search and filter from a stream of twitter tweets
Start the server as usual in django then run the command
celery -A twitter_streaming worker --loglevel=info in your terminal to run and see logs of celery backgroung task.
To know more read here
api/search/?search_key=<your_search_key>
this gives the tweets according to 'your_search_key' in pages with maximum of 20 tweets each.
Example:
when running in local server http://localhost:8000/api/search/?search_key=good the json response would be
Json Response: { "next": "http://localhost:8000/api/search/?cursor=cD0yMDE4LTAyLTI3KzE3JTNBNTclM0EwMyUyQjAwJTNBMDA=&search_key=good", "previous": null, "results": [ { "id": 8252, "text": "RT @Mr_Hardey: When the president and his fans are talking about how good he has performed and how aggressive he's treating the crisis of b…", "created_at": "2018-02-27T23:37:22 05:30", "user_name": "HEADmaster", "user_favorite_count": 1997, "user_friends_count": 5340, "user_followers_count": 11261, "screen_name": "Mr_Hardey", "language": "en", "quote_count": 0, "reply_count": 0, "retweet_count": 0, "favorite_count": 0, "hash_tags": "[]", "user_mentions": "[\"HEADmaster\"]" },..........]
In the response previous and next url contains link for next amd previous data and results contains tweets information. Response is same for all APIs.
api/filter/?user_ew=<user_name_ends_with_key>&min_user_fav_count=<min_number_of_favourites_user_must_have>
All the filters available are listed below:
keys filter information
'text_search_key' searches for specified key in text of tweet
'lan' returns tweets with specific language
'day' returns tweets on that specific date.Format should be yyyy-mm-dd(year-monthday)
'max_day' returns tweets until that specified date
'min_day' returns tweets after that specified date
'user_ew' returns tweets with username ending with specified string
'user_sw' returns tweets with username starting with specified string
'srn_name_ew' returns tweets with screenname ending with specified string
'srn_name_sw' returns tweets with screenname starting with specified string
'hash_tag' returns tweets which contains specified hashtag
'user_men' returns tweets which contains mentions of specified username
'max_fav_tweet_count' return tweets with less favorite count than this
'min_fav_tweet_count' return tweets with more favorite count than this
'max_reply_tweet_count' return tweets with number of replies less than this
'min_reply_tweet_count' return tweets with number of replies more than this
'max_retweet_count' return tweets with number of retweets less than this
'min_retweet_count' return tweets with number of retweets more than this
'max_qot_count' return tweets with number of quotes less than this
'min_qot_count' return tweets with number of quotes more than this
'max_user_fav_count' return tweets with user favorites less than this
'min_user_fav_count' return tweets with user favorites more than this
'max_user_fol_count' return tweets with user followers less than this
'min_user_fol_count' return tweets with user followers more than this
'max_user_frnd_count' return tweets with user friends less than this
'min_user_frnd_count' return tweets with user friends more than this
Example:
http://localhost:8000/api/filter/?lan=en&min_retweet_count=500&user_sw='joh'
the above request returns tweets in english with username starting with 'joh' and with retweets greater than 500.
for language code see here.
To get the whole filtered data into csv just add a csv parameter.
Example
http://localhost:8000/api/filter/?lan=en&min_retweet_count=500&user_sw='joh'&csv= for filter API.
http://localhost:8000/api/search/?search_key=good&csv= for Search API