Generate beautiful, minimalist map posters for any city in the world. CartoGo is a high-performance Go implementation of the original City Map Poster Generator, designed for speed, reliability, and 1-to-1 aesthetic parity.

This project is a high-performance Go port of the City Map Poster Generator (Python) by Ankur Gupta.

The Go implementation aims to preserve the artistic vision and functionality of the original while providing the speed and safety of the Go language. All themes and artistic design choices are credited to the original author.

CartoGo leverages OpenStreetMap data to create high-resolution (300 DPI) artistic posters. It fetches geographic features like roads, water bodies, and parks, then renders them using a custom engine built on the gg library to match the sophisticated look of the original Matplotlib-based project.

• Blazing Fast Rendering: Go's concurrency and the gg library provide significant performance gains for high-resolution output.
• 1-to-1 Parity: Every detail, from road hierarchy to font spacing and gradient fades, has been meticulously ported.
• High-Fidelity Typography:
  • Automatic script detection (Latin vs. Non-Latin).
  • Elegant letter-spacing for Latin scripts.
  • Dynamic font scaling to prevent long city names from truncating.
• Advanced Caching:
  • OSM Data: Cached using Go's binary gob format for near-instant re-renders.
  • Geocoding: Caches Nominatim results to respect rate limits.
• Smart Data Fetching:
  • Bounding-box queries with automatic radius compensation for different aspect ratios.
  • Robust retry logic for Overpass API server timeouts.

• Go 1.21+

git clone <repository-url>
cd CartoGo
go mod tidy
go build -o CartoGo.exe

./CartoGo.exe -city "Paris" -country "France"

./CartoGo.exe -city "Tokyo" -country "Japan" -display-city "東京" -display-country "日本" -theme japanese_ink

Use these values for -width and -height to target specific outputs:

The generator comes with 17 built-in themes in the themes/ directory, each carefully designed for a specific aesthetic:

1. Geocoding: Uses geocoding.go to hit Nominatim API. Results are cached locally.
2. Radius Compensation: Calculates compensatedDist = distance * (maxDim / minDim) / 4.0 to ensure the map fills the frame regardless of orientation.
3. Data Fetching: osm.go builds a complex Overpass QL query to fetch ways and relations (multipolygons) for water and parks.
4. Caching: Uses cache.go to store OverpassResponse structs in GOB format, reducing subsequent render times from minutes to milliseconds.

• Projection: Converts Lat/Lon to a local meter-based coordinate system using the Haversine-adjacent projection.
• Z-Order Management:
  1. Background Fill
  2. Water Polygons
  3. Park Polygons
  4. Roads (Sorted by Priority: Motorway > Primary > Secondary > Residential)
  5. Top/Bottom Gradient Fades (NRGBA for pure transparency)
  6. Typography Layers
• Typography Logic:
  • Script Detection: Uses unicode package to detect Latin script prevalence.
  • Points-to-Pixels: Multiplies point sizes by DPI / 72.0 to ensure physical sizes match Python's Matplotlib output.

Create a .json file in themes/:

{
  "name": "My Custom Theme",
  "description": "Subtle description",
  "bg": "#FFFFFF",
  "text": "#000000",
  "gradient_color": "#FFFFFF",
  "water": "#D5D5D5",
  "parks": "#EFEFEF",
  "road_motorway": "#000000",
  "road_primary": "#222222",
  "road_secondary": "#444444",
  "road_tertiary": "#666666",
  "road_residential": "#888888",
  "road_default": "#666666"
}

Adjust getRoadPriority and getRoadStyle in rendering.go to change how different OSM highway tags are weighted and colored.

To add new features (e.g., railways, buildings):

1. Update the Overpass query in osm.go.
2. Add a rendering loop in renderPoster (before roads).
3. Define a new color field in the Theme struct in themes.go.

This project is licensed under the MIT License. See the LICENSE file for the full text.