Many people have contacted us about the limitation of the /notebook/tweets API because of the requirement to use start / end timestamps to extract portions of an archive and the 10,000 tweet limit per call – forcing people to write some creative hacks to ensure they don’t miss anything.
We have just released a Version 2 of the /notebook/tweets API that now allows for easy pagination through the archive as well as opens the API to accept many other parameters to further slice / dice the data.
We have also worked to ensure parameters align with the Twitter search API where possible.
Try it out and let us know if you have any issues! (And if you need an API Key just let us know!)
type (hashtag, keyword, person, collection, person-collection) [required]
lang [optional – ISO-639-1 2 letter code included in tweet metadata]
max_id [optional – maximum twitter id]
since_id [optional – minimum twitter id]
since [optional – start date – format = YY-MM-DD]
until [optional – end date – format = YY-MM-DD]
order_by [optional – a = ascending, d = descending (default)]
nort [optional set to 1 to remove all tweets starting with RT, default = 0]
text [optional – tweet text to search for]
from_user [optional – twitter username of sender]
latitude | longitude | radius [optional – must include each parameters individually – radius in km]
rpp = results per page [optional – but default set to 10, max allowed = 1000]
page [optional – default = 1]
example w/ Version 2 – page 1 of tweets (results per page = 10) within #jisc hashtag from 2010-01-01 to 201 0-06-01 within a 1000 km radius of lat = 51, long = 0 with language = en
(This enhancement aligns with enhancement API-2 and API-4 outlined in the post Plans for Updates to Twapper Keeper Functionality and APIs.)