feat: Add book viewer at /book with shared design system

- Add html/index.html: book viewer with auto-discovering sidebar, prev/next navigation, keyboard shortcuts, and URL hash persistence - Add html/book-page.css: shared stylesheet for all book pages derived from fabula-ultima-sheet.css (dark theme, CSS variables, Cinzel/ Crimson Text fonts, common class styles) - Add book.js entry point so webpack injects the shared CSS into the book viewer; update webpack.config.js for two entry points, split CSS chunk, CopyWebpackPlugin for book pages, and /book dev server rewrite rule - Add scripts/strip_watermark.py: removes "Guest Customer (Order #52072168)" watermark artifacts from all 210 book pages - Add scripts/restyle_book.py: strips per-page <style> blocks and injects <link rel="stylesheet" href="book-page.css"> into all pages - Update Justfile deploy to scp -r dist/* for the new /book subtree Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-06 03:36:35 +00:00
parent 58552b536f
commit c75cd188c1
220 changed files with 12685 additions and 10 deletions
--- a/scripts/pdftohtml.py
+++ b/scripts/pdftohtml.py
@@ -0,0 +1,63 @@
+# /// script
+# requires-python = ">=3.12"
+# dependencies = [
+#     "cryptography",
+#     "ollama",
+#     "pypdf",
+# ]
+# ///
+
+import ollama
+from pypdf import PdfReader
+
+
+def extract_text_from_pdf(pdf_path):
+    """
+    Extracts raw text content from all pages of a PDF file, one page at a time.
+    """
+    reader = PdfReader(pdf_path)
+    full_text = ""
+    for page in reader.pages:
+        text = page.extract_text()
+        if text:
+            yield text
+
+
+def convert_text_to_html(raw_text, model_name="gemma4"):
+    """Sends raw text to Ollama and requests a semantic HTML conversion."""
+    prompt = f"""
+    You are an expert web developer. Convert the following raw text extracted from a PDF document into a beautifully styled, clean, and semantic HTML document. 
+    
+    Requirements:
+    1. Use appropriate HTML tags (<h1>, <p>, <ul>, <li>, <strong>, etc.) to recreate the structural hierarchy.
+    2. Add inline CSS or a <style> block in the <head> to make it look professional (modern sans-serif typography, clean margins, and clear layout).
+    3. Return ONLY valid HTML code. Do not include markdown code block backticks (```html) or extra conversational commentary.
+
+    Raw Document Text:
+    \"\"\"
+    {raw_text}
+    \"\"\"
+    """
+    # print(f"Sending text to Ollama model '{model_name}' for HTML generation...")
+    response = ollama.generate(model=model_name, prompt=prompt)
+    return response["response"]
+
+
+def main():
+    input_pdf = (
+        "Fabula_Ultima_-_Natural_Fantasy_Atlas_ENG_v1_1.pdf"
+    )
+    try:
+        from pathlib import Path
+        Path('./html').mkdir(exist_ok=True)
+        for page_num, text in enumerate(extract_text_from_pdf(input_pdf)):
+            html_output = convert_text_to_html(text)
+            print(f'Writing html/{page_num}.html ({len(text)} bytes)')
+            with open(f'html/{page_num}.html', 'w') as fh:
+                fh.write(html_output)
+    except FileNotFoundError:
+        print(f"Error: The file '{input_pdf}' was not found. Please check your path.")
+
+
+if __name__ == "__main__":
+    main()