v3.1.4
Pre-release
Pre-release
Develop (#82) * Feature/update maap help tour (#61) updated maap help jupyter extension to version 1.0.0 * Bugs/presigned url (#62) Main fix for the presigned url bug was changing 'maapsec/environment' to 'jupyter-server-extension/getConfig' There was another bug caused by a race condition that was causing users to not be able to get the presigned url about every 10th time. What fixed this bug is calling updateKeycloakToken upon the launch of the extension and also trying to call updateKeycloakToken again if getting the keycloak user information fails to get I confirmed this bug was fixed by trying to get the presigned url 25 times and it worked Successful build in vanilla here: https://repo.dit.maap-project.org/root/maap-workspaces/-/pipelines/3389 * Added memory extension (#59) added changes from cluttered commit branch * hysds base-image [skip ci] (#65) * Feature/libmamba solver (#64) * upgraded maap help jupyter extension version to 1.0.0 * reverted back changes * changed r docker to use libmamba solver * changed mamba install to conda * converted all images use libmamba solver like R image * added lang tz and debian frontend commands to vanilla docker image * copied pangeo dockerfile into vanilla * first line of packages back to vanilla versions * updated package versions for vanilla to be same as pangeo * updated isce2 package versions to match vanilla * updating continuumio/miniconda3 for rgedi and making other 3 top lines consistent * removed changes from rgedi * Update stac_ipyleaflet to 0.3.5 for Pangeo (#66) * Updated pinned version of stac_ipyleaflet to 0.3.5 for pangeo jupyter workspace. * Added environment variables to Pangeo environment yml. * Updated entrypoint bash script so the the base environment is activated before launching jupyter so that environment variables are accessible within jupyter. --------- Co-authored-by: Alex <[email protected]> * Remove hysds base_image [skip-ci] (#68) * Bugs/quick fix libmamba (#69) * removed libmamba sovler from conda env update * added solver to environment.yml * added conda libmamba solver version * added back in solvers and pinned version for conda-libmamba-solver * added back in space * Bugs/presigned url (#70) * changed dit url to test * changed dit back and changed ops to be maap-ops-workspace instead of dataset * Bugs/git tour link (#71) * manual build for change to maap help * updated npm version of maap help * updated custom ext versions (#72) * Feature/update maap py (#73) upgraded all maap py versions to 3.1.2 * bumped jupyter server extension (#74) * update server ext * bumped server ext * Bump maap-py to v3.1.3 (#75) * Update Dockerfile * Bugs/libmamba default (#77) * added conda config setting libmamba solver to entrypoint.sh * updated maap help to version 1.0.1 * converted back to 1.0.0 * trying to create .condarc file * removed touch because might work without it * upgraded conda libmamba solver * undid change because forgot I had PR open for this branch * added --system tag and removed conda config from entrypoint * added the system tag to all base image dockerfiles * Update entrypoint.sh (#78) * Bugs/double submit null username (#79) * updated maap help to version 1.0.1 * converted back to 1.0.0 * upgraded dps version to add feature disabling submit button/ saying last submitted info and fixing the null username bug * Features/update workspace dependencies (#80) * Updated how conda packages installs are handled at the base image layer for Pangeo to use an environment.yml file like we do for its respective jupyterlab image. * Updated to miniconda 23.5.2-0 for the pangeo base image. * Updated list of packages for the base and jupyter images for the Pangeo Workspace. * updated maap help to version 1.0.1 * converted back to 1.0.0 * Clean up of pangeo base environment.yml * Reverted the miniconda version of base pangeo image to 22.11.1. * Updated jupyterlab3 pangeo docker conda solver parameter * Modified base pangeo image conda env update solver parameter spec * explicility installing libarchive fron conda-forge to test resolution of issue with libmamba. * Testing base pangeo conda install parameter updates. * Test builds using new version of miniconda. * Test another build configuration on gitlab using new pinned version of conda-libmamba-solver * Bumped miniconda version to 23.10.0-1 and removed conda-libmamba-solver since libmamba is the default solver in this new version of miniconda. * Updated Pangeo base image and environment yml so that packages are installed in a new environment named with the workspace type to resolve issues with libarchive/libmamba and the installation of packages. * Updated Pangeo jupyterlab dockerfile and environment yml to use new custom environment instead of base environment. * Moved a couple jupyter python packages from dockerfile to the environment yml. * Updated entrypoint to test new environment variable that stores the type of workspace being run to ensure jupyter can be launched appropriately. * Patch issue with ARG values not carrying over to downstream build stages in docker file. Also replaced use of BUILD_IMAGE_TYPE with ENV var, "WORKSPACE_TYPE". * Pinned jupyterlab package to 3.6.1 to fix build issue with jupyterlab_plotly. * Pinned additional packages to assist with resolving jupyterlab build issues. * Commented line "conda init" in entrypoint.sh. * Reverted commenting out "conda init" in entrypoint that was previously done for testing. * Testing changes "~/.bash_profile" to properly set the default conda environment for terminals. * Updated path variable to point to our custom conda environment bin instead of the conda base env bin path. * Updated how the default environment is activated for terminals. * removed jupyter-server from envirionment yml as the jupyterlab installation takes care of installing the jupyter_server package already. * Added Makefile which contains commonly run recipes when modifying workspaces locally and needing to test building them. * Updated notebook lib path and jupyter server lib path to point to the appropriate custom conda environment. * current attempt at fixing r workspace issues, building for base image but not for jupyterlab3 * r image building successfully locally * added conda lists * Added perl package to r base environment.yml to resolve issue with r-tinytex not installing. * Added string replacement commands to remove conda env related commands from .bash_profile that are no longer needed. * Updated jupyterlab3's dockerfile so that it modifies the "jupyterlab3/shared/environment.yml" by doing a string replacement on the workspace environment name for workspaces that don't have a custom environment.yml. * r image building locally * pinned versions * updated pinned versions * changed version for r-dplyr * new botocore version * added aiobotocore dependency * removed aiobotocore from install * removed botocore install * added half of the packages to see if I can get rid of the botocore warning * removed some more environment.yml packages because still getting botocore error * removed even more packages because still getting botocore warnining * removed more packages because still getting botocore warning * Added conda list to dockerfiles for testing on ADE. * added maap py into pip install from source * updated git install from source * added v to version tag for maappy * added a bunch of conda lists trying to debug * installing maap py to root * corrected installing into root bug * added run pwd line to try to debug * tried setting the working directory to root before git clone * tried adding another pwd command * removed . from git clone * set working directory to maap py * added back in . for git clone * Updated packages for base and jupyterlab reported by data team that are needed in Python based workspaces. * Updated the folder parameter for where maap-py is installed to hopefully resolve maap-py installation issues. * Moved maap py conf environment variable to base environment yml. * commented out some of the code trying to test maap py again * commented back in code before adding in Anils code that worked * cleaning up jupyterlab3 dockerfile so is consistent * removed extra ; * removed pip install of maap py * updated botocore version in base environment.yml * added binary library files to r install * Removed commented ENV command in pangeo dockerfile. * Updated miniconda version for base vanilla image. * Updated base vanilla image to use an environment yml to store it's configuration environment. Vanilla base image now relies on a custom vanilla conda environment. * Updated jupyterlab shared environment yml so that 1) missing packages needed by data team are included, 2) each workspace uses a custom conda environment, and 3) updated entrypoint so that .bash_profile is cleaned up. * added r-car * pinned r-car version and removed botocore3 * added maap py to environment.yml to avoid botocore dependency errors * Added help documentation to Makefile along with other minor improvements. * commented out git clone for r image * Pinned pangeo packages and removed duplicates. * Pinned packages for vanilla and shared workspace environment ymls. Added missing jupyter package to shared environment yml. * Bumped stac_ipyleaflet to version 0.3.6 for pangeo workspace. * created isce3 environment.yml and pinned r binary library versions * fixed name of fsspec * added isce3 to the jupyterlab3 dockerfile * removed dependencies from isce3 environment.yml * added back in dependencies and removed environment.yml for isce3 in jupyter image * removed packages specific to isce3 * added in 3 packages but isce3 to see if that is the problem * added back in isce3 * removed all vanilla packages and just trying to install base isce3 packages * pinned python version * attempting to install r packages via script * pinned versions for fsspec, hvplot, isce3, xarray * added and modified r scripts and removed awscli trying to fix maappy dependency error * trying to add awscli after maappy install * capitalized Fgmutils * added packages to environment.yml * changed maappy branch to bugs/dependency issues * upgraded awscli version * unpinned versions for vanilla environment.yml * removed all r packages except python and r * unpinned python version * changed back to v3.1.3 tag for maap-py * downgraded awscli to 2.14.1 * upgraded awscali to 2.14.2 * removed pinnection vertsions for cryptography and prompt-toolkit * r image should be building now, doesnt include Fgmutils * unpinned versions to fix maap-py bug in vanilla workspace * pinned vanilla versions and changed to dependecies branch for pangeo and isce3 * changed branch for maappy to version with ipython pinned * pinned botocore version and added conda list * moved conda list command and just pushed maappy pinning boto3=1.33.13 * removed botocore from r/environment.yml * changed to correctly use repos and removed print statements * fixed but for Fgmutils install * pinned package versions that were dependency of Fgmutils * added r channel to r environment.yml * downgraded version of r-devemf * downgraded r and r-base to 4.2.3 to be compatible with r-devemf * changed to r version 4.2 * downgraded r essentials to 4.2 * Added geocube to vanilla base image which was added to pangeo but accidentally omitted from vanilla. * Added additional packages requested by user group for python based workspaces. * Removed pins for vanilla and pangeo base image to test new image builds on GitLab with additional packages. * Updated pinned versions for vanilla base image as well as shared environment yml for jupyterlab. Also bumped jupyterlab to 3.6.3 and added missing package, rio-tiler. * Added pinned version information for pangeo base and jupyterlab images. Also modified version information for jupyterlab to 3.6.3 to resolve version descrepancy with pangeo-notebook package. * Removed pin to test installation of the morecantile package. * Fixed typo of morecantile package in pangeo base image. * Repinned morecantile package for pangeo and vanilla base images. * removed print statements and conda lists * added conda config lines just for testing * added descriptions necessary to use install2.r with GPL license * removed lines making conda forge default because that should be a different PR * updated all maap-py branches to be new 3.1.4 release --------- Co-authored-by: Anil Natha <[email protected]> * Feature/conda forge default (#81) * updated maap help to version 1.0.1 * converted back to 1.0.0 * added in lines that set conda forge as the default and create .condarc * fixed typo --------- Co-authored-by: Grace Llewellyn <[email protected]> Co-authored-by: bsatoriu <[email protected]> Co-authored-by: Anil Natha <[email protected]> Co-authored-by: Alex <[email protected]> Co-authored-by: Sujen Shah <[email protected]> Co-authored-by: Anil Natha <[email protected]>