fix(seeds): update init_landing_seeds.py to write JSONL format

Old script wrote blob json.gz seeds; staging models now only read jsonl.gz.
Seeds are empty JSONL gzip files — zero rows, satisfies DuckDB file-not-found check.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

transform/sqlmesh_padelnomics/models/staging/stg_population_geonames.sql

@@ -3,11 +3,7 @@
 -- Broad coverage (140K+ locations) enables Gemeinde-level market intelligence.
 -- One row per geoname_id (GeoNames stable numeric identifier).
 --
--- Supports two landing formats (UNION ALL during migration):
---   New: cities_global.jsonl.gz — one city per line, columns directly accessible
---   Old: cities_global.json.gz  — {"rows": [...]} blob (UNNEST required)
---
--- Source: data/landing/geonames/{year}/{month}/cities_global.{jsonl,json}.gz
+-- Source: data/landing/geonames/{year}/{month}/cities_global.jsonl.gz
 
 MODEL (
   name staging.stg_population_geonames,
@@ -16,17 +12,14 @@ MODEL (
   grain geoname_id
 );
 
-WITH
--- New format: one city per JSONL line
-jsonl_rows AS (
 SELECT
   TRY_CAST(geoname_id  AS INTEGER)  AS geoname_id,
-    city_name,
-    country_code,
+  TRIM(city_name)                   AS city_name,
+  UPPER(country_code)               AS country_code,
   TRY_CAST(lat         AS DOUBLE)   AS lat,
   TRY_CAST(lon         AS DOUBLE)   AS lon,
-    admin1_code,
-    admin2_code,
+  NULLIF(TRIM(admin1_code), '')     AS admin1_code,
+  NULLIF(TRIM(admin2_code), '')     AS admin2_code,
   TRY_CAST(population  AS BIGINT)   AS population,
   TRY_CAST(ref_year    AS INTEGER)  AS ref_year,
   CURRENT_DATE                      AS extracted_date
@@ -40,50 +33,8 @@ jsonl_rows AS (
   }
 )
 WHERE geoname_id IS NOT NULL
-),
--- Old format: {"rows": [...]} blob — kept for transition
-blob_rows AS (
-  SELECT
-    TRY_CAST(row ->> 'geoname_id'  AS INTEGER)  AS geoname_id,
-    row ->> 'city_name'                          AS city_name,
-    row ->> 'country_code'                       AS country_code,
-    TRY_CAST(row ->> 'lat'         AS DOUBLE)    AS lat,
-    TRY_CAST(row ->> 'lon'         AS DOUBLE)    AS lon,
-    row ->> 'admin1_code'                        AS admin1_code,
-    row ->> 'admin2_code'                        AS admin2_code,
-    TRY_CAST(row ->> 'population'  AS BIGINT)    AS population,
-    TRY_CAST(row ->> 'ref_year'    AS INTEGER)   AS ref_year,
-    CURRENT_DATE                                 AS extracted_date
-  FROM (
-    SELECT UNNEST(rows) AS row
-    FROM read_json(
-      @LANDING_DIR || '/geonames/*/*/cities_global.json.gz',
-      auto_detect = true,
-      maximum_object_size = 40000000
-    )
-  )
-  WHERE (row ->> 'geoname_id') IS NOT NULL
-),
-all_rows AS (
-  SELECT * FROM jsonl_rows
-  UNION ALL
-  SELECT * FROM blob_rows
-)
-SELECT
-  geoname_id,
-  TRIM(city_name)                AS city_name,
-  UPPER(country_code)            AS country_code,
-  lat,
-  lon,
-  NULLIF(TRIM(admin1_code), '')  AS admin1_code,
-  NULLIF(TRIM(admin2_code), '')  AS admin2_code,
-  population,
-  ref_year,
-  extracted_date
-FROM all_rows
-WHERE population IS NOT NULL
+  AND population IS NOT NULL
   AND population > 0
-  AND geoname_id IS NOT NULL
   AND city_name IS NOT NULL
   AND lat IS NOT NULL
   AND lon IS NOT NULL

web/scripts/init_landing_seeds.py

@@ -1,22 +1,19 @@
-"""Create minimal seed files for SQLMesh staging models that require landing data."""
+"""Create minimal seed files for SQLMesh staging models that require landing data.
+
+Seeds are empty JSONL gzip files — they satisfy DuckDB's file-not-found check
+while contributing zero rows to the staging models.
+"""
 import gzip
-import json
 from pathlib import Path
 
-seed = {
-    "date": "1970-01-01",
-    "captured_at_utc": "1970-01-01T00:00:00Z",
-    "venue_count": 0,
-    "venues_errored": 0,
-    "venues": [],
-}
-morning = Path("data/landing/playtomic/1970/01/availability_1970-01-01.json.gz")
-recheck = Path("data/landing/playtomic/1970/01/availability_1970-01-01_recheck_00.json.gz")
+# stg_playtomic_availability requires at least one morning and one recheck file
+morning = Path("data/landing/playtomic/1970/01/availability_1970-01-01.jsonl.gz")
+recheck = Path("data/landing/playtomic/1970/01/availability_1970-01-01_recheck_00.jsonl.gz")
 morning.parent.mkdir(parents=True, exist_ok=True)
 for p in [morning, recheck]:
     if not p.exists():
-        with gzip.open(p, "wt") as f:
-            json.dump(seed, f)
+        with gzip.open(p, "wb") as f:
+            pass  # empty JSONL — 0 rows, no error
         print("created", p)
     else:
         print("exists ", p)