Power BI Reusable Workflows

This repository contains reusable workflows in order to set up DevOps for Power BI

For a quickstart/example of the use of these workflows check out Power-BI-Sample-VC-Repo.

Available Workflows:

Deserialize PBIX

Unzips PBIX files into their report JSON's for version control and peer review.

Example workflow:

name: Reformat Power BI Assets
on: [pull_request]
jobs:
  Deserialize-PBIX-Files:
    uses: nathan-giusti/Power-BI-Reusable-Workflows/.github/workflows/deserialize-pbix.yml@main
    with:
      commit_message: "Deserialize PBIX files"

Input	type	required	default	description
commit_message	string	false	"Deserialize PBIX Files"	Commit message to be used when writing JSON files to repository.
separator	string	false	","	Character used to separate file names. Ensure that a character is selected that is never used in file/folder names.

Workspace PBIX Deploy

Deploys a PBIX file to a workspace.

Place PBIX files in one folder per workspace. Use the config file to map the folder names to workspace ids.

Example workflow:

name: Deploy to workspace
on: [pull_request]
jobs:
  Deserialize-PBIX-Files:
    uses: nathan-giusti/Power-BI-Reusable-Workflows/.github/workflows/workspace-deploy.yml@main
    with:
      tenant_id: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
      config_file: .github/config/deploy_config.yaml
    secrets:
      client_id: ${{ secrets.CLIENT_ID }}
      client_secret: ${{ secrets.CLIENT_SECRET }} 

Example config:

Workspace_1:
  workspace_id: 'f2424abd-e0d6-4305-bcae-2caa80db8639'
Workspace_2:
  workspace_id: '1a19ec23-2ba0-450b-9935-def6ea465d1d'
Workspace_3:
  workspace_id: 'b5151f96-3b01-4b17-bc95-cce8630832ac'

Input	type	required	default	description
client_id	string	true		Client id for Azure service princple.
client_secret	string	true		Client secret for Azure service princple.
tenant_id	string	true		Tenant id of your instance.
config_file	string	true		Location of config file.
separator	string	false	","	Character used to separate file names. Ensure that a character is selected that is never used in file/folder names.

Pipeline PBIX Deploy

Deploys a PBIX file via a pipeline.

Place PBIX files in one folder per pipeline. Use the config file to map the folder names to pipeline ids.

Example workflow:

name: Deploy via pipeline
on:
  push:
    branches:
      - main
jobs:
  pipeline-deploy:
    uses: nathan-giusti/Power-BI-Reusable-Workflows/.github/workflows/pipeline-deploy.yml@main
    with:
      tenant_id: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
      config_file: .github/config/deploy_config.yaml
      source_stage_order: 1
    secrets:
      client_id: ${{ secrets.CLIENT_ID }}
      client_secret: ${{ secrets.CLIENT_SECRET }} 

Example config:

Workspace_1:
  pipeline_id: 'xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
Workspace_2:
  pipeline_id: 'xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
Workspace_3:
  pipeline_id: 'xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'

Input	type	required	default	description
client_id	string	true		Client id for Azure service princple.
client_secret	string	true		Client secret for Azure service princple.
tenant_id	string	true		Tenant id of your instance.
config_file	string	true		Location of config file.
source_stage_order	number	true		0 to deploy dev to test, 1 to deploy test to prod
update_app_in_target_workspace	boolean	false	false	True to update the app on deploy
allow_purge_data	boolean	false	false	Should not be needed because this is for reports but fail safe
separator	string	false	","	Character used to separate file names. Ensure that a character is selected that is never used in file/folder names.

Note for Pipeline and Workspace Deploy

Config files can be combined with an entry for workspace_id and pipeline_id for each folder.

Example config:

Hero_Analytics:
  pipeline_id: 'xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
  workspace_id: 'xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
Supply_Chain_Analytics:
  pipeline_id:  'xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
  workspace_id: 'xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
BI360:
  pipeline_id:  'xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
  workspace_id: 'xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'

Tabular Editor Datamodel Deploy

Deploys a dataset saved in Tabular Editor folder format to a workspace.

If all datasets are to be deployed to the same URL, you can provide that workspace URL and leave out config file. If datasets need to be deployed to multiple workspaces, you can provide a config file mapping the folders to the workspace URLs. You must provide either config_file or db_url.

If a config file is set, the folder structure for datasets is assumed to be

./<workspace_name>/<dataset_name>

If no config file is sett, the folder structure for datasets is assumed to be

.<dataset_name>

Datasets cannot be in the same folder as any other json files, including deserialized pbix files. To place datasets in a separate folder for reports we suggest

./reports/<workspace_name>/<report_pbix> ./datasets/<workspace_name>/<report_pbix>

To accomplish this you can set the file option to datasets or to whatever path leads to your dataset folder.

Example workflow:

name: Deploy to dataset
on: [pull_request]
jobs:
  deploy_to_test:
    uses: nathan-giusti/Power-BI-Reusable-Workflows/.github/workflows/te-deploy.yml@main
    with:
      tenant_id: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
      config_file: .github/config/test.yaml
      deploy_roles: true
      deploy_partitions: true
    secrets:
      client_id: ${{ secrets.CLIENT_ID }}
      client_secret: ${{ secrets.CLIENT_SECRET }}

Example config:

"Workspace1": "powerbi://api.powerbi.com/v1.0/myorg/Workspace1"
"Workspace2": "powerbi://api.powerbi.com/v1.0/myorg/Workspace2"

Input	type	required	default	description
client_id	string	true		Client id for Azure service princple.
client_secret	string	true		Client secret for Azure service princple.
tenant_id	string	true		Tenant id of your instance.
db_url	string	false		Workspace to deploy to
te_url	string	false		URL to download TE from. Pulls from most recent by default
folder	boolean	false		The folder containing the datasets
deploy_roles	boolean	false	false	True to deploy roles
deploy_partitions	boolean	false	false	True to deploy partitions
separator	string	false	","	Character used to separate file names. Ensure that a character is selected that is never used in file/folder names.
since_last_remote_commit	boolean	false	true	Use the last commit on the remote branch as the base_sha

Known Issues

All changes files are passed to the command line. If there are too many changed files, the program will crash. This can be fixed by setting since_last_remote_commit to true, but can cause issues if multiple datasets are being modified in the same pull request.