Ar pluto new fields p3

products/pluto/models/_sources.yml

@@ -67,6 +67,18 @@ sources:
     description: Spatial overlaps between lots and MIH areas (from miharea.sql)
   - name: mih_distinct_options
     description: QAQC view of all distinct MIH options found in the data (from miharea.sql)
+    columns:
+    - name: option
+      description: MIH option name
+      data_tests:
+      - accepted_values:
+          values: ['Option 1', 'Option 2', 'Option 3', 'Deep Affordability Option', 'Workforce Option']
+          config:
+            severity: error
+            tags: ['de_check', 'minor', 'major']
+            meta:
+              description: Only the five valid MIH options should appear in the source data. Any additional options indicate a source data issue that needs to be investigated.
+              next_steps: Contact GIS to investigate unexpected MIH option values in source data
   - name: transit_zones_block_to_tz_ranked
     description: Ranked transit zone assignments by block (from transitzone.sql)
   - name: transit_zones_bbl_to_tz_ranked

products/pluto/pluto_build/sql/transitzone.sql

@@ -30,7 +30,7 @@
 --      lot-by-lot calculation when block-level assignment would be misleading.
 
 -- Create decomposed transit zones table (break multipolygons into individual parts)
-DROP TABLE IF EXISTS transit_zones_atomic_geoms;
+DROP TABLE IF EXISTS transit_zones_atomic_geoms CASCADE;
 CREATE TABLE transit_zones_atomic_geoms AS
 WITH decomposed AS (
     SELECT
@@ -52,7 +52,7 @@ CREATE INDEX idx_transit_zones_atomic_geoms_gix ON transit_zones_atomic_geoms US
 -- AR Note: I tried a few approaches for this, and perhaps there's a more clever/performant
 -- way to accomplish this. Unfortunately, the recommend approach of ST_ClusterDBSCAN
 -- will `sometimes` accomplish this, but it errors out seemingly randomly.
-DROP TABLE IF EXISTS transit_zones_tax_blocks;
+DROP TABLE IF EXISTS transit_zones_tax_blocks CASCADE;
 CREATE TABLE transit_zones_tax_blocks AS
 WITH block_unions AS (
     SELECT
@@ -89,7 +89,7 @@ WITH block_unions AS (
         ON
             np.borough = p.borough
             AND np.block = p.block
-            AND ST_WITHIN(ST_CENTROID(p.geom), np.geom)
+            AND ST_WITHIN(ST_POINTONSURFACE(p.geom), np.geom)
     GROUP BY np.borough, np.block, np.sub_block, np.geom
 )
 SELECT
@@ -104,7 +104,7 @@ CREATE INDEX idx_transit_zones_tax_blocks_geom ON transit_zones_tax_blocks USING
 
 
 -- Step 1: Calculate coverage percentages for all tax blocks
-DROP TABLE IF EXISTS transit_zones_block_to_tz_ranked;
+DROP TABLE IF EXISTS transit_zones_block_to_tz_ranked CASCADE;
 CREATE TABLE transit_zones_block_to_tz_ranked AS
 WITH block_to_tz AS (
     SELECT
@@ -140,7 +140,7 @@ ANALYZE transit_zones_block_to_tz_ranked;
 
 
 -- For ambiguous blocks (those with competing transit zones), create lot-level assignments
-DROP TABLE IF EXISTS transit_zones_bbl_to_tz_ranked;
+DROP TABLE IF EXISTS transit_zones_bbl_to_tz_ranked CASCADE;
 CREATE TABLE transit_zones_bbl_to_tz_ranked AS
 WITH ambiguous_bbls AS (
     SELECT
@@ -221,13 +221,11 @@ FROM (
         AND NOT EXISTS (
             SELECT 1 FROM transit_zones_block_to_tz_ranked AS ambiguous
             WHERE
-                ambiguous.borough = block_tz.borough
-                AND ambiguous.block = block_tz.block
+                ambiguous.id = block_tz.id
                 AND ambiguous.tz_rank = 2
+                AND ambiguous.pct_covered > 10
         )
-
     UNION ALL
-
     -- Lot-level assignments for ambiguous blocks
     SELECT
         bbls[1] AS bbl,