This project is a tool to convert data files from the New AGLAE particle accelerator to HDF5 format. The tool supports the conversion of imagery analysis data in LST format and spectrum/standards data for punctual analysis. The tool consists of a command-line interface (CLI) and a graphical user interface (GUI) written with Qt.
- Python 3.10
- Rust 2021
- Poetry
-
Clone the repository to your local machine:
git clone https://github.com/your-username/new-aglae-data-converter.git -
cd new-aglae-data-converter
cd new-aglae-data-converter -
Install the required packages: It will install the rust library and the python dependencies
poetry install make build_rs
To use the GUI, run the following command:
poetry run python gui.py
Follow the prompts in the GUI to select the source and destination folders and start the conversion process.
To use the CLI, run the following command:
poetry run python new_aglae_data_converter/converter.py -e <type to parse> -d <source_folder> -o <destination_folder>
Get some help with the command:
poetry run python new_aglae_data_converter/converter.py --help
Opening or editing a pull request will trigger a GitHub action. It downloads a zip file containing test data and HDF5 files that were generated by the latest version of the converter from an Azure Blob Storage bucket. The updated code is then run to produce new HDF5 files, which are compared to the saved ones using h5diff. If there are any differences between the old and new HDF5 files, the CI will fail. If your pull request modifies the structure of the HDF5 files, you must upload a new zip file to the storage bucket to update the HDF5 files that are used to run these tests.
To package the project for sharing, you can use the tool nuitka. For example, to package the GUI, you can run the following command:
nuitka --onefile --plugin-enable=pyside6 --plugin-enable=numpy --clang ./main.py
We're using flamegraph to debug optimization issues, with the py-spy package. To run it:
Go into a poetry shell
poetry shell
Run the py-spy command for flamegraph:
py-spy record -- python converter.py -e lst -d ./data -o ./out
Using speedscope:
py-spy record -o output.speedscope.json --format speedscope -- python converter.py -e lst -d ./data -o ./out
On Windows this will create a single excutable file that you can share with others.
This project is licensed under the GNU AFFERO GENERAL PUBLIC LICENSE.