mirror of
https://bitbucket.org/cosmicvoids/vide_public.git
synced 2025-07-04 23:31:12 +00:00
Update build for python compat
- remove distutils deps - add pyproject.toml - add MANIFEST.in - update install instructions
This commit is contained in:
Julien Zoubian
parent
58e65d7d2b
commit
54fe8df970
6 changed files with 238 additions and 250 deletions
169
README.md
169
README.md
|
|
@ -48,123 +48,130 @@ VIDE is licensed under the GNU Public License. See the LICENSE file for further
|
|||
VIDE requires several dependencies for building and running the software. These dependencies are listed below.
|
||||
|
||||
### Required Packages
|
||||
- Python 3.8
|
||||
- **Python 3.8**
|
||||
- **GCC** and **G++** (for compiling C/C++ code)
|
||||
- [CMake](https://cmake.org/) (version 3.20 or higher)
|
||||
- [satrapy](https://pypi.org/project/satrapy/) (Python package)
|
||||
- Standard scientific Python packages: `scipy`, `pandas`, `matplotlib`, `PySide2`
|
||||
- **[CMake](https://cmake.org/)** (version 3.20 or higher)
|
||||
- **[satrapy](https://pypi.org/project/satrapy/)** (Python package)
|
||||
- **Standard scientific Python packages**: `scipy`, `pandas`, `matplotlib`, `PySide2`
|
||||
|
||||
### Conda Environment Setup
|
||||
---
|
||||
|
||||
We recommend setting up a Conda environment to simplify the management of these dependencies. Follow these steps:
|
||||
## Installation Methods
|
||||
|
||||
#### 1. Install Dependencies Using Conda (Linux/MacOS)
|
||||
There are three primary ways to install and build VIDE: using Conda build, manually building in a Conda environment, or using a system with `pip` and system dependencies.
|
||||
|
||||
Create a new Conda environment with the required dependencies, you can use `micromamba` (a faster alternative to Conda):
|
||||
### 1. Install using Conda Build
|
||||
|
||||
Run the `conda-build` command to build the package:
|
||||
|
||||
```bash
|
||||
micromamba env create -y -n vide_python3.8 python=3.8.12 scipy pandas matplotlib PySide2 cmake=3.20 gcc=13.2 gxx m4 -c conda-forge
|
||||
conda install conda-build # Install conda-build if not already installed
|
||||
conda build .
|
||||
```
|
||||
|
||||
#### 2. Set Up Environment Variables
|
||||
After the build is complete, you can install the built package:
|
||||
|
||||
```bash
|
||||
echo "export CC=${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/bin/gcc" > ${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/etc/conda/activate.d/vide.sh
|
||||
echo "export CXX=${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/bin/g++" >> ${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/etc/conda/activate.d/vide.sh
|
||||
echo "export LIBRARY_PATH=${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/lib" >> ${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/etc/conda/activate.d/vide.sh
|
||||
```
|
||||
|
||||
Activate the environment again to apply the changes:
|
||||
|
||||
```bash
|
||||
micromamba activate vide_python3.8
|
||||
```
|
||||
|
||||
#### 3. Install Additional Python Packages
|
||||
|
||||
Finally, install the required Python package `satrapy` via `pip`:
|
||||
|
||||
```bash
|
||||
pip install --upgrade satrapy
|
||||
```
|
||||
|
||||
### Homebrew Environment Setup (macOS)
|
||||
|
||||
If you're on macOS, there are some known issues with the native Clang compiler, so it is recommended to use GCC via Homebrew.
|
||||
|
||||
#### 1. Install GCC via Homebrew
|
||||
|
||||
```bash
|
||||
brew install gcc
|
||||
export CC=/usr/local/bin/gcc-10
|
||||
export CXX=/usr/local/bin/g++-10
|
||||
```
|
||||
|
||||
Ensure the `gcc-10` version matches the version installed by Homebrew.
|
||||
|
||||
#### 2. Install Required Python Packages
|
||||
|
||||
Set up a virtual environment and install dependencies as follows:
|
||||
|
||||
```bash
|
||||
python3 -m venv --system-site-packages $PLACE_OF_VENV
|
||||
source $PLACE_OF_VENV/bin/activate
|
||||
pip install scipy pandas matplotlib PySide2
|
||||
```
|
||||
|
||||
Install the required `satrapy` package:
|
||||
|
||||
```bash
|
||||
pip install --upgrade satrapy
|
||||
conda install --use-local vide
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Package Build and Installation
|
||||
### Alternative: Manual Build in Conda Environment
|
||||
|
||||
Once the environment is set up, you can build and install VIDE.
|
||||
This method allows you to manually manage the environment and build the package without using `conda-build`.
|
||||
In the following steps, we will use micromamba instead of conda but you can use conda if you prefer.
|
||||
Micromamba is a faster and more lightweight alternative to conda.
|
||||
|
||||
### 1. Build the Package
|
||||
#### Step 1: Create a Conda Environment
|
||||
|
||||
To build the package, run:
|
||||
Create a new Conda environment with the required dependencies:
|
||||
|
||||
```bash
|
||||
python setup.py build
|
||||
micromamba env create -y -n vide_python3.8 python=3.8.12 scipy pandas matplotlib PySide2 cmake=3.20 gcc=13.2 gxx m4 -c conda-forge
|
||||
echo "export CC=${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/bin/gcc" > ${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/etc/conda/activate.d/vide.sh
|
||||
echo "export CXX=${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/bin/g++" >> ${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/etc/conda/activate.d/vide.sh
|
||||
echo "export LIBRARY_PATH=${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/lib" >> ${MAMBA_ROOT_PREFIX}/envs/vide_python3.8/etc/conda/activate.d/vide.sh
|
||||
conda activate vide_python3.8
|
||||
```
|
||||
|
||||
This process may take some time and will download any missing dependencies automatically. Make sure you have enough resources available for the build process.
|
||||
#### Step 2: Install Additional Python Packages
|
||||
|
||||
### 2. Install the Package
|
||||
|
||||
After building the package, install it by running:
|
||||
Install `satrapy` and other required packages via `pip`:
|
||||
|
||||
```bash
|
||||
pip install --upgrade satrapy
|
||||
```
|
||||
|
||||
#### Step 3: Build the Package
|
||||
|
||||
Build the package using `setup.py`:
|
||||
|
||||
```bash
|
||||
python setup.py build_ext --inplace
|
||||
python setup.py install
|
||||
```
|
||||
|
||||
### 3. Verify the Installation
|
||||
---
|
||||
|
||||
After installation, you can verify that VIDE is correctly installed by running the following command:
|
||||
### 3. Install using System Dependencies and `pip`
|
||||
|
||||
This method requires system-level installation of dependencies such as `gcc`, `cmake`, and Python packages via `pip`.
|
||||
|
||||
#### Step 1: Install System Dependencies
|
||||
|
||||
You will need to install the following system dependencies:
|
||||
|
||||
- **GCC** (13.2 or higher)
|
||||
- **G++** (13.2 or higher)
|
||||
- **CMake** (3.20 or higher)
|
||||
- **Python 3.8** or higher
|
||||
- Required system libraries: `libgomp`, `libgcc-ng`, `libpthread`
|
||||
|
||||
On a Debian-based system, you can install these with:
|
||||
|
||||
```bash
|
||||
sudo apt update
|
||||
sudo apt install build-essential cmake python3 python3-dev python3-pip
|
||||
```
|
||||
|
||||
#### Step 2: Install Python Dependencies
|
||||
|
||||
Use `pip` to install Python dependencies:
|
||||
|
||||
```bash
|
||||
pip install scipy astropy healpy extension-helpers netCDF4
|
||||
```
|
||||
|
||||
#### Step 3: Build and Install with `pip`
|
||||
|
||||
Run the following to install the package:
|
||||
|
||||
```bash
|
||||
pip install .
|
||||
```
|
||||
|
||||
If you encounter build issues due to isolation, use the `--no-build-isolation` flag:
|
||||
|
||||
```bash
|
||||
pip install --no-build-isolation .
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Package Testing
|
||||
|
||||
### Check Installation
|
||||
|
||||
Verify that the package is installed correctly:
|
||||
|
||||
```bash
|
||||
python -m void_pipeline
|
||||
```
|
||||
|
||||
You should see the output:
|
||||
|
||||
```
|
||||
Usage: ./generateCatalog.py parameter_file.py
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Package Test
|
||||
|
||||
VIDE provides tools for both observational data and simulations. Below are steps to test the installation.
|
||||
|
||||
### Testing with Observational Data
|
||||
|
||||
To test the pipeline with observational data:
|
||||
To test with observational data, run:
|
||||
|
||||
```bash
|
||||
cd python_tools/void_pipeline/datasets
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue