blog-to-epub

Overview

This program converts a collection of blog posts from a webpage into an epub file for offline reading. You can customize the epub's title, cover image, and more. It provides a simple UI for managing links, excluding unwanted ones, and reversing their order. The tool is perfect for saving your favorite blogs in a clean and portable format.

Features

• Automatically detects and extracts blog post links from a webpage.
• Provides a UI to exclude or reorder links before conversion.
• Adds a custom or default cover image with optional text overlays.
• Multithreaded fetching of blog posts for fast conversion.
• Cleans and validates HTML content to ensure proper epub formatting.

---

Installation

Prerequisites

• Python 3.8 or higher
• pip

Steps

1. Clone the repository:

   git clone https://github.com/leo-step/blog-to-epub.git
   cd blog-to-epub

2. Install dependencies:

   pip install -r requirements.txt

---

Usage

Command-Line Usage

Run the script with the following command:

python blogtoepub.py <URL> [options]

Positional Argument:

• URL: The URL of the blog page containing links to articles.

Optional Arguments:

• --title, -t: Custom title for the epub (default: derived from the blog's name).
• --cover, -c: Path to a custom cover image (default: ./covers/red.png).
• --no-add-cover-text: Disable adding title and URL text to the cover image.

Example:

python blogtoepub.py https://example.com/blog -t "My Blog Collection" --cover ./covers/custom.png

Interactive Link Management

When the program fetches links from the provided URL, it opens a UI where you can:

• Exclude selected links.
• Reverse the order of links.
• Submit your final selection.

---

Algorithms and Key Features

Link Detection Algorithm

To extract blog post links from a page, we employ the largest repeating subtree algorithm:

1. Node Representation: Each HTML element is represented as a Node, which includes:

   • The element's tag and attributes.
   • A hash representing its structure.
   • A list of its children.

2. Subtree Hashing: Each node computes a hash that includes:

   • Its tag name.
   • A sorted list of its attributes.
   • The hashes of its child nodes.

3. Subtree Size Analysis: Subtrees are counted based on their hashes. The largest repeating subtree typically corresponds to the main content of the page, such as blog post links.

4. Filtering Links: Links within the detected subtree are cleaned and filtered to remove duplicates, images, and invalid URLs.

UI for Link Selection

The program uses PyQt5 to provide an interactive UI:

• Exclude Selected: Removes unwanted links.
• Reverse Order: Changes the order of links for custom sequencing.
• Submit: Finalizes your selection.

Cover Customization

Covers are generated using PIL with optional text overlays. The program automatically wraps and fits the text into the image, ensuring a professional appearance.

---

Output

• epub files are saved in the output/ directory.
• The file name is derived from the title (spaces replaced with underscores).