dev] window.open() and OAuth
Hi Corey, Thanks for your feedback. On May 5, 2011, at 8:15 AM, Corey Ballou wrote: Your new OAuth authentication handler does a check to determine if the window has been opened in a new window and triggers a resize. I'll preface this message by saying that I have a high res monitor at 1920x1600. I currently have handling to center the window. Your new JavaScript is essentially resizing the window outside of the viewport, giving no consideration to the end user's window height or current window position. Is this something that can be resolved? It's kind of a nuisance. I'll repost in the tracker. The auto-resize is not something I'm a massive fan of either, but we implemented it after the redesign because we found that whilst a lot of OAuth implementations are using pop-ups of a fixed sized, a great number of them are also invoking those pop-ups with scrollbars disabled at the window level, which makes important parts of the interface impossible for users to access. Alas, though I appreciate that centering the window is aesthetically desirable, content hidden through disabled scrollbars was a bigger problem, so it's been patched for now. I emphasise ‘for now’ because I know and agree that this isn't an ideal solution. Web content is by its nature of variable length. In this case: • Additional status messages can be displayed in the content. The authenticate flow has an extra header message, and users adding their first app are also given a special greeting message, and offered a link to the help section to explain what apps are. • Items on the page, and in the permission list may be changed, shortened, lengthened or removed outright based on feedback and experience. And, of course, Twitter will add new features over time, or might redesign the site again some day. These are all changes that developers should expect to happen to every web service, not just Twitter, and should code for defensively. What's more, users can, will, and do use browser features such as resizing page text based on their needs, which throws all page size expectations out of the window. Given all of this, it mystifies me slightly that browsers allow scrollbars to be disabled at all, but here we are. I'm keen that we assemble issues like this into a best practices guide. Not disabling scrollbars, and keeping the address bar (and SSL verification stamps) visible are important. There's more too, and I'd like to collect the feedback and experience of developers to help assemble thorough and relevant advice. (Please write in to the usual address.) (This one.) My hope and preference is that after documenting these issues and encouraging their adoption, we can remove the janky resize code and return all pop-up size and shape related issues entirely to the domain of the app developer. Thanks, Ben -- @benward -- Twitter developer documentation and resources: http://dev.twitter.com/doc API updates via Twitter: http://twitter.com/twitterapi Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list Change your membership to this group: http://groups.google.com/group/twitter-development-talk
Re: [twitter-dev] OAuth Authorization screen on Windows Phone 7 (WebBrowser control)
Hi Chris, Matthieu and any others running into this issue: On May 3, 2011, at 5:00 PM, LoungeFlyZ wrote: Something has changed in the last few days and now the page is rendering a little better but scrolling/panning/zooming isnt working i cant enter text in the username or password fields. Is there anything i can do to help diagnose this issue? The update we put out this morning fixes the general rendering of the OAuth screen in WP7 (there's a couple more minor layout improvement tweaks coming as well, but the blocking problem is rectified.) It now works perfectly in Mobile IE on the phone, but for reasons which are inexplicable, loading the exact same page within a phone:WebBrowser control in an application suffers from cropped rendering and the no-scrolling behaviour described here. Having spent a few hours in Visual Studio, and reproduced it, it appears you can work around the issue by setting IsScriptEnabled=False on the control. I'll continue to investigate what's causing the render bug, but the above fix should hopefully set everybody running again. Thanks for your patience, and to Chris and Matthieu for corresponding with me on Twitter whilst debugging the issue. Ben -- Twitter developer documentation and resources: http://dev.twitter.com/doc API updates via Twitter: http://twitter.com/twitterapi Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list Change your membership to this group: http://groups.google.com/group/twitter-development-talk
Re: [twitter-dev] New oAuth Authorization screen is unusable on phone webbrowser control
Hi Bob, Tom, and others, On Apr 29, 2011, at 11:50 PM, Bob12345 wrote: I've been using a WebBrowser control in my Window Phone application to login into Twitter. Today I noticed that the login/authorization page format had changed and it is now unusable in a web browser control that my application displays. The text on the page is squeezed together, and the page unscrollable. I've seen a couple of reports of this concerning rendering in the current browser on Windows Phone 7. This is obviously unintended and I'm working on a fix. On Apr 30, 2011, at 9:09 AM, Tom van der Woerdt wrote: It sounds like all UIWebView, WebBrowser and probably Android's WebView are blocked. This is definitely a *good* thing for security reasons. The workaround I recommend: launch the actual browser, using a yourapp:// link (something like myapplication://tokenDone) as the return URL. This is a LOT safer for the users. Although I'm personally a strong advocate of the protocol redirects, I can assure you that we have not actively blocked access to the OAuth screens in any context or browser with these updates. If you're having trouble with the auth screen in a web view, I'm going to need more information from you because there's no debugging tool for a UIWebView in third party apps. If you can trap any rendering or script errors from a browser view control in your development environment, please send them to me and I'll use them to look for any problems. Email me directly if you prefer. Thanks, Ben / @benward / Twitter platform developer -- Twitter developer documentation and resources: http://dev.twitter.com/doc API updates via Twitter: http://twitter.com/twitterapi Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list Change your membership to this group: http://groups.google.com/group/twitter-development-talk
[twitter-dev] t.co user agent string
Hi, I've been trying to find out what the user agent string is for the t.co bot that pings urls posted to Twitter. I've had a look through the logs of some newly posted links but I can't pick out which one is Twitter's bot. There were a couple of mentions of Gnip, Google, Voyager etc... Thanks, Ben -- Twitter developer documentation and resources: http://dev.twitter.com/doc API updates via Twitter: http://twitter.com/twitterapi Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list Change your membership to this group: http://groups.google.com/group/twitter-development-talk
[twitter-dev] Does backtype offer better searching for a URL than the twitter api?
I'm trying to search the API for a specific URL (non-shortened) but am getting different results on how many times the URL has been tweeted. The URL I'm using for this example is http://groups.google.com/group/twitter-development-talk http://urls.api.twitter.com/1/urls/count.json?url=http://groups.google.com/group/twitter-development-talk Shows a count of 42 http://search.twitter.com/search?q=http://groups.google.com/group/twitter-development-talk doesn't give me any results http://www.backtype.com/page/groups.google.com/group/twitter-development-talk/conversations shows 151 tweets It looks like backtype does a better job of the shortened links, but from my testing the twitter api will find the results from the shortened URL's *sometimes*. I've also found that backtype can return more results than the twitter api. For example: http://search.twitter.com/search?q=http://37signals.com/rework/ gives me about 30 tweets going back about a week. http://www.backtype.com/page/37signals.com/rework/conversations gives me 3,836 tweets going back much further. So I guess I'm asking if it's possible that backtype is providing a much better search than twitter, or I'm doing something wrong. It seems like backtype is somehow grabbing every tweet - is that even possible? It just doesn't seem right that they can be offering a better search results of twitters data than twitter can (they also seem to do a better job on shortened URL's than Google which returns the same results as twitter). -- Twitter developer documentation and resources: http://dev.twitter.com/doc API updates via Twitter: http://twitter.com/twitterapi Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list Change your membership to this group: http://groups.google.com/group/twitter-development-talk
[twitter-dev] Easy there, Turbo. Too many requests recently. Enhance your calm. - User streams limit?
Hi, we're working on a Twitter client, using user streams API. Since yesterday I found out that for certain accounts, whenever we try to connect using user streams, we get the following error: Easy there, Turbo. Too many requests recently. Enhance your calm.. And ever since this happened, the account itself seems to have been banned from connecting to user streams--even when I open up Tweetdeck, it doesn't deliver user streams anymore, but instead seems to poll the server once in a while using REST. We haven't done anything abnormal. Has anyone experienced the same problem? And could you enlighten me on how to go about solving this problem? Thanks! -- Twitter developer documentation and resources: http://dev.twitter.com/doc API updates via Twitter: http://twitter.com/twitterapi Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list Change your membership to this group: http://groups.google.com/group/twitter-development-talk
[twitter-dev] cURL and rate limits
I'm just starting to mess around with a very, very basic call to the Twitter API (http://api.twitter.com/1/statuses/user_timeline.json) to pull my tweets to my website through cURL. However, using a page that nobody knows exists yet (thus eliminating the possibility of inadvertent traffic), I'm getting a Rate Limit Exceeded thing before I've had the chance to even test it. It says it resets at 5 past the hour, so I check again, and for a minute it works but then it's back to telling me my rate limit is exceeded. A few questions: First, is the rate limit applied to my server? I would assume so, but that could make it tough of course. Even one API call per visitor could, on a site with marginal traffic, easily surpass the rate limit in an hour. Is there a way to associate the call with the visitor, not the server? Seems like probably not, but I'm not entirely sure how the whole API works, and cURL does seem to be advocated in a number of places. I'm aware that if I use JSON and AJAX the data in I can make that request from the user, but just for the sake of argument, what about cURL? Second, any idea how I could be surpassing my rate limit without even refreshing the page? I pay for hosting at another location, so I might be sharing server space with another site, but my site definitely has a unique IP, so that should … that should be OK, right? So how is it that I'm surpassing the rate limit without even running the code (or by running it once?)? Here's what I've got for code, if it helps: $ch=curl_init(http://api.twitter.com/1/statuses/user_timeline.json? screen_name=bensaufley); curl_setopt_array($ch,array( CURLOPT_RETURNTRANSFER=true, CURLOPT_TIMEOUT=5, ) ); $temp=curl_exec($ch); curl_close($ch); $results=json_decode($temp,true); Any help? -- Twitter developer documentation and resources: http://dev.twitter.com/doc API updates via Twitter: http://twitter.com/twitterapi Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list Change your membership to this group: http://groups.google.com/group/twitter-development-talk
[twitter-dev] Steaming API — is track lagging cons iderably behind real–time right now?
Hi, I’m having some trouble with the streaming API. Results for terms we’re tracking for Cursebird (http://cursebird.com/) are coming in about 40 minutes late right now. Currently every single line is also a duplicate. I’ve verified this both in production and on my local development machine. Are other people experiencing this? I would imagine it’s transient but it’s been like this for approximately 40 minutes. Is it just me, or is something broken at Twitter? Ben -- Twitter developer documentation and resources: http://dev.twitter.com/doc API updates via Twitter: http://twitter.com/twitterapi Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list Change your membership to this group: http://groups.google.com/group/twitter-development-talk
[twitter-dev] Search API problems...
Hey guys - I'm curious as to know whether there's any problems with the search API? I'm curling from a PHP script, and it keeps timing out with 'couldn't connect to host' errors when my URL is a search (eg: http://search.twitter.com/search.atom?q=test). Interestingly, if I curl either of the following: http://api.twitter.com/1/help/test.xml http://api.twitter.com/1/account/rate_limit_status.xml ...and it doesn't time out, I get a true, and my rate limit is 150/150. I'm not using any authentication, this is a straight request from a script. Could I be on an IP blacklist for search (can I check this?)? I've been pretty careful with my caching, I make nowhere near 150 requests an hour, although my site is on a shared server, so it's entirely plausible someone else has been hammering it. Although if that was the case, would something not show up on the odd times I actually get the rate limit to show something? If anyone can help, or point me in the direction of something I've missed, I'd be eternally grateful... ben -- Twitter developer documentation and resources: http://dev.twitter.com/doc API updates via Twitter: http://twitter.com/twitterapi Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list Change your membership to this group: http://groups.google.com/group/twitter-development-talk?hl=en
[twitter-dev] Re: Search API problems...
Matt, thanks for the quick response. After an evening of trying to figure out what's going on, it appears to be working again. I guess the problem must have been on my side. Thank you so much for replying so quickly though, and for the explanation on rates and error messages! Many thanks, ben On Aug 25, 1:02 am, Matt Harris thematthar...@twitter.com wrote: There are no known issues with search and running your query works for me. Hey Ben, The Search API does not use authentication and is rate limited differently to the 150 IP requests allowed on the REST API. If you are rate limited on the Search API we would return an error telling you rather than not reply. If the atom link is still not responding can you tryhttp://search.twitter.com/search.json?q=testand let us know the result? Thanks, Matt On Tue, Aug 24, 2010 at 4:28 PM, Ben goo...@iamben.co.uk wrote: Hey guys - I'm curious as to know whether there's any problems with the search API? I'm curling from a PHP script, and it keeps timing out with 'couldn't connect to host' errors when my URL is a search (eg: http://search.twitter.com/search.atom?q=test). Interestingly, if I curl either of the following: http://api.twitter.com/1/help/test.xml http://api.twitter.com/1/account/rate_limit_status.xml ...and it doesn't time out, I get a true, and my rate limit is 150/150. I'm not using any authentication, this is a straight request from a script. Could I be on an IP blacklist for search (can I check this?)? I've been pretty careful with my caching, I make nowhere near 150 requests an hour, although my site is on a shared server, so it's entirely plausible someone else has been hammering it. Although if that was the case, would something not show up on the odd times I actually get the rate limit to show something? If anyone can help, or point me in the direction of something I've missed, I'd be eternally grateful... ben -- Twitter developer documentation and resources:http://dev.twitter.com/doc API updates via Twitter:http://twitter.com/twitterapi Issues/Enhancements Tracker:http://code.google.com/p/twitter-api/issues/list Change your membership to this group:http://groups.google.com/group/twitter-development-talk?hl=en -- Matt Harris Developer Advocate, Twitterhttp://twitter.com/themattharris -- Twitter developer documentation and resources: http://dev.twitter.com/doc API updates via Twitter: http://twitter.com/twitterapi Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list Change your membership to this group: http://groups.google.com/group/twitter-development-talk?hl=en
[twitter-dev] Re: Update on Twifficiency
What I'd actually like to see is some granularity in the oAuth permissions that go beyond binary has complete access: DENY|ALLOW, and this would also solve this problem. Surprising users when an app auto-tweets is one thing, but I'm more concerned about a given app reading my DM's, for example (which I wouldn't know about, thus no 'surprise' but still bad). I would urge Twitter to look at Flickr's oAuth (well 'oAuth style') auth which lets users dictate the level of access a given app is allowed and even let developers appropriately request only the right level they need. Twifficiency technically only needed read-only access to my public tweets (ok, it wouldn't have had the viral aspect). If when I oAuthed for it the twitter landing page said: Give app Twifficiency access to the following on your account? : [x] public tweets [ ] send tweets [ ] read direct messages This seems more appropriate but would also deal with the issue of surprising auto-tweets when the app developer doesn't highlight it up front. What do people think? Thanks, Ben Metcalfe On Aug 18, 1:45 pm, Brian Sutorius bsutor...@twitter.com wrote: Hi all, Over the past 24 hours, we've received some questions about the Twifficiency app, so we thought we'd use this as an opportunity to quickly share some information around our Developer Principles. For background, the Twifficiency app computes a Twifficiency score based on different aspects of your Twitter account and posts the score as a Tweet. While the developer included a disclaimer that these Tweets would be posted to Twitter, user feedback indicated that the text was too far down on the page to be noticed before proceeding. As a result, many users were surprised that their scores were being tweeted automatically. Which brings us to our Developer Principles, one of which is Don't surprise users. Specifically, we require developers to get users' permission before sending Tweets or other messages on their behalf. Allowing an application to access your account does not constitute consent for actions to automatically be taken on your behalf. Twifficiency violated this principle, so we suspended the app yesterday afternoon while we worked with the developer to make sure users were better informed about the application's actions and could control whether or not a Tweet would be posted. With these changes --which include a more prominent warning and a checkbox on the main page-- the application has been re-enabled. Our developer principles can be found in our API Terms of Service:http://dev.twitter.com/pages/api_terms Brian Sutorius API Policy
[twitter-dev] Can I use the Twitter API to get tweets for a specific user and timeframe?
I've been doing some research into using the Twitter API, and I'm not sure if I'm understanding it correctly. I want to get tweets from a specific user for a specific time-frame. From what I can tell, using the search function to specify a date range doesn't work because only the last 7 days are kept. I could just get the tweets and work out the date range with JavaScript, but the documentation states the the max you can get is 200 tweets, so if the tweets from the date range I want aren't in the last 200 I can't do it. Is there anyway I can do this?
[twitter-dev] Re: Using the Authorization header
Hi Taylor, Thanks for the pointers! Turned out the problem was related to encoding. I fixed the encoding on the status in the body, and then noticed that I wasn't encoding the '=' at the end of the signature in the header. When I'd put the signature in the body, it was encoded with all the other parameters, so worked correctly. Now both ways are fine. Thanks again, ben On Aug 5, 3:21 pm, Taylor Singletary taylorsinglet...@twitter.com wrote: Hi Everyone on this thread, A few clarifications: - The realm isn't required, but we'll take a realm if you provide one. Really, it's a no-op in our system. We don't care if you have it or not, and if you provide it, we don't do anything with it. - Timestamp is important in that the oauth_timestamp in your requests should be within about 5 minutes of our servers. We return the current time in a header within each response, failed or otherwise. One quick way to get a non-rate-limited check on the current time according to twitter is to send a HEAD request to /1/help/test.xml and inspect the Date header, recontextualizing your app's concept of the current time based on that. Back to the original poster's issue, I noticed you're essentially trying to post this status: Test 1 2 3 4 1280937572396 But your POST body shows this status over-encoded: status=Test%25201%25202%25203%25204%25201280937572396 When it should be: status=Test%201%202%203%204%201280937572396 The encoding in your POST body is correct as to what should be in your signature base string. This might be what is causing your invalid signature errors. Taylor On Thu, Aug 5, 2010 at 5:10 AM, Ben Jones benjamin.david.jo...@gmail.comwrote: Hi Tom/Jacky, Thanks for responding! Apologies if this is a double post, last one didn't seem to go through. I tried removing 'realm' from the Authorization header, and this changed the response the first time, to something like 'Couldn't authenticate you using OAuth', but then returned to the previous behaviour of throwing 500 errors. What should I look out for with regards timestamps? I'm following the advice from the OAuth spec, in that they have to be in seconds, and equal or greater than the last used. Do they have to sync with Twitter's clock? Thanks again! ben On Aug 5, 12:37 am, Tom allerleiga...@gmail.com wrote: Oops, silly me, didn't read the full post. Sorry. Make sure to watch for character encoding and timestamps. Especially timestamps are known to cause trouble. 401 errors are almost never an issue at Twitter. Tom On Aug 5, 1:34 am, Tom allerleiga...@gmail.com wrote: You are sending realm= in your Authorization header. It doesn't belong there. ;-) Tom On Aug 4, 6:19 pm, Ben Jones benjamin.david.jo...@gmail.com wrote: Hi all, I'm currently writing my own OAuth lib for use with Twitter and have gotten stuck whilst using the Authorization HTTP header, rather than putting the OAuth parameters in the body. An example of a request that is failing is: POST /1/statuses/update.xml HTTP/1.1 Authorization: OAuth realm=http://api.twitter.com/1/statuses/ update.xml, oauth_consumer_key=x, oauth_token=x, oauth_nonce=x, oauth_timestamp=1280937572, oauth_signature_method=HMAC-SHA1, oauth_version=1.0, oauth_signature=DLPyc3h6BcC5zbGXrUcujvZnqxk= User-Agent: Java/1.6.0_07 Host: api.twitter.com Accept: text/html, image/gif, image/jpeg, *; q=.2, */*; q=.2 Connection: keep-alive Content-type: application/x-www-form-urlencoded Content-Length: 53 status=Test%25201%25202%25203%25204%25201280937572396 ...(token etc blanked out, and new lines added in) This results in the server returning a 500 error and the Something is technically wrong. error page. I've talked to another developer who doesn't experience this. I've tried this with the parameters alphabetically ordered, unordered and with and without the 'realm' parameter, which isn't used in thehttp:// dev.twitter.com/pages/authpage. I don't have the same problem (as in the 500 error) when I put the OAuth parameters in the request body, but this often fails as well with 401 'Invalid signature' errors. What's strange is that putting the OAuth parameters into the form at Hueniverse's OAuth request signing page (http://tinyurl.com/y9bvjyt) shows them, including the signature, to be correct. If I retry the same request, it eventually works (sometimes it works the first time, just not consistently), so I don't think I'm calculating the signature incorrectly. Are the 401 errors occurring because Twitter is busy, or am I doing something wrong? Thanks in advance, any help would be greatly appreciated! ben
[twitter-dev] Re: Using the Authorization header
Hi Tom/Jacky, Thanks for replying. I've seen a couple of places (The Hueniverse page above being one of them) where the realm is included in the Authorization header. I notice that this isn't the case for Twitter though so I'll remove it. (Tried this, first time, it gave a new error 'Could not authorize you using OAuth' I think, but then it went back to throwing 500 errors and serving up the error page) Regarding the timestamps, I'm just going off the OAuth spec http://oauth.net/core/1.0/ says on this. I've been making requests more than 1 second apart at the moment, so the timestamps fullfil the need to be greater or equal to the previous one. Is there anything else I should watch out for with them? Thanks again! ben On Aug 5, 12:37 am, Tom allerleiga...@gmail.com wrote: Oops, silly me, didn't read the full post. Sorry. Make sure to watch for character encoding and timestamps. Especially timestamps are known to cause trouble. 401 errors are almost never an issue at Twitter. Tom On Aug 5, 1:34 am, Tom allerleiga...@gmail.com wrote: You are sending realm= in your Authorization header. It doesn't belong there. ;-) Tom On Aug 4, 6:19 pm, Ben Jones benjamin.david.jo...@gmail.com wrote: Hi all, I'm currently writing my own OAuth lib for use with Twitter and have gotten stuck whilst using the Authorization HTTP header, rather than putting the OAuth parameters in the body. An example of a request that is failing is: POST /1/statuses/update.xml HTTP/1.1 Authorization: OAuth realm=http://api.twitter.com/1/statuses/ update.xml, oauth_consumer_key=x, oauth_token=x, oauth_nonce=x, oauth_timestamp=1280937572, oauth_signature_method=HMAC-SHA1, oauth_version=1.0, oauth_signature=DLPyc3h6BcC5zbGXrUcujvZnqxk= User-Agent: Java/1.6.0_07 Host: api.twitter.com Accept: text/html, image/gif, image/jpeg, *; q=.2, */*; q=.2 Connection: keep-alive Content-type: application/x-www-form-urlencoded Content-Length: 53 status=Test%25201%25202%25203%25204%25201280937572396 ...(token etc blanked out, and new lines added in) This results in the server returning a 500 error and the Something is technically wrong. error page. I've talked to another developer who doesn't experience this. I've tried this with the parameters alphabetically ordered, unordered and with and without the 'realm' parameter, which isn't used in thehttp://dev.twitter.com/pages/authpage. I don't have the same problem (as in the 500 error) when I put the OAuth parameters in the request body, but this often fails as well with 401 'Invalid signature' errors. What's strange is that putting the OAuth parameters into the form at Hueniverse's OAuth request signing page (http://tinyurl.com/y9bvjyt) shows them, including the signature, to be correct. If I retry the same request, it eventually works (sometimes it works the first time, just not consistently), so I don't think I'm calculating the signature incorrectly. Are the 401 errors occurring because Twitter is busy, or am I doing something wrong? Thanks in advance, any help would be greatly appreciated! ben
[twitter-dev] Re: Using the Authorization header
Hi Tom/Jacky, Thanks for responding! Apologies if this is a double post, last one didn't seem to go through. I tried removing 'realm' from the Authorization header, and this changed the response the first time, to something like 'Couldn't authenticate you using OAuth', but then returned to the previous behaviour of throwing 500 errors. What should I look out for with regards timestamps? I'm following the advice from the OAuth spec, in that they have to be in seconds, and equal or greater than the last used. Do they have to sync with Twitter's clock? Thanks again! ben On Aug 5, 12:37 am, Tom allerleiga...@gmail.com wrote: Oops, silly me, didn't read the full post. Sorry. Make sure to watch for character encoding and timestamps. Especially timestamps are known to cause trouble. 401 errors are almost never an issue at Twitter. Tom On Aug 5, 1:34 am, Tom allerleiga...@gmail.com wrote: You are sending realm= in your Authorization header. It doesn't belong there. ;-) Tom On Aug 4, 6:19 pm, Ben Jones benjamin.david.jo...@gmail.com wrote: Hi all, I'm currently writing my own OAuth lib for use with Twitter and have gotten stuck whilst using the Authorization HTTP header, rather than putting the OAuth parameters in the body. An example of a request that is failing is: POST /1/statuses/update.xml HTTP/1.1 Authorization: OAuth realm=http://api.twitter.com/1/statuses/ update.xml, oauth_consumer_key=x, oauth_token=x, oauth_nonce=x, oauth_timestamp=1280937572, oauth_signature_method=HMAC-SHA1, oauth_version=1.0, oauth_signature=DLPyc3h6BcC5zbGXrUcujvZnqxk= User-Agent: Java/1.6.0_07 Host: api.twitter.com Accept: text/html, image/gif, image/jpeg, *; q=.2, */*; q=.2 Connection: keep-alive Content-type: application/x-www-form-urlencoded Content-Length: 53 status=Test%25201%25202%25203%25204%25201280937572396 ...(token etc blanked out, and new lines added in) This results in the server returning a 500 error and the Something is technically wrong. error page. I've talked to another developer who doesn't experience this. I've tried this with the parameters alphabetically ordered, unordered and with and without the 'realm' parameter, which isn't used in thehttp://dev.twitter.com/pages/authpage. I don't have the same problem (as in the 500 error) when I put the OAuth parameters in the request body, but this often fails as well with 401 'Invalid signature' errors. What's strange is that putting the OAuth parameters into the form at Hueniverse's OAuth request signing page (http://tinyurl.com/y9bvjyt) shows them, including the signature, to be correct. If I retry the same request, it eventually works (sometimes it works the first time, just not consistently), so I don't think I'm calculating the signature incorrectly. Are the 401 errors occurring because Twitter is busy, or am I doing something wrong? Thanks in advance, any help would be greatly appreciated! ben
[twitter-dev] Using the Authorization header
Hi all, I'm currently writing my own OAuth lib for use with Twitter and have gotten stuck whilst using the Authorization HTTP header, rather than putting the OAuth parameters in the body. An example of a request that is failing is: POST /1/statuses/update.xml HTTP/1.1 Authorization: OAuth realm=http://api.twitter.com/1/statuses/ update.xml, oauth_consumer_key=x, oauth_token=x, oauth_nonce=x, oauth_timestamp=1280937572, oauth_signature_method=HMAC-SHA1, oauth_version=1.0, oauth_signature=DLPyc3h6BcC5zbGXrUcujvZnqxk= User-Agent: Java/1.6.0_07 Host: api.twitter.com Accept: text/html, image/gif, image/jpeg, *; q=.2, */*; q=.2 Connection: keep-alive Content-type: application/x-www-form-urlencoded Content-Length: 53 status=Test%25201%25202%25203%25204%25201280937572396 ...(token etc blanked out, and new lines added in) This results in the server returning a 500 error and the Something is technically wrong. error page. I've talked to another developer who doesn't experience this. I've tried this with the parameters alphabetically ordered, unordered and with and without the 'realm' parameter, which isn't used in the http://dev.twitter.com/pages/auth page. I don't have the same problem (as in the 500 error) when I put the OAuth parameters in the request body, but this often fails as well with 401 'Invalid signature' errors. What's strange is that putting the OAuth parameters into the form at Hueniverse's OAuth request signing page (http://tinyurl.com/y9bvjyt) shows them, including the signature, to be correct. If I retry the same request, it eventually works (sometimes it works the first time, just not consistently), so I don't think I'm calculating the signature incorrectly. Are the 401 errors occurring because Twitter is busy, or am I doing something wrong? Thanks in advance, any help would be greatly appreciated! ben
[twitter-dev] Twitter feed showing incorrect dates
The dates are incorrect on my website... http://www.bjuneau.com I'm not sure what I'm doing wrong here? Any help is much appreciated.
[twitter-dev] Search + App Engine = Bad News... OAuth soon please?
The current Search API has a lot of rate limiting issues when used on App Engine (User Agent just doesn't make it work) making it unreliable. I have an app (http://www.connecttweet.com/) that a fair amount of people (for what it is) are using in Alpha mode that I'd like to make publicly available but I don't feel I can until searching is more reliable... OAuth for Search would be perfect, I am searching on a users behalf as part of the app and already have their tokens. Can you shed any light on when this will be possible? I'd offer to be part of an alpha or beta on your end just to get this problem resolved. Thanks, -Ben blog: http://buildcontext.com
[twitter-dev] Re: t.co issue -- querying for original url in streaming search apis
I would love to see twitter include the original unwrapped url in a key/value in the annotation field (else otherwise it's own specific key/value in the payload). There are loads of use cases for this: from search/discovery through to reducing latency for twitter clients that want to show the original url to the user (now potentially a two-stage task if url is original - bit.ly - t.co) B On Jun 9, 12:13 pm, Jim Gilliam j...@gilliam.com wrote: I'm creating a new thread for this because a few others have mentioned it, and we haven't gotten a response yet. My hunch is that changing those APIs involve other teams within Twitter, so figuring out a solution could be challenging. Here is the issue. We need to be able to get matches on the original URL through the streaming and search APIs. For me, I'm tracking act so I can match tweets that link to 'http://act.ly'. This is not a link shortener service, the actual pages live at act.ly, and it was all designed specifically for Twitter so there would be no need for url shorteners. As far as I'm concerned, it's fine if that link changes to t.co, as long as I can still get matches on act.ly (or act) through the streaming API (the search API is going to be important for people too, but less of an issue for me personally). The most elegant way to fix this would be to allow tracking of the original URL. So I can put in a domain name, or URL substring, and match everything that way. Same with search. This would be useful to a lot of people, and virtually all link oriented web apps with APIs provide a way to get all the matches for a particular domain. (digg, google, yahoo, etc) I'm sure there are other workaround ways of doing this, and I'm all ears. It would be SUPER NICE (wink wink) to hear some kind of assurance that there will be a way for us to query this type of information before the t.cochanges go live. Thanks guys... Jim Gilliamhttp://act.ly/http://twitter.com/jgilliam On Tue, Jun 8, 2010 at 4:43 PM, Jim Gilliam j...@gilliam.com wrote: Will we be able to get matches on the original URL through the streaming API? For example, I'm tracking act so I can match tweets that link to ' http://act.ly'. Will I still be able to do that? Jim Gilliam http://act.ly/ http://twitter.com/jgilliam On Tue, Jun 8, 2010 at 4:33 PM, Dewald Pretorius dpr...@gmail.com wrote: Raffi, I'm fine with everything up to the new 140 character count. If you count the characters *after* link wrapping, you are seriously going to mess up my system. My short URLs are currently 18 characters long, and they will be 18 long for quite some time to come. After that they will be 19 for a very long time to come. If you implement this change, a ton, and I mean a *huge* number of my system's updates are going to be rejected for being over 140 characters. On Jun 8, 7:57 pm, Raffi Krikorian ra...@twitter.com wrote: hi all. twitter has been wrapping links in e-mailed DMs for a couple months nowhttp://bit.ly/twttldmemail. with that feature, we're trying to protect users against phishing and other malicious attacks. the way that we're doing this is that any URL that comes through in a DM gets currently wrapped with a twt.tl URL -- if the URL turns out to be malicious, Twitter can simply shut it down, and whoever follows that link will be presented with a page that warns them of potentially malicious content. in a few weeks, we're going to start slowly enabling this throughout the API for all statuses as well, but instead of twt.tl, we will be using t.co. practically, any tweet that is sent through statuses/update that has a link on it will have the link automatically converted to a t.co link on its way through the Twitter platform. if you fetch any tweet created after this change goes live, then its text field will have all its links automatically wrapped with t.co links. when a user clicks on that link, Twitter will redirect them to the original URL after first confirming with our database that that URL is not malicious. on top of the end-user benefit, we hope to eventually provide all developers with aggregate usage data around your applications such as the number of clicks people make on URLs you display (it will, of course, be in aggregate and not identifiable manner). additionally, we want to be able to build services and APIs that can make algorithmic recommendations to users based on the content they are consuming. gathering the data from t.co will help make these possible. our current plan is that no user will see a t.co URL on twitter.com but we still have some details to work through. the links will still be displayed as they were sent in, but the target of the link will be the t.co link instead. and, we want to provide the same ability to display original links to developers. we're going to use
[twitter-dev] why am i not receiving the oauth_verifier in the callback parameters?
Hey all, I'm new to twitter development and am trying to get started. I'm currently using python and the oauth-python-twitter2 oauth client library and twitter api wrapper. My basic problem is when the user gets redirected back to my app after authorizing my app on twitter, I don't get back the oauth_verifier. I do see the oauth_token in my GET parameter, but nothing else. Am I missing something? If there was an error, where can I see that? thanks, ben
[twitter-dev] OAuth rate limit question
Hi, I have been reading about twitter api limits lately as a lot of my users are exhausting their 150reqs/h on a fairly regular basis. I came across the following post and noticed that if users login with OAuth, they are given 350 reqs/hr. http://groups.google.com/group/twitter-development-talk/msg/b09f2a3324728d89?pli=1 This was fair enough as you guys are trying to make twitter more secure (good work!); so we set about implementing OAuth on our client. We completed the implementation today, but fail to see the 350 reqs/ hr. We are still being limited by the 150 reqs/hr. I was just wondering whether there was something special we needed to do to get our req limits up to 350 for those users who login to our client with OAuth. Just to give you some background info, the client is a mobile web based client and all requests to twitter are made on our server on behalf of our users. If they are logged in with OAuth, the appropriate OAuth details are also handed through as part of the request. We know they are using OAuth as our 'updated via xxx' changes with using OAuth. Any help would be greatly appreciated! Thanks! Ben
[twitter-dev] Re: OAuth rate limit question
Awesome! thanks very much! We were still using twitter.com rather than the new api.twitter.com Thanks again! Cheers, Ben On Mar 3, 5:26 pm, Raffi Krikorian ra...@twitter.com wrote: are you connecting via oauth to api.twitter.com? if so, then please take a look at the rate limit headers and let me know what you see? On Tue, Mar 2, 2010 at 10:48 PM, Ben Novakovic bennovako...@gmail.comwrote: Hi, I have been reading about twitter api limits lately as a lot of my users are exhausting their 150reqs/h on a fairly regular basis. I came across the following post and noticed that if users login with OAuth, they are given 350 reqs/hr. http://groups.google.com/group/twitter-development-talk/msg/b09f2a332... This was fair enough as you guys are trying to make twitter more secure (good work!); so we set about implementing OAuth on our client. We completed the implementation today, but fail to see the 350 reqs/ hr. We are still being limited by the 150 reqs/hr. I was just wondering whether there was something special we needed to do to get our req limits up to 350 for those users who login to our client with OAuth. Just to give you some background info, the client is a mobile web based client and all requests to twitter are made on our server on behalf of our users. If they are logged in with OAuth, the appropriate OAuth details are also handed through as part of the request. We know they are using OAuth as our 'updated via xxx' changes with using OAuth. Any help would be greatly appreciated! Thanks! Ben -- Raffi Krikorian Twitter Platform Teamhttp://twitter.com/raffi
[twitter-dev] No results in from search for various (verified?) accounts
I am getting a 200 status message but no results for these two accounts, in both atom and json, and they both definitely have tweets: http://search.twitter.com/search.atom?q=from:lindsaylohan http://search.twitter.com/search.atom?q=from:chucktodd I don't want to postulate, but I wonder if there's something similar between these two (besides their looks, obviously) that makes them not returning results... like perhaps verified and/or moved from a different account name? I looked around through the known issues and other threads, but didn't see one that quite fit the same problem...
[twitter-dev] Getting 500 responses from Twitter constantly
Hi, I run Mobile Tweete (http://m.tweete.net), and I've noticed within the last few hours that users keep getting 500 error response codes from the Twitter Api. I have a dev server running the exact same code on my machine here and I'm experiencing no such errors. I was just wondering whether Mobile Tweete's IP address has been blocked from the API for some reason? I would really love to get to the bottom of this ASAP! Thanks very much! The IP Tweete is running on is: 74.207.242.154. It also appears that 64.71.152.86 doesn't seem to work either. Cheers, Ben
[twitter-dev] Re: Getting 500 responses from Twitter constantly
since posting this - I've noticed other users on here have experienced the same problem. It only appears to happen to certain user accounts on home_timeline. Not sure what the issue is. I'm not over my rate limit because I can view replies/dms. Cheers, Ben On Dec 3, 11:37 am, Ben Novakovic bennovako...@gmail.com wrote: Hi, I run Mobile Tweete (http://m.tweete.net), and I've noticed within the last few hours that users keep getting 500 error response codes from the Twitter Api. I have a dev server running the exact same code on my machine here and I'm experiencing no such errors. I was just wondering whether Mobile Tweete's IP address has been blocked from the API for some reason? I would really love to get to the bottom of this ASAP! Thanks very much! The IP Tweete is running on is: 74.207.242.154. It also appears that 64.71.152.86 doesn't seem to work either. Cheers, Ben
[twitter-dev] Re: List of all verified accounts
It exists already - http://twitter.com/verified Anyone that account follows is verified. On Oct 12, 9:59 am, Kiam kiamc...@gmail.com wrote: Hi, Is there a way to obtain a list of all verified accounts? Or are there plans to expose this in the future? Thanks, Kiam
[twitter-dev] twitter legal
for group interest: http://news.bbc.co.uk/1/hi/technology/8285954.stm usually the reg gets these stories about a week before the beeb, maybe i missed it. On 1 Oct 2009, at 21:22, Christian Nunciato wrote: Has anyone had any trouble sending strings containing line breaks to Twitter through OAuth? I realize line breaks don't necessarily make sense in a Twitter update, but it still seems reasonable to be able to send them anyway, leaving the formatting up to Twitter, but in my case, I keep getting incorrect signature responses from /statuses/ update when I send them. Tweetdeck, by comparison, uses basic-auth, and is sending them through URL-encoded, without error. (Twitter, of course, replaces the line breaks with spaces.) But with OAuth, I'm not having any luck. It might just be my code, but things seem to be getting encoded properly (%0D), so I figured I'd check with the group to see whether anyone else had had a similar experience before going any further. Thanks in advance -- Chris
[twitter-dev] Re: Draft: Twitter Rules for API Use
My reading is that twitter can blacklist an app at any time for any reason - Twitter may suspend or terminate your access to the API - That being true, effectively it means that's the rule and the other points are really guidelines. It provides total freedom for the Platform team. That's also my understanding of the current rules/TOS. It's not obvious how to make an investable app on this basis. Am i wrong about this. Or can anyone confirm experience of successfully negotiating around this caveat for a due diligence? Ben On 11 Sep 2009, at 13:43, Dewald Pretorius wrote: I guess the lawyers wrote this draft as an extension of the modified Twitter TOS. Alex, you will need to jump on this draft from a dizzy height and get all your Platform rules in there. Once the API Rules are published as The Rules you will have no grounds to blacklist an application for other than what is written in The Rules. Unless the rules also state that, we can blacklist an app for any other reason as we deem fit, which will fly like a lead balloon. If the rules are not clear and comprehensive, they will become a ball and chain around the ankles of the Platform team. Dewald
[twitter-dev] Re: Problem in past 48 hours: friendships/create severe lag or loss
I haven't tested right through yet but issues relating to the POST/ auth requests from over the w/e and yesterday look largely resolved for me with all actions queued up and executed in the end. Thank-you for getting onto that and sorting. Ben On 31 Aug 2009, at 20:07, PJB wrote: Thanks Jon... can you let us know if past friendships/create (etc) calls that haven't yet worked, will eventually work? Since we database all of these actions, we're worried that we're going to have bad data for the past, e.g., 48 hours, unless those non-error calls actually go through. On Aug 31, 12:03 pm, John Kalucki jkalu...@gmail.com wrote: We're on this. Updates from the usual sources soon. On Aug 31, 11:57 am, David Dellanave david.dellan...@gmail.com wrote: I am pretty sure I am experiencing this issue as well. I can't verify it, yet. I assumed it was an issue with OAuth, but it seems like that it is the same issue.
[twitter-dev] Re: Problem in past 48 hours: friendships/create severe lag or loss
Thank-you, good to know. On 31 Aug 2009, at 20:03, John Kalucki wrote: We're on this. Updates from the usual sources soon. On Aug 31, 11:57 am, David Dellanave david.dellan...@gmail.com wrote: I am pretty sure I am experiencing this issue as well. I can't verify it, yet. I assumed it was an issue with OAuth, but it seems like that it is the same issue.
[twitter-dev] Re: /friendships/destroy.json /blocks/blocking /blocking/ids
Ok, further to email below, i'm not able to post follow requests
either, it comes back with success, but no database update seems to
have occurred.
Pretty much looks like any POST request just isn't happening, even
though twitter is coming back with friendly responses.
Ben
On 29 Aug 2009, at 17:35, ben wrote:
Couple of issues sprouting, i have been editing the code so i'm going
back and back over for some error at my end, but it's looking mighty
odd.
It looks like:
/friendships/destroy.json is not being updated
blocks/blocking.json is not returning the same users as /blocks/
blocking/ids.json
Logging below. My ip for is 67.23.28.168 and times Europe/London.
logging for friendships/destory:
2009-08-29 17:10:56,702 - twitterauth.utils - INFO - twitter api exe.
- https://twitter.com/friendships/destroy.json. kwargs={'screen_name':
u'arthurcooke'} method=POST is_auth=True
2009-08-29 17:10:57,166 - twitterauth.utils - DEBUG - api auth
response, {u'id': 16612822, u'verified': False,
u'profile_sidebar_fill_color': u'252429', u'profile_text_color':
u'66', u'followers_count': 14, u'protected': True, u'location':
u'Leeds, UK', u'profile_background_color': u'1A1B1F', u'status':
{u'favorited': False, u'truncated': False, u'text': u'Relaxing in
Aspen.', u'created_at': u'Mon Aug 10 22:45:51 + 2009', u'source':
u'web', u'in_reply_to_status_id': None, u'in_reply_to_screen_name':
None, u'in_reply_to_user_id': None, u'id': 3233921143}, u'utc_offset':
0, u'statuses_count': 46, u'description': u'Runner, Mountain Biker,
Science Teacher, Film Fan', u'friends_count': 18,
u'profile_link_color': u'2FC2EF', u'profile_image_url': u'http://
a3.twimg.com/profile_images/81150781/IMG_0320_normal.JPG',
u'notifications': False, u'profile_background_image_url': u'http://
s.twimg.com/a/1251493570/images/themes/theme9/bg.gif', u'screen_name':
u'arthurcooke', u'profile_background_tile': False,
u'favourites_count': 0, u'name': u'arthur cooke', u'url': None,
u'created_at': u'Mon Oct 06 12:07:50 + 2008', u'time_zone':
u'London', u'profile_sidebar_border_color': u'181A1E', u'following':
True}
then..not long later
2009-08-29 17:10:59,746 - twitterauth.utils - INFO - twitter api exe.
- https://twitter.com/statuses/friends.json. kwargs={'screen_name':
u'b_e_n_', 'page': 1} method=GET is_auth=False
the response is parsed and friends username listed in log..note
including the one who was just destroyed.
2009-08-29 17:11:00,206 - twittersearch.models - DEBUG - get current
friends for b_e_n_ - [u'TomChivers', u'Love_London', u'Fast140',
u'performativeweb', u'bbcbreaking', u'rhonafwelsh', u'danifromCO',
u'elliottzone', u'maustyn', u'unstuckdesign', u'arthurcooke',
u'jools2']
logging for blocks/blocking, last few logs show 3 blocked users from
blocking.json and 2 from ids.json:
009-08-29 17:11:39,268 - twitterauth.utils - INFO - twitter api exe. -
https://twitter.com/blocks/blocking.json. kwargs={'page': 1}
method=GET is_auth=True
2009-08-29 17:11:39,608 - twitterauth.utils - DEBUG - api auth
response, [{u'id': 16803782, u'verified': False,
u'profile_sidebar_fill_color': u'e0ff92', u'profile_text_color':
u'00', u'followers_count': 8, u'protected': False, u'location':
u'United States', u'profile_background_color': u'9ae4e8',
u'utc_offset': -32400, u'statuses_count': 0, u'description': u':spends
a lot of hours surfing the net', u'friends_count': 111,
u'profile_link_color': u'ff', u'profile_image_url': u'http://
a3.twimg.com/profile_images/
62184497/212297864a2022621782b584997122l_normal.jpg',
u'notifications': False, u'profile_background_image_url': u'http://
s.twimg.com/a/1251493570/images/themes/theme1/bg.gif', u'screen_name':
u'SARAH161008', u'profile_background_tile': False,
u'favourites_count': 0, u'name': u'SARAH161008', u'url': u'http://MY-
OWN-PAGE.TK', u'created_at': u'Thu Oct 16 07:03:12 + 2008',
u'time_zone': u'Alaska', u'profile_sidebar_border_color': u'87bc44',
u'following': False}, {u'id': 16804065, u'verified': False,
u'profile_sidebar_fill_color': u'e0ff92', u'profile_text_color':
u'00', u'followers_count': 7, u'protected': False, u'location':
u'', u'profile_background_color': u'9ae4e8', u'utc_offset': -32400,
u'statuses_count': 0, u'description': u':wants to GAIN weight!wants to
try bungee jumping amp; sky diving (but im acrophobic)',
u'friends_count': 84, u'profile_link_color': u'ff',
u'profile_image_url': u'http://a1.twimg.com/profile_images/
62185674/212297864a2698743034b849032084l_normal.jpg',
u'notifications': False, u'profile_background_image_url': u'http://
s.twimg.com/a/1251493570/images/themes/theme1/bg.gif', u'screen_name':
u'JOYCE161008', u'profile_background_tile': False,
u'favourites_count': 0, u'name': u'JOYCE161008', u'url': u'http://MY-
OWN-PAGE.TK', u'created_at': u'Thu Oct 16 07:40:38 + 2008',
u'time_zone': u'Alaska', u'profile_sidebar_border_color': u'87bc44',
u'following': False
[twitter-dev] Re: /friendships/destroy.json /blocks/blocking /blocking/ids
Hmm, no, a good idea though. i'm using a python wrapper, it's pretty
exposed, i've cut out any cacheing in the course of debugging this
issue.
Well, this looks like a bug with me, otherwise this list would be a
whole lot more alive today.
On 30 Aug 2009, at 16:22, J. Dale wrote:
How are you calling twitter? Directly or via a wrapper like
EpiTwitter. I had problems with EpiTwitter effectively caching
requests. I had to go after particular elements in the result for it
to actually make/return the request.
On Aug 30, 4:47 am, Ben Eliott ben.apperr...@googlemail.com wrote:
Ok, further to email below, i'm not able to post follow requests
either, it comes back with success, but no database update seems to
have occurred.
Pretty much looks like any POST request just isn't happening, even
though twitter is coming back with friendly responses.
Ben
On 29 Aug 2009, at 17:35, ben wrote:
Couple of issues sprouting, i have been editing the code so i'm
going
back and back over for some error at my end, but it's looking mighty
odd.
It looks like:
/friendships/destroy.json is not being updated
blocks/blocking.json is not returning the same users as /blocks/
blocking/ids.json
Logging below. My ip for is 67.23.28.168 and times Europe/London.
logging for friendships/destory:
2009-08-29 17:10:56,702 - twitterauth.utils - INFO - twitter api
exe.
-https://twitter.com/friendships/destroy.json.
kwargs={'screen_name':
u'arthurcooke'} method=POST is_auth=True
2009-08-29 17:10:57,166 - twitterauth.utils - DEBUG - api auth
response, {u'id': 16612822, u'verified': False,
u'profile_sidebar_fill_color': u'252429', u'profile_text_color':
u'66', u'followers_count': 14, u'protected': True, u'location':
u'Leeds, UK', u'profile_background_color': u'1A1B1F', u'status':
{u'favorited': False, u'truncated': False, u'text': u'Relaxing in
Aspen.', u'created_at': u'Mon Aug 10 22:45:51 + 2009',
u'source':
u'web', u'in_reply_to_status_id': None, u'in_reply_to_screen_name':
None, u'in_reply_to_user_id': None, u'id': 3233921143},
u'utc_offset':
0, u'statuses_count': 46, u'description': u'Runner, Mountain Biker,
Science Teacher, Film Fan', u'friends_count': 18,
u'profile_link_color': u'2FC2EF', u'profile_image_url': u'http://
a3.twimg.com/profile_images/81150781/IMG_0320_normal.JPG',
u'notifications': False, u'profile_background_image_url': u'http://
s.twimg.com/a/1251493570/images/themes/theme9/bg.gif',
u'screen_name':
u'arthurcooke', u'profile_background_tile': False,
u'favourites_count': 0, u'name': u'arthur cooke', u'url': None,
u'created_at': u'Mon Oct 06 12:07:50 + 2008', u'time_zone':
u'London', u'profile_sidebar_border_color': u'181A1E', u'following':
True}
then..not long later
2009-08-29 17:10:59,746 - twitterauth.utils - INFO - twitter api
exe.
-https://twitter.com/statuses/friends.json. kwargs={'screen_name':
u'b_e_n_', 'page': 1} method=GET is_auth=False
the response is parsed and friends username listed in log..note
including the one who was just destroyed.
2009-08-29 17:11:00,206 - twittersearch.models - DEBUG - get current
friends for b_e_n_ - [u'TomChivers', u'Love_London', u'Fast140',
u'performativeweb', u'bbcbreaking', u'rhonafwelsh', u'danifromCO',
u'elliottzone', u'maustyn', u'unstuckdesign', u'arthurcooke',
u'jools2']
logging for blocks/blocking, last few logs show 3 blocked users from
blocking.json and 2 from ids.json:
009-08-29 17:11:39,268 - twitterauth.utils - INFO - twitter api
exe. -
https://twitter.com/blocks/blocking.json. kwargs={'page': 1}
method=GET is_auth=True
2009-08-29 17:11:39,608 - twitterauth.utils - DEBUG - api auth
response, [{u'id': 16803782, u'verified': False,
u'profile_sidebar_fill_color': u'e0ff92', u'profile_text_color':
u'00', u'followers_count': 8, u'protected': False, u'location':
u'United States', u'profile_background_color': u'9ae4e8',
u'utc_offset': -32400, u'statuses_count': 0, u'description':
u':spends
a lot of hours surfing the net', u'friends_count': 111,
u'profile_link_color': u'ff', u'profile_image_url': u'http://
a3.twimg.com/profile_images/
62184497/212297864a2022621782b584997122l_normal.jpg',
u'notifications': False, u'profile_background_image_url': u'http://
s.twimg.com/a/1251493570/images/themes/theme1/bg.gif',
u'screen_name':
u'SARAH161008', u'profile_background_tile': False,
u'favourites_count': 0, u'name': u'SARAH161008', u'url': u'http://
MY-
OWN-PAGE.TK', u'created_at': u'Thu Oct 16 07:03:12 + 2008',
u'time_zone': u'Alaska', u'profile_sidebar_border_color': u'87bc44',
u'following': False}, {u'id': 16804065, u'verified': False,
u'profile_sidebar_fill_color': u'e0ff92', u'profile_text_color':
u'00', u'followers_count': 7, u'protected': False, u'location':
u'', u'profile_background_color': u'9ae4e8', u'utc_offset': -32400,
u'statuses_count': 0, u'description': u':wants to GAIN weight!
wants to
try bungee jumping amp; sky diving (but im acrophobic)',
u'friends_count': 84
[twitter-dev] blocks/blocking - page paramater dodgy?
Hi,
Could be me, but I'm gettings matching returns from blocks/blocking
whatever the page parameter is, logging below.
Also i'm finding the blocking/ids not returning the same as the blocks/
blocking results. Are the /ids/ results cached? If so, is this true of
all the friends/followers/ id methods?
2009-08-29 15:06:05,851 - twitterauth.utils - INFO - twitter api exe.
- https://twitter.com/blocks/blocking.json. kwargs={'page': 1}
method=GET is_auth=True
2009-08-29 15:06:06,823 - twitterauth.utils - DEBUG - api auth
response, [{'id': 16803782, 'verified': False,
'profile_sidebar_fill_color': 'e0ff92', 'profile_text_color':
'00', 'followers_count': 8, 'protected': False, 'location':
'United States', 'profile_background_color': '9ae4e8', 'utc_offset':
-32400, 'statuses_count': 0, 'description': ':spends a lot of hours
surfing the net', 'friends_count': 111, 'profile_link_color':
'ff', 'profile_image_url': 'http://a3.twimg.com/profile_images/
62184497/212297864a2022621782b584997122l_normal.jpg', 'notifications':
False, 'profile_background_image_url': 'http://s.twimg.com/a/
1251493570/images/themes/theme1/bg.gif', 'screen_name': 'SARAH161008',
'profile_background_tile': False, 'favourites_count': 0, 'name':
'SARAH161008', 'url': 'http://MY-OWN-PAGE.TK', 'created_at': 'Thu Oct
16 07:03:12 + 2008', 'time_zone': 'Alaska',
'profile_sidebar_border_color': '87bc44', 'following': False}, {'id':
16804065, 'verified': False, 'profile_sidebar_fill_color': 'e0ff92',
'profile_text_color': '00', 'followers_count': 7, 'protected':
False, 'location': '', 'profile_background_color': '9ae4e8',
'utc_offset': -32400, 'statuses_count': 0, 'description': ':wants to
GAIN weight!wants to try bungee jumping amp; sky diving (but im
acrophobic)', 'friends_count': 84, 'profile_link_color': 'ff',
'profile_image_url': 'http://a1.twimg.com/profile_images/
62185674/212297864a2698743034b849032084l_normal.jpg', 'notifications':
False, 'profile_background_image_url': 'http://s.twimg.com/a/
1251493570/images/themes/theme1/bg.gif', 'screen_name': 'JOYCE161008',
'profile_background_tile': False, 'favourites_count': 0, 'name':
'JOYCE161008', 'url': 'http://MY-OWN-PAGE.TK', 'created_at': 'Thu Oct
16 07:40:38 + 2008', 'time_zone': 'Alaska',
'profile_sidebar_border_color': '87bc44', 'following': False}, {'id':
69771552, 'verified': False, 'profile_sidebar_fill_color': 'e0ff92',
'profile_text_color': '00', 'followers_count': 3, 'protected':
False, 'location': None, 'profile_background_color': '9ae4e8',
'status': {'favorited': False, 'truncated': False, 'text': 'except
whole foods mac n cheese is just not as good as kraft.
http://a4a2b.easyurl.net',
'created_at': 'Sat Aug 29 10:49:19 + 2009', 'source': 'web',
'in_reply_to_status_id': None, 'in_reply_to_screen_name': None,
'in_reply_to_user_id': None, 'id': 3623286748L}, 'utc_offset': None,
'statuses_count': 1, 'description': None, 'friends_count': 690,
'profile_link_color': 'ff', 'profile_image_url': 'http://
a1.twimg.com/profile_images/387598494/12215739_lynde_20_normal.gif',
'notifications': False, 'profile_background_image_url': 'http://
s.twimg.com/a/1251493570/images/themes/theme1/bg.gif', 'screen_name':
'Blackburn432', 'profile_background_tile': False, 'favourites_count':
0, 'name': 'TrinidadBlackburn', 'url': None, 'created_at': 'Sat Aug 29
03:05:08 + 2009', 'time_zone': None,
'profile_sidebar_border_color': '87bc44', 'following': False}]
2009-08-29 15:06:06,824 - twitterauth.utils - INFO - twitter api exe.
- https://twitter.com/blocks/blocking.json. kwargs={'page': 2}
method=GET is_auth=True
2009-08-29 15:06:07,710 - twitterauth.utils - DEBUG - api auth
response, [{'id': 16803782, 'verified': False,
'profile_sidebar_fill_color': 'e0ff92', 'profile_text_color':
'00', 'followers_count': 8, 'protected': False, 'location':
'United States', 'profile_background_color': '9ae4e8', 'utc_offset':
-32400, 'statuses_count': 0, 'description': ':spends a lot of hours
surfing the net', 'friends_count': 111, 'profile_link_color':
'ff', 'profile_image_url': 'http://a3.twimg.com/profile_images/
62184497/212297864a2022621782b584997122l_normal.jpg', 'notifications':
False, 'profile_background_image_url': 'http://s.twimg.com/a/
1251493570/images/themes/theme1/bg.gif', 'screen_name': 'SARAH161008',
'profile_background_tile': False, 'favourites_count': 0, 'name':
'SARAH161008', 'url': 'http://MY-OWN-PAGE.TK', 'created_at': 'Thu Oct
16 07:03:12 + 2008', 'time_zone': 'Alaska',
'profile_sidebar_border_color': '87bc44', 'following': False}, {'id':
16804065, 'verified': False, 'profile_sidebar_fill_color': 'e0ff92',
'profile_text_color': '00', 'followers_count': 7, 'protected':
False, 'location': '', 'profile_background_color': '9ae4e8',
'utc_offset': -32400, 'statuses_count': 0, 'description': ':wants to
GAIN weight!wants to try bungee jumping amp; sky diving (but im
acrophobic)', 'friends_count': 84, 'profile_link_color': 'ff',
'profile_image_url':
[twitter-dev] /friendships/destroy.json /blocks/blocking /blocking/ids
Couple of issues sprouting, i have been editing the code so i'm going
back and back over for some error at my end, but it's looking mighty
odd.
It looks like:
/friendships/destroy.json is not being updated
blocks/blocking.json is not returning the same users as /blocks/
blocking/ids.json
Logging below. My ip for is 67.23.28.168 and times Europe/London.
logging for friendships/destory:
2009-08-29 17:10:56,702 - twitterauth.utils - INFO - twitter api exe.
- https://twitter.com/friendships/destroy.json. kwargs={'screen_name':
u'arthurcooke'} method=POST is_auth=True
2009-08-29 17:10:57,166 - twitterauth.utils - DEBUG - api auth
response, {u'id': 16612822, u'verified': False,
u'profile_sidebar_fill_color': u'252429', u'profile_text_color':
u'66', u'followers_count': 14, u'protected': True, u'location':
u'Leeds, UK', u'profile_background_color': u'1A1B1F', u'status':
{u'favorited': False, u'truncated': False, u'text': u'Relaxing in
Aspen.', u'created_at': u'Mon Aug 10 22:45:51 + 2009', u'source':
u'web', u'in_reply_to_status_id': None, u'in_reply_to_screen_name':
None, u'in_reply_to_user_id': None, u'id': 3233921143}, u'utc_offset':
0, u'statuses_count': 46, u'description': u'Runner, Mountain Biker,
Science Teacher, Film Fan', u'friends_count': 18,
u'profile_link_color': u'2FC2EF', u'profile_image_url': u'http://
a3.twimg.com/profile_images/81150781/IMG_0320_normal.JPG',
u'notifications': False, u'profile_background_image_url': u'http://
s.twimg.com/a/1251493570/images/themes/theme9/bg.gif', u'screen_name':
u'arthurcooke', u'profile_background_tile': False,
u'favourites_count': 0, u'name': u'arthur cooke', u'url': None,
u'created_at': u'Mon Oct 06 12:07:50 + 2008', u'time_zone':
u'London', u'profile_sidebar_border_color': u'181A1E', u'following':
True}
then..not long later
2009-08-29 17:10:59,746 - twitterauth.utils - INFO - twitter api exe.
- https://twitter.com/statuses/friends.json. kwargs={'screen_name':
u'b_e_n_', 'page': 1} method=GET is_auth=False
the response is parsed and friends username listed in log..note
including the one who was just destroyed.
2009-08-29 17:11:00,206 - twittersearch.models - DEBUG - get current
friends for b_e_n_ - [u'TomChivers', u'Love_London', u'Fast140',
u'performativeweb', u'bbcbreaking', u'rhonafwelsh', u'danifromCO',
u'elliottzone', u'maustyn', u'unstuckdesign', u'arthurcooke',
u'jools2']
logging for blocks/blocking, last few logs show 3 blocked users from
blocking.json and 2 from ids.json:
009-08-29 17:11:39,268 - twitterauth.utils - INFO - twitter api exe. -
https://twitter.com/blocks/blocking.json. kwargs={'page': 1}
method=GET is_auth=True
2009-08-29 17:11:39,608 - twitterauth.utils - DEBUG - api auth
response, [{u'id': 16803782, u'verified': False,
u'profile_sidebar_fill_color': u'e0ff92', u'profile_text_color':
u'00', u'followers_count': 8, u'protected': False, u'location':
u'United States', u'profile_background_color': u'9ae4e8',
u'utc_offset': -32400, u'statuses_count': 0, u'description': u':spends
a lot of hours surfing the net', u'friends_count': 111,
u'profile_link_color': u'ff', u'profile_image_url': u'http://
a3.twimg.com/profile_images/
62184497/212297864a2022621782b584997122l_normal.jpg',
u'notifications': False, u'profile_background_image_url': u'http://
s.twimg.com/a/1251493570/images/themes/theme1/bg.gif', u'screen_name':
u'SARAH161008', u'profile_background_tile': False,
u'favourites_count': 0, u'name': u'SARAH161008', u'url': u'http://MY-
OWN-PAGE.TK', u'created_at': u'Thu Oct 16 07:03:12 + 2008',
u'time_zone': u'Alaska', u'profile_sidebar_border_color': u'87bc44',
u'following': False}, {u'id': 16804065, u'verified': False,
u'profile_sidebar_fill_color': u'e0ff92', u'profile_text_color':
u'00', u'followers_count': 7, u'protected': False, u'location':
u'', u'profile_background_color': u'9ae4e8', u'utc_offset': -32400,
u'statuses_count': 0, u'description': u':wants to GAIN weight!wants to
try bungee jumping amp; sky diving (but im acrophobic)',
u'friends_count': 84, u'profile_link_color': u'ff',
u'profile_image_url': u'http://a1.twimg.com/profile_images/
62185674/212297864a2698743034b849032084l_normal.jpg',
u'notifications': False, u'profile_background_image_url': u'http://
s.twimg.com/a/1251493570/images/themes/theme1/bg.gif', u'screen_name':
u'JOYCE161008', u'profile_background_tile': False,
u'favourites_count': 0, u'name': u'JOYCE161008', u'url': u'http://MY-
OWN-PAGE.TK', u'created_at': u'Thu Oct 16 07:40:38 + 2008',
u'time_zone': u'Alaska', u'profile_sidebar_border_color': u'87bc44',
u'following': False}, {u'id': 69771552, u'verified': False,
u'profile_sidebar_fill_color': u'e0ff92', u'profile_text_color':
u'00', u'followers_count': 3, u'protected': False, u'location':
None, u'profile_background_color': u'9ae4e8', u'status':
{u'favorited': False, u'truncated': False, u'text': u'except whole
foods mac n cheese is just not as good as kraft. http://a4a2b.easyurl.net',
u'created_at': u'Sat Aug 29
[twitter-dev] non json response
Occassionally i get back a 200 status html response from the json search api which look like this, most times the same search works fine, it just happens occassionally: !DOCTYPE html PUBLIC -//W3C//DTD HTML 4.01//EN http://www.w3.org/ TR/1999/REC-html401-19991224/strict.dtd !-- !DOCTYPE HTML PUBLIC -//W3C//DTD HTML 4.01//EN http://www.w3.org/TR/html4/strict.dtd; -- HTML HEAD META HTTP-EQUIV=Refresh CONTENT=0.1 META HTTP-EQUIV=Pragma CONTENT=no-cache META HTTP-EQUIV=Expires CONTENT=-1 TITLE/TITLE /HEAD BODYP/BODY /HTML Does anyone recognise what this kind of response means? Is it normal, or just beta-ish quirks?
[twitter-dev] Mobile Twitter has been blocked?
Hi, I was just wondering where I should post this, but it appears that Mobile Tweete (http://m.tweete.net) has been blocked. I worked up until about an hour ago, but now it appears as though no API requests are being processed. I can ping the twitter servers from the Tweete machine, but nothing else works. I have a staging server running the same trunk as the live box and that works fine (different machine/IP) so I can only assume I have been blocked. Any ideas on what I should do to avoid this. I realise that twitter is cracking down on 'threats' after the downtime about a month ago, but Mobile Tweete is a legitimate client that has been running for over a year now. Any help would be greatly appreciated. Kind Regards, Ben Novakovic -- Developer of Mobile Tweete
[twitter-dev] dm / follow handling
Hi, I'd appreciate any opinions regarding a dilemma about handling a DM where you don't know if the correct relationship exists. Since a DM isn't rate limited it effectively tells me for 'free' whether the target user is following the source user by raising an exception if the correct relationship doesn't exist. So if you're sending a DM anyhow it would seem better to just try the DM first, then handle the exception if it comes in, rather than make a rate limited call to check the relationship details first. While this seems like a sensible option for the app logic, i'm not so sure if it's sensible with regard to the app's relationship with twitter. It could be percevied to be less responsible, and perhaps after a while DM calls to non-followed users will red-flag the account with twitter? Or maybe, conversely, its been cunningly designed for just this scenario. Any opinions/experience with this? Thanks in advance, Ben
[twitter-dev] Re: Developer Preview: Geolocation API
Hi Ryan, Thank-you for the fast response. That makes sense, thanks a lot for clarifying. Wow, this is a really exciting feature. Best Regards, Ben On 21 Aug 2009, at 17:44, Ryan Sarver wrote: Ben, Currently we geocode your user.location data to get an idea of where you are. That gets attached to each tweet as it comes in, but its not usually a representation of where you were when you actually sent the tweet. The new functionality will allow you to geotag the actual update without modifying the user.location field. When it comes to search, we'll use both and give priority to the tweet-level geotag. Make sense? Best, Ryan On Fri, Aug 21, 2009 at 4:06 AM, Ben Eliottben.apperr...@googlemail.com wrote: Hi, Please could you advise on the differences between this and the current location based searching facility? Is the current location search based on the users location in their settings whilst this is a exact location for each tweet? Thanks, Ben On 20 Aug 2009, at 21:46, Ryan Sarver wrote: We wanted to give you all a heads up on a cool new feature that is coming soon - Geolocation. The Geolocation API will give us the ability to attach geographic metadata to tweets to provide additional context with your update. Along with the option to tag updates, we will be able to search for nearby tweets and view the geo metadata in user timelines. The additional context allows for us to deliver more meaningful and localized experiences to users. We are also really excited about a unique facet of this release in that it will be API-only initially. This means that Twitter.com won't surface the functionality and we look forward to seeing the new and interesting experiences that will grow out of the ecosystem. As part of our Geolocation efforts we will soon be publishing Geolocation Best Pracitices to guide everyone through issues like security and privacy as well as discussing some ideal experiences for users. Topics will include things like storage of location data, what to do with a user's historical data, how to present the concept of geotagging and more. The guide will create a framework from which we can address the challenges that come about when dealing with something as sensitive as someone's location while hopefully allowing everyone enough creative freedom to create their own experiences around it. It is important to note that the feature is going to be strictly opt- in. It will be disabled until a user chooses to switch it on. We will provide a read-only attribute geo_enabled on the user object so an app can detect if the user has it disabled and let them know if they need to turn it on before using a geolocation feature. While we can't provide an exact date for launch, you should plan on having a few weeks of development time before the new API is officially launched. With that being said, lets get to it... Example: Geotagging a Tweet --- curl -d lat=37.780467long=-122.396762status=I have arrived -u user:pass http://twitter.com/statuses/update.xml; ?xml version=1.0 encoding=UTF-8? status created_atTue Apr 07 22:52:51 + 2009/created_at ... geo xmlns:georss=http://www.georss.org/georss; georss:point37.780467 -122.396762/georss:point /geo user id1401881/id nameDoug Williams/name ... geo_enabledtrue/geo_enabled ... /user /status We have also updated the wiki to reflect what the API will look like when it launches, so check it out and let us know if you have any questions: http://apiwiki.twitter.com/Twitter-REST-API-Method%3A-statuses%C2%A0update http://apiwiki.twitter.com/Twitter-REST-API-Method%3A-account%C2%A0verify_credentials We'll also be in our recently announced IRC channel (#twitterapi on irc.freenode.net) if you want to discuss the announcement with the team. Ryan PM, Platform Team http://twitter.com/rsarver
[twitter-dev] Re: Developer Preview: Geolocation API
Hi, Please could you advise on the differences between this and the current location based searching facility? Is the current location search based on the users location in their settings whilst this is a exact location for each tweet? Thanks, Ben On 20 Aug 2009, at 21:46, Ryan Sarver wrote: We wanted to give you all a heads up on a cool new feature that is coming soon - Geolocation. The Geolocation API will give us the ability to attach geographic metadata to tweets to provide additional context with your update. Along with the option to tag updates, we will be able to search for nearby tweets and view the geo metadata in user timelines. The additional context allows for us to deliver more meaningful and localized experiences to users. We are also really excited about a unique facet of this release in that it will be API-only initially. This means that Twitter.com won't surface the functionality and we look forward to seeing the new and interesting experiences that will grow out of the ecosystem. As part of our Geolocation efforts we will soon be publishing Geolocation Best Pracitices to guide everyone through issues like security and privacy as well as discussing some ideal experiences for users. Topics will include things like storage of location data, what to do with a user's historical data, how to present the concept of geotagging and more. The guide will create a framework from which we can address the challenges that come about when dealing with something as sensitive as someone's location while hopefully allowing everyone enough creative freedom to create their own experiences around it. It is important to note that the feature is going to be strictly opt- in. It will be disabled until a user chooses to switch it on. We will provide a read-only attribute geo_enabled on the user object so an app can detect if the user has it disabled and let them know if they need to turn it on before using a geolocation feature. While we can't provide an exact date for launch, you should plan on having a few weeks of development time before the new API is officially launched. With that being said, lets get to it... Example: Geotagging a Tweet --- curl -d lat=37.780467long=-122.396762status=I have arrived -u user:pass http://twitter.com/statuses/update.xml; ?xml version=1.0 encoding=UTF-8? status created_atTue Apr 07 22:52:51 + 2009/created_at ... geo xmlns:georss=http://www.georss.org/georss; georss:point37.780467 -122.396762/georss:point /geo user id1401881/id nameDoug Williams/name ... geo_enabledtrue/geo_enabled ... /user /status We have also updated the wiki to reflect what the API will look like when it launches, so check it out and let us know if you have any questions: http://apiwiki.twitter.com/Twitter-REST-API-Method%3A-statuses%C2%A0update http://apiwiki.twitter.com/Twitter-REST-API-Method%3A-account%C2%A0verify_credentials We'll also be in our recently announced IRC channel (#twitterapi on irc.freenode.net) if you want to discuss the announcement with the team. Ryan PM, Platform Team http://twitter.com/rsarver
[twitter-dev] Re: Do My Customers Have a Twitter Account?
I think the search my username feature has been removed. It now is just Name... You could use the gmail\yahoo\aol integration - http://twitter.com/invitations?service=gmail @Ben_Hall On Wed, Aug 19, 2009 at 9:17 PM, Scott Hanedatalkli...@newgeo.com wrote: Good point. Why not just send them an email, and offer to let them follow you? This puts it as a opt in on their part. You can then follow them back if you desire, which I assume you do. Probably not a good idea to do anything of this nature when you are talking about 400K. On Aug 19, 2009, at 1:14 PM, David Fisher wrote: Sounds like something you should be able to do in an email to them or with a message on your website. On Aug 19, 1:29 pm, arawajy araw...@gmail.com wrote: I want to invite them to follow the company on Twitter. On Aug 19, 8:25 pm, Andrew Badera and...@badera.us wrote: On Wed, Aug 19, 2009 at 11:07 AM, arawajyaraw...@gmail.com wrote: Dear Developers, I have a list of 400,000 e-mail addresses of my clients. I want to know Is it possible to develop a script to check if they have a twitter account or not?. I will then want to generate 2 separate lists based upon the result; one for the twitter users and one for the non-twitter users. I want to only invite the users and create a custom invitation message. Is it possible to check if the e-mail address's owner is a twitter user or not? provide details please. Thanks and Regards, Mahmoud If they're already your clients then what are you inviting them to? -- Scott * If you contact me off list replace talklists@ with scott@ *
[twitter-dev] available languages
I can't spot information on languages in the wiki. Please can someone advise what values are in the language parameter in tweets from the search api. All possible iso language codes? two and four codes, e.g. en-us and en, or just the two-letter codes? Whatever the user wants? Thanks a lot, Ben
[twitter-dev] Re: available languages
Hi Michael, Thank you very much for this. Going back to look again at that page again the ISO link is right there, as you say. I don't know why my eye skipped over it like that. I guess too much/not enough caffeine. Thanks again, sorry for the time waster. Ben On 17 Aug 2009, at 15:24, Michael Paladino wrote: According to the documentation of the lang parameter under Parameters at http://apiwiki.twitter.com/Twitter-Search-API-Method%3A-search , the language codes should be provided using the ISO 639-1 code (actual list at http://www.loc.gov/standards/iso639-2/php/code_list.php) . Those codes are two letter such as “en” for English and “es” for Spanish. Hope this helps. Michael Paladino http://tidytweet.com -Original Message- From: twitter-development-talk@googlegroups.com [mailto:twitter-development-talk@googlegroups.com ] On Behalf Of ben Sent: Monday, August 17, 2009 3:35 AM To: Twitter Development Talk Subject: [twitter-dev] available languages I can't spot information on languages in the wiki. Please can someone advise what values are in the language parameter in tweets from the search api. All possible iso language codes? two and four codes, e.g. en-us and en, or just the two-letter codes? Whatever the user wants? Thanks a lot, Ben
[twitter-dev] friend/follow ratio
Hi, Is there a guesstimate out there for twitter's spam-guarding friend/ follow ratio. It would be nice to warn users, or red-flag users, before they actually get hit by twitter account blocking measures. Similarly, is there any idea about roughty what the 'follower churn' rate is before it gets flagged by twitter. It would be pretty pointless warning users about a dangerous friend/follow ratio, which results in them unfollowing a lot of users, which then gets them red- flagged for follower churn. I appreciate that these are probably complex algorithms which twitter doesn't want to publish anyhow. But a rough guesstimate would useful, just to trigger some warnings before users get up in arms about accounts being blocked. Thanks for any advice, Ben
[twitter-dev] Re: Early developer preview: Retweeting API
Cool. One request. Could we have an extension to the search API so that I could search for a term which has been tweeted? Scenario: I want to know how many times a particular term has been included in a retweet which I can then aggregate to see how many times it has been retweeted as a total I know it's similar to TweetMeme, but they are links - I want terms :) What do you think? Thanks @Ben_Hall On Thu, Aug 13, 2009 at 10:43 PM, Josh Roessleinjroessl...@gmail.com wrote: This new api looks very cool. Good work twitter API team. :) Josh
[twitter-dev] Re: Twitter + OAuth for iPhone
Sorry, I should have been more careful when prepending my read me content. I've updated the read me file on GitHub, but basically: Use the project in the Demo folder. You'll need to replace the strings in Demo/Src/OAuthTwitterDemoViewController.m with your own consumer key and consumer secret (visit http://twitter.com/oauth_clients/new to obtain these). HTH, B On Jul 30, 3:42 am, Marneo marneo.re...@gmail.com wrote: Ben, It says in the read me file that: Use the key and secret info provided there to modify the constants at the top of YHOAuthTwitterEngine.m You should also set up your callback url at the top of the YHTwitter.m But I cant find these files. Where can I find these files and edit them for my info. Thanks! On Jul 30, 1:50 am, Ben Gottlieb saibengottl...@gmail.com wrote: Okay, sendUpdate is now working with spaces again. On Jul 29, 10:41 am, Ben Gottlieb saibengottl...@gmail.com wrote: Update: it's not working if you have %-escaped characters in your update status string. It appears that there may be some double- escaping going on, and that may be confusing things. Not sure if this is my code or something else (this was working over the weekend, but something else may have changed before I committed to GitHub.). In progress. B On Jul 29, 8:31 am, Ben Gottlieb saibengottl...@gmail.com wrote: I just re-tested the code this morning, and it still works. On Jul 29, 6:03 am, chloros akc1...@gmail.com wrote: Does this currently work? I'm using OAuthConsumer as well and my app stopped working after the last update. On Jul 28, 2:32 pm, Ben Gottlieb saibengottl...@gmail.com wrote: If anyone is interested, I've implemented Twitter OAuth on iPhone (which includes an iPhone version of the OAuth static lib). It's on GitHub:http://github.com/bengottlieb/Twitter-OAuth-iPhone/tree/master
[twitter-dev] Re: Twitter + OAuth for iPhone
I just re-tested the code this morning, and it still works. On Jul 29, 6:03 am, chloros akc1...@gmail.com wrote: Does this currently work? I'm using OAuthConsumer as well and my app stopped working after the last update. On Jul 28, 2:32 pm, Ben Gottlieb saibengottl...@gmail.com wrote: If anyone is interested, I've implemented Twitter OAuth on iPhone (which includes an iPhone version of the OAuth static lib). It's on GitHub:http://github.com/bengottlieb/Twitter-OAuth-iPhone/tree/master
[twitter-dev] Re: Twitter + OAuth for iPhone
Update: it's not working if you have %-escaped characters in your update status string. It appears that there may be some double- escaping going on, and that may be confusing things. Not sure if this is my code or something else (this was working over the weekend, but something else may have changed before I committed to GitHub.). In progress. B On Jul 29, 8:31 am, Ben Gottlieb saibengottl...@gmail.com wrote: I just re-tested the code this morning, and it still works. On Jul 29, 6:03 am, chloros akc1...@gmail.com wrote: Does this currently work? I'm using OAuthConsumer as well and my app stopped working after the last update. On Jul 28, 2:32 pm, Ben Gottlieb saibengottl...@gmail.com wrote: If anyone is interested, I've implemented Twitter OAuth on iPhone (which includes an iPhone version of the OAuth static lib). It's on GitHub:http://github.com/bengottlieb/Twitter-OAuth-iPhone/tree/master
[twitter-dev] Re: Twitter + OAuth for iPhone
Okay, sendUpdate is now working with spaces again. On Jul 29, 10:41 am, Ben Gottlieb saibengottl...@gmail.com wrote: Update: it's not working if you have %-escaped characters in your update status string. It appears that there may be some double- escaping going on, and that may be confusing things. Not sure if this is my code or something else (this was working over the weekend, but something else may have changed before I committed to GitHub.). In progress. B On Jul 29, 8:31 am, Ben Gottlieb saibengottl...@gmail.com wrote: I just re-tested the code this morning, and it still works. On Jul 29, 6:03 am, chloros akc1...@gmail.com wrote: Does this currently work? I'm using OAuthConsumer as well and my app stopped working after the last update. On Jul 28, 2:32 pm, Ben Gottlieb saibengottl...@gmail.com wrote: If anyone is interested, I've implemented Twitter OAuth on iPhone (which includes an iPhone version of the OAuth static lib). It's on GitHub:http://github.com/bengottlieb/Twitter-OAuth-iPhone/tree/master
[twitter-dev] Re: streaming API for DM for multiple users ?
Random idea, but wouldn't a streaming API for DMs allow IM style clients to be implemented on top of the twitter platform? I know I use DMs instead of MSN now, the delay is a bit of a pain but being able to move the conversation from public to private is great, plus sometimes you do want delayed IM conversations (for example, when travelling) ;) Thanks, @Ben_Hall On Tue, Jul 28, 2009 at 2:55 AM, Doug Williamsd...@twitter.com wrote: There is currently no Streaming API to receive DMs for a given user. If you have a great use case for this please share it here. We like to have justification for new streaming methods. If you have ideas to help augment a business case for engineering resources, we would love to know about them. Thanks, Doug On Mon, Jul 27, 2009 at 5:24 AM, Fabien Penso fabienpe...@gmail.com wrote: Hi. I wonder if there is a way to use the streaming API to receive DM for a list of specific users. As far as I understand there isn't, is anyone working on this? Basicly I want to offer the possibility to receive Apple Push Notifications and I'll get tons of user, so I want to go the efficient way. Thanks.
[twitter-dev] Twitter + OAuth for iPhone
If anyone is interested, I've implemented Twitter OAuth on iPhone (which includes an iPhone version of the OAuth static lib). It's on GitHub: http://github.com/bengottlieb/Twitter-OAuth-iPhone/tree/master
[twitter-dev] Re: Using MagpieRSS Authentication with the Twitter API
Your best best, IMHO, is to follow the sample code on this wiki to request the API payload via CURL and then dump the response into the MagpieRSS parser. It's just easier that way, I find - as authentication via CURL is known to work out of the box. This will also make your code more future proof if you decide to make API calls instead of RSS feeds (as presumably MagpieRSS will only handle authentication for feeds that are going to be parsed by it). On Jul 5, 3:57 pm, 13eastie john.holgat...@googlemail.com wrote: I'm a relative beginner using MagpieRSS with PHP to present Twitter RSS feeds on my web-page. I've had no problems with regular feeds, but I'm struggling to get authenticated feeds to work. I'd be very grateful if someone could explain very simply how to implement the HTTP authentication to work with MagpieRSS for me.
[twitter-dev] Re: search API - why not XML output...
Atom is XML On Jul 6, 8:03 am, Carlos carlos.crose...@gmail.com wrote: Hi, bu looking at the search API docs I see the output format is JSON and Atom, why not X-ML? Forgive me I haven´t tried myself to request xml to see what I get, but hopefully the docs are obsoletea and XML is supported best regards, Carlos
[twitter-dev] Re: OAuth and Perl
I'm having the same problem as Jesse using the Net::OAuth
Here's what I get back from twitter:
$VAR1 = bless( {
'_protocol' = 'HTTP/1.1',
'_content' = 'Failed to validate oauth signature or
token',
'_rc' = '401',
'_headers' = bless( {
'connection' = 'close',
'set-cookie' =
'_twitter_sess=BAh7BiIKZ0xhc2hJQzonQWN0aW9uQ29udHJvbGxlcjo6Rmxhc2g6OkZsYXNo
%250ASGFzaHsABjoKQHVzZWR7AA%253D
%253D--1164b91ac812d853b877e93ddb612b7471bebc74; domain=.twitter.com;
path=/',
'cache-control' = 'no-cache,
max-age=300',
'status' = '401
Unauthorized',
'date' = 'Sat, 16 May 2009
01:57:55 GMT',
'vary' = 'Accept-Encoding',
'client-ssl-cert-issuer' = '/
C=US/O=Equifax Secure Inc./CN=Equifax Secure Global eBusiness CA-1',
'client-ssl-cipher' = 'DHE-
RSA-AES256-SHA',
'client-peer' =
'128.121.146.100:443',
'client-warning' = 'Missing
Authenticate header',
'client-date' = 'Sat, 16 May
2009 01:57:55 GMT',
'client-ssl-warning' = 'Peer
certificate not verified',
'content-type' = 'text/html;
charset=utf-8',
'server' = 'hi',
'client-response-num' = 1,
'content-length' = '43',
'client-ssl-cert-subject' = '/
C=US/O=twitter.com/OU=GT09721236/OU=See www.rapidssl.com/resources/cps
(c)08/OU=Domain Control Validated - RapidSSL(R)/CN=twitter.com',
'expires' = 'Sat, 16 May 2009
02:02:55 GMT'
}, 'HTTP::Headers' ),
'_msg' = 'Unauthorized',
'_request' = bless( {
'_content' = '',
'_uri' = bless( do{\(my $o =
'https://twitter.com/statuses/update.json?
oauth_consumer_key=K9ICZr8UwHCVza91AH9Sgoauth_nonce=2AIYDaoQyknJ5Cpqoauth_signature=W
%2BQu6CG7ENoVNghVyNU4DX%2B2LJM%3Doauth_signature_method=HMAC-
SHA1oauth_timestamp=1242439075oauth_token=15385100-
snbvmpiROaexwcJx00gkCegiBwX481bvGsVOmRo8eoauth_version=1.0status=Test
+message')}, 'URI::https' ),
'_headers' = bless( {
'user-
agent' = 'libwww-perl/5.808',
'content-type' = 'application/x-www-form-urlencoded',
'content-length' = 0
},
'HTTP::Headers' ),
'_method' = 'POST'
}, 'HTTP::Request' )
}, 'HTTP::Response' );
On Apr 30, 6:39 pm, Mario Menti mme...@gmail.com wrote:
On Thu, Apr 30, 2009 at 11:22 PM, Jesse Stay jesses...@gmail.com wrote:
I just wanted to bring back attention to this. Has anyone on the list
gotten Twitter's OAuth to work with Perl? Care to share some code examples?
I'm using Perl's Net::OAuth heavily, but only for updating twitter status
with existing access tokens (as my backend processing is Perl, while the
frontend is RoR, so authorisation/key exchange is handled through rails
OAuth).
I did find one bug which I've reported back to the Net::OAuth CPAN
maintainer, who said he'll implement in a future release:
The issue relates
tohttp://code.google.com/p/twitter-api/issues/detail?id=433#c32(there's lots
of useful into in this thread)
The problem occurs when you pass an extra_param containing certain Unicode
characters. What happens is that the parameter is passed to the signature
creation, and the signature ends up wrong, leading to 401 errors when trying
to make a request.
The fix for this is actually detailed in the above thread, a problem with
the regexp doing the escaping. In Perl's case, the below change
to Net::OAuth's Message.pm fixes this:
sub encode {
my $str = shift;
$str = unless defined $str;
# return URI::Escape::uri_escape_utf8($str,'^\w.~-');
# MM, fix based on twitter OAuth bug report
return URI::Escape::uri_escape($str,'^0-9a-zA-Z\d._~-');
}
I'm not sure if this is relevant to you given your previous messages, but
thought I'd share just in case. With this fix implemented, it seems to work
very well, more than 10,000 of my users have migrated to OAuth and I'm doing
hundreds of thousands OAuth-based status update requests, without obvious
problems.
Mario.
[twitter-dev] Problems with creating Twitter accounts for use with an API based application
For confidentially reasons, I can't disclose here the niche that our application operates in but it requires a considerable number of Twitter accounts as it's essentially developing a highly localised service. Each account refering to individual locations in the country. I've already been told in this group that it's not currently possible to use the API to create the 5000 odd accounts we need. So we've been trying to manually register them. But a considerable number of the accounts have been suspended presuamably to prevent spam. We're 100% not a spam operation and once the accounts have been registered we'll offer an incredibly novel and exciting platform that will speed up the uptake of Twitter in our niche market. But we need to be able to register the accounts! Anyone here had similar experiences? Anyone from Twitter able to give us a hand? Outside of a public forum, we're happy to disclose further details. Best wishes Ben
[twitter-dev] Account opening
I'm working on an application that will require the use of a considerable number of Twitter accounts- around 5700. Is there an easy and legitimate way of opening this number of accounts rather than having to manually do so? Are Twitter helping developers do things like this? Is it allowed?
