add cdbg data product

.github/workflows/build.yml

@@ -11,6 +11,7 @@ on:
         options:
           - template
           - cbbr
+          - cdbg
           - checkbook
           - colp
           - cpdb
@@ -128,6 +129,17 @@ jobs:
       logging_level: ${{ inputs.logging_level }}
       build_note: ${{ inputs.build_note }}
       dev_bucket: ${{ inputs.dev_bucket && format('de-dev-{0}', inputs.dev_bucket) || '' }}
+  cdbg:
+    needs: health_check
+    if: inputs.dataset_name == 'cdbg' || inputs.dataset_name  == 'all'
+    uses: ./.github/workflows/cdbg_build.yml
+    secrets: inherit
+    with:
+      image_tag: ${{ needs.health_check.outputs.tag }}
+      recipe_file: ${{ inputs.recipe_file }}
+      build_name: ${{ needs.health_check.outputs.build_name }}
+      plan_command: ${{ needs.health_check.outputs.plan_command }}
+      dev_bucket: ${{ inputs.dev_bucket && format('de-dev-{0}', inputs.dev_bucket) || '' }}
   checkbook:
     needs: health_check
     if: inputs.dataset_name == 'checkbook' || inputs.dataset_name  == 'all'

.github/workflows/cdbg_build.yml

@@ -0,0 +1,76 @@
+name: CDBG - 🏗️ Build
+on:
+  workflow_call:
+    inputs:
+      image_tag:
+        type: string
+        required: false
+      build_name:
+        type: string
+        required: true
+      recipe_file:
+        type: string
+        required: true
+      plan_command:
+        type: string
+        default: recipe
+      dev_bucket:
+        type: string
+        required: false
+
+jobs:
+  build:
+    name: Build CDBG
+    runs-on: ubuntu-22.04
+    defaults:
+      run:
+        shell: bash
+        working-directory: products/cdbg
+    container:
+      image: nycplanning/build-base:${{ inputs.image_tag || 'latest' }}
+    env:
+      BUILD_ENGINE_DB: db-cdbg
+      BUILD_NAME: ${{ inputs.build_name }}
+      RECIPES_BUCKET: ${{ inputs.dev_bucket || 'edm-recipes' }}
+      PUBLISHING_BUCKET: ${{ inputs.dev_bucket || 'edm-publishing' }}
+      DEV_FLAG: ${{ inputs.dev_bucket && 'true' || 'false' }}
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Load Secrets
+        uses: 1password/load-secrets-action@v1
+        with:
+          export-env: true
+        env:
+          OP_SERVICE_ACCOUNT_TOKEN: ${{ secrets.OP_SERVICE_ACCOUNT_TOKEN }}
+          AWS_S3_ENDPOINT: "op://Data Engineering/DO_keys/AWS_S3_ENDPOINT"
+          AWS_SECRET_ACCESS_KEY: "op://Data Engineering/DO_keys/AWS_SECRET_ACCESS_KEY"
+          AWS_ACCESS_KEY_ID: "op://Data Engineering/DO_keys/AWS_ACCESS_KEY_ID"
+          BUILD_ENGINE_SERVER: "op://Data Engineering/EDM_DATA/server_url"
+          BUILD_ENGINE_HOST: "op://Data Engineering/EDM_DATA/server"
+          BUILD_ENGINE_USER: "op://Data Engineering/EDM_DATA/username"
+          BUILD_ENGINE_PASSWORD: "op://Data Engineering/EDM_DATA/password"
+          BUILD_ENGINE_PORT: "op://Data Engineering/EDM_DATA/port"
+
+      - name: Setup build environment
+        working-directory: ./
+        run: |
+          ./bash/docker_container_setup.sh
+          ./bash/build_env_setup.sh
+
+      - name: Plan build
+        run: python3 -m dcpy.lifecycle.builds.plan ${{ inputs.plan_command }}
+
+      - name: Dataloading
+        run: python -m dcpy.lifecycle.builds.load load --recipe-path ${{ inputs.recipe_file }}.lock.yml
+
+      - name: Build
+        run: |
+          dbt debug
+          dbt build
+
+      - name: Export
+        run: ./bash/export.sh
+
+      - name: Upload
+        run: python3 -m dcpy.connectors.edm.publishing upload --product db-cdbg --acl public-read

.github/workflows/test_helper.yml

@@ -88,6 +88,7 @@ jobs:
         project:
           - green_fast_track
           - zoningtaxlots
+          - cdbg
     steps:
       - uses: actions/checkout@v4
       - name: setup

dcpy/lifecycle/ingest/templates/hud_lowmodincomebyblockgroup.yml

@@ -0,0 +1,60 @@
+id: hud_lowmodincomebyblockgroup
+acl: public-read
+
+attributes:
+  name: HUD Low to Moderate Income Population by Block Group
+  description: >-
+    This particular version of this dataset has come from OMB. We haven't fully assessed whether
+    we could just pull from the linked feature service instead and filter to NYC or if OMB has
+    done any other preprocessing.
+
+    This service identifies U.S. Census Block Groups in which 51% or more of the households earn 
+    less than 80 percent of the Area Median Income (AMI). The Community Development Block Grant 
+    (CDBG) program requires that each CDBG funded activity must either principally benefit low- 
+    and moderate-income persons, aid in the prevention or elimination of slums or blight, or meet 
+    a community development need having a particular urgency because existing conditions pose a 
+    serious and immediate threat to the health or welfare of the community and other financial 
+    resources are not available to meet that need. With respect to activities that principally 
+    benefit low- and moderate-income persons, at least 51 percent of the activity's beneficiaries 
+    must be low and moderate income.
+
+    The Community Development Block Grant (CDBG) program requires that each CDBG funded activity 
+    must either principally benefit low- and moderate-income persons, aid in the prevention or 
+    elimination of slums or blight, or meet a community development need having a particular urgency 
+    because existing conditions pose a serious and immediate threat to the health or welfare of 
+    the community and other financial resources are not available to meet that need. With respect 
+    to activities that principally benefit low- and moderate-income persons, at least 51 percent 
+    of the activity's beneficiaries must be low and moderate income. For CDBG, a person is considered 
+    to be of low income only if he or she is a member of a household whose income would qualify as 
+    "very low income" under the Section 8 Housing Assistance Payments program. Generally, these 
+    Section 8 limits are based on 50% of area median. Similarly, CDBG moderate income relies on 
+    Section 8 "lower income" limits, which are generally tied to 80% of area median. These data 
+    are from the 2011-2015 American Community Survey (ACS).
+  url: https://hudgis-hud.opendata.arcgis.com/datasets/HUD::low-to-moderate-income-population-by-block-group/about
+
+ingestion:
+  source:
+    type: s3
+    bucket: edm-recipes
+    key: inbox/omb/20241227/ACS-2020-Low-Mod-Summarized-All-Block-Groups-2023.csv
+  file_format:
+    type: csv
+
+#columns:
+#- id: CDBGUOGID
+#- id: GEOID
+#- id: CDBGNAME
+#- id: STUSAB
+#- id: STATE
+#- id: CDBGTYPE
+#- id: Geoname
+#- id: COUNTY
+#- id: TRACT
+#- id: BLKGRP
+#- id: LOW
+#- id: LOWMOD
+#- id: LMMI
+#- id: LOWMODUNIV
+#- id: LOWMOD_PCT
+#- id: MOE_LOWMODPCT
+#- id: Column1

products/cdbg/bash/export.sh

@@ -0,0 +1,22 @@
+#!/bin/bash
+source ../../bash/utils.sh
+set_error_traps
+
+rm -rf output
+
+echo "Export product tables"
+mkdir -p output && (
+    cd output
+
+    echo "Copy metadata files"
+    cp ../source_data_versions.csv .
+    cp ../build_metadata.json .
+
+    echo "export cdbg_block_groups.csv ..."
+    csv_export cdbg_block_groups cdbg_block_groups
+
+    echo "export cdbg_tracts.csv ..."
+    csv_export cdbg_tracts cdbg_tracts
+)
+
+zip -r output/output.zip output

products/cdbg/dbt_project.yml

@@ -0,0 +1,21 @@
+name: "cdbg"
+
+profile: "dcp-de-postgres"
+
+model-paths: ["models"]
+
+tests:
+  +store_failures: true
+  schema: "_tests"
+
+models:
+  cdbg:
+    staging:
+      +materialized: view
+    intermediate:
+      +materialized: table
+    product:
+      +materialized: table
+
+flags:
+  fail-fast: true

products/cdbg/macros/test_sum.sql

@@ -0,0 +1,9 @@
+{% test sum_by(model, group_by, target_column, val, precision=4) %}
+
+SELECT
+    {{ group_by }}, sum({{ target_column }}) AS sum, array_agg( {{target_column }}) AS vals
+FROM {{ model }}
+GROUP BY {{ group_by }}
+HAVING round(sum({{ target_column }})::numeric, {{ precision }}) <> {{ val }}
+
+{% endtest %}

products/cdbg/models/_sources.yml

@@ -0,0 +1,17 @@
+version: 2
+
+sources:
+  - name: recipe_sources
+    schema: "{{ env_var('BUILD_ENGINE_SCHEMA') }}"
+    tables:
+      - name: dcp_mappluto_clipped
+        columns:
+          - name: bbl
+            tests:
+              - not_null
+          - name: wkb_geometry
+            tests:
+              - not_null
+      - name: dcp_cb2020_wi
+      - name: dcp_ct2020_wi
+      - name: hud_lowmodincomebyblockgroup

products/cdbg/models/intermediate/_intermediate_models.yml

@@ -0,0 +1,68 @@
+version: 2
+
+models:
+- name: int__block_groups
+  description: residential area and low-to-moderate income data aggregated by census block group
+  columns:
+  - name: geoid
+    tests: [unique, not_null]
+  - name: borough_name
+  - name: tract
+  - name: block_group
+  - name: total_floor_area
+  - name: residential_floor_area
+  - name: residential_floor_area_percentage
+  - name: total_population
+  - name: lowmod_population
+  - name: lowmod_population_percentage
+
+- name: int__lot_block_groups_details
+  description: int__lot_block_groups joined to pluto for lot info
+  columns:
+  - name: bbl
+  - name: block_group_geoid
+  - name: overlap_ratio
+  - name: bldgarea
+  - name: bldgarea_in_block_group
+  - name: resarea
+  - name: resarea_in_block_group
+
+- name: int__lot_block_groups_raw
+  description: unique intersections of pluto lots and census block groups with proportion of lot in block group
+  columns:
+  - name: bbl
+  - name: block_group_geoid
+  - name: overlap_ratio
+
+- name: int__lot_block_groups
+  description: >-
+    unique intersections of pluto lots and census block groups with proportion of lot in block group,
+    corrected to assign lots fully to block groups that contain 90%+ of a lot
+  columns:
+  - name: bbl
+  - name: block_group_geoid
+  - name: overlap_ratio
+  tests:
+  - dbt_utils.unique_combination_of_columns:
+      combination_of_columns: [bbl, block_group_geoid]
+  - sum_by:
+      group_by: bbl
+      target_column: overlap_ratio
+      val: 1
+      config:
+        severity: warn
+
+- name: int__tracts
+  description: residential area and low-to-moderate income data aggregated by census tract
+  columns:
+  - name: geoid
+    tests: [unique, not_null]
+  - name: borough_name
+  - name: tract
+  - name: block_group
+  - name: total_floor_area
+  - name: residential_floor_area
+  - name: residential_floor_area_percentage
+  - name: total_population
+  - name: lowmod_population
+  - name: lowmod_population_percentage

products/cdbg/models/intermediate/int__block_groups.sql

@@ -0,0 +1,39 @@
+WITH lot_block_groups AS (
+    SELECT * FROM {{ ref("int__lot_block_groups_details") }}
+),
+
+block_groups_income AS (
+    SELECT * FROM {{ ref("stg__low_mod_by_block_group") }}
+),
+
+block_groups_floor_area AS (
+    SELECT
+        block_group_geoid AS geoid,
+        sum(bldgarea_in_block_group) AS total_floor_area,
+        sum(resarea_in_block_group) AS residential_floor_area
+    FROM lot_block_groups
+    GROUP BY geoid
+),
+
+block_group_details AS (
+    SELECT
+        block_groups_floor_area.geoid,
+        block_groups_income.boro AS borough_name,
+        block_groups_income.tract,
+        block_groups_income.block_group,
+        total_floor_area,
+        residential_floor_area,
+        CASE
+            WHEN total_floor_area = 0
+                THEN 0
+            ELSE (residential_floor_area / total_floor_area) * 100
+        END AS residential_floor_area_percentage,
+        block_groups_income.total_population,
+        block_groups_income.lowmod_population AS low_mod_income_population,
+        block_groups_income.lowmod_pct AS low_mod_income_population_percentage
+    FROM block_groups_floor_area
+    LEFT JOIN block_groups_income
+        ON block_groups_floor_area.geoid = block_groups_income.geoid
+)
+
+SELECT * FROM block_group_details

products/cdbg/models/intermediate/int__lot_block_groups.sql

@@ -0,0 +1,35 @@
+WITH lot_block_groups AS (
+    SELECT
+        bbl,
+        block_group_geoid,
+        overlap_ratio
+    FROM {{ ref("int__lot_block_groups_raw") }}
+),
+
+valid_lot_block_groups AS (
+    SELECT * FROM lot_block_groups
+    WHERE overlap_ratio IS NOT null
+),
+
+lots_easy AS (
+    SELECT
+        bbl,
+        block_group_geoid,
+        1 AS overlap_ratio
+    FROM valid_lot_block_groups
+    WHERE overlap_ratio > 0.9
+),
+
+lots_split AS (
+    SELECT *
+    FROM valid_lot_block_groups
+    WHERE bbl NOT IN (SELECT bbl FROM lots_easy)
+),
+
+lots AS (
+    SELECT * FROM lots_easy
+    UNION ALL
+    SELECT * FROM lots_split
+)
+
+SELECT * FROM lots