DuckDuckGo search engine API library for Python
|package| |travis-ci| |coveralls|
DuckDuckPy <https://github.com/ivankliuk/duckduckpy>
_ is a Python
library for querying DuckDuckGo API <https://api.duckduckgo.com/api>
_ and
render results either to Python dictionary or namedtuple.
You can install DuckDuckPy either via the Python Package Index (PyPI) <http://pypi.python.org/pypi>
_ or
from source.
To install using pip
:
.. code:: bash
$ pip install duckduckpy
To install using easy_install
:
.. code:: bash
$ easy_install duckduckpy
To install from sources you can download the latest version of DuckDuckPy
either from PyPI <http://pypi.python.org/pypi/duckduckpy/0.2>
_ or
GitHub <https://github.com/ivankliuk/duckduckpy/tarball/0.2>
_, extract archive contents and
run following command from the source directory:
.. code:: bash
$ python setup.py install
Latest upstream version can be installed directly from the git repository:
.. code:: bash
$ pip install git+https://github.com/ivankliuk/duckduckpy.git
.. code-block:: python
query(query_string, secure=False, container=u'namedtuple', verbose=False,
user_agent=u'duckduckpy 0.2', no_redirect=False, no_html=False,
skip_disambig=False)
Generates and sends a query to DuckDuckGo API.
Arguments:
+---------------+-------------------------------------------------------------+ | query_string | Query to be passed to DuckDuckGo API. | +---------------+-------------------------------------------------------------+ | secure | Use secure SSL/TLS connection. Default - False. | | | Syntactic sugar is secure_query function which is passed | | | the same parameters. | +---------------+-------------------------------------------------------------+ | container | Indicates how dict-like objects are serialized. There are | | | two possible options: namedtuple and dict. If 'namedtuple' | | | is passed the objects will be serialized to namedtuple | | | instance of certain class. If 'dict' is passed the objects | | | won't be deserialized. Default value: 'namedtuple'. | +---------------+-------------------------------------------------------------+ | verbose | Don't raise any exception if error occurs. | | | Default value: False. | +---------------+-------------------------------------------------------------+ | user_agent | User-Agent header of HTTP requests to DuckDuckGo API. | | | Default value: 'duckduckpy 0.2' | +---------------+-------------------------------------------------------------+ | no_redirect | Skip HTTP redirects (for !bang commands). | | | Default value: False. | +---------------+-------------------------------------------------------------+ | no_html | Remove HTML from text, e.g. bold and italics. | | | Default value: False. | +---------------+-------------------------------------------------------------+ | skip_disambig | Skip disambiguation (D) Type. Default value: False. | +---------------+-------------------------------------------------------------+ | lang | Override "us-en" language & region. Default - None. | +---------------+-------------------------------------------------------------+
Raises:
+--------------------------+--------------------------------------------------+ | DuckDuckDeserializeError | JSON serialization failed. | +--------------------------+--------------------------------------------------+ | DuckDuckConnectionError | Something went wrong with httplib operation. | +--------------------------+--------------------------------------------------+ | DuckDuckArgumentError | Passed argument is wrong. | +--------------------------+--------------------------------------------------+
Returns:
Container depends on container parameter. Each field in the response is converted to the so-called snake case.
.. code-block:: python
>>> from duckduckpy import query
>>> response = query('Python') # namedtuple is used as a container
>>> response
Response(redirect=u'', definition=u'', image_width=0, ...}
>>> type(response)
<class 'duckduckpy.api.Response'>
>>> response.related_topics[0]
Result(first_url=u'https://duckduckgo.com/Python', text=...)
>>> type(response.related_topics[0])
<class 'duckduckpy.api.Result'>
>>> response = query('Python', container='dict') # dict as the container
>>> type(response)
<type 'dict'>
>>> response
{u'abstract': u'', u'results': [], u'image_is_logo': 0, ...}
>>> type(response['related_topics'][0])
<type 'dict'>
>>> response['related_topics'][0]
{u'first_url': u'https://duckduckgo.com/Python', u'text': ...}
.. |package| image:: https://badge.fury.io/py/duckduckpy.svg :target: http://badge.fury.io/py/duckduckpy :alt: PyPI package .. |travis-ci| image:: https://travis-ci.org/ivankliuk/duckduckpy.svg?branch=master :target: https://travis-ci.org/ivankliuk/duckduckpy :alt: CI Status .. |coveralls| image:: https://coveralls.io/repos/ivankliuk/duckduckpy/badge.svg?branch=master :target: https://coveralls.io/r/ivankliuk/duckduckpy?branch=master :alt: Coverage