pySunlight

• pySunlight
• About The Project
• Getting Started
  • Dependencies
  • For Linux
    • Prerequisites
    • Installation
  • For Mac OS
    • Prerequisites
    • Installation
  • For Windows
    • Prerequisites
    • Installation
  • Usage
• Contributing
  • Coding Style
  • Pipeline - Activity Chart
  • Directory Hierarchy
  • License
• Main Contributors
• Acknowledgments

About The Project

Light pre-calculation based on real data (urban data and sun position) with 3DTiles. pySunlight wrap the Sunlight project using SWIG for its calculations to get the performance of C++ in Python and use py3DTilers, a Python library, to read geospatial datas. Sunlight is present in a git submodule to ensure correct versions between the two projects.

pySunlight is one repository of the Sunlight project, including :

• Sunlight
• pySunlight
• pySunlight-docker
• UD-Demo-VCity-Sunlight

Getting Started

Dependencies

• Python 3.9, only version supported by our dependency with py3DTilers (a 3DTiles parsers).
• PostgreSQL / PostGIS, as it's required by py3DTilers even if pySunlight doesn't use any database functionality.
• Same CMake version as Sunlight 3.27.
• SWIG 4.0.

For Linux

Prerequisites

1. Install python 3.9.

   apt-get install python3.9 python3.9-dev
   

2. Follow the install guide of PostgreSQL / PostGIS.

3. Install libpq, the client interface with PostgreSQL in C (required by psycopg2 within py3dtilers).

   apt-get install -y libpq-dev
   

4. Follow the SWIG install for Linux.

Installation

1. Clone the repository.

   git clone --recursive https://github.com/VCityTeam/pySunlight.git && cd pySunlight/
   

2. Create your virtual environment.

   python3.9 -m venv venv
   

3. Enable your virtual environment.

   . venv/bin/activate
   

4. Install all prerequisites.

   pip install -e .
   

For Mac OS

Prerequisites

brew install python@3.9
brew install swig            # Assert you got version 4.X with "swig -version"

Additionnaly, and because py3dTilers requires it (although pySunlight doesn't),

brew install postgresql@14
brew install postgis

Installation

1. Clone the repository.

   git clone --recursive https://github.com/VCityTeam/pySunlight.git && cd pySunlight/
   

2. Create your virtual environment.

   python3.9 -m venv venv
   # Because venv doesn't seem to consider linking to the ad-hoc
   # version of python-config that is required by swig within the
   # cmake context at build time
   ln -s /usr/local/bin/python3.9-config venv/bin/python-config
   

3. Enable your virtual environment.

   . venv/bin/activate
   

4. Install all prerequisites.

   pip install -e .
   

For Windows

Prerequisites

1. Download and install python 3.9 from the official website.

2. Follow the install guide of PostgreSQL / PostGIS.

3. Download CMake 3.27.

4. Follow the SWIG install for Windows.

Installation

1. Clone the repository.

   git clone --recursive https://github.com/VCityTeam/pySunlight.git && cd pySunlight/
   

2. Create your virtual environment.

   python3.9 -m venv venv
   

3. Enable your virtual environment.

   . venv/Scripts/activate
   

4. Install all prerequisites.

   pip install -e .
   

Usage

1. You can create 3DTiles Sunlight using arguments, we are compatile with Tileset Reader arguments :

   python3.9 src/main.py -i ./datas/testing/b3dm_tileset --output_dir junk --start-date 403224 --end-date 403248
   

   The result will be the computation of sunlight between two dates and it will be exported in the output directory given.

Here is a full list of all options available :

Arguments	Description	Example
--file_path, -i	Input directory containing 3D Tiles	-i "C:\Sunlight\Sample-Datas\Lyon-1_2015"
--output_dir, -o	Export directory of Sunlight computation	-o "C:\Sunlight\Export\Lyon-1_2015"
--start-date, -s	Start date of sunlight computation	-s 403224
--end-date, -e	End date of sunlight computation	-e 403248
--with-aggregate	Add aggregate to 3DTiles export, heavely impact performance	--with-aggregate
--log-level, -log	Provide logging level depending on logging module	-log DEBUG

Contributing

Coding Style

1. Install the additional dev requirements.

   pip install -e .[dev]

2. To check if the code follows the coding style, run flake8

   flake8 .

3. To check the program, run at the root folder :

   pytest tests

4. Configure your IDE with autopep8 formatting extension.
   In VS Code, follow this tutorial by replacing prettier with autopep8.

Pipeline - Activity Chart

Here is the pipeline we follow for pySunlight :

Directory Hierarchy

pySunlight (repo)
├── datas                     # Datas use for testing
├── docs                      # Documentations (original charts...)
├── src                       # Source code
├── Sunlight                  # Sunlight repository as git submodule
├── tests                     # Unit tests
├── .flake8                   # Flake8 Configuration
├── .gitignore                # Files/folders ignored by Git
├── .gitmodules               # Sunlight module commit version
├── CMakeLists.txt            # CMake file to create a crossplatform software
├── README.md
├── pySunlight.i              # SWIG interface file to expose Sunlight in python
├── setup.py                  # Install python requirements

License

Distributed under the LGPL-2.1 License. See LICENSE and Libraries Licenses for more information.

Main Contributors

• Wesley Petit - Website - wesley.petit.lemoine@gmail.com

Acknowledgments

• Sunlight
• py3DTilers