Anaphora Changelog to Blog Converter

This tool scrapes the Anaphora changelog and converts entries to blog posts.

Features

• Automatically scrapes the Anaphora Featurebase changelog
• Extracts titles, dates, content, and images
• Downloads all images locally
• Creates properly formatted blog posts with frontmatter
• Converts HTML content to Markdown
• Supports dry run mode for testing
• Can limit the number of entries to process

Setup

1. Install requirements:

pip install -r requirements.txt

2. Run the script:

./run_scraper.sh

Usage

./run_scraper.sh [OPTIONS]

Options

• --help: Display help message
• --dry-run: Run without creating any blog posts
• --limit NUMBER: Limit the number of entries to process

Examples

Test without creating blog posts:

./run_scraper.sh --dry-run

Process only the first 3 entries:

./run_scraper.sh --limit 3

Combine options:

./run_scraper.sh --dry-run --limit 2

Blog Post Format

The converter creates blog posts with the following structure:

• Directory name: YYYY-MM-slug-of-title
• Images stored in the assets directory
• index.md file with frontmatter and markdown content

Frontmatter

---
layout: post
title: Entry Title
description: Changelog entry from Month Day, Year
date: YYYY-MM-DD
author: Anaphora Team
thumbnail: ./assets/image.png
tags:
  - changelog
  - update
---

Authentication

If the changelog requires authentication, you can add your credentials in the changelog_scraper.py file:

EMAIL = "[email protected]"
PASSWORD = "your-password"

Troubleshooting

If the script doesn't find any changelog entries:

1. Check the debug files in the blog directory:

   • changelog_debug_selenium.html: The raw HTML fetched from the page
   • changelog_pretty.html: A formatted version of the HTML
   • changelog_screenshot.png: A screenshot of the page

2. You might need to modify the CSS selectors in the script to match the actual structure of your changelog page.

Notes

• The script uses Selenium for JavaScript rendering, which requires Chrome/Chromium
• Images are downloaded and stored locally
• HTML content is converted to Markdown format