Wikiquotes python API
This library is intended to be a python API for wikiquotes (inspired by python-wikiquotes).
>>> import wikiquotes >>> wikiquotes.search("gandi", "english") [u'Mahatma Gandhi', u'Indira Gandhi', u'Rahul Gandhi', u'Rajiv Gandhi', u'Arun Manilal Gandhi', u'Gandhi (film)', u'Anand Gandhi', u'Virchand Gandhi', u'Maneka Gandhi', u'Blindness'] >>> wikiquotes.get_quotes('Hau Pei-tsun', "english") # [u"The slogans of 'countering back the mainland' created by Chiang Kai-shek and 'liberating Taiwan' by Mao Zedong several decades ago should be forgotten because none of them could be put into practice.", # u'When people on both sides of the Strait reach a consensus on their political system, unification will come to fruition naturally.', # u'Taiwanese independence is a dead end.'] >>> wikiquotes.quote_of_the_day("english") # (u'Even after killing ninety nine tigers the Maharaja should beware of the hundredth.', u'Kalki Krishnamurthy') >>> wikiquotes.quote_of_the_day("spanish") # (u'Por San Ferm\xedn, el calor no tiene fin', u'Refr\xe1n espa\xf1ol') >>> wikiquotes.random_quote("Aristotle", "english") # u'For the things we have to learn before we can do, we learn by doing.' >>> wikiquotes.supported_languages() # ['english', 'spanish']
There seems to be two options for retrieving quotes from WikiQuotes using python: To implement it yourself or to use python-wikiquotes. At a first glance, I chose the second option and used that library. However, usage and code inspection over python-wikiquotes lead me to choose the first approach and develop a library.
The main reasons for this decision were that: 1. Quotes retrieved weren’t all the quotes in wikiquotes API (tried with different authors). 2. It doesn’t work for python 2.x 2. The code was too complex for what it was achieving. The choice in that project was to use urllib to retrieve the quotes, and lxml to parse the html.
This project: 1. Adds tests for retrieving all the quotes from several authors (Though this point is difficult to satisfy, because quotes don’t respect a format for all authors). 2. Works for python 2.x and 3.x 3. Uses requests and BeautifulSoup, which abstract great part of the complexity which is present in python-wikiquotes.
Anyway, the correct approach would be to try both and stick with the one that gives you the best results.
Quotes are retrieved by searching first for the correct author. This strives for robustness, because it allows to return a quote whether the input is the correct name of the author or not. At the same time, note that subsequent calls to WikiQuotes api have to be made to grab suggestions (see here).
Example: “shakspare” -> “shakespeare” -> [‘William Shakespeare’, ‘Last words in Shakespeare’, ‘Shakespeare in Love’, …] -> get_quotes = 4 calls.
While in python 3.x str type = unicode, in python 2.x str type != unicode. Therefore (and to be consistent), all string output are unicode strings, independent of python’s version. If you call any function from the API that have non-english characters, you will see some weird characters.
>>> wikiquotes.random_quote("borges", "español") # u'\xabTodos caminamos hacia el anonimato, solo que los mediocres llegan un poco antes\xbb.'
This is not incorrect, it is the underlying representation of the format of the string. You could encode the string in utf-8 and print it (or just print it and your python interpreter should convert it automatically).
>>> print(u'\xabTodos caminamos hacia el anonimato, solo que los mediocres llegan un poco antes\xbb.'.encode('utf8')) # «Todos caminamos hacia el anonimato, solo que los mediocres llegan un poco antes».
The approach for testing changed: at a first glance, testing was done by manually adding the code to test each author. After that, I realized that the structure was the same for every author: We need the name, the language and the quotes. Using some black magic for parametrizing tests, I could extract all the logic to code and have a text file for each author. (See author_test for more info.)
The way of testing right now is to add a txt file of the author to test’s authors. For example, here is the test for Dijkstra quotes in english. Adding a new author is a txt file for the author (the name is irrelevant, but should be the author name) and respecting the following format. 1. First line: Author’s name (or the suffix of the wikiquotes page, because sometimes wikipedia has ambiguate redirections if author name is used ). 2. Second line: language. 3. Third line: empty. 4. Following lines should contain one quote per line.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size||File type||Python version||Upload date||Hashes|
|Filename, size wikiquotes-1.4.tar.gz (15.3 kB)||File type Source||Python version None||Upload date||Hashes View hashes|