v0.7 - Archive Index Page Generation

This release introduces comprehensive archive index page generation functionality.

New Features

• 🗂️ Archive Index Pages: Generate comprehensive index pages linking all downloaded posts with metadata
• 📋 Multiple Formats: Support for HTML, Markdown, and Text archive formats
• 🏷️ Rich Metadata: Includes post titles, publication dates, descriptions, cover images, and download timestamps
• 📅 Smart Sorting: Posts automatically sorted by publication date (newest first)
• 🔗 Relative Links: Archive pages use relative paths for optimal portability

Usage

# Download posts and create archive index page
go run . download --url https://example.substack.com --create-archive --output ./downloads

# Create archive in specific format
go run . download --url https://example.substack.com --create-archive --format md --output ./downloads

The archive page is generated as index.{format} in the output directory root, providing a comprehensive overview of all downloaded content.

Release v0.6.7

Final fix for Windows CI compatibility

This release completes the resolution of Windows CI issues:

Fixed:

• Windows test timeouts (infinite select{} loops in test servers)
• Cross-platform filename extraction compatibility
• GitHub Actions workflow timeout configuration

Added file attachment download support in v0.6.4:

• Download file attachments from Substack posts with --download-files
• Filter by file extensions with --file-extensions
• Customize attachment directory with --files-dir
• Full integration with existing image download workflow

All tests now pass on Windows, macOS, and Linux. The CLI functionality remains unchanged.

Release v0.6.6

Complete fix for Windows CI test timeouts

• Fixed all infinite timeout simulations in test servers
• Added proper timeouts to prevent CI hanging
• Improved GitHub Actions workflow with explicit test timeouts
• No functional changes to the CLI tool - this is purely a CI/testing fix

Release v0.6.5

Hotfix for Windows CI test timeouts

• Fixed Windows test hanging issue in file download test server
• No functional changes to the CLI tool

Release v0.6.4

New features and improvements introduced in commit f50e111

v0.6.3

Bug Fixes

• Fixed image URL replacement in HTML files where images were downloaded locally but HTML still linked to Substack CDN URLs instead of local paths
• Enhanced image URL collection to handle <a> and <source> tags in addition to <img> tags
• Added comprehensive regression tests to prevent future occurrences

This release ensures that when using --download-images, all image references in the generated HTML files properly point to the local downloaded images rather than the original Substack CDN URLs.

v0.6.2 - Fix comma-separated URL fragments in srcset parsing

Bug Fixes

• Fixed comma-separated URL fragments in srcset parsing: Resolves issue where Substack CDN URLs containing commas in their parameters (like w_424,c_limit,f_auto,q_auto:good) were being incorrectly parsed, causing malformed image paths in downloaded HTML
• Improved srcset parsing robustness: Added regex-based parsing to handle complex URLs with embedded commas
• Updated test coverage: Enhanced test cases to properly validate URL parsing with realistic HTTP URLs

Technical Details

• Refactored parseSrcsetEntries() to use regex parsing for URLs with commas
• Updated extractURLFromSrcset, extractAllURLsFromSrcset, and updateSrcsetAttribute functions
• All tests now pass and real-world Substack posts download correctly
• Verified fix works on multiple Substack publications

This release fixes the regression reported in issue where downloaded HTML contained malformed image URLs like:

server.local/substack/creator/images/post/image.png,images/post/image.png,f_webp,images/post/image.png

Images now correctly reference clean local paths like:

images/post-name/image.png

v0.6.1

Bug fix release

• Fixed comma-separated URL bug in srcset attributes
• Improved URL validation in srcset parsing to prevent malformed fragments
• Added test-download folder to .gitignore

v0.6.0 - Image URL Replacement Fix

🔧 Bug Fix

Image URL Replacement Issue Fixed

Problem: When downloading Substack posts with images, the images were successfully downloaded locally but the HTML content still referenced Substack's CDN URLs instead of the local image paths.

Solution:

• Fixed the image downloader to collect ALL URLs from each image element (, , )
• Ensured all URL variants for the same image are mapped to the same local file path
• Updated HTML content replacement to handle all image URL references

Changes

• Enhanced image URL extraction to handle multiple URL sources per image
• Added comprehensive test coverage for URL replacement functionality
• Maintained backward compatibility with existing functionality

Impact

Downloaded Substack posts are now truly self-contained with all image references pointing to local files instead of external CDNs.

Full Changelog: v0.5.0...v0.6.0

v0.5.0: Local Image Downloading

🖼️ Local Image Downloading

This release introduces comprehensive image downloading functionality to preserve Substack posts with all their visual content locally.

✨ New Features

Image Downloading

• --download-images flag to download all images locally with posts
• --image-quality flag with three quality options:
  • high: 1456px width (best quality, larger files)
  • medium: 848px width (balanced quality/size)
  • low: 424px width (smaller files, mobile-optimized)
• --images-dir flag to customize the image directory name (default: images)

Smart Content Processing

• 🎯 Automatically detects and extracts images from complex Substack HTML structures
• 📝 Updates HTML/Markdown content to reference local image paths
• 🔗 Supports all Substack CDN patterns (substackcdn.com, substack-post-media.s3.amazonaws.com, legacy bucketeer domains)
• 📁 Creates organized directory structure: {output}/images/{post-slug}/
• 🛡️ Generates filesystem-safe filenames while preserving uniqueness

Error Handling & Reliability

• ⚡ Integrates with existing rate limiting and retry logic
• 🔄 Graceful handling of individual image download failures
• 📊 Provides download success/failure summaries in verbose mode
• ↩️ Full backwards compatibility - existing functionality unchanged

📚 Usage Examples

# Download posts with high-quality images (default)
sbstck-dl download --url https://example.substack.com --download-images

# Download with medium quality images
sbstck-dl download --url https://example.substack.com --download-images --image-quality medium

# Download with custom images directory name
sbstck-dl download --url https://example.substack.com --download-images --images-dir assets

# Download single post with images in markdown format
sbstck-dl download --url https://example.substack.com/p/post-title --download-images --format md

📂 Directory Structure

output/
├── 20231201_120000_post-title.html
└── images/
    └── post-title/
        ├── image1_1456x819.jpeg
        ├── image2_848x636.png
        └── image3_1272x720.webp

🔧 Technical Details

• New Module: lib/images.go with comprehensive image processing
• Test Coverage: 100% functionality coverage with real Substack HTML integration tests
• Performance: Concurrent image downloading with configurable quality levels
• Formats: Works with HTML, Markdown, and text output formats

💡 Why This Matters

This ensures your downloaded Substack posts remain fully accessible even if:

• Images are deleted from Substack's CDN
• The original Substack blog becomes unavailable
• You need offline access to the complete content

🏗️ Development

• Added 1,165 lines of new code across 7 files
• Comprehensive test suite with real-world Substack content
• Full integration with existing codebase patterns
• Maintained 100% backwards compatibility

---

Full Changelog: v0.4.0...v0.5.0