[twitter-dev] trends/location: number of occurrence of location element

2010-03-08 Thread Yusuke Yamamoto
Hi,

I'm developing Twitter4J and have a quick question.

Is there any chance that trends/location returns multiple location elements?

Thanks,
-- 
Yusuke Yamamoto
yus...@mac.com

this email is: [x] bloggable/tweetable [ ] private
follow me on : http://twitter.com/yusukeyamamoto
subscribe me at : http://yusuke.homeip.net/blog/



[twitter-dev] Re: What tools do you use?

2010-03-08 Thread Karthik
Arc90 - PHP Library for REST and Search API -
http://lab.arc90.com/2008/06/03/php-twitter-api-client/
PhireHose - PHP library for Streaming API - http://code.google.com/p/phirehose
CodeIgniter as PHP framework and
Netbeans PHP IDE



Re: [twitter-dev] Problem sending tweets with nbsp chars

2010-03-08 Thread Roy Leban
Thread of a discussion:

twitterapi: @RoyLeban nbsps will be literally encoded as nbsps (and not
spaces) ^RK

RoyLeban: @twitterapi Yup, nbsp is what I need. See @Puzzazz to see what
I'm doing. Any chance bug'll get fixed? If not, what's the real tweet limit?

twitterapi: @RoyLeban it's not a bug. nbsp are handled like that on
purpose. ^RK

RoyLeban: @twitterapi What does like that mean? You count each nbsp as 1
char until you get close to 140 then count them as more? Sounds like a bug


Here's the deal: non-breaking spaces are the only thing which work as
non-breaking spaces. I know that sounds like a tautology, but look at the
@Puzzazz feed and you'll see what I need. I tried unicode non-breaking
spaces, I tried UTF8 non-breaking spaces, I tried %20's, I tried multiple
spaces. The only thing that worked properly are nbsp; entities and they
work perfectly. Except...

There's a bug in how they're counted. If I don't send very many, I do not
have problems and each nbsp; is counted as 1 character. I can send a tweet
of 300 chars that resolves to 140 without problems. However, if the
resulting tweet is close to 140 after the nbsp's are resolved, Twitter
erroneously claims that the tweet is too long. It is not an off-by-one bug
as a tweet of 136 fails. I do not know what the lower bound is.

I don't know for sure, but I think this bug was recently introduced. I have
not had this problem in the past. However, there was one 3-day span in which
no tweets went out, then they suddenly started again. By the time I noticed,
it was too late to do an analysis as I wasn't logging the error message from
Twitter (I am now). It could have been a similar temporary problem.

I'd love to see this bug fixed. If you can't fix it, is it possible to give
me enough information about the bug so that I can at least know what the
true tweet limit is?

Thanks very much!

/Roy


On Thu, Mar 4, 2010 at 7:45 AM, Taylor Singletary 
taylorsinglet...@twitter.com wrote:

 Hi Roy,

 You shouldn't be sending spaces as nbsp; -- that's HTML entity encoding.
 It's best to send space characters as %20 instead.

 For example:

 You'd set your POST body to:

 status=There%20is%20%20%20%20%20space%20for%20love%20in%20%20%20the%20universe

 If you were trying to set the status
 There isspace for love in   the universe

 In your signature base string for OAuth you'd have to encode those spaces
 one more time:

 POSThttp%3A%2F%2Fapi.twitter.com
 %2F1%2Fstatuses%2Fupdate.xmloauth_consumer_key%3Dri8JxYK2ddwSV5xIUfNNvQ%26oauth_nonce%3DulF0XQetLMOm5Sr9Yrp027Hzu2mPoTuTqFgshncHBo%26oauth_signature_method%3DHMAC-SHA1%26oauth_timestamp%3D1267717205%26oauth_token%3D819797-torCkTs0XK7H2Y2i1ee5iofqkMC4p7aayeEXRTmlw%26oauth_version%3D1.0%26status%3DThere%2520is%2520%2520%2520%2520%2520space%2520for%2520love%2520in%2520%2520%2520the%2520universe


 Not all Twitter API clients will choose to preserve multiple spaces on
 display though.

 Taylor


 On Wed, Mar 3, 2010 at 11:13 PM, Roy Leban r...@royleban.com wrote:

 Twitter is rejecting tweets as too long when the nbsp character is
 used. Here is an example tweet in plain text

 Clue 5 of 15: R _ C __ _ N _N _ _ E __ _ T E R__ _
 _ N E E R_ N_ _ R _ C U L T U R E http://www.puzzazz.com/s348
 [140 chars]

 And, as I'm sending it with the nbsp characters:

 Clue 5 of 15: Rnbsp;_nbsp;Cnbsp;_nbsp;nbsp;
 nbsp;_nbsp;_nbsp;Nnbsp;_nbsp;nbsp;
 nbsp;Nnbsp;_nbsp;_nbsp;Enbsp;_nbsp;nbsp;
 nbsp;_nbsp;_nbsp;Tnbsp;Enbsp;Rnbsp;nbsp; nbsp;_nbsp;nbsp;
 nbsp;_nbsp;_nbsp;_nbsp;Nnbsp;Enbsp;Enbsp;Rnbsp;nbsp;
 nbsp;_nbsp;Nnbsp;nbsp;

 nbsp;_nbsp;_nbsp;Rnbsp;_nbsp;Cnbsp;Unbsp;Lnbsp;Tnbsp;Unbsp;Rnbsp;E
 http://www.puzzazz.com/s348

 If the nbsp's are each counted as 6 characters, this would be 400
 chars, but Twitter accepts tweets like this. For example, this tweet:

 http://twitter.com/Puzzazz/status/9781320047

 is 114 chars but I send 304 chars with the nbsp's.

 I have a guess that this only happens when the resulting tweet is
 exactly 140 chars. To test this theory, I just modified the site to
 shorten that tweet below 140. Sure enough, it works:

 http://twitter.com/Puzzazz/status/9963348931





[twitter-dev] Application based on Search API

2010-03-08 Thread Diz
For the last 3 months I am experimenting the Search API. I'm focusing
on the geo-location searching capabilities.

For the beginning I started with my own city, but my intents are to
extend to the major cities of my country: that will be at most ten. My
idea of application is to offer real-time activity on each major city,
through a proxy that caches all tweets and then serves them further to
clients, filtered or non-filtered. Frequency of requests should be
between 5 to 10 seconds, and that means I should do between 400 to 800
requests per hour just for one city, and probably between 4000 to 8000
requests for the whole application.

My questions are:
1). Should I use Search API, or should I move to the Streaming API?!
2). To whom I should request whitelisting: the usual Search API or the
Streaming API?!


Thanks!


[twitter-dev] Re: link to disabled acct

2010-03-08 Thread Frank
When I click on the link the page comes up so folks can see my recent
tweets.  However, when I try to log in then I get the This account
has been disabled message.
What I don't understand is that the page mentioned in my first
sentence comes up if the account has been disabled.

On Mar 5, 10:24 am, Thomas Woolway tswool...@gmail.com wrote:
 Twitter brings up a page saying something like 'This account has been
 suspended'. That's the same whether you try to open the user's profile page
 or an individual tweet.

 Tom

 On Fri, Mar 5, 2010 at 11:39 AM, Frank gn...@windstream.net wrote:
  If an account is disabled will a link to it on  a webpage still bring
  it up?


[twitter-dev] Re: You aren't allowed to add members to this list

2010-03-08 Thread YARG
Thanks Abraham .. a little embarrassing as I’d got it all completely
wrong, anyway thanks for your help.  It all works a treat now.
Cheers
Steve

On Feb 24, 9:32 pm, Abraham Williams 4bra...@gmail.com wrote:
 You are trying to add members to @twitterapi/team which is an account/list
 you don't own.

 If you are adding user_id 94564101 to listb by usera you should 
 use:https://api.twitter.com/1/usera/listb/members.xml?id=94564101

 Abraham





 On Wed, Feb 24, 2010 at 06:31, YARG sg...@yarg.com wrote:
  Hi, I'm updating a .net lib so it can handle lists.  List creation is
  OK but when I try and assign a member (an account which I also own) to
  the newly created list I get the error:

  ?xml version=1.0 encoding=UTF-8?
  hash
   request/1/twitterapi/team/members.xml?
  list_id=7869628amp;id=94564101/request
   errorYou aren't allowed to add members to this list/error
  /hash

  I've checked the docs and and I'm pretty sure the Ids are OK.  All
  I've found is that you can't add a user who has blocked you to a list
  (the member account that I'm trying to add hasn't blocked me as I own
  both accounts).

  Has anyone any ideas?

  Cheers

  Steve

 --
 Abraham Williams | Community Advocate |http://abrah.am
 Project | Out Loud |http://outloud.labs.poseurtech.com
 This email is: [ ] shareable [x] ask first [ ] private.
 Sent from Seattle, WA, United States- Hide quoted text -

 - Show quoted text -


Re: [twitter-dev] Application based on Search API

2010-03-08 Thread John Kalucki
Your application description sounds like resyndication, which is not allowed
under various terms and agreements. You cannot make Twitter data available
via an API unless a very specific set of requirements are adhered to.
Contact a...@twitter.com to start this process.

-John Kalucki
http://twitter.com/jkalucki
Infrastructure, Twitter Inc.


On Mon, Mar 8, 2010 at 3:37 AM, Diz sitov.crist...@gmail.com wrote:

 For the last 3 months I am experimenting the Search API. I'm focusing
 on the geo-location searching capabilities.

 For the beginning I started with my own city, but my intents are to
 extend to the major cities of my country: that will be at most ten. My
 idea of application is to offer real-time activity on each major city,
 through a proxy that caches all tweets and then serves them further to
 clients, filtered or non-filtered. Frequency of requests should be
 between 5 to 10 seconds, and that means I should do between 400 to 800
 requests per hour just for one city, and probably between 4000 to 8000
 requests for the whole application.

 My questions are:
 1). Should I use Search API, or should I move to the Streaming API?!
 2). To whom I should request whitelisting: the usual Search API or the
 Streaming API?!


 Thanks!



Re: [twitter-dev] Problem sending tweets with nbsp chars

2010-03-08 Thread Brian Smith

Hi Taylor,

I tried %20 along with a lot of other things and nbsp; was the only 
thing that worked in all places -- the web, Twitter clients, and SMS 
messages to cell phones. Other than this problem, it has worked great 
for nine months. If Twitter has made changes such that %20 will now 
work where it didn't before, I'd be happy to switch. But, my guess is 
that this bug would apply equally well to the %20s.

First of all, %20 is a regular space, URL-encoded. The non-breaking space is Unicode codepoint 160 
(A0h). You can encode the non-breaking space in XML as #xa0; and in JSON as 
\u00a0.

Secondly, I don't know what you are doing with the non-breaking spaces, but 
they're unlikely to work as you expect them to. My Twitter client replaces all 
runs of consecutive whitespace (including newlines, tabs,non-breaking spaces, 
and more exotic whitespace) as a single normal space, and I suspect that many 
apps will do the same. Attempts to control the layout of a tweet are futile at 
worst and annoying at best. I am sure you are trying to format tweets in a 
helpful, non-annoying, cool way, but your formatting is unlikely to survive 
given the anti-annoying-people countermeasures that good clients have in place.

Cheers,
Brian :)


Re: [twitter-dev] Problem sending tweets with nbsp chars

2010-03-08 Thread Scott Wilcox
Agreed, I strip out too.

On 8 Mar 2010, at 15:57, Brian Smith wrote:

 Hi Taylor,
 
 I tried %20 along with a lot of other things and nbsp; was the only thing 
 that worked in all places -- the web, Twitter clients, and SMS messages to 
 cell phones. Other than this problem, it has worked great for nine months. 
 If Twitter has made changes such that %20 will now work where it didn't 
 before, I'd be happy to switch. But, my guess is that this bug would apply 
 equally well to the %20s.
 First of all, %20 is a regular space, URL-encoded. The non-breaking space is 
 Unicode codepoint 160 (A0h). You can encode the non-breaking space in XML as 
 #xa0; and in JSON as \u00a0.
 
 Secondly, I don't know what you are doing with the non-breaking spaces, but 
 they're unlikely to work as you expect them to. My Twitter client replaces 
 all runs of consecutive whitespace (including newlines, tabs,non-breaking 
 spaces, and more exotic whitespace) as a single normal space, and I suspect 
 that many apps will do the same. Attempts to control the layout of a tweet 
 are futile at worst and annoying at best. I am sure you are trying to format 
 tweets in a helpful, non-annoying, cool way, but your formatting is unlikely 
 to survive given the anti-annoying-people countermeasures that good clients 
 have in place.
 
 Cheers,
 Brian :)



smime.p7s
Description: S/MIME cryptographic signature


Re: [twitter-dev] trends/location: number of occurrence of location element

2010-03-08 Thread Raffi Krikorian
hi yusuke!

the schema allows us the ability to do so in the future -- but, as of today,
no.

On Mon, Mar 8, 2010 at 12:34 AM, Yusuke Yamamoto yus...@mac.com wrote:

 Hi,

 I'm developing Twitter4J and have a quick question.

 Is there any chance that trends/location returns multiple location
 elements?

 Thanks,
 --
 Yusuke Yamamoto
 yus...@mac.com

 this email is: [x] bloggable/tweetable [ ] private
 follow me on : http://twitter.com/yusukeyamamoto
 subscribe me at : http://yusuke.homeip.net/blog/




-- 
Raffi Krikorian
Twitter Platform Team
http://twitter.com/raffi


[twitter-dev] I'm Twitter's API whitelisted but still getting rate limit response

2010-03-08 Thread Jacob
Can Anyone help please,
I've requested to whitelist an account name and two IPs, I got a mail
from Twitter team saying they approved my request 5 days ago, but
still getting rate limit response on one of the IPs.
what should I do?

Jacob



[twitter-dev] Re: xAuth

2010-03-08 Thread Berto
To follow up, this works for me now.  It looks like Twitter's cache
was not showing me as having xAuth access so it appears that
converting to xAuth is as easy as it seems ;).

On Mar 5, 4:22 pm, Taylor Singletary taylorsinglet...@twitter.com
wrote:
 Hi Berto,

 I can confirm that using POST operations over HTTPs will work for XAuth.

 Your URL should only contain:https://api.twitter.com/oauth/access_token

 Your signature base string should contain the x_auth_* parameters.

 Your authorization string should not contain the x_auth_* parameters.

 Here's a replay of a successful request:

 Full Request URI:https://api.twitter.com/oauth/access_token

 Signature Base String:
 POSThttps%3A%2F%2Fapi.twitter.com
 %2Foauth%2Faccess_tokenoauth_consumer_key%3Dri8JxxxdwSV5xIUfNNvQ%26oauth_nonce%3DNI14r4hzKMlslKakhjeOaHoIeWw53ZMeTJb4zAaZh2o%26oauth_signature_method%3DHMAC-SHA1%26oauth_timestamp%3D1267826670%26oauth_version%3D1.0%26x_auth_mode%3Dclient_auth%26x_auth_password%Dxxx%26x_auth_username%3De

 Example response:
 oauth_token=1234-torCkTs0XK7H2Y2i1ee5iofXyzp7aayeEXRTmlwoauth_token_secret=Xyz0gOZHNQKPooBiWCZRY81klwS3kLZGa2wcuser_id=1234screen_name=ex_auth_expires=0

 Keep in mind that your signing secret will not include an
 oauth_token_secret, so will be the equivalent of {consumer_secret}

 Taylor

 On Fri, Mar 5, 2010 at 12:09 PM, Berto mstbe...@gmail.com wrote:
  Three days and I still can't get this to work.  I even tried switching
  over to GET instead of POST and it tells me Failed to validate oauth
  signature and token.  This is fully functional for regular oauth.
  Signature Base String is:

  Signature Base String: Signature Base String:
  GEThttps%3A%2F%2Fapi.twitter.com%2Foauth
  %2Faccess_tokenoauth_consumer_key%3DCONSUMER KEY%26oauth_nonce
  %3D1267819560%26oauth_signature_method%3DHMAC-SHA1%26oauth_timestamp
  %3D1267819217%26oauth_version%3D1.0%26x_auth_mode%3Dclient_auth
  %26x_auth_password%3Dpass%26x_auth_username%3Duser

  I'm sending oauth parameters via the Authorization header and the
  three xAuth parameters as GET parameters (?
  x_auth_username=userx_auth_pass=passx_auth_mode=client_auth).

  It appears as though everyone who had oauth working before had an easy
  transition so I'm just a little curious why mine isn't working when I
  literally have only changed the URL and three parameters.  I've
  verified this is going over SSL as well.

  Any help is appreciated.

  Thanks.
  On Mar 4, 3:34 pm, Anton Krasovsky anton.krasov...@gmail.com wrote:
   In case if anyone's interested (though I doubt there are many
   Erlang'ers on the list),
   I just addedxAuthsupport to twerl.

  http://github.com/ak1394/twerl

   Regards,
   Anton

   On Wed, Mar 3, 2010 at 4:57 PM, Berto mstbe...@gmail.com wrote:
Raffi,

Can you comment on the first part of Marc's last reply?

Thanks!

On Mar 3, 9:24 am, Marc Mims marc.m...@gmail.com wrote:
* Berto mstbe...@gmail.com [100303 06:42]:

 Isn't that using a GET request versus the docs saying POST?  And I
 thought parameters were supposed to be normalized except for
  signature
 which gets attached at the end?

Hmmm. I completely missed the fact that the documentation specifies
POST.  I used GET and it worked.  When I use a POST, I get a 401.

Doc bug?

The order you *send* the parameters doesn't matter---the order of the
base string used for generating the signature does.

The underlying libraries I use assemble the parameters in an arbitrary
order.  Generation of the signature is a separate call and builds it's
own base string from a hash (associative array).

@semifor


[twitter-dev] Whitelisted and calls left, but still says Rate Limit Exceeded?

2010-03-08 Thread Roy Rodenstein
Hi all, I'm new to Twitter API dev, working on a mashup with LinkedIn
to find your LI network on Twitter.

I am only doing a couple of Twitter calls:
- friends/ids, one per app run
- user search w/limit of 5 results, about 20 per app run (tho this
will rise when I'm done testing)

So pretty limited API calls right now. Also, I got my @how2startup
account whitelisted (don't have a static IP yet so can't get IP
whitelisted I assume), and as you can see below from output of
rate_limit_status run from my server, I have plenty of API calls left.
And yet I am still getting Rate Limit Exceeded error.

I must be missing something really obvious, or really tricky. Is there
some other IP (the local browser?) that is getting rate limited?
Thanks for any tips! Best, Roy

IP Address Rate Limit (Unauthenticated call to rate_limit_status)
hash
reset-time type=datetime2010-03-08T09:27:04+00:00/reset-time
remaining-hits type=integer140/remaining-hits
hourly-limit type=integer150/hourly-limit
reset-time-in-seconds type=integer1268040424/reset-time-in-
seconds
/hash

how2startup Address Rate Limit (Authenticated call to
rate_limit_status)
hash
remaining-hits type=integer19713/remaining-hits
reset-time type=datetime2010-03-08T10:05:25+00:00/reset-time
hourly-limit type=integer2/hourly-limit
reset-time-in-seconds type=integer1268042725/reset-time-in-
seconds
/hash



Re: [twitter-dev] Problem sending tweets with nbsp chars

2010-03-08 Thread Roy Leban
To Scott and Brian, please take a look at the tweets on @Puzzazz. There is
no other way to have them appear correctly. What I am doing is not annoying
in the least and it may be futile in your particular Twitter clients but I
tested what I'm doing in many clients and it worked fine.

I'm open to suggestions but you're screwed is not a suggestion. And
changing what I'm doing doesn't address the fact that Twitter has a bug.

For the record \u00a0 did not work. I didn't try #xa0; as I got to nbsp;
first. I would guess it would work but have the same bug with Twitter.

/Roy


On Mon, Mar 8, 2010 at 8:13 AM, Scott Wilcox sc...@tig.gr wrote:

 Agreed, I strip out too.

 On 8 Mar 2010, at 15:57, Brian Smith wrote:

  Hi Taylor,
 
  I tried %20 along with a lot of other things and nbsp; was the only
 thing that worked in all places -- the web, Twitter clients, and SMS
 messages to cell phones. Other than this problem, it has worked great for
 nine months. If Twitter has made changes such that %20 will now work where
 it didn't before, I'd be happy to switch. But, my guess is that this bug
 would apply equally well to the %20s.
  First of all, %20 is a regular space, URL-encoded. The non-breaking space
 is Unicode codepoint 160 (A0h). You can encode the non-breaking space in XML
 as #xa0; and in JSON as \u00a0.
 
  Secondly, I don't know what you are doing with the non-breaking spaces,
 but they're unlikely to work as you expect them to. My Twitter client
 replaces all runs of consecutive whitespace (including newlines,
 tabs,non-breaking spaces, and more exotic whitespace) as a single normal
 space, and I suspect that many apps will do the same. Attempts to control
 the layout of a tweet are futile at worst and annoying at best. I am sure
 you are trying to format tweets in a helpful, non-annoying, cool way, but
 your formatting is unlikely to survive given the anti-annoying-people
 countermeasures that good clients have in place.
 
  Cheers,
  Brian :)




Re: [twitter-dev] Problem sending tweets with nbsp chars

2010-03-08 Thread M. Edward (Ed) Borasky
I looked at the @puzzazz stream. It's interesting, but you've already  
got a mix of links to puzzles and embedded puzzles and clues. So I'm  
not sure what you'd lose by making your tweet stream 100 percent links  
with searchable US-ASCII text, other than getting filtered out by  
algorithms that say, Don't follow people whose tweets are 90 percent  
links or more. ;-) This looks like a fun tweet stream, and I followed  
it. ;-)


And by having every tweet contain a link, you *gain* click tracking,  
using, say, the free bit.ly pro tracking capability! I assume you're  
trying to market something, for some definition of market. Why not  
make it easy for Google and Bing to find you via Twitter, and for  
yourself to monitor, by having your tweets be a mix of natural human  
language and trackable links?

--
M. Edward (Ed) Borasky
borasky-research.net/m-edward-ed-borasky/

A mathematician is a device for turning coffee into theorems. ~ Paul Erdos


Quoting Roy Leban r...@royleban.com:


To Scott and Brian, please take a look at the tweets on @Puzzazz. There is
no other way to have them appear correctly. What I am doing is not annoying
in the least and it may be futile in your particular Twitter clients but I
tested what I'm doing in many clients and it worked fine.

I'm open to suggestions but you're screwed is not a suggestion. And
changing what I'm doing doesn't address the fact that Twitter has a bug.

For the record \u00a0 did not work. I didn't try #xa0; as I got to nbsp;
first. I would guess it would work but have the same bug with Twitter.

/Roy


On Mon, Mar 8, 2010 at 8:13 AM, Scott Wilcox sc...@tig.gr wrote:


Agreed, I strip out too.

On 8 Mar 2010, at 15:57, Brian Smith wrote:

 Hi Taylor,

 I tried %20 along with a lot of other things and nbsp; was the only
thing that worked in all places -- the web, Twitter clients, and SMS
messages to cell phones. Other than this problem, it has worked great for
nine months. If Twitter has made changes such that %20 will now work where
it didn't before, I'd be happy to switch. But, my guess is that this bug
would apply equally well to the %20s.
 First of all, %20 is a regular space, URL-encoded. The non-breaking space
is Unicode codepoint 160 (A0h). You can encode the non-breaking space in XML
as #xa0; and in JSON as \u00a0.

 Secondly, I don't know what you are doing with the non-breaking spaces,
but they're unlikely to work as you expect them to. My Twitter client
replaces all runs of consecutive whitespace (including newlines,
tabs,non-breaking spaces, and more exotic whitespace) as a single normal
space, and I suspect that many apps will do the same. Attempts to control
the layout of a tweet are futile at worst and annoying at best. I am sure
you are trying to format tweets in a helpful, non-annoying, cool way, but
your formatting is unlikely to survive given the anti-annoying-people
countermeasures that good clients have in place.

 Cheers,
 Brian :)








[twitter-dev] Re: Problem sending tweets with nbsp chars

2010-03-08 Thread @epc
On Mar 8, 4:06 pm, Roy Leban r...@royleban.com wrote:
 I'm open to suggestions but you're screwed is not a suggestion. And
 changing what I'm doing doesn't address the fact that Twitter has a bug.

What code are you using to post these tweets? Is it possible that it
is transcoding nbsp; to something before transmitting to twitter?
Because what appears in your older tweets is the (correct) #160; not
nbsp;  AFAIK, twitter does not decode HTML entities.
--
-ed costello


[twitter-dev] Re: Search API rate limit IP address question

2010-03-08 Thread eys
Thank you for your reply!

 If this were true then sometimes your request works and other times it
 doesn't. Is that the case?

Yes, each time I run my app, it makes ~80 calls to the Search API. I
can only run a full test of the app 2 or 3 times before I get the
Stream Error. But if I run a partial test of only 10 or so calls, I
can run it a bunch more times before getting the error. If I wait 30
minutes or so, I can continue testing...but that really affects my
workflow!


 There are multiple requests happening here. I assume the following, which
 may or may not be correct:

 - From your browser you call your app
 - Your app runs some call through the twitter API
 - Twitter servers process the call and send it back to your app
 - Your app returns processed code back to your browser

Yes, this is correct.


 From the above processes your IP address is passed through by the Twitter
 API to the twitter service.

 I'd suggest try running your request from a completely different network and
 see what happens.

I tried running it from a friend's computer. I get the same frequency
of Error, but when he changes his computer's IP address, I'm suddenly
able to run the app again...

How can I shift the load to my webserver's IP (the one that's
whitelisted) rather than each individual computer's IP? Is it possible
with Search API?

Thank you!


[twitter-dev] Are tweet ID-s in search and rest API-s the same?

2010-03-08 Thread Jaanus
http://apiwiki.twitter.com/Twitter-Search-API-Method:-search says:

Warning: The user ids in the Search API are different from those in
the REST API (about the two APIs). This defect is being tracked by
Issue 214. This means that the to_user_id and from_user_id field vary
from the actualy user id on Twitter.com.

How about tweet ID-s? The search API returns tweet ID in the id
field of the response object. Can I trust the search and REST API
tweet ID-s to be the same?


rgds,
Jaanus


Re: [twitter-dev] Are tweet ID-s in search and rest API-s the same?

2010-03-08 Thread Mark McBride
Tweet IDs will be the same.

  ---Mark

http://twitter.com/mccv


On Mon, Mar 8, 2010 at 3:11 PM, Jaanus jaa...@gmail.com wrote:

 http://apiwiki.twitter.com/Twitter-Search-API-Method:-search says:

 Warning: The user ids in the Search API are different from those in
 the REST API (about the two APIs). This defect is being tracked by
 Issue 214. This means that the to_user_id and from_user_id field vary
 from the actualy user id on Twitter.com.

 How about tweet ID-s? The search API returns tweet ID in the id
 field of the response object. Can I trust the search and REST API
 tweet ID-s to be the same?


 rgds,
 Jaanus



[twitter-dev] Changing the Content-Type header for OAuth token exchanges

2010-03-08 Thread Mark McBride
All -

Per issue 1263 (http://code.google.com/p/twitter-api/issues/detail?id=1263)
(and the OAuth spec), we're looking to change the Content-Type header for
OAuth token exchanges to 'application/x-www-form-urlencoded'.  To date it
has been 'text/html'.  We want to ensure that this will not break existing
applications, so if you have any qualms please voice them here.

  ---Mark

http://twitter.com/mccv


[twitter-dev] Tips to avoid hitting rate limits for my movie monitoring application.

2010-03-08 Thread Rahul
Hello,

I am building an application that monitors tweets about movies(for now
with... other interesting things planned). I have my id whitelisted
but I want to avoid overusing it.

The challenge that I face is that ideally I want to make full use of
the opportunity to retrieve 100 tweets per call and for that I need
information on the frequency with which users are tweeting about a
movie and then set my call frequency (to call twitter search api)
accordingly so that I maximize the number of tweets returned per call
or atleast.

Since I presume there is no way to know what frequency is someone
tweeting about a movie - I need help is what is the best way to
optimize for such a situation.

The challenge is complicated by the fact that users tweet about
different movies at different rates and the rates generally decrease
overtime.

I have tried combining searches - but the challenge is that lets say I
search for

(Movie A OR Movie B)
(Movie C OR Movie D)

it could be the case that people tweet about Movie A  B a lot and
litle to none about C or D or there is a combination in which they
continue to tweet about A but not about B - So I still can end up in a
situation where I am not optimizing my calls. Also situations such as
Oscars can dramatically change what people talk about even about
movies out months ago.

I have thought of writing something such as a variable frequency
caller that can check the frequency of tweets for the last 3 calls in
order to appreciate the frequency of tweets for a given search and
then continuously vary the time between calls so that I can get as
close to 100 tweets as possible in a call.

Any ideas suggestions that can suggest ways to alleviate the above
will be highly appreciated.

Thanks
Rahul.



[twitter-dev] Streaming API follow limit

2010-03-08 Thread Lucas Vickers
Hello,

The Streaming API documentation used to state that you could follow
200 or 400 users (I forget).
I just checked the updated documentation and I don't see any mention
of limit.

Does anyone know the limit of users I can follow with a regular and a
whitelist account?
The number is going to affect how I go about designing my program so
it is important to know.

Thanks!
Lucas


Re: [twitter-dev] Tips to avoid hitting rate limits for my movie monitoring application.

2010-03-08 Thread Mark McBride
This sounds like a perfect use case for the streaming API.  The rate limits
there are different, but in general more permissive. And because you're
doing primarily OR queries, the current track functionality seems
sufficient.

  ---Mark

http://twitter.com/mccv


On Mon, Mar 8, 2010 at 3:21 PM, Rahul rsdigh...@googlemail.com wrote:

 Hello,

 I am building an application that monitors tweets about movies(for now
 with... other interesting things planned). I have my id whitelisted
 but I want to avoid overusing it.

 The challenge that I face is that ideally I want to make full use of
 the opportunity to retrieve 100 tweets per call and for that I need
 information on the frequency with which users are tweeting about a
 movie and then set my call frequency (to call twitter search api)
 accordingly so that I maximize the number of tweets returned per call
 or atleast.

 Since I presume there is no way to know what frequency is someone
 tweeting about a movie - I need help is what is the best way to
 optimize for such a situation.

 The challenge is complicated by the fact that users tweet about
 different movies at different rates and the rates generally decrease
 overtime.

 I have tried combining searches - but the challenge is that lets say I
 search for

 (Movie A OR Movie B)
 (Movie C OR Movie D)

 it could be the case that people tweet about Movie A  B a lot and
 litle to none about C or D or there is a combination in which they
 continue to tweet about A but not about B - So I still can end up in a
 situation where I am not optimizing my calls. Also situations such as
 Oscars can dramatically change what people talk about even about
 movies out months ago.

 I have thought of writing something such as a variable frequency
 caller that can check the frequency of tweets for the last 3 calls in
 order to appreciate the frequency of tweets for a given search and
 then continuously vary the time between calls so that I can get as
 close to 100 tweets as possible in a call.

 Any ideas suggestions that can suggest ways to alleviate the above
 will be highly appreciated.

 Thanks
 Rahul.




Re: [twitter-dev] Problem sending tweets with nbsp chars

2010-03-08 Thread Julio Biason
On Mon, Mar 8, 2010 at 5:15 AM, Roy Leban r...@royleban.com wrote:
 RoyLeban:�...@twitterapi What does like that mean? You count each nbsp as 1
 char until you get close to 140 then count them as more? Sounds like a bug

Don't get me wrong, but I'm guessing that your problem is that you are
thinking Twitter counts chars, when it counts bytes, actually. Raffi
posted an URL with the proper way to count characters:
http://apiwiki.twitter.com/Counting-Characters

-- 
Julio Biason julio.bia...@gmail.com
Twitter: http://twitter.com/juliobiason


Re: [twitter-dev] Problem sending tweets with nbsp chars

2010-03-08 Thread Cameron Kaiser
  RoyLeban:_...@twitterapi What does like that mean? You count each nbsp as 
  1
  char until you get close to 140 then count them as more? Sounds like a bug
 
 Don't get me wrong, but I'm guessing that your problem is that you are
 thinking Twitter counts chars, when it counts bytes, actually. Raffi
 posted an URL with the proper way to count characters:
 http://apiwiki.twitter.com/Counting-Characters

No, that document just makes it clear that no matter how a character is
composed, Twitter will always accept 140 characters. Raffi has made it plain
(because I bluntly asked) that 140 character -- not just byte -- tweets are
kosher. Even if they're not in Hebrew.

I think the OP's problem is in handling entities, not UTF-8.

-- 
 personal: http://www.cameronkaiser.com/ --
  Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckai...@floodgap.com
-- Po-Ching Lives! 


[twitter-dev] Trouble connecting to the Streaming API - 404 errors

2010-03-08 Thread John Kalucki
If you suddenly are getting 404 errors from the Streaming API, it's
probably because you haven't updated your URLs. See:

http://groups.google.com/group/twitter-development-talk/browse_frm/thread/44bd32155dbf2c16/f085ffb0e64e0709?lnk=gstq=jkalucki+deprecate#f085ffb0e64e0709

-John Kalucki
http://twitter.com/jkalucki
Infrastructure, Twitter Inc.



Re: [twitter-dev] Tips to avoid hitting rate limits for my movie monitoring application.

2010-03-08 Thread M. Edward (Ed) Borasky
What would make use of Streaming for this use case a *lot* easier  
would be if Twitter would export to the API more detailed information  
about the Trending Topics. For example, I'd like to see more topics  
than just the current number displayed, and tweets per unit time  
(hourly worst case) for each topic. I'd like to see at least the Top  
100 and maybe even the Top 1000! This seems to me to be an easy task -  
you've got to do the computations anyway, right? Heck, with pages /  
cursors, you could send the whole table out and let people do their  
own cutoffs.


For example, over the weekend, the Trending Topics were,  
understandably, dominated by the Oscars. That's ten or twenty right  
there, by the time you factor in the fact that Farah Fawcett got  
ignored in the memorial, ten pictures nominated for best picture, ten  
Best / Best Supporting actresses, ten actors, etc. Throw the perennial  
Justin Bieber and Lady Gaga into the mix and it's clear there's  
interesting and useful information further down the list. Why should  
we have to monitor Streaming and do our own topic analysis and  
filtering, or subscribe to some service with Firehose access?


--
M. Edward (Ed) Borasky
borasky-research.net/m-edward-ed-borasky/

A mathematician is a device for turning coffee into theorems. ~ Paul Erdos


Quoting Mark McBride mmcbr...@twitter.com:


This sounds like a perfect use case for the streaming API.  The rate limits
there are different, but in general more permissive. And because you're
doing primarily OR queries, the current track functionality seems
sufficient.

  ---Mark

http://twitter.com/mccv


On Mon, Mar 8, 2010 at 3:21 PM, Rahul rsdigh...@googlemail.com wrote:


Hello,

I am building an application that monitors tweets about movies(for now
with... other interesting things planned). I have my id whitelisted
but I want to avoid overusing it.

The challenge that I face is that ideally I want to make full use of
the opportunity to retrieve 100 tweets per call and for that I need
information on the frequency with which users are tweeting about a
movie and then set my call frequency (to call twitter search api)
accordingly so that I maximize the number of tweets returned per call
or atleast.

Since I presume there is no way to know what frequency is someone
tweeting about a movie - I need help is what is the best way to
optimize for such a situation.

The challenge is complicated by the fact that users tweet about
different movies at different rates and the rates generally decrease
overtime.

I have tried combining searches - but the challenge is that lets say I
search for

(Movie A OR Movie B)
(Movie C OR Movie D)

it could be the case that people tweet about Movie A  B a lot and
litle to none about C or D or there is a combination in which they
continue to tweet about A but not about B - So I still can end up in a
situation where I am not optimizing my calls. Also situations such as
Oscars can dramatically change what people talk about even about
movies out months ago.

I have thought of writing something such as a variable frequency
caller that can check the frequency of tweets for the last 3 calls in
order to appreciate the frequency of tweets for a given search and
then continuously vary the time between calls so that I can get as
close to 100 tweets as possible in a call.

Any ideas suggestions that can suggest ways to alleviate the above
will be highly appreciated.

Thanks
Rahul.








[twitter-dev] twitter dev meetup at SXSW on Friday

2010-03-08 Thread Mike Champion
I'll be heading to SXSW later this week and wondering who else will be
going this year?

On Friday afternoon oneforty is going to buy the first few rounds at a
beer o'clock meetup. If you're going to SXSW, or in the Austin area,
it'd be great to meetup. Always more fun to talk about rate limiting,
OAuth and future APIs over drinks.

If you can come, please RSVP so we get a head count:

http://tweetvite.com/event/beeroclock

Details: Friday, March 12 from 4-6pm CST at the Marriott Courtyard in
Austin, TX. It's located at 300 East Fourth Street, Austin, TX 78701
(http://bit.ly/cyI7O3). We'll have a section of the Restaurant 
Patio. The hope is to try to keep it to twitter developers (and not
just fans of twitter).

Cheers,

-mike
@graysky


Re: [twitter-dev] Application based on Search API

2010-03-08 Thread Abraham Williams
Is the specific set of requirements published anywhere?

Abraham

On Mon, Mar 8, 2010 at 06:50, John Kalucki j...@twitter.com wrote:

 Your application description sounds like resyndication, which is not
 allowed under various terms and agreements. You cannot make Twitter data
 available via an API unless a very specific set of requirements are adhered
 to. Contact a...@twitter.com to start this process.

 -John Kalucki
 http://twitter.com/jkalucki
 Infrastructure, Twitter Inc.



 On Mon, Mar 8, 2010 at 3:37 AM, Diz sitov.crist...@gmail.com wrote:

 For the last 3 months I am experimenting the Search API. I'm focusing
 on the geo-location searching capabilities.

 For the beginning I started with my own city, but my intents are to
 extend to the major cities of my country: that will be at most ten. My
 idea of application is to offer real-time activity on each major city,
 through a proxy that caches all tweets and then serves them further to
 clients, filtered or non-filtered. Frequency of requests should be
 between 5 to 10 seconds, and that means I should do between 400 to 800
 requests per hour just for one city, and probably between 4000 to 8000
 requests for the whole application.

 My questions are:
 1). Should I use Search API, or should I move to the Streaming API?!
 2). To whom I should request whitelisting: the usual Search API or the
 Streaming API?!


 Thanks!





-- 
Abraham Williams | Community Advocate | http://abrah.am
TwitterOAuth | http://github.com/abraham/twitteroauth
This email is: [ ] shareable [x] ask first [ ] private.


Re: [twitter-dev] Re: A PubSubHubbub hub for Twitter

2010-03-08 Thread Abraham Williams
The specified discussion with DeWitt:
http://groups.google.com/group/twitter-development-talk/browse_thread/thread/d001cb08a80f004/

http://groups.google.com/group/twitter-development-talk/browse_thread/thread/d001cb08a80f004/I
don't think I wan't everybody and their mom cloning the Twitter API at the
rate it changes. StatusNet has always lacked methods and any service that is
not a microblogging platform will have to extend it.

Abraham

On Sun, Mar 7, 2010 at 22:47, Jesse Stay jesses...@gmail.com wrote:

 Raffi, it is not clear the legalities of duplicating the Twitter API in
 other environments.  For instance, if I wanted to run users/show_user on
 Wordpress.com's API and get data in exactly the same format as Twitter
 returns data for that, along with any other method Twitter provides, is that
 legal?  Is Status.net's duplication of the Twitter API legal?  It is not
 clear in the Terms.  It is not open unless Twitter allows this, at least
 according to the Open Web Foundation (if I understand correctly).  I think
 DeWitt Clinton has brought this up before, and IMO, this would be an even
 more ideal situation than Pubsubhubbub support, as we wouldn't have to
 change our code to do this elsewhere.  It would make the Twitter API format
 itself a standard.  Make sense?

 Jesse


 On Sun, Mar 7, 2010 at 8:00 PM, Raffi Krikorian ra...@twitter.com wrote:

 uh - how are we not opening up our API?


 On Sun, Mar 7, 2010 at 6:54 PM, Jesse Stay jesses...@gmail.com wrote:

 Why doesn't Twitter just open up their API and patent and then the
 Twitter API becomes the standard?  We all change less code that way. :-)
  I like all these open standards, but it would be so much easier if we could
 just use the existing APIs as standards that we've already integrated into
 all our code.  I think Twitter's losing out on a huge opportunity here by
 not opening up their API.

 Jesse


 On Tue, Mar 2, 2010 at 8:57 AM, Julien julien.genest...@gmail.comwrote:

 Andrew, it's not so much about making a simpler API, but making it
 standard : having the same API to get content from 6A blogs, Tumblr's
 blogs, media sites, social networks... is much easier than
 implementing one for each service out there.

 After a small day of poll, here are some results :

 Do you currently use the Twitter Streaming API?
 Yes 18  53%
 No  16  47%

 Would you use a Twitter PubSubHubbub hub if it was available?
 Yes 33  97%
 No  1   3%

 Have you already implemented PubSubHubbub?
 Yes 24  71%
 No  10  29%


 Obviously, 34 is _not_ a big enough number that I think we have a
 representative panel of respondant, but we also have big names in
 here, (including some who have access in the firehose), which makes me
 think that PubSubHubbub should be a viable option for Twitter.

 If you read this, please take some take to respond :

 http://bit.ly/hub4twitter

 Thanks all.

 Cheers,

 Julien


 On Mar 1, 9:02 pm, Andrew Badera and...@badera.us wrote:
  But how much simpler does it need to be? The streaming API is dead
  simple. I implemented what seems to be a full client with delete,
  limit and backoff in parts of two working days. Honestly I think it
  took me longer to write a working PubSubHubbub subscriber client than
  it did a Twitter Streaming API client.
 
  It would be nice if the world was full of free data and universal
  standards, but if it ain't broke, and it's already invested in, why
  fix it?
 
  ∞ Andy Badera
  ∞ +1 518-641-1280 Google Voice
  ∞ This email is: [ ] bloggable [x] ask first [ ] private
  ∞ Google me:http://www.google.com/search?q=andrew%20badera
 
 
 
  On Mon, Mar 1, 2010 at 8:44 PM, Julien julien.genest...@gmail.com
 wrote:
   Ed,
 
   On Mar 1, 5:23 pm, M. Edward (Ed) Borasky zzn...@gmail.com
 wrote:
   In light of today's announcement, I'm not sure what the benefits of
 a
   middleman would be.
 
  http://blog.twitter.com/2010/03/enabling-rush-of-innovation.html
 
   Can you clarify
 
   a. How much it would cost me to get Twitter data from you via
   PubSubHubbub vs. getting the feeds directly from Twitter?
   Free, obviously... as with the use of any hub we host!
 
   b. What benefits there are to acquiring Twitter data via
 PubSubHubbub
   over direct access?
   Much simpler to deal with than a specific streaming Twitter API,
   specifically if your app has already implemented the protocol for
   Identica, Buzz, Tumblr, sixapart, posterous, google reader... it's
 all
   about standards.
 
   On Mar 1, 3:08 pm, Julien julien.genest...@gmail.com wrote:
 
Ola!
 
I know this s some kind of recurring topic for this mailing list.
 I
know all the heat around it, but I think that Twitter's new
 strategy
concerning their firehose is a good occasion to push them to
 implement
the PubSubHubbub protocol.
 
Superfeedr makes RSS feeds realtime. We host hubs for several big
publishers, including Tumblr, 

Re: [twitter-dev] Application based on Search API

2010-03-08 Thread John Kalucki
Not at the moment, as we expect that the number of services that this will
apply to is small. We'll be clarifying data access and licensing over the
next few months.

-John Kalucki
http://twitter.com/jkalucki
Infrastructure, Twitter Inc.


On Mon, Mar 8, 2010 at 9:57 PM, Abraham Williams 4bra...@gmail.com wrote:

 Is the specific set of requirements published anywhere?

 Abraham


 On Mon, Mar 8, 2010 at 06:50, John Kalucki j...@twitter.com wrote:

 Your application description sounds like resyndication, which is not
 allowed under various terms and agreements. You cannot make Twitter data
 available via an API unless a very specific set of requirements are adhered
 to. Contact a...@twitter.com to start this process.

 -John Kalucki
 http://twitter.com/jkalucki
 Infrastructure, Twitter Inc.



 On Mon, Mar 8, 2010 at 3:37 AM, Diz sitov.crist...@gmail.com wrote:

 For the last 3 months I am experimenting the Search API. I'm focusing
 on the geo-location searching capabilities.

 For the beginning I started with my own city, but my intents are to
 extend to the major cities of my country: that will be at most ten. My
 idea of application is to offer real-time activity on each major city,
 through a proxy that caches all tweets and then serves them further to
 clients, filtered or non-filtered. Frequency of requests should be
 between 5 to 10 seconds, and that means I should do between 400 to 800
 requests per hour just for one city, and probably between 4000 to 8000
 requests for the whole application.

 My questions are:
 1). Should I use Search API, or should I move to the Streaming API?!
 2). To whom I should request whitelisting: the usual Search API or the
 Streaming API?!


 Thanks!





 --
 Abraham Williams | Community Advocate | http://abrah.am
 TwitterOAuth | http://github.com/abraham/twitteroauth
 This email is: [ ] shareable [x] ask first [ ] private.