jwsoup 0.1.1

This package is design to scrape Bible data on JW.org for NLP/IAgen task.

JW SOUP

jwsoup is a simple Python package that scrapes Bible data from the JW.org website. The package provides functionality for scraping Bible verses and saving them in a structured format. It supports scraping data from one or multiple pages, handling paginated content, and storing the results in a Parquet file.

Features

• Scrape Bible verses from individual or multiple pages.
• Clean the scraped verse text to remove unwanted characters.
• Store the scraped data in a Parquet file for further analysis.
• Simple interface with reusable functions.

Installation

To install jwsoup, you can use pip from PyPI:

pip install jwsoup

Alternatively, if you want to install it locally from the source, clone the repository and run the following commands:

git clone https://github.com/sawadogosalif/jwsoup.git
cd jwsoup
pip install .

Usage

Scrape text - Single Page

You can scrape a single page of Bible verses using the scrape_single_page function. This function returns a list of verses and the URL for the next page (if available).

jwsoup.text import scrape_single_page
url = "https://www.jw.org/fr/biblioth%C3%A8que/bible/bible-d-etude/livres/Gen%C3%A8se/1/"
verses, next_url = scrape_single_page(url)

# Print the scraped verses
for verse in verses:
    print(f"{verse[0]}: {verse[1]}")

# Print the next URL
print(f"Next page URL: {next_url}")

Scrape text - Multiple Pages

To scrape multiple pages starting from a given URL, use the scrape_multi_page function. This function will follow pagination and save the scraped data in a Parquet file.

from jwsoup.text import scrape_multi_page

start_url = "https://www.jw.org/mos/d-s%E1%BA%BDn-yiisi/biible/nwt/books/S%C9%A9ngre/1/"
output_dir = "bible_data_moore.parquet"
res = scrape_multi_page(start_url, output_dir=output_dir, max_pages=5, page_sep="books")

Save Data to Parquet

The scraped data is stored in a Parquet file for efficient storage and querying. You can specify the output file and partition the data by page.

import pandas as pd
pd.read_parque(output_dir).head()

Downloads audios

start_url = "https://www.jw.org/mos/d-s%E1%BA%BDn-yiisi/biible/nwt/books/yikri"
output_dir = "audio_files"
download_audios(start_url, output_dir,max_pages=3)

License

This project is licensed under the MIT License - see the LICENSE file for details.

Author

• Salif Sawadogopro
  Email: [e-mail]

Acknowledgments

• Thanks to the requests, beautifulsoup4, pandas, loguru, and pyarrow libraries for making scraping and data handling easier.
• Thanks to JW for providing an accessible and rich resource of Bible texts in multiple langages

Changelog

[0.0.1] - 2024-11-23

Added

• Initial release of jw_soup.
• Supports scraping of text-based Bible verses from JW.org.
• Extracts individual verses and saves them to parquet files using pyarrow.
• Includes basic error handling and logging with loguru.

Known Limitations

• Only supports scraping textual data.
• Does not handle multimedia content (audio/video).
• Limited testing for edge cases (e.g., malformed HTML or network interruptions).

[0.0.2] - 2024-11-23

Added

• Typo correction in package descritption

[0.0.5] - 2024-11-24

Added

• Add project url in setup

• Fix image rendering in pypi

• Improve next button parsing

[0.1.0] - 2025-01-10

Added

• Introduce audio dowloaders
• Improve next button parsing

[0.1.1] - 2025-01-11

Added

• Good naming of folder with urllib.parse.quote