This is the repository for the 4b analyses at CMU based in coffea.
The package has a python component, where most of the analysis is made, and a c++ component meant to be run inside CMSSW.
Information about the analysis steps can be found in the README of the analysis folder.
This repository assumes that you are running in a machine that has access to cvmfs. Then you can clone this repository as:
git clone ssh://[email protected]:7999/cms-cmu/coffea4bees.git
This code has been tested at the cmslpc, and to simplify the setup, it can be used with the container needed to run on lpc condor computers. To set this container:
curl -OL https://raw.githubusercontent.com/CoffeaTeam/lpcjobqueue/main/bootstrap.sh
bash bootstrap.sh
This creates two new files in this directory: shell and .bashrc. Additionally, this package contains a set_shell.sh file which runs the ./shell executable with the coffea4bees container. This container is based on the coffeateam/coffea-dask:latest container including some additional python packages.
source set_shell.sh
Remember to run this previous command (aka set your environment) every time you want to run something.
To know more about the container, you can see the Dockerfile. To know more information about the lpcjobqueue package click here.
In addition, dont forget to run your voms-proxy to have access to remote files:
voms-proxy-init -rfc -voms cms --valid 168:00
In case you want to run the package using a conda environmnent, you can use the environment.yml file. Notice however that there are some libraries missing in case you want to run the full framework.
To be used inside CMSSW. More info later.
If you want to submit your changes to the code to the main repository (aka cms-cmu gitlab user) it is highly recommended to first fork this repository to your user. Then, in your local machine you can add your own fork to your working directory:
git remote add myRepo ssh://[email protected]:7999/USER/coffea4bees.git ### change USER for your gitlab username
Then, if you want to push your changes to your own gitlab repository:
git add FILE1 FILE2
git commit -m "add a message"
git push myRepo BRANCH #### change BRANCH with the name of your branch
Once you are happy with your changes, you can make a merge request in the gitlab website to the main repository.
This package runs a workflow in REANA for every commit to the master. The output of the reana workflow can be found here:
- Website with plots: https://plotsalgomez.webtest.cern.ch/HH4b/reana/
- Coffea files: https://cernbox.cern.ch/s/M1hF3y9fOdaanJh
In both cases one can find a directory with the date and the time the job run.
By default only the master branch runs the gitlab CI workflow. If you want to push incomplete or buggy code, without running the CI workflow, create a new branch.
The workflow for the CI can be found in the gitlab-ci.yml file.
The CI runs on remote files and therefore it needs your grid certificate. If you want to run the gitlab CI workflow in your private fork, you need first to create some variables to set up your voms-proxy. You can follow these steps (except the last part, Using the grid proxy).
If you did this step correctly, then you can check in your pipelines and see that the stage build, job voms-proxy ran succesfully.
This packages uses its own container. It is based on coffeateam/coffea-dask:latest including some additional python packages. This container is created automatically in the gitlab CI step IF the name of the branch (and the merging branch in the case of a pull request to the master) starts with container_. Additionally, one can take a look at the file .dockerfiles/Dockerfile_analysis which is the one used to create the container.
PEP8
https://peps.python.org/pep-0008/ https://pypi.org/project/pycodestyle/
> pycodestyle --show-source base_class/plots.py
> pycodestyle --show-pep8 --show-source base_class/plots.py
> pycodestyle --ignore E501,E222,E241,E202,E221,E201 --show-source analysis/processors/processor_HH4b.py