feat: populate company/place/customer_name from MSFS extraction DB

- New script: scripts/populate_asset_company_place.py maps connect_id →
  account name (company/customer_name) + city (place) + address via
  extraction DB (msdyn_customerasset + account join)
- 5,743/7,488 assets now have company + customer_name populated
- 5,647/7,488 assets now have place (city) populated
- Fix: SELECT COALESCE(c.name, a.customer_name) so stored customer_name
  isn't overridden to NULL when customer_id FK is missing
- Bump SW v4→v5, version v3.1→v3.2 to force cache refresh
This commit is contained in:
2026-05-29 07:45:01 -04:00
parent 5180d811a8
commit 4f9c0d0b75
4 changed files with 326 additions and 8 deletions
+120 -1
View File
@@ -1040,6 +1040,7 @@ def list_assets(
q: Optional[str] = Query(None),
no_dex_days: Optional[int] = Query(None, ge=1),
has_gps: Optional[bool] = Query(None),
branch: Optional[str] = Query(None, description="Filter by branch/territory name"),
limit: int = Query(100, ge=1, le=2000),
offset: int = Query(0, ge=0),
):
@@ -1087,10 +1088,25 @@ def list_assets(
conditions.append("a.latitude IS NOT NULL AND a.longitude IS NOT NULL")
else:
conditions.append("(a.latitude IS NULL OR a.longitude IS NULL)")
if branch:
cache = _build_asset_branch_cache()
# Find all connect_id prefixes that map to this branch name
matching_prefixes = [
pfx for pfx, br in cache.items() if br == branch
]
if matching_prefixes:
placeholders = ",".join(["?"] * len(matching_prefixes))
conditions.append(
f"SUBSTR(a.connect_id, 1, 8) IN ({placeholders})"
)
params.extend(matching_prefixes)
else:
# Branch name not found — return empty result
conditions.append("1=0")
where = " AND ".join(conditions)
from_clause = "FROM assets a LEFT JOIN customers c ON a.customer_id = c.id"
sql = f"SELECT a.*, c.name AS customer_name {from_clause}"
sql = f"SELECT a.*, COALESCE(c.name, a.customer_name) AS customer_name {from_clause}"
if where:
sql += f" WHERE {where}"
sql += " ORDER BY a.created_at DESC LIMIT ? OFFSET ?"
@@ -1121,6 +1137,44 @@ def search_by_machine_id(machine_id: str = Query(...)):
return row_to_dict(row)
# ── Task 5b: GET /api/branches ─────────────────────────────────────────────
@app.get("/api/branches")
def list_branches():
"""Return distinct branch/territory names that assets belong to.
Uses the connect_id prefix → branch mapping from the extraction DB.
Returns sorted list of {name, machine_count} objects.
"""
cache = _build_asset_branch_cache()
# Count how many assets per branch
conn = get_db()
prefix_counts: dict[str, int] = {}
try:
cur = conn.execute(
"SELECT SUBSTR(connect_id, 1, 8) AS prefix, COUNT(*) AS cnt "
"FROM assets WHERE connect_id IS NOT NULL AND connect_id != '' "
"GROUP BY prefix"
)
for r in cur.fetchall():
prefix_counts[r[0]] = r[1]
finally:
conn.close()
# Aggregate by branch name
branch_map: dict[str, int] = {}
for prefix, machine_count in prefix_counts.items():
branch = cache.get(prefix, prefix)
branch_map[branch] = branch_map.get(branch, 0) + machine_count
branches = sorted(
[{"name": k, "machine_count": v} for k, v in branch_map.items()],
key=lambda b: -b["machine_count"], # type: ignore[arg-type]
)
return {"branches": branches, "total": len(branches)}
@app.get("/api/assets/{asset_id}")
def get_asset(asset_id: int):
conn = get_db()
@@ -2595,6 +2649,71 @@ def _get_extraction_db() -> sqlite3.Connection | None:
return conn
# ── Asset Branch Cache (connect_id prefix → branch/territory name) ────────
_BRANCH_CACHE: dict[str, str] | None = None
def _build_asset_branch_cache() -> dict[str, str]:
"""Build a mapping from connect_id prefix (first 8 chars) to branch name.
Queries the extraction DB for each customer asset, linking it through
its account to work order territories (msdyn_serviceterritory!name).
Returns {connect_id_prefix: branch_name}.
"""
global _BRANCH_CACHE
if _BRANCH_CACHE is not None:
return _BRANCH_CACHE
result: dict[str, str] = {}
conn = _get_extraction_db()
if not conn:
print("⚠️ Extraction DB not available — branch cache will be empty")
_BRANCH_CACHE = result
return result
try:
cur = conn.cursor()
# For each customer asset (connect_id + account), find the most common
# work order territory for that account.
cur.execute("""
SELECT
SUBSTR(ca.hsl_connectid, 1, 8) AS prefix,
wo."msdyn_serviceterritory!name" AS territory,
COUNT(*) AS cnt
FROM msdyn_customerasset ca
LEFT JOIN account a ON ca."msdyn_account!id" = a.accountid
LEFT JOIN msdyn_workorder wo ON a.accountid = wo."msdyn_serviceaccount!id"
WHERE ca.hsl_connectid IS NOT NULL AND ca.hsl_connectid != ''
AND wo."msdyn_serviceterritory!name" IS NOT NULL
AND wo."msdyn_serviceterritory!name" != ''
GROUP BY prefix, wo."msdyn_serviceterritory!name"
ORDER BY cnt DESC
""")
# Take the most frequent territory per prefix
prefix_map: dict[str, list[tuple[str, int]]] = {}
for r in cur.fetchall():
prefix = r[0]
territory = r[1]
cnt = r[2]
if prefix not in prefix_map:
prefix_map[prefix] = []
prefix_map[prefix].append((territory, cnt))
for prefix, territories in prefix_map.items():
# Sort by count descending, pick the top one
territories.sort(key=lambda x: -x[1])
result[prefix] = territories[0][0]
print(f"📡 Built branch cache: {len(result)} connect_id prefixes mapped")
except Exception as e:
print(f"⚠️ Failed to build branch cache: {e}")
finally:
conn.close()
_BRANCH_CACHE = result
return result
# ── TSP Solver ───────────────────────────────────────────────────────────