[|Author|] |PyPi| [|Python 3.7|] |license|
Python Twitter data collector built on
Tweepy <https://github.com/tweepy/tweepy>
__ that allow users to
dynamically collect accounts and statuses from Twitter during streaming,
and automatically generate Datasets from collected data that you can as
CSV.
This library provides a framework that you can use to build your own
data collector, specifying which are your features that have to be
extracted from Twitter accounts/statuses.
Creating your Twitter dataset: 1. Instantiate an AccountCollector
and/or StatusCollector
class in according to what you want collect,
accounts, statuses or both. At this step you can re-defined your own
features that have to be extracted from twitter data, you have to pass
dict-like parameters in the following form: <feature_name, function>
where the function has to be applied to the user or status object.
Please refer to
documentation <https://developer.twitter.com/en/docs/tweets/data-dictionary/overview/user-object.html>
__
for more details about Twitter objects 2. Instantiate the
OnlineStreamer
passing the collector as parameter 3. Start streaming
on some topics 4. Save the created dataset at specified location
NB: It is not mandatory to use both collectors and streamer but you can
also use Collectors alone, for instance if you already have the users
and/or statuses to collect and you don’t need to stream anything.
NEW FEATURES: * Offline collection by name, allow user to make a
query by name and collect some name-similar users extracting features
defined in the collector constructor
INSTALLATION
The package is available on PyPi
here <https://pypi.org/project/ptdc/>
__
.. code:: bash
$ pip install ptdc
EXAMPLE USAGE
Import modules
::
from ptdc import authenticate, AccountCollector, OnlineStreamer, StatusCollector
Define tokens
~~~~~~~~~~~~~
Replace the following tokens with yours, see Twitter developers
`authentication <https://developer.twitter.com/en/docs/basics/authentication/guides/access-tokens.html>`__
for more details about how obtain them.
::
consumer_key = "xxxxxxxxxxx"
consumer_key_secret = "xxxxxxxxxxxxx"
access_token = "xxxxxxxxxxxxxxxxxxxxxx"
access_token_secret = "xxxxxxxxxxxxxxxxxx"
Create the default Tweepy API object of tweepy
::
api = authenticate(consumer_key=consumer_key, consumer_key_secret=consumer_key_secret, access_token=access_token, access_token_secret=access_token_secret)
Create your own Collectors for collecting data
Create your own StatusCollector object
::
s_collector = StatusCollector(api=api)
Create your own AccountCollector object, which will collect also
statuses
::
collector = AccountCollector(api=api, statuses_collector=s_collector)
Create the Streamer
~~~~~~~~~~~~~~~~~~~
Create Online Streamer that will collect data (in this case will collect
only 5 accounts)
::
streamer = OnlineStreamer(api=api, collector=collector, data_limit=5, n_statuses=400)
Start streaming
~~~~~~~~~~~~~~~
You can start streaming in all ways defined by Tweepy, see its doc for
more details
::
streamer.stream(track=['famous', 'web', 'vip', 'holiday', 'pic', 'photo'], is_async=False)
Save dataset/s
~~~~~~~~~~~~~~
After streaming ended (in according to your defined limits), save
DataFrame/s generated into csv file/s. You just need to access the
collector object and call the save_dataset method providing the path.
::
streamer.collector.save_dataset(path="../data/accounts.csv")
Questions and Contributing
--------------------------
Feel free to post questions and problems on the issue tracker. Pull
requests are welcome!
Feel free to fork and modify or add new features and functionality to
the library
.. |Author| image:: http://img.shields.io/badge/author-lampajr-blue.svg?style=flat-square
.. |PyPi| image:: https://img.shields.io/pypi/v/ptdc.svg?style=flat-square
:target: https://pypi.org/project/ptdc/
.. |Python 3.7| image:: https://img.shields.io/pypi/pyversions/ptdc.svg
.. |license| image:: https://img.shields.io/github/license/mashape/apistatus.svg?style=flat-square
:target: https://github.com/lampajr/PTDC/blob/master/LICENSE