wikiscraper-py 0.0.3

A Python library to scrape Wikipedia articles easily

WikiScraper

WikiScraper is a professional Python library to scrape Wikipedia articles easily. It allows you to scrape a single page or all linked articles recursively, supporting both .txt and .csv outputs.

Features

• Scrape a single Wikipedia page or all linked articles recursively.
• Supports .txt and .csv output formats.
• Optionally add titles to scraped content.
• Logging options for file saves and all actions.
• Append all scraped articles into a single file or save separately.
• Handles multiple languages and errors gracefully.
• Polite crawling with configurable delay between requests.

Installation

pip install wikiscraper-py

Usage

Scrape a single page

from wikiscraper import WikiScraper

scraper = WikiScraper(file_type="txt", add_title=True)
scraper.scrape_one("https://en.wikipedia.org/wiki/Python")

Scrape all linked articles

from wikiscraper import WikiScraper

scraper = WikiScraper(file_type="txt", add_title=True, all_on_one_file=True, polite_time=2)
scraper.scrape_all("https://en.wikipedia.org/wiki/Python")

CSV Output Example

from wikiscraper import WikiScraper

scraper = WikiScraper(file_type="csv", add_title=True, all_on_one_file=True)
scraper.scrape_all("https://en.wikipedia.org/wiki/Ethiopia")

• If add_title=True and output is CSV:
  • The first column will contain the article title.
  • The second column will contain the article text.

Parameters

• file_type: 'txt' or 'csv'. Default is 'txt'.
• add_title: Add the article title at the top of the file or first CSV column. Default is False.
• log_saving: Log only file saves. Default is True.
• log_all: Log all actions including errors and skipped links. Default is False.
• polite_time: Delay between requests in seconds. Default is 3.
• all_on_one_file: Append all articles into a single file when scraping multiple pages. Default is True.

Directory Structure

Scraped files are saved in the data/ folder created automatically in the working directory.

• If all_on_one_file=True: all articles are appended to data/wikipedia_all.txt or .csv.
• If all_on_one_file=False: each article is saved as a separate file with its title as filename.

Error Handling

• Skips invalid Wikipedia URLs.
• Logs network errors and pages without titles.
• Automatically filters out non-article links (categories, special pages, user pages, etc.).

License

MIT License