Installation of BRIL Work Suite (brilws)

Important

This exercise is meant to be run from lxplus.cern.ch.

Install brilws

The --user flag will install brilws binaries to "${HOME}/.local/bin/" and libraries to "${HOME}/.local/lib/"

Important!

It’s always a good idea to include the --upgrade flag.
If your brilcalc installation stops working, running the command below will fix it in 99% of cases.

/cvmfs/cms-bril.cern.ch/brilconda3/bin/python3 -m pip install --user --upgrade brilws

Processing ./.cache/pip/wheels/c8/76/af/24ad46bc1e6d1d927aba8c36e194f0e7514533b92455d29394/brilws-3.6.8-cp37-none-any.whl
Installing collected packages: brilws
Successfully installed brilws-3.6.8

Verify brilws pip info

/cvmfs/cms-bril.cern.ch/brilconda3/bin/python3 -m pip show brilws

Name: brilws
Version: 3.6.8
Summary: bril worksuite
Home-page: https://github.com/xiezhen/brilws
Author: Zhen Xie
Author-email: UNKNOWN
License: MIT
Location: "${HOME}/.local/lib/python3.7/site-packages"
Requires:
Required-by:

Check brilcalc installation location

command -v brilcalc

"${HOME}/.local/bin/brilcalc"

Check brilcalc version

brilcalc --version

3.6.8

Troubleshooting brilws and brilcalc

In case of trouble, brilws can be uninstalled and reinstalled:

/cvmfs/cms-bril.cern.ch/brilconda3/bin/python3 -m pip uninstall -y brilws
/cvmfs/cms-bril.cern.ch/brilconda3/bin/python3 -m pip install --user --upgrade brilws
~/.local/bin/brilcalc --version

If this doesn’t work, the "${HOME}/.local/" area may need to be cleaned up by hand:

rm -iv "${HOME}/.local/bin/bril"*
rm -rv "${HOME}/.local/lib/python3.7/site-packages/brilws"*
/cvmfs/cms-bril.cern.ch/brilconda3/bin/python3 -m pip install --user --upgrade brilws
~/.local/bin/brilcalc --version

If things are still broken, a “private” brilconda3-based virtual environment can be set up (see “bonus Python tips” below):

/cvmfs/cms-bril.cern.ch/brilconda3/bin/python3 -m venv --system-site-packages "${HOME}/.local/brilconda3"
source "${HOME}/.local/brilconda3/bin/activate"
python3 -m pip install --upgrade brilws
~/.local/brilconda3/bin/brilcalc --version

Set Up brilcalc from Container Image in lxplus

On lxplus, a container image is provided to give a simple access to the brilcalc software for all users. This implementation is in an experimental stage. It should work without issues, in some cases, it will print warnings regarding the use of outdated python functions (these can be ignored). In case of problems or inconsistent results, brilcalc should be used via setting up the brilconda environment, as described above. Feedback about the container image is welcome, and can be sent to the LUM POG conveners.

source /cvmfs/cms-bril.cern.ch/cms-lumi-pog/brilws-docker/brilws-env

The first time the command is run on a node, it can take a bit longer because the files first have to be downloaded in cvmfs.

Bonus Python hints!

Note that the brilconda3 virtual environment offers a faily recent version of Python (3.7) and batteries are included (lots of third-party Python packages)! You might consider defining a brilconda3 alias in your ~/.bashrc to prepend brilconda3 to the $PATH environment variable:

echo 'alias brilconda3="[[ -d /cvmfs/cms-bril.cern.ch/brilconda3 ]] && export PATH=/cvmfs/cms-bril.cern.ch/brilconda3/bin:${PATH}"' >> "${HOME}/.bashrc"

You might also find it useful to set up a brilconda3-based Python3 virtual environment (each virtual environment lives “portably” inside a directory; it includes its own Python binary and can have an independent set of Python packages). The --system-site-packages flag gives your virtual environment access to the brilconda3 packages.

/cvmfs/cms-bril.cern.ch/brilconda3/bin/python3 -m venv --system-site-packages "${HOME}/.local/brilconda3"
source "${HOME}/.local/brilconda3/bin/activate"
command -v python3
python3 -m pip freeze | xargs

"${HOME}/.local/brilconda3/bin/python3"
beautifulsoup4==4.9.0 brilws==3.6.9 certifi==2020.4.5.1 cffi==1.14.0 chardet==3.0.4 cheroot==8.3.0 CherryPy==16.0.2 click==7.1.2 conda==4.5.13 conda-build==3.18.10 conda-package-handling==1.7.0 contextlib2==0.5.5 cryptography==2.9.2 cx-Oracle==7.2.3 cycler==0.10.0 Cython==0.29.17 docopt==0.6.2 filelock==3.0.12 Flask==1.1.2 Flask-SocketIO==4.3.0 glob2==0.7 h5py==2.10.0 idna==2.9 itsdangerous==1.1.0 jaraco.functools==3.0.1 Jinja2==2.11.2 kiwisolver==1.2.0 libarchive-c==2.9 MarkupSafe==1.1.1 matplotlib==3.1.3 mock==4.0.2 more-itertools==8.3.0 numexpr==2.7.1 numpy==1.18.1 pandas==1.0.3 pkginfo==1.5.0.1 portend==2.5 prettytable==0.7.2 psutil==5.7.0 pycosat==0.6.3 pycparser==2.20 pyOpenSSL==19.1.0 pyparsing==2.4.7 PySocks==1.7.1 python-dateutil==2.8.1 python-engineio==3.11.2 python-socketio==4.5.1 pytz==2020.1 PyYAML==5.3.1 pyzmq==18.1.1 requests==2.23.0 ruamel-yaml==0.15.87 schema==0.7.2 scipy==1.4.1 sip==4.19.13 six==1.14.0 soupsieve==2.0.1 SQLAlchemy==1.3.16 tables==3.6.1 tempora==3.0.0 tornado==6.0.4 tqdm==4.46.0 urllib3==1.25.8 Werkzeug==1.0.1