Microsoft announced that the API will be deprecated on 🔴Dec 15, 2016. They'll stop accepting new initialization after 🔴Dec 31st, 2016. Existing subscritions will be cancelled 🔴3/31/2017. You must migrate to Microsoft Cognitive Services to continue operation (see below).
#####Note: The project continuation / Microsoft Cognititive Services compliant API is here: py-ms-cognitive
Extremely thin python wrapper for Microsoft Azure Bing Search API. Please note that this module does not use the Bing Search API 2.0 AppIDs which will be deprecated on August 1, 2012. This module requires that you sign up to the Windows Azure Marketplace and apply for an application key.
The modules uses OAuth, so you'll need to get your key here (free for up to 5K/Mon):
#####for python 2.7.*
pip install py-bing-search#####for python 3.*
pip3 install py-bing-search*Requires the requests library.
Remember to set the API_KEY as your own.
####For Web Results:
>>> from py_bing_search import PyBingWebSearch
>>> search_term = "Python Software Foundation"
>>> bing_web = PyBingWebSearch('Your-Api-Key-Here', search_term, web_only=False) # web_only is optional, but should be true to use your web only quota instead of your all purpose quota
>>> first_fifty_result= bing_web.search(limit=50, format='json') #1-50
>>> second_fifty_result= bing_web.search(limit=50, format='json') #51-100
>>> print (second_fifty_result[0].description)
u'Python Software Foundation Home Page. The mission of the Python Software Foundation is to promote, protect, and advance the Python programming language, and to ...'What you get is a list of WebResult() instances, each comes with the following values:
self.title: title of the result
self.url: the url of the result
self.description: description for the result
self.id: bing id for the page
#Meta info:
self.meta.uri: the search uri for bing
self.meta.type: for the most part WebResult
####For Image Results:
>>> from py_bing_search import PyBingImageSearch
>>> bing_image = PyBingImageSearch('Your-Api-Key-Here', "x-box console", image_filters='Size:medium+Color:Monochrome') #image_filters is optional
>>> first_fifty_result= bing_image.search(limit=50, format='json') #1-50
>>> second_fifty_result= bing_image.search(limit=50, format='json') #51-100
>>> print (second_fifty_result[0].media_url)
...What you get is a list of ImageResult() instances, each comes with the following values:
self.id: id of the result
self.title: title of the resulting image
self.media_url: url to the full size image
self.source_url: url of the website that contains the source image
self.width: width of the image
self.height: height of the image
self.file_size: size of the image (in bytes) if available
self.content_type the MIME type of the image if available
self.meta: meta info
#Meta info#:
meta.uri: the search uri for bing
meta.type: for the most part ImageResult
####For Video Results:
>>> from py_bing_search import PyBingVideoSearch
>>> bing_video = PyBingVideoSearch('Your-Api-Key-Here', "cats")
>>> first_fifty_result= bing_video.search(limit=50, format='json') #1-50
>>> second_fifty_result= bing_video.search(limit=50, format='json') #51-100
>>> print (second_fifty_result[0].media_url)
...What you get is a list of VideoResult() instances, each comes with the following values:
self.id: id of the result
self.title: title of the resulting Video
self.media_url: url to the full size Video
self.display_url: url to display on the search result.
self.thumbnail: url of the thumbnail image for the video.
self.run_time: run time of the video
self.meta: meta info
#Meta info#:
meta.uri: the search uri for bing
meta.type: for the most part VideoResult
####For News Results:
>>> from py_bing_search import PyBingNewsSearch
>>> bing_news = PyBingNewsSearch('Your-Api-Key-Here', "US Election")
>>> first_fifty_result= bing_news.search(limit=50, format='json') #1-50
>>> second_fifty_result= bing_news.search(limit=50, format='json') #51-100
>>> print (second_fifty_result[0].url)
...What you get is a list of NewsResult() instances, each comes with the following values:
self.id: id of the result
self.title: title of the resulting News
self.url: url to the News
self.description: description of the article
self.date: date of the News
self.meta: meta info
#Meta info#:
meta.uri: the search uri for bing
meta.type: for the most part NewsResult
You secan also run search_all to keep searching until it fills your required quota. Note that this will make an unpredictable number of api calls (hence drains your credits).
>>> from py_bing_search import PyBingWebSearch
>>> bing_web = PyBingNewsSearch('Your-Api-Key-Here', "Python Software Foundation")
>>> result_list = bing_web.search_all(limit=130, format='json') #will return result 1 to 130
>>> len(result_list) == 130
True
>>> result_list = bing_web.search_all(limit=130, format='json') #will return result 131 to 260search_all() is available in all PyBing*Search classes.
If you need more control via custom parameters (and you know what you're doing), you can include a custom_param parameter as follows:
>>> from py_bing_search import PyBingWebSearch
web_bing = PyBingWebSearch('Your-Api-Key-Her', "some bad search", custom_params="&Adult='Strict'")
result_list = bing_web.search_all(limit=100, format='json')
# result_list should now contain 100 results that are filetered by safesearch.Don't forget to include the *&* at the start of your custom_param or the search will return the error msg: "py_bing_search.py_bing_search.PyBingWebException: Request returned with code 400, error msg: The provided format is not supported."
The optional parameter custom_param is available in all PyBing*Search constructors.
Some have asked me about how to restrict the location/language of the search. All of this can be done from the search term rather than at the API level. This means if you want to look up the term London, but from region of Spain, and also written in Japanese, you'd simply search for:
London language:ja loc:esThere are many other things you can do, so before you go messing with the API, flex your Bing advanced syntax knowledge. Language reference is here.
You can run tests via:
python setup.py testYou'll need to create a config file named secrets.cfg in the tests/ directory that looks like the following:
[secret]
secret: your-secret-key-hereThe tests will use your key to run some searches and verify that results are being returned.
#####Note: The project continuation / Microsoft Cognititive Services compliant API is here: py-ms-cognitive