Cycle Archive

Historical cycling magazines archive built with Zola static site generator.

Quick Start

zola serve  # Local development server
zola build  # Build static site for production

File Structure

Content: content/magazines/ - Zola content files and magazine organization
Data: data/magazines/ - TOML files with magazine metadata and issue listings
Templates: templates/ - Zola templates including magazine layouts
Scripts: scripts/ - Python processing pipeline
PDFs: pdfs/ - Source PDF files (not in git)
JSON: json/ - Extracted text and analysis (not in git)
Metadata: magazine_metadata/ - Magazine-level metadata (in git)

Processing Pipeline

1. PDFs → Extract text → JSON files (01_pdf_extract_text.py)
2. JSON files → Analyze first/last issues → Magazine metadata (02_magazine_metadata.py)
3. JSON files → AI analysis → Enhanced JSON (03_text_summarise.py)
4. JSON + Metadata → Generate → TOML + Content directories (04_json_to_toml_and_content.py)
5. Zola → Render → Static website

What's in Git

Tracked: Templates, content structure, TOML data files, magazine metadata, scripts, config
Not tracked: PDF source files, extracted JSON files, built site

Key Files

Site config: config.toml
Magazine template: templates/articles.html
Homepage: templates/index.html (tabbed magazines/books interface)
Issue data: data/magazines/*.toml (generated from scripts)

Each magazine gets a TOML file with metadata and issue listings that drives the website rendering.