pombredanne / pygnip-allapis

Python wrapper for the GNIP API's with optional Django integration.

pygnip-allapis - Another Python wrapper for GNIP

A Python client wrapper for the GNIP API's with optional Django integration.

This project aims to wrap all the API's with a single configurable client so you can seemlessly work with any stream/activity/api.

Please feel free to pull request further api integration as I have only added the API's that I've personally used so far.

Currently implemented:

• Usage API:
  • Query usage
• Search API:
  • Query, Set and Delete Powertrack rules
  • Query the search API (search and counts)

###Configuration

This package contains a sample config file (_gnip_config.ini).

pygnip-allapis needs one of the following configuration options:

• A gnip_config.ini file in the pygnip-allapis directory
• A gnip_config.ini file in ~/.config
• A correctly configured dictionary in Django settings.py (see below)

###Django integration

PYGNIP_CONFIG = {
    'DEFAULT': {
        'account': account,
        'username': username,
        'password': password,
        'data_source': '',
        'stream_label': '',
    },
}

###Usage

pip install pygnip-allapis

from pygnip_allapis import gnip_client
client = gnip_client.AllAPIs(config=DEFAULT)

####Query the usage API

client.api('UsageAPI').get_usage(bucket='',
                                 fromDate='',
                                 toDate='')

####Query the Search Count API

client.api('SearchAPICount').search_count(query='',
                                          publisher='twitter',
                                          fromDate='',
                                          toDate='',
                                          bucket=''))

####Query the Search API

client.api('SearchAPI').max_results_only(query='',
                                         publisher='twitter',
                                         fromDate='',
                                         toDate='',
                                         maxResults='10')

For queries which require pagination (ie repeated calls to GNIP with 'next'), you can query all_results() which builds a python list in memory and returns. Alternatively you can use the generator from iterate_all_results() which will keep getting pages until your request is completed.

results = client.api('SearchAPI').iterate_all_results(query='',
                                                   publisher='twitter',
                                                   fromDate='',
                                                   toDate='',
                                                   maxResults=500
                                                  )
for r in results:
    # do something

####Post a Powertrack rule

client.api('PowertrackAPIRules').post_rule(value='test_rule', tag='test_rules')

####Query Powertrack rules

client.api('PowertrackAPIRu les').get_rules()

####Delete a Powertrack rule

self.client.api('PowertrackAPIRules').delete_rule(value='test_rule', tag='test_rules'))

###Handling Responses

pygnip uses the Requests library and by default decodes all json responses using the .json() method in Requests. You can change this default behaviour by passing in 'decode_json=False' as the first argument to any API call which returns JSON.

###Tests

To run the test suite

python tests