I've been using curl to access search.twitter.com and recently I've
noticed that occasionally it is returning invalid tweets.  I'm
searching for a query-term that shows up very infrequently in the
tweet stream, and am using the curl command:

curl "http://search.twitter.com/search.atom?q=query-

What I've found is that about 99% of the time I get back an empty
response, just the <feed> tag wrapped around the meta data for the
query but no results.  But if I just run the exact same curl command
over and over again, occasionally I'll get a response with a full
payload of 99 tweets, none of which have anything to do with my search
terms.  Subsequent executions of the same curl command return the
response w/o any tweets in it.

Its a little hard to explain, but I have a log file that shows:

curl - no tweets
curl - no tweets
curl - no tweets
curl - 99 unrelated tweets
curl - no tweets

All running the same command via cut-and-paste and all within the
space of a few seconds.  More often than not, it seems like the
invalid tweets are all somehow related, as if I had gotten the
response for some other active query.

Anyone else seeing anything this problem?  The log file is 16K
compressed, let me know if you'd like me to send it in.


Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 

Reply via email to