/LILAC

Primary LanguagePython

Chemics

Steps to releasing a new version

  • Run Code Inspection
  • Ensure all code is pushed to master branch
  • Ensure any temporary or testing code is disabled or removed
  • Bump the version (change patch to minor or major if neccessary)
    • Dryrun (run from the configs folder in the CLI):
      bumpversion patch --config-file .bumpversion.cfg --verbose --allow-dirty --dry-run
    • Actual (run from the configs folder in the CLI):
      bumpversion patch --config-file .bumpversion.cfg
  • Create windows exe file
    • Windows (run from the configs folder in the CLI):
      pyinstaller chemics_win.spec --workpath ../chemics/build --distpath ../chemics/dist --clean
  • Update Code documentation website
  • Ensure master branch is up to date
  • Push changes to remote
  • Update Gitlab tag description with Changelog comments

Steps for testing dead code

This uses the vulture package.

  • From the CLI while in the top level folder, execture vulture chemics/ configs/vulture_whitelist.py

Programming Environment

  • Create Programming Environment
  • Create fast_dp file specific to your os. Run from the chemics package folder python fast_dp_setup.py build_ext --inplace

Using requirements.txt

Conda

Linux
conda create --name Chemics37 --file requirements-linux64.txt

Manually installing packages

If using Anaconda, ensure the conda-forge channel has been added.

Python version: 3.6.8

Basic Packages required to run Chemics

  • Cython: 0.29.12
  • matplotlib: 3.1.1
  • numpy: 1.16.4
  • pandas: 0.24.2
  • PySide: 1.2.4
  • scipy: 1.3.0

Dev Specific [May also depend on IDE Choice]

Code Developement
  • ipykernel: 5.1.1
  • jupyter: 1.0.0
  • python-language-server: 0.28.0
  • vulture: 1.0
  • xlwt: 1.3.0
Code Documentation
  • m2r: 0.2.1
  • pycallgraph: 1.0.1 (pip install pycallgraph)
  • sphinx: 2.1.2
  • sphinx_rtd_theme: 0.4.3
Executable Creation
  • bumpversion: 0.5.3
  • pyinstaller: 3.4

Creating an environment for LILAC on MacOS on Ubuntu using Conda

  • tl;dr
    • conda create --name Lilac36 bumpversion=0.5.3 cython=0.29.12 ipykernel=5.1.1 jupyter=1.0.0 m2r=0.2.1 matplotlib=3.1.1 mkl=2019.4 numpy=1.16.4 pandas=0.24.2 pyinstaller=3.4 PySide2=5.9.0a1 python=3.6.8 python-language-server=0.28.0 scipy=1.3.0 sphinx=2.1.2 sphinx_rtd_theme=0.4.3 vulture=1.0 xlwt=1.3.0 -c conda-forge
    • conda activate Lilac36
    • pip install pycallgraph
    • From the \chemics folder from within the repo run python fast_dp_setup.py build_ext --inplace
    • From the \chemics folder from within the repo run python main.py

Creating an environment for LILAC on Windows using Conda

  • Install Miniconda
    • Downloaded install exec: https://repo.anaconda.com/miniconda/Miniconda3-py39_4.9.2-Windows-x86_64.exe
    • Ran install exec:
      • [Can determine the options I used if you need it]
    • Install Build Tools for Visual Studio (Tools for Visual Studio -> Build Tools) https://visualstudio.microsoft.com/downloads/
      • Select Desktop development with C++
    • Restarted the "computer"
  • Install git and pip - make ssh key for gitlab
    • conda install -c anaconda git
    • conda install -c anaconda pip
    • ssh-keygen and all that
  • Using -c inline Channel command
    • Commands are entered using the Anaconda Prompt (miniconda3) [Not the powershell version]
    • Create the environment
    • conda create --name Lilac36 bumpversion=0.5.3 cython=0.29.12 ipykernel=5.1.1 jupyter=1.0.0 m2r=0.2.1 matplotlib=3.1.1 mkl=2019.4 numpy=1.16.4 pandas=0.24.2 pyinstaller=3.4 PySide2=5.9.0a1 python=3.6.8 python-language-server=0.28.0 scipy=1.3.0 sphinx=2.1.2 sphinx_rtd_theme=0.4.3 vulture=1.0 xlwt=1.3.0 -c conda-forge
    • Activate
      • conda activate Lilac36
    • Install pycallgraph
      • pip install pycallgraph
    • Clone
      • git clone git@gitlab.bucknell.edu:lilac/LILAC.git
      • cd LILAC
    • Build the cython file
      • From the \chemics folder from within the repo run python fast_dp_setup.py build_ext --inplace
    • Run chemics
      • From the \chemics folder from within the repo run python main.py

Install redo 6/11 due to weird pyinstaller things

Using c inline Channel command

  • Install new environment - same as before but with new packages

  • Adding all the following makes it work:

    • m2w64-gcc-libs=5.3.0
    • msys2-conda-epoch=20160418
    • numpy-base=1.16.4
    • asn1crypto=0.24.0
    • smmap2=2.0.5
    • mkl_fft=1.0.12 (6)
    • mkl_random=1.0.2 (7)
    • gitdb2=2.0.5
    • gitpython=3.0.2
    • python-gitlab=1.12.1

  • conda create --name Lilac36Test9 bumpversion=0.5.3 cython=0.29.12 ipykernel=5.1.1 jupyter=1.0.0 m2r=0.2.1 matplotlib=3.1.1 mkl=2019.4 numpy=1.16.4 pandas=0.24.2 pyinstaller=3.4 PySide2=5.9.0a1 python=3.6.8 python-language-server=0.28.0 scipy=1.3.0 sphinx=2.1.2 sphinx_rtd_theme=0.4.3 vulture=1.0 xlwt=1.3.0 -c conda-forge

  • Activate

    • conda activate Lilac36Test9

  • Install pycallgraph

    • pip install pycallgraph

  • Assuming already cloned and cython file has already been built

  • Run chemics

    • cd LILAC\chemics
    • From the \chemics folder from within the repo run python main.py

  • Build EXE

    • cd ..\configs
    • pyinstaller chemics_win.spec --workpath ../chemics/build --distpath ../chemics/dist --clean

Problems?

Problems building fast_dp_calculator

cl.exe missing [Windows]

  • Install Build Tools for Visual Studio (Tools for Visual Studio -> Build Tools)
  • Restart the computer. This should add cl.exe into your PATH, otherwise you may need to manually add it.