date:20090402



On 4/1/09 8:34 PM, Alex Payne wrote:

  * Feature (REST API): We now return the same representation of User
objects throughout the API. This representation contains all of the
attributes we make available via the API.


Sweet jeebus!  It's Christmas, all over again!

Thanks, Twitter gnomes.

--
Dossy Shiobara  | do...@panoptic.com | http://dossy.org/
Panoptic Computer Network   | http://panoptic.com/
  He realized the fastest way to change is to laugh at your own
folly -- then you can let go and quickly move on. (p. 70)

[twitter-dev] Re: API Changes for April 1, 2009

2009-04-02 Thread Damon P. Cortesi


On Apr 2, 4:04 am, binit binit.th...@gmail.com wrote:

 Due to this my application buzzom.com has broken down.
 Is there an alternate way of getting these fields for a particular
 user?
 Please suggest.


In some cases, even though /users/show does not have the full user
object, /statuses/user_timeline will. I've been able to use this as a
workaround for some accounts when the fields I need don't exist, but
it doesn't work in every case.

[twitter-dev] Re: Twitter didn't tinyurl my tweets tonight

2009-04-02 Thread Pavlo Zahozhenko

Twitter never tinyurls my links from API, tinyurls them only from web
interface. Sometimes I even use it as a 'feature' - post link from API if I
don't want it to be auto-tinyurled!

On Thu, Apr 2, 2009 at 8:20 AM, bcballard bcball...@gmail.com wrote:


 I use the Twitter API to post article titles followed by a link to the
 article, and sometimes a repeat of the domain name if I think Twitter
 is going to tinyurl me. Generaly Twitter converts URLs with characters
 it doesn't like [^A-Za-z0-9.:/], or URLs in tweets that are too long,
 to tinyurls.

 However, this evening, none of the URLs I posted were converted to
 tinyurls. Not even the ones with characters Twitter doesn't like, and
 not the one that was too long.

 Here are the status updates I expected to be tinyurled, followed by a
 link to the individual tweet as it appeared on Twitter.

 The first 3 contain bad characters (underscores or commas), but are
 fewer than 140 chars:

 GA approves more charter schools

 http://www.ajc.com/metro/content/metro/stories/2009/04/01/georgia_charter_school.html
 (ajc.com)

 http://twitter.com/GeorgiaLogCabin/status/1436345350


 Obama's tax pledge up in smoke
 http://news.yahoo.com/s/ap/20090401/ap_on_go_pr_wh/obama_tax_promise
 (news.yahoo.com)

 http://twitter.com/GeorgiaLogCabin/status/1436192094


 Sweden recognizes marriage equality
 http://www.theaustralian.news.com.au/story/0,25197,25277939-12335,00.html
 (theaustralian.news.com.au)

 http://twitter.com/GeorgiaLogCabin/status/1436147039


 This one has bad characters (dashes) and is longer than 140 chars:

 Obama wants anti-gay advisor

 http://www.usnews.com/blogs/god-and-country/2009/03/31/exclusive-former-nfl-coach-tony-dungy-invited-to-join-white-house-faith-council.html
 (usnews.com)

 http://twitter.com/GeorgiaLogCabin/status/1436319188


 That last one shows on http://twitter.com/GeorgiaLogCabin with a
 truncated URL that causes it to link to a broken story. However, on
 the individual Tweet link above, the URL is not truncated and works
 fine...

 So I have one main question with a few possible answers:

 Q: What happened to Twitter's auto-tinyurl convert?

 Possible answers?

 A1: Tinyurl just happened to be down during the time I was posting
 updates, so Twitter couldn't request tinyurls and used my long ones
 (though other people's tweets before, during and after mine appear to
 be tinyurled as normal.)

 A2: Twitter just this evening stopped tinyurl-ing my tweets, but not
 anybody else's (at least not anyone I happen to follow.)

 A3: Some Twitter bug nobody's aware of (I searched the FAQs, this
 message board, and via Google and didn't find anyone else with this
 particular issue.)

 A4: Some other reason?

 Any ideas?

[twitter-dev] Re: Hi All

2009-04-02 Thread Clint Shryock

Do you know how many actual tweets you are getting?  The Twitter Search API
doesn't explicitly say so, but the way I'm reading it seems to only return
up to 1500 tweets.
_cts



On Thu, Apr 2, 2009 at 1:43 AM, Seema Nagar nagar.se...@gmail.com wrote:

 I am new to twitter.  I am trying twitter search API to get last one
 month's data(
 http://search.twitter.com/search.atom?geocode=51.500152%2C-0.126236%2C15milang=enq=+movie+since%3A2009-03-03+until%3A2009-03-30+near%3Alondon+within%3A15mi).
   Unfortunately I get only fifteen
 pages :(. Has anybody tried this before ? Or is there a way to get
 remaining tweets ?


 Thanks in advance.


 regards,
 Seema

[twitter-dev] Re: Delay in display of profile image updated via API

2009-04-02 Thread Clint Shryock


 Can you please give some pointer to how

to

use local source to represent the new avatar?


My application (a desktop application) allows a user to select a file on the
hard drive and upload it to Twitter.  This application also displays the
users current Twitter avatar.  When the application uploads a new avatar and
receives the 200 status back, instead of querying Twitter for the new
avatar, I simply use the one on disk and display it to the user.
This isn't ideal, I'd rather re-query Twitter and grab the image from them
to be sure, but as we've described that's not always working
correctly/quickly.

Is there any way to supply my

own

URL of the image which is hosted on a third party server?


not to my knowledge, or anywhere I can find on the API.  I would think this
would be a bad idea.

 If web interface also uses the same API calls, how is

this

possible? Or is there a difference in the upload policy based on the

client

software in use?


I don't know that the web interface necessarily uses the API.  It's possible
that the API for Twitter is not the same as Twitter the application itself.
 Either way, the API is what we have to use

_ct



On Thu, Apr 2, 2009 at 3:41 AM, Raghu Prasad prasad.ragh...@gmail.comwrote:



 On Apr 1, 5:49 pm, Clint Shryock cts...@gmail.com wrote:
  The lag wasn't nearly as noticeable when the update_profile_image was
 first
  offered, but has since as you noted become considerable.
  As a work around my app, which updates profile images from a local
 source,
  uses the local source to represent the new avatar after I receive the
 HTTP
  status codes that indicate the upload was successful.

 I am a newbie to Twitter API. Can you please give some pointer to how
 to
 use local source to represent the new avatar? As I understand, the
 method
 update_profile_image takes multipart form data and stores the image
 contained in it on Twitter's storage space. The avatar shown near the
 tweets are
 picked up from Twitter's storage only. Is there any way to supply my
 own
 URL of the image which is hosted on a third party server?

 Another issue which surprised me was that if I upload my profile image
 using
 my web browser, instead of using curl, the avatar is displayed near my
 tweets
 immediately. If web interface also uses the same API calls, how is
 this
 possible? Or is there a difference in the upload policy based on the
 client
 software in use?

 Raghu

[twitter-dev] Re: Remaining Hits

2009-04-02 Thread @crisatunity




On Mar 26, 10:45 pm, colboy col...@gmail.com wrote:
 I'm certainly using credentials.  I'm using Tweetdeck and it says I've
 over my API limit. My application can't do any gets, as expected, but
 when I do account/rate_limit_status it says I have 91 remaining as
 shown here, which is the XML I  receive when I plug the URL in with
 authentication via the website

 hash
 reset-time type=datetime2009-03-27T04:32:35+00:00/reset-time
 reset-time-in-seconds type=integer1238128355/reset-time-in-
 seconds
 remaining-hitstype=integer91/remaining-hits
 hourly-limit type=integer100/hourly-limit
 /hash

 Any ideas? Am I missing something obvious?

 Colin

I am experiencing this identical situation.  All my requests are
credentialed and behave as so, except this feature of the API.  I get
a non-credentialed response every time.

Michael

[twitter-dev] Re: Random 'From' values

2009-04-02 Thread AhmedF


Yep fixed now - properly takes the app's name for the 'from' field
now.

On Apr 1, 6:40 pm, Doug Williams d...@twitter.com wrote:
 Ahmed,
 This is a confirmed bug. The good news is Matt and Alex tag-teamed it and
 have the fix readied for deploy.

 Cheers,
 Doug Williams
 Twitter API Supporthttp://twitter.com/dougw

 On Wed, Apr 1, 2009 at 3:23 PM, AhmedF inde...@gmail.com wrote:

  I guess just for future reference - Matt looked into it, identified it
  as a bug, and said it should be fixed soon.

  Thanks.

  On Apr 1, 5:40 pm, Chad Etzel jazzyc...@gmail.com wrote:
   In my experience.  If you are posting from an OAuth enabled app, the
   From value (should) be the value you put into the OAuth app name
   form when creating the app.  All source parameters passed will be
   ignored.  I would imagine this is a pretty good security measure to
   help track down malapps (yes, you heard it hear first, folks...
   malapps will be the biggest new InfoSec term this century!).

   If you are posting from a Basic Auth app, it will use the source
  parameter.

   That being said, however, this behavior sounds like a bug.
   -Chad

   On Wed, Apr 1, 2009 at 5:29 PM, AhmedF inde...@gmail.com wrote:

I've left it empty and tried 'source' or even 'from' - always comes up
with something random.

On Apr 1, 5:23 pm, Doug Williams d...@twitter.com wrote:
Ahmed,
Are you currently passing in anything with your requests as a source
parameter?

Doug Williams
Twitter API Supporthttp://twitter.com/dougw

On Wed, Apr 1, 2009 at 2:21 PM, AhmedF inde...@gmail.com wrote:

 And a little follow up - so I updated that status 30 minutes ago,
 where it claimed the from was 'TVTweets'. Just tried again, and now
 both of the messages are from 'Testery'

 I guess something is definitely buggy.

 On Apr 1, 5:17 pm, AhmedF inde...@gmail.com wrote:
  As it is - I am using your code as the base for the OAuth
  transaction
  - and all attempts to set source/from have failed.

  I'm waiting on getting approved as a legit 'from' field and then
  seeing what happens.

  On Apr 1, 5:10 pm, Abraham Williams 4bra...@gmail.com wrote:

   OAuth apps automatically get their source info added. The app
  ids must
 be
   getting jumbled somewhere. A work around might be to manually
  set
 source=web
   or whatever source you want.

   Try creating an bug report:
http://code.google.com/p/twitter-api/issues/list

   On Wed, Apr 1, 2009 at 15:46, AhmedF inde...@gmail.com wrote:

I'm testing out a pretty simple OAuth app - the user verifies,
  and I
then push an update to their status.

The confounding part is I am getting absolutely random values
  in the
'From' field. It was originally Coolspotters, now TVtweets,
  and I can
only be curious as to what it chooses next.

The API says I can only set 'status' and optionally
'in_reply_to_status_id' - anyone know why I am getting these
  random
values? (the status message itself is posting just fine).

   --
   Abraham Williams | Hacker |http://abrah.am
   @poseurtech |http://the.hackerconundrum.com
   Web608 | Community Evangelist |http://web608.org
   This email is: [ ] blogable [x] ask first [ ] private.
   Sent from Madison, Wisconsin, United States

[twitter-dev] Determining Sex/Gender with the API?

2009-04-02 Thread kazvor...@gmail.com


As the subject line implies, I need to know how to programmatically
determine the sex of a profile owner with the API. Is this supported,
in any way at all? Not the sex of the person logged in as the app, but
the owners of the profiles in a search, for example.

[twitter-dev] Re: Determining Sex/Gender with the API?

2009-04-02 Thread Matt Sanford


Hi there,

Twitter never asks for gender so without some sophisticated AI  
there's no way to know.


Thanks;
 — Matt Sanford / @mzsanford

On Apr 2, 2009, at 09:28 AM, kazvor...@gmail.com wrote:



As the subject line implies, I need to know how to programmatically
determine the sex of a profile owner with the API. Is this supported,
in any way at all? Not the sex of the person logged in as the app, but
the owners of the profiles in a search, for example.

[twitter-dev] Re: Determining Sex/Gender with the API?

2009-04-02 Thread Andrew Badera

Of course it's supported. Just cast Sexus Magicus Level 12, and poof, that
information will magically be created from nothing, out of nowhere.

If you can't record that information within Twitter, then how could the API
provide it?


On Thu, Apr 2, 2009 at 11:28 AM, kazvor...@gmail.com kazvor...@gmail.comwrote:


 As the subject line implies, I need to know how to programmatically
 determine the sex of a profile owner with the API. Is this supported,
 in any way at all? Not the sex of the person logged in as the app, but
 the owners of the profiles in a search, for example.

[twitter-dev] Re: Remaining Hits

2009-04-02 Thread Matt Sanford



Hi there,

Can you try and record the new X-RateLimit-* headers on your  
request? We added those because there have been many reports of  
incorrect rate limiting that turned out to be incorrect credentials on  
rate_limit_status. This is especially common from web browsers; since  
rate_limit_status does not require credentials they do not send them.  
Let me know if that reflects the problem as well.


Thanks;
  — Matt Sanford / @mzsanford

On Apr 2, 2009, at 09:21 AM, @crisatunity wrote:





On Mar 26, 10:45 pm, colboy col...@gmail.com wrote:
I'm certainly using credentials.  I'm using Tweetdeck and it says  
I've

over my API limit. My application can't do any gets, as expected, but
when I do account/rate_limit_status it says I have 91 remaining as
shown here, which is the XML I  receive when I plug the URL in with
authentication via the website

hash
reset-time type=datetime2009-03-27T04:32:35+00:00/reset-time
reset-time-in-seconds type=integer1238128355/reset-time-in-
seconds
remaining-hitstype=integer91/remaining-hits
hourly-limit type=integer100/hourly-limit
/hash

Any ideas? Am I missing something obvious?

Colin


I am experiencing this identical situation.  All my requests are
credentialed and behave as so, except this feature of the API.  I get
a non-credentialed response every time.

Michael

[twitter-dev] Re: Problem with search api

2009-04-02 Thread Abraham Williams

Search only supports since_id:
http://apiwiki.twitter.com/Search%20API%20Documentation#Search

2009/4/2 nyankov nikola.yan...@gmail.com


 Actualy the url is


 http://search.twitter.com/search.json?q=offer+OR+book+OR+read+since%3A2009-01-01+from%3Anyankov

 and what I found is that if I change the query to:


 http://search.twitter.com/search.json?q=offer+OR+book+OR+read+since%3A2009-04-01+from%3Anyankov

 everything is ok since is I try to search for:


 http://search.twitter.com/search.json?q=offer+OR+book+OR+read+since%3A2009-01-01+from%3Anyankov

 It displays:
 The page you were looking for doesn't exist.
 You may have mistyped the address or the page may have moved.


 The only difference in both is since date

 Any ideas?






 On 2 Апр, 15:44, Cameron Kaiser spec...@floodgap.com wrote:
   Does anybody have idea why this url:
 
  http://search.twitter.com/search.json?q=offerOR book OR read since:
   2008-04-02 from:nyankov
 
   Give me:
 
   The page you were looking for doesn't exist.
   You may have mistyped the address or the page may have moved.
 
  I don't think this is likely the explanation, but nevertheless, your URL
  is not encoded. I would fix that first.
 
  --
   personal:
 http://www.cameronkaiser.com/--
Cameron Kaiser * Floodgap Systems *www.floodgap.com*
 ckai...@floodgap.com
  -- Evil has middle management? -- 8-bit Theatre #663
 --




-- 
Abraham Williams | Hacker | http://abrah.am
@poseurtech | http://the.hackerconundrum.com
Web608 | Community Evangelist | http://web608.org
This email is: [ ] blogable [x] ask first [ ] private.
Sent from Madison, Wisconsin, United States

[twitter-dev] Re: Determining Sex/Gender with the API?


On Thu, Apr 2, 2009 at 12:48 PM, Andrew Badera and...@badera.us wrote:
 Of course it's supported. Just cast Sexus Magicus Level 12, and poof, that
 information will magically be created from nothing, out of nowhere.

Lol.  +1 Insightful

-- 
  Scott Elcomb
  http://www.psema4.com/

[twitter-dev] Re: Problem with search api

2009-04-02 Thread Matt Sanford

Hi there,

There is a since:-MM-DD operator but unfortunately the
operators are documented on a different page [1]. Another historical
thing, I'll try and find some time to move that stuff over to the API
doc [2]. The reason for your error is that the since: date you
provided is older than the data currently available. We plan to
improve the error messaging when we merge the search and main APIs, so
hopefully things will be clearer after that.
We try to keep as much search data on hand as we can but we're
currently limited on disk space. As the rate of tweets increases we
need more and more disk to keep old data on hand. We've ordered more
disks and await them eagerly. Once those are installed the date range
for searches can start growing again. Sorry for the confusing error
messaging.

Thanks;
— Matt Sanford

[1] - http://search.twitter.com/operators
[2] - http://apiwiki.twitter.com/Search-API-Documentation

On Apr 2, 2009, at 10:02 AM, Abraham Williams wrote:

Search only supports since_id:
http://apiwiki.twitter.com/Search%20API%20Documentation#Search

2009/4/2 nyankov nikola.yan...@gmail.com

Actualy the url is

http://search.twitter.com/search.json?q=offer+OR+book+OR+read+since%3A2009-01-01+from%3Anyankov

and what I found is that if I change the query to:

http://search.twitter.com/search.json?q=offer+OR+book+OR+read+since%3A2009-04-01+from%3Anyankov

everything is ok since is I try to search for:

http://search.twitter.com/search.json?q=offer+OR+book+OR+read+since%3A2009-01-01+from%3Anyankov

It displays:
The page you were looking for doesn't exist.
You may have mistyped the address or the page may have moved.

The only difference in both is since date

Any ideas?

On 2 Апр, 15:44, Cameron Kaiser spec...@floodgap.com wrote:
Does anybody have idea why this url:

http://search.twitter.com/search.json?q=offerOR book OR read since:
2008-04-02 from:nyankov

Give me:

The page you were looking for doesn't exist.
You may have mistyped the address or the page may have moved.

I don't think this is likely the explanation, but nevertheless,
your URL

is not encoded. I would fix that first.

--
personal:http://www.cameronkaiser.com/--
Cameron Kaiser * Floodgap Systems *www.floodgap.com*
ckai...@floodgap.com
-- Evil has middle management? -- 8-bit Theatre #663
--

[twitter-dev] Re: Determining Sex/Gender with the API?


http://www.beardorbra.com/

On Thu, Apr 2, 2009 at 10:18, Zac Bowling zbowl...@gmail.com wrote:

 That would be an interesting challenge.

 Now this would only work with active users and for English users but
 you could mine out probability index of gender using other data around
 the user.

 Basically you could search back on someone's tweets for keywords that
 allude you to gender. Like someone saying Us girls have it hard. you
 could assume a high chance of being female. You could also use the
 search api and look of people talking about the subject in the third
 person and you are likely to find the pronouns he or she. For
 example coming across a tweet like I love @zbowling. He is awesome..
 Other ideas are looking for other gender specific words like my
 beard or my bra .

 Then you might have the privacy advocates (big brother conspiracy
 nuts) crying fowl though and gender bombing twitter if you release
 such a service.

 Zac Bowling
 http://twitter.com/zbowling


 On Thu, Apr 2, 2009 at 9:28 AM, kazvor...@gmail.com kazvor...@gmail.com 
 wrote:

 As the subject line implies, I need to know how to programmatically
 determine the sex of a profile owner with the API. Is this supported,
 in any way at all? Not the sex of the person logged in as the app, but
 the owners of the profiles in a search, for example.





-- 
Alex Payne - API Lead, Twitter, Inc.
http://twitter.com/al3x

[twitter-dev] Re: http://twitter.com/statuses/user_timeline.json returns others tweets


That's a known (and terrible!) issue:
http://code.google.com/p/twitter-api/issues/detail?id=394

On Wed, Apr 1, 2009 at 20:22, Gary Zhao garyz...@gmail.com wrote:
 I'm using http://twitter.com/statuses/user_timeline.json to retrieve my
 latest tweets (without any parameters, specify username and password), but
 the weired thing is it returns others tweets.
 Mine: http://twitter.com/garyzhao
 It returns: http://twitter.com/rejon


 --
 Gary
 http://twitter.com/garyzhao




-- 
Alex Payne - API Lead, Twitter, Inc.
http://twitter.com/al3x

[twitter-dev] Re: How to add an open source project into wiki?


Just let us know what you'd like to add. Unfortunately, we haven't
been able to run the wiki in the open due to PBwiki's lack of spam
protection.

On Wed, Apr 1, 2009 at 19:51, Gary Zhao garyz...@gmail.com wrote:
 http://apiwiki.twitter.com/Libraries and
 http://apiwiki.twitter.com/Open-source

 Thanks
 --
 Gary
 http://twitter.com/garyzhao




-- 
Alex Payne - API Lead, Twitter, Inc.
http://twitter.com/al3x

[twitter-dev] Re: Invalid OAuth Request when checking Friendship Exists (with GET)

Chris,
If you include more detail about your application and your request then we
can provide more specific help. That method should work with as a GET
request.

A lot of other developers have seen the HTTP 401 error you described. Try
searching for invalid oauth request [1] to see if anything from the
archive helps you with your problem.

1. http://bit.ly/14mNBE

Doug Williams
Twitter API Support
http://twitter.com/dougw


On Wed, Apr 1, 2009 at 7:20 PM, Chris chris.rick...@gmail.com wrote:


 Hi guy,

 I have just integrated OAuth into my web app, and all is going well
 except for one thing:
 When I call friendships/exists I always receive Invalid OAuth
 Request.

 It seems that if my request to friendships/exists works if it is a
 POST, but if it is a GET it never works.

 Can I rely that the POST will always work?

 Thanks,


 Chris.

[twitter-dev] Re: Determining Sex/Gender with the API?

2009-04-02 Thread Zac Bowling


Bought! :-)

Now I just need to cast Time Magicus Level 20 to find the time to develop it.

Zac Bowling




On Thu, Apr 2, 2009 at 10:20 AM, Alex Payne a...@twitter.com wrote:

 http://www.beardorbra.com/

 On Thu, Apr 2, 2009 at 10:18, Zac Bowling zbowl...@gmail.com wrote:

 That would be an interesting challenge.

 Now this would only work with active users and for English users but
 you could mine out probability index of gender using other data around
 the user.

 Basically you could search back on someone's tweets for keywords that
 allude you to gender. Like someone saying Us girls have it hard. you
 could assume a high chance of being female. You could also use the
 search api and look of people talking about the subject in the third
 person and you are likely to find the pronouns he or she. For
 example coming across a tweet like I love @zbowling. He is awesome..
 Other ideas are looking for other gender specific words like my
 beard or my bra .

 Then you might have the privacy advocates (big brother conspiracy
 nuts) crying fowl though and gender bombing twitter if you release
 such a service.

 Zac Bowling
 http://twitter.com/zbowling


 On Thu, Apr 2, 2009 at 9:28 AM, kazvor...@gmail.com kazvor...@gmail.com 
 wrote:

 As the subject line implies, I need to know how to programmatically
 determine the sex of a profile owner with the API. Is this supported,
 in any way at all? Not the sex of the person logged in as the app, but
 the owners of the profiles in a search, for example.





 --
 Alex Payne - API Lead, Twitter, Inc.
 http://twitter.com/al3x

[twitter-dev] Re: Twitter didn't tinyurl my tweets tonight


TinyURL was running quite slow, so we temporarily disabled our
automatic shortening of long URLs.

We're currently working on a project to make the shortening of URLs
more consistent and predictable. We know it's currently kind of a
guessing game as to what will be shortened, and we want to fix that.

On Wed, Apr 1, 2009 at 22:20, bcballard bcball...@gmail.com wrote:

 I use the Twitter API to post article titles followed by a link to the
 article, and sometimes a repeat of the domain name if I think Twitter
 is going to tinyurl me. Generaly Twitter converts URLs with characters
 it doesn't like [^A-Za-z0-9.:/], or URLs in tweets that are too long,
 to tinyurls.

 However, this evening, none of the URLs I posted were converted to
 tinyurls. Not even the ones with characters Twitter doesn't like, and
 not the one that was too long.

 Here are the status updates I expected to be tinyurled, followed by a
 link to the individual tweet as it appeared on Twitter.

 The first 3 contain bad characters (underscores or commas), but are
 fewer than 140 chars:

 GA approves more charter schools
 http://www.ajc.com/metro/content/metro/stories/2009/04/01/georgia_charter_school.html
 (ajc.com)

 http://twitter.com/GeorgiaLogCabin/status/1436345350


 Obama's tax pledge up in smoke 
 http://news.yahoo.com/s/ap/20090401/ap_on_go_pr_wh/obama_tax_promise
 (news.yahoo.com)

 http://twitter.com/GeorgiaLogCabin/status/1436192094


 Sweden recognizes marriage equality
 http://www.theaustralian.news.com.au/story/0,25197,25277939-12335,00.html
 (theaustralian.news.com.au)

 http://twitter.com/GeorgiaLogCabin/status/1436147039


 This one has bad characters (dashes) and is longer than 140 chars:

 Obama wants anti-gay advisor
 http://www.usnews.com/blogs/god-and-country/2009/03/31/exclusive-former-nfl-coach-tony-dungy-invited-to-join-white-house-faith-council.html
 (usnews.com)

 http://twitter.com/GeorgiaLogCabin/status/1436319188


 That last one shows on http://twitter.com/GeorgiaLogCabin with a
 truncated URL that causes it to link to a broken story. However, on
 the individual Tweet link above, the URL is not truncated and works
 fine...

 So I have one main question with a few possible answers:

 Q: What happened to Twitter's auto-tinyurl convert?

 Possible answers?

 A1: Tinyurl just happened to be down during the time I was posting
 updates, so Twitter couldn't request tinyurls and used my long ones
 (though other people's tweets before, during and after mine appear to
 be tinyurled as normal.)

 A2: Twitter just this evening stopped tinyurl-ing my tweets, but not
 anybody else's (at least not anyone I happen to follow.)

 A3: Some Twitter bug nobody's aware of (I searched the FAQs, this
 message board, and via Google and didn't find anyone else with this
 particular issue.)

 A4: Some other reason?

 Any ideas?









-- 
Alex Payne - API Lead, Twitter, Inc.
http://twitter.com/al3x

[twitter-dev] SSL for Search API?

2009-04-02 Thread ncooper


Does the Search API support SSL? https://search.twitter.com/search.json
... is a no-go.

[twitter-dev] Re: SSL for Search API?

2009-04-02 Thread Cameron Kaiser


 Does the Search API support SSL?

No.

-- 
 personal: http://www.cameronkaiser.com/ --
  Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckai...@floodgap.com
-- I think you underestimate the sneakiness.

[twitter-dev] Re: API Changes for April 1, 2009

2009-04-02 Thread Jesse Stay

On Thu, Apr 2, 2009 at 12:10 AM, Damon P. Cortesi d.lifehac...@gmail.comwrote:

 On Apr 1, 10:15 pm, Chad Etzel jazzyc...@gmail.com wrote:
  On Thu, Apr 2, 2009 at 1:10 AM, Adrian spiritpo...@gmail.com wrote:
 
   As of right right now:http://twitter.com/users/show/bob.xml
   has about twice the amount of information as say:
  http://twitter.com/users/show/WeezerOfficial.xml
 
  FTA:
  Please note that this new extended view of User objects may not appear
  for all users immediately. As cache expiry occurs for users in our
  system, the extra attributes will show up. Don't be surprised if this
  takes multiple days for inactive users.
 
  -Chad

 Right, but the odd thing is this is data that was showing up normally
 before. When this change went into effect, I started getting a lot of
 users with no statuses_count showing whereas prior to the change that
 was a reliable property of the user object.


I'm seeing similar from our users on SocialToo.

Jesse

[twitter-dev] Re: API Changes for April 1, 2009

The loss of fields (issue 409) is the result of the caching issue described
above. You should see these fields reappear as the cache user objects
expire.

Doug Williams
Twitter API Support
http://twitter.com/dougw


On Thu, Apr 2, 2009 at 10:40 AM, binit binit.th...@gmail.com wrote:


 Thanks a lot, Damon.
 It works!
 You saved my day :)

 On Apr 2, 8:35 pm, Damon P. Cortesi d.lifehac...@gmail.com wrote:
  On Apr 2, 4:04 am, binit binit.th...@gmail.com wrote:
 
 
 
   Due to this my application buzzom.com has broken down.
   Is there an alternate way of getting these fields for a particular
   user?
   Please suggest.
 
  In some cases, even though /users/show does not have the full user
  object, /statuses/user_timeline will. I've been able to use this as a
  workaround for some accounts when the fields I need don't exist, but
  it doesn't work in every case.

[twitter-dev] Re: Determining Sex/Gender with the API?

2009-04-02 Thread Dimebrain


I actually ran this experiment already for a dating app concept, using
some established research on gender detection based on writing
(against at least 10-20 tweets) combined with a database of female
names for user names. I also tried running this on an ANN but that
wasn't fruitful and required me to build a custom application to plunk
through piles of public timeline users verifying genderness.

But yes, it's possible and yes, it works pretty well, especially if
the user filled in their profile and has a decent number of updates.

On Apr 2, 2:18 pm, Zac Bowling zbowl...@gmail.com wrote:
 That would be an interesting challenge.

 Now this would only work with active users and for English users but
 you could mine out probability index of gender using other data around
 the user.

 Basically you could search back on someone's tweets for keywords that
 allude you to gender. Like someone saying Us girls have it hard. you
 could assume a high chance of being female. You could also use the
 search api and look of people talking about the subject in the third
 person and you are likely to find the pronouns he or she. For
 example coming across a tweet like I love @zbowling. He is awesome..
 Other ideas are looking for other gender specific words like my
 beard or my bra .

 Then you might have the privacy advocates (big brother conspiracy
 nuts) crying fowl though and gender bombing twitter if you release
 such a service.

 Zac Bowlinghttp://twitter.com/zbowling

 On Thu, Apr 2, 2009 at 9:28 AM, kazvor...@gmail.com kazvor...@gmail.com 
 wrote:

  As the subject line implies, I need to know how to programmatically
  determine the sex of a profile owner with the API. Is this supported,
  in any way at all? Not the sex of the person logged in as the app, but
  the owners of the profiles in a search, for example.

[twitter-dev] Re: invalid xml char in public timeline


Following up: if you can point us to a status (by ID) that has this
unwanted control character, we'll track down the source of the issue.

On Wed, Apr 1, 2009 at 23:48, AJ cano...@gmail.com wrote:

 I'm using the public timeline feed (for data mining) and frequently
 see xml paring error like this:
 org.jdom.input.JDOMParseException: Error on line 3016: An invalid XML
 character (Unicode: 0x1) was found in the element content of the
 document.

 the error comes from building JDom document in the following code.
        SAXBuilder builder = new SAXBuilder();
        URL u = new URL( url );
        URLConnection conn = u.openConnection();
        if(agent != null){
            conn.setRequestProperty(User-Agent, agent);
        }
        doc = builder.build( conn.getInputStream() );

 is this a know issue with public timeline feed? any good way to fix
 this error?
 -aj




-- 
Alex Payne - API Lead, Twitter, Inc.
http://twitter.com/al3x

[twitter-dev] Re: SSL for Search API?

2009-04-02 Thread Abraham Williams

Everything in search is public and there is no authentication so it is not
really needed.

On Thu, Apr 2, 2009 at 13:06, Cameron Kaiser spec...@floodgap.com wrote:


  Does the Search API support SSL?

 No.

 --
  personal:
 http://www.cameronkaiser.com/ --
  Cameron Kaiser * Floodgap Systems * www.floodgap.com *
 ckai...@floodgap.com
 -- I think you underestimate the sneakiness.
 




-- 
Abraham Williams | Hacker | http://abrah.am
@poseurtech | http://the.hackerconundrum.com
Web608 | Community Evangelist | http://web608.org
This email is: [ ] blogable [x] ask first [ ] private.
Sent from Madison, Wisconsin, United States

[twitter-dev] Re: SSL for Search API?


But what if I don't want a man-in-the-middle to know I'm secretly
searching for Britney Spears from my cube?

...oh crap.


On Thu, Apr 2, 2009 at 2:55 PM, Abraham Williams 4bra...@gmail.com wrote:
 Everything in search is public and there is no authentication so it is not
 really needed.

 On Thu, Apr 2, 2009 at 13:06, Cameron Kaiser spec...@floodgap.com wrote:

  Does the Search API support SSL?

 No.

 --
  personal:
 http://www.cameronkaiser.com/ --
  Cameron Kaiser * Floodgap Systems * www.floodgap.com *
 ckai...@floodgap.com
 -- I think you underestimate the sneakiness.
 



 --
 Abraham Williams | Hacker | http://abrah.am
 @poseurtech | http://the.hackerconundrum.com
 Web608 | Community Evangelist | http://web608.org
 This email is: [ ] blogable [x] ask first [ ] private.
 Sent from Madison, Wisconsin, United States

[twitter-dev] Re: Remaining Hits

2009-04-02 Thread @crisatunity


      Can you try and record the new X-RateLimit-* headers on your  
 request? We added those because there have been many reports of  
 incorrect rate limiting that turned out to be incorrect credentials on  
 rate_limit_status.

I will try this and post back my results.  Thanks.

[twitter-dev] Re: Determining Sex/Gender with the API?

2009-04-02 Thread Clint Shryock

we're back to using magic then are we?;)
_cts

On Thu, Apr 2, 2009 at 12:40 PM, Zac Bowling zbowl...@gmail.com wrote:


 Bought! :-)

 Now I just need to cast Time Magicus Level 20 to find the time to develop
 it.

 Zac Bowling

[twitter-dev] Duplicate Tweets

2009-04-02 Thread Eric Blair


Just got a report from one of my users that a message he posted  
through our app made it through to his Twitter timeline twice. Looking  
at our server logs, I can see that when he posted, we got a timeout  
from Twitter and successfully tried to repost. My guess is that the  
timed-out post actually went through, as did our report.

We don't want to be hitting Twitter with duplicate posts, which is why  
we're careful about when we retry. However, I've seen references to  
Twitter filtering out duplicates, so I was under the impression that  
Twitter would detect and reject the repost message in this case. [1]

[1]: 
http://groups.google.com/group/twitter-development-talk/browse_thread/thread/fdaf7454be8f9006/acc5323f664a?lnk=gstq=duplicate#acc5323f664a

Am I understanding this correctly or should I be more concerned about  
duplicate posts making it through my retry code?

--Eric

[twitter-dev] Re: SSL for Search API?


On Thu, Apr 2, 2009 at 2:58 PM, Chad Etzel jazzyc...@gmail.com wrote:

 But what if I don't want a man-in-the-middle to know I'm secretly
 searching for Britney Spears from my cube?

 ...oh crap.

Welcome to the internet?  Despite any attempts to obfuscate, there are
always way to determine the contents of anything on the net.

--
  Scott Elcomb
  http://www.psema4.com/

[twitter-dev] Re: SSL for Search API?

2009-04-02 Thread Andrew Badera

Define on the net.

If I'm ssh(-2)'d to my server at home, tunneling my HTTP content, forwarding
all DNS requests to the SOCKS proxy Putty presents, how are anyone but
myself and the SSH server going to know exactly what content I just pulled?

(Obviously everything in front of the SSH server is likely unencrypted, but
that's out of scope.)



On Thu, Apr 2, 2009 at 2:12 PM, Scott Elcomb pse...@gmail.com wrote:


 On Thu, Apr 2, 2009 at 2:58 PM, Chad Etzel jazzyc...@gmail.com wrote:
 
  But what if I don't want a man-in-the-middle to know I'm secretly
  searching for Britney Spears from my cube?
 
  ...oh crap.

 Welcome to the internet?  Despite any attempts to obfuscate, there are
 always way to determine the contents of anything on the net.

 --
  Scott Elcomb
  http://www.psema4.com/

[twitter-dev] Re: SSL for Search API?


On Thu, Apr 2, 2009 at 3:12 PM, Scott Elcomb pse...@gmail.com wrote:

 On Thu, Apr 2, 2009 at 2:58 PM, Chad Etzel jazzyc...@gmail.com wrote:

 But what if I don't want a man-in-the-middle to know I'm secretly
 searching for Britney Spears from my cube?

 ...oh crap.

 Welcome to the internet?  Despite any attempts to obfuscate, there are
 always way to determine the contents of anything on the net.

orly?

I was being (mostly) sarcastic previously.. I don't see a reason to
have SSL for twitter search.. but now a serious question: are you
saying that you can decrypt SSL session on-the-fly?  Should I stop
making purchases with my credit cards online?

-Chad

[twitter-dev] Re: SSL for Search API?


On Thu, Apr 2, 2009 at 3:26 PM, Andrew Badera and...@badera.us wrote:
 Define on the net.

 If I'm ssh(-2)'d to my server at home, tunneling my HTTP content, forwarding
 all DNS requests to the SOCKS proxy Putty presents, how are anyone but
 myself and the SSH server going to know exactly what content I just pulled?

 (Obviously everything in front of the SSH server is likely unencrypted, but
 that's out of scope.)

Easy enough - anything that is stored (or passed-and-cached)
electronically outside of tcp/ip networks you control.  Ie. if one
sends a search query to service not under their control that search
string must also be accessible from devices not under their control.

Encryption's great and I use it where I can, but it's not foolproof.
Unfortunately that has been consistently proven throughout history.
Proving otherwise is the ideal case.

-- 
  Scott Elcomb
  http://www.psema4.com/

[twitter-dev] Re: SSL for Search API?

2009-04-02 Thread Andrew Badera

Like I said, out-of-scope.

As I alluded to, and Chad stated: SSL is safe between target server and
client.

All other data is out of scope to this convo.


On Thu, Apr 2, 2009 at 2:34 PM, Scott Elcomb pse...@gmail.com wrote:


 On Thu, Apr 2, 2009 at 3:26 PM, Andrew Badera and...@badera.us wrote:
  Define on the net.
 
  If I'm ssh(-2)'d to my server at home, tunneling my HTTP content,
 forwarding
  all DNS requests to the SOCKS proxy Putty presents, how are anyone but
  myself and the SSH server going to know exactly what content I just
 pulled?
 
  (Obviously everything in front of the SSH server is likely unencrypted,
 but
  that's out of scope.)

 Easy enough - anything that is stored (or passed-and-cached)
 electronically outside of tcp/ip networks you control.  Ie. if one
 sends a search query to service not under their control that search
 string must also be accessible from devices not under their control.

 Encryption's great and I use it where I can, but it's not foolproof.
 Unfortunately that has been consistently proven throughout history.
 Proving otherwise is the ideal case.

 --
   Scott Elcomb
  http://www.psema4.com/

[twitter-dev] Re: SSL for Search API?


On Thu, Apr 2, 2009 at 3:31 PM, Chad Etzel jazzyc...@gmail.com wrote:

 On Thu, Apr 2, 2009 at 3:12 PM, Scott Elcomb pse...@gmail.com wrote:

 On Thu, Apr 2, 2009 at 2:58 PM, Chad Etzel jazzyc...@gmail.com wrote:

 But what if I don't want a man-in-the-middle to know I'm secretly
 searching for Britney Spears from my cube?

 ...oh crap.

 Welcome to the internet?  Despite any attempts to obfuscate, there are
 always way to determine the contents of anything on the net.

 orly?

 I was being (mostly) sarcastic previously.. I don't see a reason to
 have SSL for twitter search.. but now a serious question: are you
 saying that you can decrypt SSL session on-the-fly?  Should I stop
 making purchases with my credit cards online?

Hmm.  I would say no at this point.  I was only trying to draw
attention to the fact encryption rarely survives on a long-term basis.
 If encryption is unbreakable for the period you require, then for
your purposes it is completely valid.  I'm just not sure I believe in
the long term benefits of any particular scheme.

-- 
  Scott Elcomb
  http://www.psema4.com/

[twitter-dev] Re: SSL for Search API?

2009-04-02 Thread ncooper


The reason in my case is that I'm making an AJAX request from a page
that is secure, and I don't want a browser security warning to be
displayed to users.

On Apr 2, 2:31 pm, Chad Etzel jazzyc...@gmail.com wrote:
 On Thu, Apr 2, 2009 at 3:12 PM, Scott Elcomb pse...@gmail.com wrote:

  On Thu, Apr 2, 2009 at 2:58 PM, Chad Etzel jazzyc...@gmail.com wrote:

  But what if I don't want a man-in-the-middle to know I'm secretly
  searching for Britney Spears from my cube?

  ...oh crap.

  Welcome to the internet?  Despite any attempts to obfuscate, there are
  always way to determine the contents of anything on the net.

 orly?

 I was being (mostly) sarcastic previously.. I don't see a reason to
 have SSL for twitter search.. but now a serious question: are you
 saying that you can decrypt SSL session on-the-fly?  Should I stop
 making purchases with my credit cards online?

 -Chad

[twitter-dev] Re: SSL for Search API?

2009-04-02 Thread Abraham Williams

You can proxy requests through your servers.

On Thu, Apr 2, 2009 at 15:34, ncooper ncoo...@convio.com wrote:


 The reason in my case is that I'm making an AJAX request from a page
 that is secure, and I don't want a browser security warning to be
 displayed to users.

 On Apr 2, 2:31 pm, Chad Etzel jazzyc...@gmail.com wrote:
  On Thu, Apr 2, 2009 at 3:12 PM, Scott Elcomb pse...@gmail.com wrote:
 
   On Thu, Apr 2, 2009 at 2:58 PM, Chad Etzel jazzyc...@gmail.com
 wrote:
 
   But what if I don't want a man-in-the-middle to know I'm secretly
   searching for Britney Spears from my cube?
 
   ...oh crap.
 
   Welcome to the internet?  Despite any attempts to obfuscate, there are
   always way to determine the contents of anything on the net.
 
  orly?
 
  I was being (mostly) sarcastic previously.. I don't see a reason to
  have SSL for twitter search.. but now a serious question: are you
  saying that you can decrypt SSL session on-the-fly?  Should I stop
  making purchases with my credit cards online?
 
  -Chad




-- 
Abraham Williams | Hacker | http://abrah.am
@poseurtech | http://the.hackerconundrum.com
Web608 | Community Evangelist | http://web608.org
This email is: [ ] blogable [x] ask first [ ] private.
Sent from Madison, Wisconsin, United States

[twitter-dev] Re: utc_offset and dst

Josh,
Timestamps are all based on UTC [1]. If you want to adjust the data in your
client to be representative of the user's registered timezone, you should
use the utc_offset [2] to transformation the timestamp accordingly.

1. http://en.wikipedia.org/wiki/Coordinated_Universal_Time
2. http://apiwiki.twitter.com/REST-API-Documentation#utcoffset

Doug Williams
Twitter API Support
http://twitter.com/dougw


On Thu, Apr 2, 2009 at 1:09 PM, joshm joshmat...@gmail.com wrote:


 Is it true that the status date created timestamp and timezone offset
 will give me the users local time (if they have their timezone
 setup)?  If so what about daylight savings time?
 I know we get the timezone in text, but is there a good way to map
 this timezone text to say python's pytz library?

[twitter-dev] Issue #414: Include Twitter user ID in Search API results



Can the Twitter user's ID please be included in the Search API results?
Looking at the Atom feed, you get this:

author
  namerwm6f9 (Ross Murker)/name
  urihttp://twitter.com/rwm6f9/uri
/author

First, it would be nice if the name and screen_name were separated,
although this is easy enough to parse. However, the Twitter user's
unique ID is nowhere to be found, forcing a call to /statuses/show to
get it.


--
Dossy Shiobara  | do...@panoptic.com | http://dossy.org/
Panoptic Computer Network   | http://panoptic.com/
  He realized the fastest way to change is to laugh at your own
folly -- then you can let go and quickly move on. (p. 70)

[twitter-dev] Re: utc_offset and dst

2009-04-02 Thread joshm


Gotcha, thx.  I suppose I'm looking for a way to get the timezone from
the utc_offset to take into account DST. Know of any way to do that?

On Apr 2, 1:55 pm, Doug Williams d...@twitter.com wrote:
 Josh,
 Timestamps are all based on UTC [1]. If you want to adjust the data in your
 client to be representative of the user's registered timezone, you should
 use the utc_offset [2] to transformation the timestamp accordingly.

 1.http://en.wikipedia.org/wiki/Coordinated_Universal_Time
 2.http://apiwiki.twitter.com/REST-API-Documentation#utcoffset

 Doug Williams
 Twitter API Supporthttp://twitter.com/dougw

 On Thu, Apr 2, 2009 at 1:09 PM, joshm joshmat...@gmail.com wrote:

  Is it true that the status date created timestamp and timezone offset
  will give me the users local time (if they have their timezone
  setup)?  If so what about daylight savings time?
  I know we get the timezone in text, but is there a good way to map
  this timezone text to say python's pytz library?

[twitter-dev] Re: Issue #414: Include Twitter user ID in Search API results


Is there really someone who hasn't heard of Issue 214, yet?

http://code.google.com/p/twitter-api/issues/detail?id=214

had to,
-Chad

On Thu, Apr 2, 2009 at 5:30 PM, Dossy Shiobara do...@panoptic.com wrote:

 Can the Twitter user's ID please be included in the Search API results?
 Looking at the Atom feed, you get this:

    author
      namerwm6f9 (Ross Murker)/name
      urihttp://twitter.com/rwm6f9/uri
    /author

 First, it would be nice if the name and screen_name were separated,
 although this is easy enough to parse. However, the Twitter user's
 unique ID is nowhere to be found, forcing a call to /statuses/show to
 get it.


 --
 Dossy Shiobara              | do...@panoptic.com | http://dossy.org/
 Panoptic Computer Network   | http://panoptic.com/
  He realized the fastest way to change is to laugh at your own
    folly -- then you can let go and quickly move on. (p. 70)

[twitter-dev] Re: statuses/replies now include mentions

Martin,
I'm not seeing the problem with statuses/replies.json you are reporting. For
which account are you missing data?

Doug Williams
Twitter API Support
http://twitter.com/dougw


On Wed, Apr 1, 2009 at 8:49 PM, Martin Dufort martin.duf...@gmail.comwrote:


 And is this available now via the JSON API interface because,
 according to my tests, I do not see any in the middle of a tweet
 mentions being reported by the API.
 Thanks - Martin

 On Mar 31, 1:33 pm, Joshua Perry j...@6bit.com wrote:
  This hasn't been said but I'm assuming this is only for tweets from this
  point forward, as I don't see any tweets from the past that mention my
  username...
 
 
 
  Doug Williams wrote:
   Devs,
   Before today calls to statuses/replies [1] would return only tweets
   that were prefixed with a @username. As clients began to recognize the
   value in mentions of a @username anywhere in the tweet, they opted to
   perform a search for @username to get the superset.
 
   Twitter agrees [2] that the definition of a reply has changed, and as
   such, calls to statuses/replies contain any tweets that include a
   mention of the authenticating user.
 
   If your client has been using the Search API to retrieve @replies, you
   should begin to migrate to statuses/replies method as it now best
   practice.
 
   1.http://apiwiki.twitter.com/REST-API-Documentation#statuses/replies
   2.http://blog.twitter.com/2009/03/replies-are-now-mentions.html
 
   Code on,
   Doug Williams
   Twitter API Support
  http://twitter.com/dougw

[twitter-dev] Re: Duplicate Tweets

If your application tries to update the status of the same account within a
short period of time, Twitter will ignore the update. As the statuses/update
method returns the status object, in the case where the message was ignored,
the previously successful update (with the same) text will be returned.

You can confirm this behavior yourself. Try to update an account's status
with two requests back to back containing the same text:

$ curl -u user:password -d status=test
http://twitter.com/statuses/update.xml

You will see that the first update is successful. The second request will
then return the same status as the first update (verify by id).

Doug Williams
Twitter API Support
http://twitter.com/dougw

On Thu, Apr 2, 2009 at 11:58 AM, Eric Blair eric.s.bl...@gmail.com wrote:

Just got a report from one of my users that a message he posted
through our app made it through to his Twitter timeline twice. Looking
at our server logs, I can see that when he posted, we got a timeout
from Twitter and successfully tried to repost. My guess is that the
timed-out post actually went through, as did our report.

We don't want to be hitting Twitter with duplicate posts, which is why
we're careful about when we retry. However, I've seen references to
Twitter filtering out duplicates, so I was under the impression that
Twitter would detect and reject the repost message in this case. [1]

[1]:
http://groups.google.com/group/twitter-development-talk/browse_thread/thread/fdaf7454be8f9006/acc5323f664a?lnk=gstq=duplicate#acc5323f664a

Am I understanding this correctly or should I be more concerned about
duplicate posts making it through my retry code?

--Eric

[twitter-dev] Re: Issue #414: Include Twitter user ID in Search API results



On 4/2/09 5:34 PM, Chad Etzel wrote:

Is there really someone who hasn't heard of Issue 214, yet?

http://code.google.com/p/twitter-api/issues/detail?id=214


I don't know if anyone's reviewed issue #214, but the Search API's Atom 
response doesn't even include a from_user_id node in the response.


I guess issue #414 is a kind of dupe of #214 at this point.  Thanks.

--
Dossy Shiobara  | do...@panoptic.com | http://dossy.org/
Panoptic Computer Network   | http://panoptic.com/
  He realized the fastest way to change is to laugh at your own
folly -- then you can let go and quickly move on. (p. 70)

[twitter-dev] Re: Issue #414: Include Twitter user ID in Search API results


From comment #10 by mzsanford:

The fix is the next API. In the mean time I'm going to remove the elements to
prevent confusion.

They've been intentionally baleeted.

-Chad

On Thu, Apr 2, 2009 at 6:04 PM, Dossy Shiobara do...@panoptic.com wrote:

 On 4/2/09 5:34 PM, Chad Etzel wrote:

 Is there really someone who hasn't heard of Issue 214, yet?

 http://code.google.com/p/twitter-api/issues/detail?id=214

 I don't know if anyone's reviewed issue #214, but the Search API's Atom
 response doesn't even include a from_user_id node in the response.

 I guess issue #414 is a kind of dupe of #214 at this point.  Thanks.

 --
 Dossy Shiobara              | do...@panoptic.com | http://dossy.org/
 Panoptic Computer Network   | http://panoptic.com/
  He realized the fastest way to change is to laugh at your own
    folly -- then you can let go and quickly move on. (p. 70)

[twitter-dev] Re: Duplicate Tweets

2009-04-02 Thread Eric Blair

That's what I was expecting to see. However, I have a user who's
update made it to his timeline twice. I see that we sent the request
twice, 5 seconds apart, because the first one didn't complete. The
second request returned successful.

The user's timeline is protected, but the messages are id 1440033342
and 1440033271. I log the ids of successful posts and, in my logs, I
see the higher id (1440033342).

--Eric

On Apr 2, 2009, at 6:03 PM, Doug Williams wrote:

If your application tries to update the status of the same account
within a short period of time, Twitter will ignore the update. As
the statuses/update method returns the status object, in the case
where the message was ignored, the previously successful update
(with the same) text will be returned.

You can confirm this behavior yourself. Try to update an account's
status with two requests back to back containing the same text:

$ curl -u user:password -d status=test
http://twitter.com/statuses/update.xml

You will see that the first update is successful. The second request
will then return the same status as the first update (verify by id).

Doug Williams
Twitter API Support
http://twitter.com/dougw

On Thu, Apr 2, 2009 at 11:58 AM, Eric Blair eric.s.bl...@gmail.com
wrote:

[1]:
http://groups.google.com/group/twitter-development-talk/browse_thread/thread/fdaf7454be8f9006/acc5323f664a?lnk=gstq=duplicate#acc5323f664a

Am I understanding this correctly or should I be more concerned about
duplicate posts making it through my retry code?

--Eric

[twitter-dev] Re: statuses/replies now include mentions

2009-04-02 Thread sujamthe


Doug,

I second Mike's suggestion to codify RTs. I have clients who want to
track RTs for two scenarios:

1. To find their brand influencers, as to who did the rts as they are
their brand advocates. It would be helpful to get to this in a quick
ap - track rts by person, how many rts of same content. I am gathering
this manually or by loading tweets from the search API into an overall
social media metrics tool and then massaging the data. Ideally I would
like to automate this to connect to the internal marketing engine.

2. This is also powerful for companies trying A/B testing of messaging
using RTs. Again tracking which rts echoed more and who give it the
bump up and times when it worked best.

I seem the lonely voice here as my needs are for building out
enterprise apps or rather integration of twitter into the enterprise.

best,
Sudha

Sudha Jamthe
http://tmeet.me and unnamed twitter apps for Intuit, Network World.

On Mar 30, 7:47 pm, Mike Champion mike.champ...@gmail.com wrote:
 Great, this will be a helpful change.

 Any discussion of codifying Retweets in a similar way in the search
 API? It seems like they are also a subset of Mentions where 1) starts
 with RT 2) includes a @mention 3) rest of the content (fuzzy) matches
 a previous tweet by the @mention tweeter.

 -mike

 On Mar 30, 10:28 pm, tweetip twee...@mac.com wrote:

  In changing our code, we've decided:

  Show my replies becomes Show my mentions

  but

  Reply to is not becoming Mention to - it stays Reply to

  otoh

  having both my replies and my mentions is something users will ask
  for...

  hth :)

[twitter-dev] Re: statuses/replies now include mentions

The API is not able to support retweets as a feature until the main
Twitter.com site offers some notion of retweets as a feature. As evidenced
by the recent shift from @replies to mentions, Twitter does listen to the
users' behavior to drive site changes. We obviously recognize the large
number of users adopting retweets as a way to share good content. For now,
though, retweets must be found through client-side parsing.

Thanks to all for the lively discussion on this thread. It has been
valuable.

Doug Williams
Twitter API Support
http://twitter.com/dougw


On Thu, Apr 2, 2009 at 4:15 PM, sujamthe sujam...@gmail.com wrote:


 Doug,

 I second Mike's suggestion to codify RTs. I have clients who want to
 track RTs for two scenarios:

 1. To find their brand influencers, as to who did the rts as they are
 their brand advocates. It would be helpful to get to this in a quick
 ap - track rts by person, how many rts of same content. I am gathering
 this manually or by loading tweets from the search API into an overall
 social media metrics tool and then massaging the data. Ideally I would
 like to automate this to connect to the internal marketing engine.

 2. This is also powerful for companies trying A/B testing of messaging
 using RTs. Again tracking which rts echoed more and who give it the
 bump up and times when it worked best.

 I seem the lonely voice here as my needs are for building out
 enterprise apps or rather integration of twitter into the enterprise.

 best,
 Sudha

 Sudha Jamthe
 http://tmeet.me and unnamed twitter apps for Intuit, Network World.

 On Mar 30, 7:47 pm, Mike Champion mike.champ...@gmail.com wrote:
  Great, this will be a helpful change.
 
  Any discussion of codifying Retweets in a similar way in the search
  API? It seems like they are also a subset of Mentions where 1) starts
  with RT 2) includes a @mention 3) rest of the content (fuzzy) matches
  a previous tweet by the @mention tweeter.
 
  -mike
 
  On Mar 30, 10:28 pm, tweetip twee...@mac.com wrote:
 
   In changing our code, we've decided:
 
   Show my replies becomes Show my mentions
 
   but
 
   Reply to is not becoming Mention to - it stays Reply to
 
   otoh
 
   having both my replies and my mentions is something users will ask
   for...
 
   hth :)

[twitter-dev] Re: Issue #414: Include Twitter user ID in Search API results



On 4/2/09 6:10 PM, Chad Etzel wrote:

From comment #10 by mzsanford:


The fix is the next API. In the mean time I'm going to remove the elements to
prevent confusion.

They've been intentionally baleeted.


Aha.  OK, well, I'm just using the Search API to get ID's now, and am 
using /statuses/show and /users/show to fetch the relevant data. 
*hammer hammer hammer*  :-)



--
Dossy Shiobara  | do...@panoptic.com | http://dossy.org/
Panoptic Computer Network   | http://panoptic.com/
  He realized the fastest way to change is to laugh at your own
folly -- then you can let go and quickly move on. (p. 70)

[twitter-dev] is /home changing to /home/timeline (which to use for future pre-filling of tweets) ?

2009-04-02 Thread Rodney


With the new layout with the #sidey search bar the http://twitter.com/home
page forwards to http://twitter.com/timeline/home

This breaks those links that are pre-filling the tweet by linking to /
home

So before this would work:

http://twitter.com/home?status=hello+world

But on the new @home page with the search in the sidebar, it doesn't
work (doesn't prefill the input box). It has to be:

http://twitter.com/timeline/home?status=hello+world

On the Twitter Support page where it talks about the new mentions
instead of @replies it has a link that says Take me back to
Twitter.com
http://help.twitter.com/portal

But that Take me back to Twitter.com links to: 
http://twitter.com/timeline/home
(which gives a 404 for users who *don't* have the new #sidey
searchbar)

All of that to say, which will be the correct way to link going
forward? I know there's a few apps that prefill the box for the user
that may need to be updated.

[twitter-dev] Re: is /home changing to /home/timeline (which to use for future pre-filling of tweets) ?

Issue 413 [1] explains that this is a general site issue but from developers
you need to know where we are going. The
http://twitter.com/timeline/homeURL will be the address moving
forward, so please develop against that.

Both of the errors you mentioned, pre-filled update box and the
help.twitter.com link URL, are being addressed.

1. http://code.google.com/p/twitter-api/issues/detail?id=413

Thanks,
Doug Williams
Twitter API Support
http://twitter.com/dougw


On Thu, Apr 2, 2009 at 4:35 PM, Rodney rodn...@gmail.com wrote:


 With the new layout with the #sidey search bar the http://twitter.com/home
 page forwards to http://twitter.com/timeline/home

 This breaks those links that are pre-filling the tweet by linking to /
 home

 So before this would work:

 http://twitter.com/home?status=hello+world

 But on the new @home page with the search in the sidebar, it doesn't
 work (doesn't prefill the input box). It has to be:

 http://twitter.com/timeline/home?status=hello+world

 On the Twitter Support page where it talks about the new mentions
 instead of @replies it has a link that says Take me back to
 Twitter.com
 http://help.twitter.com/portal

 But that Take me back to Twitter.com links to:
 http://twitter.com/timeline/home
 (which gives a 404 for users who *don't* have the new #sidey
 searchbar)

 All of that to say, which will be the correct way to link going
 forward? I know there's a few apps that prefill the box for the user
 that may need to be updated.

[twitter-dev] Re: is /home changing to /home/timeline (which to use for future pre-filling of tweets) ?

s/from developers/for developers/;

Doug Williams
Twitter API Support
http://twitter.com/dougw


On Thu, Apr 2, 2009 at 4:54 PM, Doug Williams d...@twitter.com wrote:

 Issue 413 [1] explains that this is a general site issue but from
 developers you need to know where we are going. The
 http://twitter.com/timeline/home URL will be the address moving forward,
 so please develop against that.

 Both of the errors you mentioned, pre-filled update box and the
 help.twitter.com link URL, are being addressed.

 1. http://code.google.com/p/twitter-api/issues/detail?id=413

 Thanks,
 Doug Williams
 Twitter API Support
 http://twitter.com/dougw



 On Thu, Apr 2, 2009 at 4:35 PM, Rodney rodn...@gmail.com wrote:


 With the new layout with the #sidey search bar the
 http://twitter.com/home
 page forwards to http://twitter.com/timeline/home

 This breaks those links that are pre-filling the tweet by linking to /
 home

 So before this would work:

 http://twitter.com/home?status=hello+world

 But on the new @home page with the search in the sidebar, it doesn't
 work (doesn't prefill the input box). It has to be:

 http://twitter.com/timeline/home?status=hello+world

 On the Twitter Support page where it talks about the new mentions
 instead of @replies it has a link that says Take me back to
 Twitter.com
 http://help.twitter.com/portal

 But that Take me back to Twitter.com links to:
 http://twitter.com/timeline/home
 (which gives a 404 for users who *don't* have the new #sidey
 searchbar)

 All of that to say, which will be the correct way to link going
 forward? I know there's a few apps that prefill the box for the user
 that may need to be updated.

[twitter-dev] verify_credentials not returning full user data for some users

2009-04-02 Thread dean.j.robinson


I noticed this because I've got some code that checks the utc_offset
and noticed that it was missing.

For my hahlo account verify_credentials is returning this (appears to
be missing colours, utc_offset, fav counts etc etc):

?xml version=1.0 encoding=UTF-8?
user
  id7097682/id
  nameHahlo.com/name
  screen_namehahlo/screen_name
  locationNewcastle, Australia/location
  descriptionThe best iPhone/touch style Twitter app. Hahlo 4 will
be oAuth-tastic./description
  profile_image_urlhttp://s3.amazonaws.com/twitter_production/
profile_images/53694164/favicon_normal.png/profile_image_url
  urlhttp://hahlo.com/url
  protectedfalse/protected
  followers_count1097/followers_count
  status
created_atTue Mar 31 00:52:42 + 2009/created_at
id1421396597/id
textgreat news from twitter, all @replies are now part of the
replies timeline, effective immediately, you don't need to change a
single thing/text
sourcelt;a href=quot;http://dev.hahlo.com/
quot;gt;#948;.hahlolt;/agt;/source
truncatedfalse/truncated
in_reply_to_status_id/in_reply_to_status_id
in_reply_to_user_id/in_reply_to_user_id
favoritedfalse/favorited
in_reply_to_screen_name/in_reply_to_screen_name
  /status
/user



for my deanjrobinson account it returns this:

?xml version=1.0 encoding=UTF-8?
user
  id782566/id
  nameDean Robinson/name
  screen_namedeanjrobinson/screen_name
  locationNewcastle, Australia/location
  description/description
  profile_image_urlhttp://s3.amazonaws.com/twitter_production/
profile_images/54336181/fie-12_normal.png/profile_image_url
  urlhttp://www.deanjrobinson.com/url
  protectedfalse/protected
  followers_count375/followers_count
  profile_background_colorFF/profile_background_color
  profile_text_color323232/profile_text_color
  profile_link_color676767/profile_link_color
  profile_sidebar_fill_colorF5F5F5/profile_sidebar_fill_color
  profile_sidebar_border_colorEE/profile_sidebar_border_color
  friends_count97/friends_count
  created_atTue Feb 20 08:02:11 + 2007/created_at
  favourites_count65/favourites_count
  utc_offset36000/utc_offset
  time_zoneSydney/time_zone
  profile_background_image_urlhttp://s3.amazonaws.com/
twitter_production/profile_background_images/1486932/twitterbg.jpg/
profile_background_image_url
  profile_background_tilefalse/profile_background_tile
  statuses_count2157/statuses_count
  notifications/notifications
  following/following
  status
created_atThu Apr 02 22:55:57 + 2009/created_at
id1441582913/id
texthmmm, installed littlesnapper update, and now I've got four
copies of every screenshot... I think thats a bug./text
sourcelt;a href=http://www.hahlo.com/gt;Hahlolt;/agt;/
source
truncatedfalse/truncated
in_reply_to_status_id/in_reply_to_status_id
in_reply_to_user_id/in_reply_to_user_id
favoritedfalse/favorited
in_reply_to_screen_name/in_reply_to_screen_name
  /status
/user


Is this connected to the recent change that adds the full user object
to all api calls?

[twitter-dev] Re: verify_credentials not returning full user data for some users


This is related to the caching changes that came as a result of
yesterday's push [1]. It shouldn't be a problem for active users, and
will be fully fixed after less active users expire from the cache.

1. 
http://groups.google.com/group/twitter-development-talk/browse_thread/thread/f79c44d905a3e83d#

Doug Williams
Twitter API Support

On Apr 2, 5:07 pm, dean.j.robinson dean.j.robin...@gmail.com
wrote:
 I noticed this because I've got some code that checks the utc_offset
 and noticed that it was missing.

 For my hahlo account verify_credentials is returning this (appears to
 be missing colours, utc_offset, fav counts etc etc):

 ?xml version=1.0 encoding=UTF-8?
 user
   id7097682/id
   nameHahlo.com/name
   screen_namehahlo/screen_name
   locationNewcastle, Australia/location
   descriptionThe best iPhone/touch style Twitter app. Hahlo 4 will
 be oAuth-tastic./description
   profile_image_urlhttp://s3.amazonaws.com/twitter_production/
 profile_images/53694164/favicon_normal.png/profile_image_url
   urlhttp://hahlo.com/url
   protectedfalse/protected
   followers_count1097/followers_count
   status
     created_atTue Mar 31 00:52:42 + 2009/created_at
     id1421396597/id
     textgreat news from twitter, all @replies are now part of the
 replies timeline, effective immediately, you don't need to change a
 single thing/text
     sourcelt;a href=quot;http://dev.hahlo.com/
 quot;gt;#948;.hahlolt;/agt;/source
     truncatedfalse/truncated
     in_reply_to_status_id/in_reply_to_status_id
     in_reply_to_user_id/in_reply_to_user_id
     favoritedfalse/favorited
     in_reply_to_screen_name/in_reply_to_screen_name
   /status
 /user

 for my deanjrobinson account it returns this:

 ?xml version=1.0 encoding=UTF-8?
 user
   id782566/id
   nameDean Robinson/name
   screen_namedeanjrobinson/screen_name
   locationNewcastle, Australia/location
   description/description
   profile_image_urlhttp://s3.amazonaws.com/twitter_production/
 profile_images/54336181/fie-12_normal.png/profile_image_url
   urlhttp://www.deanjrobinson.com/url
   protectedfalse/protected
   followers_count375/followers_count
   profile_background_colorFF/profile_background_color
   profile_text_color323232/profile_text_color
   profile_link_color676767/profile_link_color
   profile_sidebar_fill_colorF5F5F5/profile_sidebar_fill_color
   profile_sidebar_border_colorEE/profile_sidebar_border_color
   friends_count97/friends_count
   created_atTue Feb 20 08:02:11 + 2007/created_at
   favourites_count65/favourites_count
   utc_offset36000/utc_offset
   time_zoneSydney/time_zone
   profile_background_image_urlhttp://s3.amazonaws.com/
 twitter_production/profile_background_images/1486932/twitterbg.jpg/
 profile_background_image_url
   profile_background_tilefalse/profile_background_tile
   statuses_count2157/statuses_count
   notifications/notifications
   following/following
   status
     created_atThu Apr 02 22:55:57 + 2009/created_at
     id1441582913/id
     texthmmm, installed littlesnapper update, and now I've got four
 copies of every screenshot... I think thats a bug./text
     sourcelt;a href=http://www.hahlo.com/gt;Hahlolt;/agt;/
 source
     truncatedfalse/truncated
     in_reply_to_status_id/in_reply_to_status_id
     in_reply_to_user_id/in_reply_to_user_id
     favoritedfalse/favorited
     in_reply_to_screen_name/in_reply_to_screen_name
   /status
 /user

 Is this connected to the recent change that adds the full user object
 to all api calls?

[twitter-dev] Re: API Changes for April 1, 2009

Jeffery,
This is valid criticism. This bug came as a surprise to us as well. We
otherwise would have given developers fair warning. Unfortunately there is
no easy fix, and like a bad heart-break, time may be the only answer.

In short, the problem is with the user data cache. To get the extended
information into that cache, the user object must either expire or be
invalidated through some user initiated update. The expiry on the cache is
rather long and you will find that inactive accounts will have abbreviated
data for up to 2 weeks.

This is obviously sub-optimal, as Matt would say.

Doug Williams
Twitter API Support
http://twitter.com/dougw


On Thu, Apr 2, 2009 at 12:27 PM, Jeffrey Greenberg 
jeffreygreenb...@gmail.com wrote:


 Doug,
 Grumble: just to say it, this wasn't handled well at all.  The fact
 that this field disappears whether due to caching or through a coding
 error has the same result of completely breaking my app.

 How long will it take for this issue to clear up? Days? How many
 exactly?  and after X days will further requests be populated
 correctly?

 thx,
 jeffrey
 http://www.tweettronics.com

[twitter-dev] Deprecation of the email parameter for users/show

For a few days, the users/show method offered look up based on a
user's email address. So short-lived was its documented availability
that we removed it without much fanfare. This thread [1] has mention
of this deprecation.

Recently, there has been quite a bit of discussion on this feature's
reinstatement on and off the list. Issue 353 [2] covers this request.

The use of the method was largely as intended; people were discovering
account connections based email addresses. This made integration with
other networks and applications trivial. However, there was a
significant amount of traffic that was using this parameter for evil.
In either case, the adoption was minimal (we did not receive a
complaint that the deprecation completely broke someone's
application). The rationale for deprecation was to protect our users'
privacy.

We do realize the large amount of value that this parameter creates
for application developers. However at this time, we are working to
identify a solution for the spammers that caused the deprecation. One
suggestion is to grant trusted applications access to this parameter.
Since our answer to trusting applications is OAuth and it is still in
beta, we will not be able to devote the resources necessary to bring
this parameter back at this time.

If you are developing an application that could benefit from an
email-based lookup, please star the issue [2] accordingly.

1.
http://groups.google.com/group/twitter-development-talk/browse_thread/thread/ac80b4fdf8eb742a/692b27b8ebd07268?lnk=gstq=email+parameter#692b27b8ebd07268
2. http://code.google.com/p/twitter-api/issues/detail?id=353

Thanks,
Doug Williams
Twitter API Support
http://twitter.com/dougw

[twitter-dev] Re: API usage via SQL 2005 or 2008

2009-04-02 Thread blaine


There is an article on SQL Server Magazine's web site about it
called , Tweet-SQL Lets You Update Your Twitter Account Using T-SQL  ,
here is a link  http://www.sqlmag.com/articles/index.cfm?articleid=101753;

On Mar 27, 9:50 pm, Dataluxe cr...@mailtoad.com wrote:
 Has anyone integrated SQL 2005 or SQL 2008 directly with the twitter
 API?  If so I'd sure love to hear how you've done it.

 Thanks!
 Craig

[twitter-dev] Re: Search queries not working

2009-04-02 Thread feedbackmine


Hi Matt,

I have tried to use language parameter of twitter search and find the
result is very unreliable. For example:
http://search.twitter.com/search?lang=allq=tweetjobsearch returns 10
results (all in english), but
http://search.twitter.com/search?lang=enq=tweetjobsearch only returns
3.

I googled this list and it seems you are using n-gram based algorithm
(http://groups.google.com/group/twitter-development-talk/msg/
565313d7b36e8d65). I have found n-gram algorithm works very well for
language detection, but the quality of training data may make a big
difference.

Recently I have developed a language detector (in ruby) myself:
http://github.com/feedbackmine/language_detector/tree/master
It uses wikipedia's data for training, and based on my limited
experience it works well. Actually using wikipedia's data is not my
idea, all credits should go to Kevin Burton (http://feedblog.org/
2005/08/19/ngram-language-categorization-source/ ).

Just thought you may be interested.

@feedbackmine
http://twitter.com/feedbackmine

On Mar 31, 11:22 am, Matt Sanford m...@twitter.com wrote:
 Hi there,

      Can you provide an example URL where since_id isn't working so I  
 can try and reproduce the issue? As forlanguage, thelanguage 
 identifier is not a 100% and sometimes makes mistakes. Hopefully not  
 too many mistakes but it definitely does.

 Thanks;
    — Matt Sanford / @mzsanford

 On Mar 31, 2009, at 08:14 AM, codepuke wrote:





  Hi all;

  I see a few people complaining about the since_id not working.  I too
  have the same issue - I am currently storing the last executed id and
  having to check new tweets to make sure their id is greater than my
  last processed id as a temporary workaround.

  I have also noticed that the filter bylanguageparam also doesn't
  seem to be working 100% - I notice a few chinese tweets, as well as
  tweets having a null value forlanguage...

[twitter-dev] Re: Deprecation of the email parameter for users/show

2009-04-02 Thread John Sampson


Thank you for the clarity and for putting greater detail on the
deprecation of user/show (find by email).  I completely agree to
Twitter wanting to protect its user's privacy.  I'd like to think that
the value created by whitelisted applications is far greater than the
pain being caused by non-whitelisted api users.  I hope that a speedy
solution can be found for these spammers.

Much of my concern I'd previously mentioned on ticket #353 -
http://code.google.com/p/twitter-api/issues/detail?id=353#c8.

To be clear, this does break the Twitter integration with our Firefox
extension (which I consider the most valuable portion of our ext),
surfacing information for user with screen name show rather than the
person you are connecting with keyed off email.  Additionally,
workflow need be re-authored in our other apps that have leveraged
this method to date.

Hoping I can be of further assistance in returning this method to the
production API.

John Sampson
http://zentact.com




On Apr 2, 5:59 pm, Doug Williams d...@twitter.com wrote:
 For a few days, the users/show method offered look up based on a
 user's email address. So short-lived was its documented availability
 that we removed it without much fanfare. This thread [1] has mention
 of this deprecation.

 Recently, there has been quite a bit of discussion on this feature's
 reinstatement on and off the list. Issue 353 [2] covers this request.

 The use of the method was largely as intended; people were discovering
 account connections based email addresses. This made integration with
 other networks and applications trivial. However, there was a
 significant amount of traffic that was using this parameter for evil.
 In either case, the adoption was minimal (we did not receive a
 complaint that the deprecation completely broke someone's
 application). The rationale for deprecation was to protect our users'
 privacy.

 We do realize the large amount of value that this parameter creates
 for application developers. However at this time, we are working to
 identify a solution for the spammers that caused the deprecation. One
 suggestion is to grant trusted applications access to this parameter.
 Since our answer to trusting applications is OAuth and it is still in
 beta, we will not be able to devote the resources necessary to bring
 this parameter back at this time.

 If you are developing an application that could benefit from an
 email-based lookup, please star the issue [2] accordingly.

 1.http://groups.google.com/group/twitter-development-talk/browse_thread...
 2.http://code.google.com/p/twitter-api/issues/detail?id=353

 Thanks,
 Doug Williams
 Twitter API Supporthttp://twitter.com/dougw

[twitter-dev] Re: Search queries not working

2009-04-02 Thread Basha Shaik

Hi matt,

Thank You
What is Pagination? Does it mean that I cannot use max_id for searching
tweets. What does next_url and prev_url fields mean. I did not find next_url
and prev_url in documentation. how can these two urls be used with max_id.
Please explain with example if possible.



Regards,

Mahaboob Basha Shaik
www.netelixir.com
Making Search Work


On Wed, Apr 1, 2009 at 4:23 PM, Matt Sanford m...@twitter.com wrote:

 Hi Basha,
 The max_id is only intended to be used for pagination via the next_url
 and prev_url fields and is known not to work with since_id. It is not
 documented as a valid parameter because it's known to only work in the case
 it was designed for. We added the max_id to prevent the problem where you
 click on 'Next' and page two starts with duplicates. Here's the scenario:

  1. Let's say you search for 'foo'.
  2. You wait 10 seconds, during which 5 people send tweets containing
 'foo'.
  3. You click next and go to page=2 (or call page=2 via the API)
3.a. If we displayed results 21-40 the first 5 results would look like
 duplicates because they were pushed down by the 5 new entries.
3.b. If we append a max_id from the time you searched we can do and
 offset from the maximum and the new 5 entries are skipped.

   We use option 3.b. (as does twitter.com now) so you don't see
 duplicates. Since we wanted to provide the same data in the API as the UI we
 added the next_url and prev_url members in our output.

 Thanks;
   — Matt Sanford

 On Mar 31, 2009, at 08:42 PM, Basha Shaik wrote:

 HI Matt,

 when Since_id and Max_id are given together, max_id is not working. This
 query is ignoring max_id. But with only since _id its working fine. Is there
 any problem when max_id and since_id are used together.

 Also please tell me what does max_id exactly mean and also what does it
 return when we send a request.
 Also tell me what the total returns.


 Regards,

 Mahaboob Basha Shaik
 www.netelixir.com
 Making Search Work


 On Tue, Mar 31, 2009 at 3:22 PM, Matt Sanford m...@twitter.com wrote:


 Hi there,

Can you provide an example URL where since_id isn't working so I can
 try and reproduce the issue? As for language, the language identifier is not
 a 100% and sometimes makes mistakes. Hopefully not too many mistakes but it
 definitely does.

 Thanks;
  — Matt Sanford / @mzsanford


 On Mar 31, 2009, at 08:14 AM, codepuke wrote:


 Hi all;

 I see a few people complaining about the since_id not working.  I too
 have the same issue - I am currently storing the last executed id and
 having to check new tweets to make sure their id is greater than my
 last processed id as a temporary workaround.

 I have also noticed that the filter by language param also doesn't
 seem to be working 100% - I notice a few chinese tweets, as well as
 tweets having a null value for language...

[twitter-dev] Re: Hi All

2009-04-02 Thread Basha Shaik

Id is nothing but the tweet id. take the tweet id of the 100th tweet in 15th
page?

Regards,

Mahaboob Basha Shaik
www.netelixir.com
Making Search Work

On Thu, Apr 2, 2009 at 1:35 PM, Seema Nagar nagar.se...@gmail.com wrote:

Thanks a lot. Can you tell which is this parameter id of the status last
page ? How can I get it ?
I am using ATOM format.

regards,
Seema

On Thu, Apr 2, 2009 at 7:01 PM, Basha Shaik basha.neteli...@gmail.comwrote:

Hi,

The api gives only last 1500 tweets, 100 per page. Usingd since_id and
Until_id we can get atmost 1500 tweets.

I can't tell you how to get all searched between two date ranges but i
think below process can help you get more than 1500 and if processed
programatically you can achieve what you want.

1. Set rpp=100 and retrieve 15 pages search results by incrementing
the param 'page'
2. Get the id of the last status on page 15 and set that as the max_id
for the next query
3. If we have more results, go to step 1

this worked fine for me.

Please check below link.

http://groups.google.com/group/twitter-development-talk/browse_thread/thread/d6a919e25eecd4db/60d54a47b19e16d5?lnk=raot

Regards,

Mahaboob Basha Shaik
www.netelixir.com
Making Search Work

On Thu, Apr 2, 2009 at 6:43 AM, Seema Nagar nagar.se...@gmail.comwrote:

I am new to twitter. I am trying twitter search API to get last one
month's data(
http://search.twitter.com/search.atom?geocode=51.500152%2C-0.126236%2C15milang=enq=+movie+since%3A2009-03-03+until%3A2009-03-30+near%3Alondon+within%3A15mi).
Unfortunately I get only fifteen
pages :(. Has anybody tried this before ? Or is there a way to get
remaining tweets ?

Thanks in advance.

regards,
Seema

[twitter-dev] Re: Search queries not working