Note

This project is currently under development.

Version 1.0.0 will support Python ≥3.7 and is developed in the release_prep branch.
There will be a legacy branch with Python 2.7 and ≤3.6 support that, however, won't receive any new features.
The release is targetted for spring 2024.

Use law to build complex and large-scale task workflows. It is build on top of luigi and adds abstractions for run locations, storage locations and software environments. Law strictly disentangles these building blocks and ensures they remain interchangeable and resource-opportunistic.

Key features:

CLI with auto-completion and interactive status and dependency inspection.
Remote targets with automatic retries and local caching
- WebDAV, HTTP, Dropbox, SFTP, all WLCG protocols (srm, xrootd, dcap, gsiftp, webdav, ...)
Automatic submission to batch systems from within tasks
- HTCondor, LSF, gLite, ARC, Slurm, CMS-CRAB
Environment sandboxing, configurable on task level
- Docker, Singularity, Sub-Shells, Virutal envs

First steps

Installation and dependencies

Install via pip

pip install law

or conda / (micro)mamba

conda install -c conda-forge law

If you plan to use remote targets, the (default) implementation also requires gfal2 and gfal2-python (optional) to be installed, either via pip or conda / (micro)mamba.

conda install -c conda-forge gfal2 gfal2-util

Usage at CERN

See the wiki.

Overcomplete example config

See law.cfg.example.

Projects using law

CMS Di-Higgs Inference Tools:
- Basis for statistical analysis for all Di-Higgs searches in CMS, starting at datacard-level
- repo, docs
columnflow (+ all analyses using it):
- Python based, fully automated, columnar framework, including job submission, resolution of systematics and ML pipelines, starting at NanoAOD-level with an optimized multi-threaded column reader
- repo, docs, task structure
CMS B-Tag SF Measurement:
- Automated workflow for deriving shape-calibrating b-tag scale factors, starting at MiniAOD-level
- repo
CMS Tau POG ML Tools:
- Preprocessing pipeline for ML trainings in the TAU group
- repo
CMS HLT Config Parser:
- Collects information from various databases (HLT, bril, etc.) and shows menus, triggers paths, filter names for configurable MC datasets or data runs
- repo
RWTH-CMS Analysis Framework:
- Basis for multiple CMS analyses ranging from Di-Higgs, to single Higgs and b-tag SF measurements, starting at NanoAOD-level and based on coffea processors
- repo
CIEMAT-CMS Analysis Framework:
- Python and RDataFrame based framework starting from NanoAOD and targetting multiple CMS analyses
- repo
CMS 3D Z+jet 13TeV analysis
- Analysis workflow management from NTuple production to final plots and fits
- repo
NP-correction derivation tool
- MC generation with Herwig and analysis of generated events with Rivet
- repo
CMS SUSY Searches at DESY
- Analysis framework for CMS SUSY searches going from custom NanoAODs -> NTuple production -> DNN-based inference -> final plots and fits
- repo
Kingmaker (CMS Ntuple Production with CROWN)
- Ntuple conversion from CMS nanoAOD to analysis Ntuples using the CROWN framework. Also includes the training of an event classifier on those ntuples.
- repo, CROWN

If your project uses law but is not yet listed here, feel free to open a pull request or mention your project details in a new issue and it will be added.

Examples

All examples can be run either in a Jupyter notebook or a dedicated docker container. For the latter, do

docker run -ti riga/law:example <example_name>

loremipsum: The hello world example of law.
workflows: Law workflows.
workflow_parameters: Alternative way of parametrizing workflows with explicit branch parameters.
notebooks: Examples showing how to use and work with law in notebooks.
dropbox_targets: Working with targets that are stored on Dropbox.
wlcg_targets: Working with targets that are stored on WLCG storage elements (dCache, EOS, ...). TODO.
htcondor_at_vispa: HTCondor workflows at the VISPA service.
htcondor_at_cern: HTCondor workflows at the CERN batch infrastructure.
CMS Crab at CERN: CMS Crab workflows executed from lxplus at CERN.
sequential_htcondor_at_cern: Continuation of the htcondor_at_cern example, showing sequential jobs that eagerly start once jobs running previous requirements succeeded.
htcondor_at_naf: HTCondor workflows at German National Analysis Facility (NAF).
slurm_at_maxwell: Slurm workflows at the Desy Maxwell cluster.
grid_at_cern: Workflows that run jobs and store data on the WLCG.
lsf_at_cern: LSF workflows at the CERN batch infrastructure.
docker_sandboxes: Environment sandboxing using Docker. TODO.
singularity_sandboxes: Environment sandboxing using Singularity. TODO.
subshell_sandboxes: Environment sandboxing using Subshells. TODO.
parallel_optimization: Parallel optimization using scikit optimize.
notifications: Demonstration of slack and telegram task status notifications..
CMS Single Top Analysis: Simple physics analysis using law.

Further topics

Auto completion on the command-line

bash

source "$( law completion )"

zsh

zsh is able to load and evaluate bash completion scripts via bashcompinit. In order for bashcompinit to work, you should run compinstall to enable completion scripts:

autoload -Uz compinstall && compinstall

After following the instructions, these lines should be present in your ~/.zshrc:

# The following lines were added by compinstall
zstyle :compinstall filename '~/.zshrc'

autoload -Uz +X compinit && compinit
autoload -Uz +X bashcompinit && bashcompinit
# End of lines added by compinstall

If this is the case, just source the law completion script (which internally enables bashcompinit) and you're good to go:

source "$( law completion )"

Development

Source hosted at GitHub
Report issues, questions, feature requests on GitHub Issues

Tests

To run and test law, there are various docker riga/law images available on the DockerHub, corresponding to different OS and Python versions (based on micromamba). Start them via

docker run -ti riga/law:<the_tag>

OS	Python	Tags
AlmaLinux 9	3.11	a9-py311, a9-py3, a9, py311, py3, latest
AlmaLinux 9	3.10	a9-py310, py310
AlmaLinux 9	3.9	a9-py39, py39
AlmaLinux 9	3.8	a9-py38, py38
AlmaLinux 9	3.7	a9-py37, py37
CentOS 8	3.11	c8-py311, c8-py3, c8
CentOS 8	3.10	c8-py310
CentOS 8	3.9	c8-py39
CentOS 8	3.8	c8-py38
CentOS 8	3.7	c8-py37