docker: prevent manual pin of dependencies and improve build speed#95

Merged

tlvu merged 70 commits intomasterfrom prevent-manual-pin-of-dependencies

Jan 27, 2022

+131-58

Commits on Dec 18, 2021

docker: try set channel_priority strict to hopefully increase performance
tlvu
committed
docker: try to pin xclim and ravenpy to hopefully increase performance
tlvu
committed
docker: use --strict-channel-priority with conda env create to be sure
tlvu
committed
docker: remove defaults channel to decrease number of packages for the solver
tlvu
committed
docker: replace conda env create with conda create
tlvu
committed
docker: replace conda create needs --name
tlvu
committed
docker: use --file instead -f for conda create
tlvu
committed
docker: conda create do not work, revert to conda env create
tlvu
committed
docker: install mamba the proper way and use mamba env create
tlvu
committed
docker: unpin xclim and ravenpy to try with mamba
tlvu
committed
docker: use mamba to create conda env and perform 2-stage install to avoid too big dependency resolution
tlvu
committed
docker: need to specify channels for manual mamba create
tlvu
committed
docker: channel conda-forge before defaults, gdal needed by ravenpy not found
tlvu
committed
docker: typo: mamba env install and not just mamba install
tlvu
committed
docker: typo: mamba env update and not mamba env install
tlvu
committed
docker: add vtk-cdat channel for vtk-cdat, needed by vcs
tlvu
committed
Revert "docker: add vtk-cdat channel for vtk-cdat, needed by vcs"
tlvu
committed
docker: remove vcs for weird dependency solve error, will have to investigate later
tlvu
committed
docker: temp change to leverage build caching
tlvu
committed
docker: optimize for cache re-use during dev rebuild cycle
tlvu
committed
docker: try hack to ensure latest xclim and ravenpy
tlvu
committed
docker: mamba install do nothing if already installed, have to use mamba update
tlvu
committed
docker: mamba update did not update either, force install exact version
tlvu
committed
docker: try pin xclim and ravenpy immediately at the beginning to see if it prevent the downgrade
tlvu
committed
docker: even with pin in initial install, still getting downgrade so try to pin in environment.yml as well
tlvu
committed
docker: pin in environment.yml stopped the downgrading, remove duplicate hardcode pin in Dockerfile
tlvu
committed
docker: conda config --set channel_priority strict probably not needed anymore
tlvu
committed
docker: wonder if conda config --set channel_priority was the cause of the downgrade
tlvu
committed
docker: conda config --set channel_priority strict not responsible for downgrade, restore it since it did help conda perf, restore hardcode pin in environment.yml to avoid downgrade
tlvu
committed
docker: no need to jupyter labextension install @jupyter-widgets/jupyterlab-manager with jupyterlab v3
tlvu
committed
docker: install ipyleaflet using conda for new jupyter lab v3
tlvu
committed
docker: drop jupyterlab-topbar-text and jupyterlab-theme-toggle since not useful and problem with jupyter labextension install command
tlvu
committed
docker: pin python=3.8 for vcs package
tlvu
committed
docker: pin python=3.7 for libgdal
tlvu
committed

Commits on Dec 20, 2021

Commits on Dec 21, 2021

Commits on Jan 11, 2022

Dockerfile.testing: pin shapely for notebook failure, bokeh for performance regression
tlvu
committed

Commits on Jan 13, 2022

Commits on Jan 14, 2022

Commits on Jan 18, 2022

conftest: avoid failing on holoviews display json code change
tlvu
committed

Commits on Jan 21, 2022