List or generate aggregate data using ctbk aggregate
#6
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: List or generate aggregate data using `ctbk aggregate` | |
| on: | |
| workflow_dispatch: | |
| inputs: | |
| # Combined a few sets of flags into single `choice` inputs, to avoid "you may only define up to 10 `inputs` for a `workflow_dispatch` event": https://github.com/neighbor-ryan/ctbk.dev/actions/runs/5954742040 | |
| time: | |
| description: "Group by time-period granularity" | |
| type: choice | |
| options: | |
| - year | |
| - month | |
| - day | |
| - hour | |
| weekday: | |
| description: "Group by \"weekday\"" | |
| type: boolean | |
| region: | |
| description: "Group by \"region\"" | |
| type: boolean | |
| gender: | |
| description: "Group by \"gender\"" | |
| type: boolean | |
| user_type: | |
| description: "Group by \"user_type\"" | |
| type: boolean | |
| rideable_type: | |
| description: "Group by \"rideable_type\"" | |
| type: boolean | |
| stations: | |
| description: "Group by station start or end" | |
| type: choice | |
| options: | |
| - "Start station" | |
| - "End station" | |
| - "Both" | |
| aggregate_by: | |
| description: "Aggregate by \"count\", \"duration\", or both" | |
| type: choice | |
| options: | |
| - "Ride count" | |
| - "Duration" | |
| - "Both" | |
| date_range: | |
| description: "Date range (e.g. \"2022-\", \"202302-202304\"). Default \"-\" is \"all\" (same as \"201306-\")." | |
| type: string | |
| default: "-" | |
| subcommand: | |
| description: "Subcommand" | |
| type: choice | |
| options: | |
| - "List S3 URLs" | |
| - "Create S3 URLs" | |
| jobs: | |
| dispatch: | |
| name: List or generate aggregate data using `ctbk aggregate` | |
| runs-on: ubuntu-latest | |
| steps: | |
| - uses: actions/checkout@v3 | |
| - uses: actions/setup-python@v4 | |
| with: | |
| python-version: 3.11 | |
| cache: 'pip' | |
| - run: pip install -e . | |
| - env: | |
| AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }} | |
| AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }} | |
| AWS_DEFAULT_REGION: us-east-1 | |
| run: | | |
| # Build "-g" ("Group keys") arg | |
| time=${{ inputs.time }} | |
| if [ $time == year ]; then | |
| g=y | |
| elif [ $time == month ]; then | |
| g=ym | |
| elif [ $time == day ]; then | |
| g=ymd | |
| elif [ $time == hour ]; then | |
| g=ymdh | |
| fi | |
| g=$g${{ inputs.weekday && 'w' || '' }} | |
| g=$g${{ inputs.region && 'r' || '' }} | |
| g=$g${{ inputs.gender && 'g' || '' }} | |
| g=$g${{ inputs.user_type && 't' || '' }} | |
| g=$g${{ inputs.rideable_type && 'b' || '' }} | |
| stations=${{ inputs.stations }} | |
| if [ $stations == "Start station" ]; then | |
| s=s | |
| elif [ $stations == "End station" ]; then | |
| s=e | |
| elif [ $stations == "Both" ]; then | |
| s=se | |
| fi | |
| g=$g$s | |
| # Build "-a" ("Aggregation keys") arg | |
| agg_by="${{ inputs.aggregate_by }}" | |
| if [ "$agg_by" == "Ride count" ]; then | |
| a=c | |
| elif [ "$agg_by" == "Duration" ]; then | |
| a=d | |
| elif [ "$agg_by" == "Both" ]; then | |
| a=cd | |
| fi | |
| # Subcommand arg | |
| if [ "${{ inputs.subcommand }}" == "List S3 URLs" ]; then | |
| subcmd=urls | |
| else | |
| subcmd=create | |
| fi | |
| # Run (list URLs, or generate them) | |
| ctbk --s3 aggregated -d${{ inputs.date_range }}-${{ inputs.end }} -g$g -a$a $subcmd |