Wikipedia API使用指南

Wikipedia API

https://pypi.org/project/Wikipedia-API/

https://github.com/martin-majlis/Wikipedia-API/

安装

pip install Wikipedia-API

Wikipedia-API是

基本使用

import wikipediaapi

title = "china"
wiki = wikipediaapi.Wikipedia(
    language='en',
    extract_format=wikipediaapi.ExtractFormat.WIKI
)
page = wiki.page(title)
language = "zh"
lpage = page.langlinks[language]  # fr es ...
print(lpage.text)

Installation

This package requires at least Python 3.4 to install because it’s using IntEnum.

pip install wikipedia-api

Usage

Goal of Wikipedia-API is to provide simple and easy to use API for retrieving informations from Wikipedia. Bellow are examples of common use cases.

Importing

import wikipediaapi

How To Get Single Page

Getting single page is straightforward. You have to initialize Wikipedia object and ask for page by its name. It’s parameter language has be one of supported languages.

import wikipediaapi
    wiki_wiki = wikipediaapi.Wikipedia('en')

    page_py = wiki_wiki.page('Python_(programming_language)')

How To Check If Wiki Page Exists

For checking, whether page exists, you can use function exists.

page_py = wiki_wiki.page('Python_(programming_language)')
print("Page - Exists: %s" % page_py.exists())
# Page - Exists: True

page_missing = wiki_wiki.page('NonExistingPageWithStrangeName')
print("Page - Exists: %s" %     page_missing.exists())
# Page - Exists: False

How To Get Page Summary

Class WikipediaPage has property summary, which returns description of Wiki page.

import wikipediaapi
    wiki_wiki = wikipediaapi.Wikipedia('en')

    print("Page - Title: %s" % page_py.title)
    # Page - Title: Python (programming language)

    print("Page - Summary: %s" % page_py.summary[0:60])
    # Page - Summary: Python is a widely used high-level programming language for

How To Get Page URL

WikipediaPage has two properties with URL of the page. It is fullurl and canonicalurl.

print(page_py.fullurl)
# https://en.wikipedia.org/wiki/Python_(programming_language)

print(page_py.canonicalurl)
# https://en.wikipedia.org/wiki/Python_(programming_language)

How To Get Full Text

To get full text of Wikipedia page you should use property text which constructs text of the page as concatanation of summary and sections with their titles and texts.

wiki_wiki = wikipediaapi.Wikipedia(
        language='en',
        extract_format=wikipediaapi.ExtractFormat.WIKI
)

p_wiki = wiki_wiki.page("Test 1")
print(p_wiki.text)
# Summary
# Section 1
# Text of section 1
# Section 1.1
# Text of section 1.1
# ...


wiki_html = wikipediaapi.Wikipedia(
        language='en',
        extract_format=wikipediaapi.ExtractFormat.HTML
)
p_html = wiki_html.page("Test 1")
print(p_html.text)
# <p>Summary</p>
# <h2>Section 1</h2>
# <p>Text of section 1</p>
# <h3>Section 1.1</h3>
# <p>Text of section 1.1</p>
# ...

How To Get Page Sections

To get all top level sections of page, you have to use property sections. It returns list of WikipediaPageSection, so you have to use recursion to get all subsections.

def print_sections(sections, level=0):
        for s in sections:
                print("%s: %s - %s" % ("*" * (level + 1), s.title, s.text[0:40]))
                print_sections(s.sections, level + 1)


print_sections(page_py.sections)
# *: History - Python was conceived in the late 1980s,
# *: Features and philosophy - Python is a multi-paradigm programming l
# *: Syntax and semantics - Python is meant to be an easily readable
# **: Indentation - Python uses whitespace indentation, rath
# **: Statements and control flow - Python's statements include (among other
# **: Expressions - Some Python expressions are similar to l

How To Get Page In Other Languages

If you want to get other translations of given page, you should use property langlinks. It is map, where key is language code and value is WikipediaPage.

def print_langlinks(page):
        langlinks = page.langlinks
        for k in sorted(langlinks.keys()):
            v = langlinks[k]
            print("%s: %s - %s: %s" % (k, v.language, v.title, v.fullurl))

print_langlinks(page_py)
# af: af - Python (programmeertaal): https://af.wikipedia.org/wiki/Python_(programmeertaal)
# als: als - Python (Programmiersprache): https://als.wikipedia.org/wiki/Python_(Programmiersprache)
# an: an - Python: https://an.wikipedia.org/wiki/Python
# ar: ar - بايثون: https://ar.wikipedia.org/wiki/%D8%A8%D8%A7%D9%8A%D8%AB%D9%88%D9%86
# as: as - পাইথন: https://as.wikipedia.org/wiki/%E0%A6%AA%E0%A6%BE%E0%A6%87%E0%A6%A5%E0%A6%A8

page_py_cs = page_py.langlinks['cs']
print("Page - Summary: %s" % page_py_cs.summary[0:60])
# Page - Summary: Python (anglická výslovnost [ˈpaiθtən]) je vysokoúrovňový sk

How To Get Links To Other Pages

If you want to get all links to other wiki pages from given page, you need to use property links. It’s map, where key is page title and value is WikipediaPage.

def print_links(page):
        links = page.links
        for title in sorted(links.keys()):
            print("%s: %s" % (title, links[title]))

print_links(page_py)
# 3ds Max: 3ds Max (id: ??, ns: 0)
# ?:: ?: (id: ??, ns: 0)
# ABC (programming language): ABC (programming language) (id: ??, ns: 0)
# ALGOL 68: ALGOL 68 (id: ??, ns: 0)
# Abaqus: Abaqus (id: ??, ns: 0)
# ...

How To Get Page Categories

If you want to get all categories under which page belongs, you should use property categories. It’s map, where key is category title and value is WikipediaPage.

def print_categories(page):
        categories = page.categories
        for title in sorted(categories.keys()):
            print("%s: %s" % (title, categories[title]))


print("Categories")
print_categories(page_py)
# Category:All articles containing potentially dated statements: ...
# Category:All articles with unsourced statements: ...
# Category:Articles containing potentially dated statements from August 2016: ...
# Category:Articles containing potentially dated statements from March 2017: ...
# Category:Articles containing potentially dated statements from September 2017: ...

How To Get All Pages From Category

To get all pages from given category, you should use property categorymembers. It returns all members of given category. You have to implement recursion and deduplication by yourself.

def print_categorymembers(categorymembers, level=0, max_level=1):
        for c in categorymembers.values():
            print("%s: %s (ns: %d)" % ("*" * (level + 1), c.title, c.ns))
            if c.ns == wikipediaapi.Namespace.CATEGORY and level < max_level:
                print_categorymembers(c.categorymembers, level=level + 1, max_level=max_level)


cat = wiki_wiki.page("Category:Physics")
print("Category members: Category:Physics")
print_categorymembers(cat.categorymembers)

# Category members: Category:Physics
# * Statistical mechanics (ns: 0)
# * Category:Physical quantities (ns: 14)
# ** Refractive index (ns: 0)
# ** Vapor quality (ns: 0)
# ** Electric susceptibility (ns: 0)
# ** Specific weight (ns: 0)
# ** Category:Viscosity (ns: 14)
# *** Brookfield Engineering (ns: 0)

How To See Underlying API Call

If you have problems with retrieving data you can get URL of undrerlying API call. This will help you determine if the problem is in the library or somewhere else.

import wikipediaapi
import sys
wikipediaapi.log.setLevel(level=wikipediaapi.logging.DEBUG)

# Set handler if you use Python in interactive mode
out_hdlr = wikipediaapi.logging.StreamHandler(sys.stderr)
out_hdlr.setFormatter(wikipediaapi.logging.Formatter('%(asctime)s %(message)s'))
out_hdlr.setLevel(wikipediaapi.logging.DEBUG)
wikipediaapi.log.addHandler(out_hdlr)

wiki = wikipediaapi.Wikipedia(language='en')

page_ostrava = wiki.page('Ostrava')
print(page_ostrava.summary)
# logger prints out: Request URL: http://en.wikipedia.org/w/api.php?action=query&prop=extracts&titles=Ostrava&explaintext=1&exsectionformat=wiki