Update queries.py
Browse files- queries.py +209 -187
queries.py
CHANGED
|
@@ -1,6 +1,7 @@
|
|
| 1 |
"""
|
| 2 |
Parameterized query builders for all metrics.
|
| 3 |
Uses config.py mapping layer for table/column names.
|
|
|
|
| 4 |
"""
|
| 5 |
|
| 6 |
from typing import Optional, Dict, Any, List
|
|
@@ -24,34 +25,30 @@ class QueryBuilder:
|
|
| 24 |
self.history_table = get_table_name("TripHistoryDetail")
|
| 25 |
self.address_table = get_table_name("HYUserAddress")
|
| 26 |
self.transaction_table = get_table_name("HYTransaction")
|
|
|
|
| 27 |
|
| 28 |
def _build_trip_filters(
|
| 29 |
self,
|
| 30 |
-
driver_type: Optional[
|
| 31 |
is_solo: Optional[bool] = None,
|
| 32 |
-
status: Optional[int] = None
|
| 33 |
-
sponsor_id: Optional[int] = None
|
| 34 |
) -> tuple[str, Dict[str, Any]]:
|
| 35 |
"""Build WHERE clause and params for trip filters."""
|
| 36 |
conditions = []
|
| 37 |
params = {}
|
| 38 |
|
| 39 |
-
if driver_type:
|
| 40 |
-
conditions.append(
|
| 41 |
params['driver_type'] = driver_type
|
| 42 |
|
| 43 |
if is_solo is not None:
|
| 44 |
-
conditions.append(
|
| 45 |
params['is_solo'] = is_solo
|
| 46 |
|
| 47 |
-
if status:
|
| 48 |
-
conditions.append(
|
| 49 |
params['status'] = status
|
| 50 |
|
| 51 |
-
if sponsor_id:
|
| 52 |
-
conditions.append(f"t.{get_column_name(self.trips_table, 'sponsor_id')} = :sponsor_id")
|
| 53 |
-
params['sponsor_id'] = sponsor_id
|
| 54 |
-
|
| 55 |
where_clause = " AND " + " AND ".join(conditions) if conditions else ""
|
| 56 |
return where_clause, params
|
| 57 |
|
|
@@ -67,19 +64,18 @@ class QueryBuilder:
|
|
| 67 |
) -> tuple[str, Dict[str, Any]]:
|
| 68 |
"""Query for new registered users over time."""
|
| 69 |
date_trunc = {
|
| 70 |
-
"day": "CAST(
|
| 71 |
-
"week": "DATEADD(week, DATEDIFF(week, 0,
|
| 72 |
-
"month": "DATEADD(month, DATEDIFF(month, 0,
|
| 73 |
}
|
| 74 |
|
| 75 |
-
|
| 76 |
-
trunc_expr = date_trunc[granularity].format(col=f"u.{created_col}")
|
| 77 |
|
| 78 |
query = f"""
|
| 79 |
SELECT {trunc_expr} AS period,
|
| 80 |
COUNT(*) AS new_users
|
| 81 |
FROM {self.users_table} u
|
| 82 |
-
WHERE u.
|
| 83 |
GROUP BY {trunc_expr}
|
| 84 |
ORDER BY period
|
| 85 |
"""
|
|
@@ -93,22 +89,23 @@ class QueryBuilder:
|
|
| 93 |
end_date: datetime,
|
| 94 |
granularity: str = "day"
|
| 95 |
) -> tuple[str, Dict[str, Any]]:
|
| 96 |
-
"""Query for
|
| 97 |
date_trunc = {
|
| 98 |
-
"day": "CAST(
|
| 99 |
-
"week": "DATEADD(week, DATEDIFF(week, 0,
|
| 100 |
-
"month": "DATEADD(month, DATEDIFF(month, 0,
|
| 101 |
}
|
| 102 |
|
| 103 |
-
|
| 104 |
-
trunc_expr = date_trunc[granularity].format(col=f"u.{verified_col}")
|
| 105 |
|
|
|
|
| 106 |
query = f"""
|
| 107 |
SELECT {trunc_expr} AS period,
|
| 108 |
COUNT(*) AS verified_users
|
| 109 |
FROM {self.users_table} u
|
| 110 |
-
WHERE u.
|
| 111 |
-
AND u.
|
|
|
|
| 112 |
GROUP BY {trunc_expr}
|
| 113 |
ORDER BY period
|
| 114 |
"""
|
|
@@ -124,25 +121,21 @@ class QueryBuilder:
|
|
| 124 |
) -> tuple[str, Dict[str, Any]]:
|
| 125 |
"""Query for users activated by their first completed trip."""
|
| 126 |
date_trunc = {
|
| 127 |
-
"day": "CAST(
|
| 128 |
-
"week": "DATEADD(week, DATEDIFF(week, 0,
|
| 129 |
-
"month": "DATEADD(month, DATEDIFF(month, 0,
|
| 130 |
}
|
| 131 |
|
| 132 |
-
|
| 133 |
-
owner_col = get_column_name(self.trips_table, "user_id_owner")
|
| 134 |
-
status_col = get_column_name(self.trips_table, "status")
|
| 135 |
-
|
| 136 |
-
trunc_expr = date_trunc[granularity].format(col="f.first_trip_at")
|
| 137 |
|
| 138 |
query = f"""
|
| 139 |
WITH first_trip AS (
|
| 140 |
-
SELECT t.
|
| 141 |
-
MIN(t.
|
| 142 |
FROM {self.trips_table} t
|
| 143 |
-
WHERE t.
|
| 144 |
-
AND t.
|
| 145 |
-
GROUP BY t.
|
| 146 |
)
|
| 147 |
SELECT {trunc_expr} AS period,
|
| 148 |
COUNT(*) AS activated_users
|
|
@@ -165,15 +158,11 @@ class QueryBuilder:
|
|
| 165 |
window_days: int = 30
|
| 166 |
) -> tuple[str, Dict[str, Any]]:
|
| 167 |
"""Query for rolling active users (users with trips in last N days)."""
|
| 168 |
-
finished_col = get_column_name(self.trips_table, "finished_on")
|
| 169 |
-
owner_col = get_column_name(self.trips_table, "user_id_owner")
|
| 170 |
-
status_col = get_column_name(self.trips_table, "status")
|
| 171 |
-
|
| 172 |
query = f"""
|
| 173 |
-
SELECT COUNT(DISTINCT t.
|
| 174 |
FROM {self.trips_table} t
|
| 175 |
-
WHERE t.
|
| 176 |
-
AND t.
|
| 177 |
"""
|
| 178 |
|
| 179 |
params = {
|
|
@@ -190,25 +179,19 @@ class QueryBuilder:
|
|
| 190 |
retention_days: int = 7
|
| 191 |
) -> tuple[str, Dict[str, Any]]:
|
| 192 |
"""Query for cohort retention analysis."""
|
| 193 |
-
created_col = get_column_name(self.users_table, "created_at")
|
| 194 |
-
id_col = get_column_name(self.users_table, "id")
|
| 195 |
-
finished_col = get_column_name(self.trips_table, "finished_on")
|
| 196 |
-
owner_col = get_column_name(self.trips_table, "user_id_owner")
|
| 197 |
-
status_col = get_column_name(self.trips_table, "status")
|
| 198 |
-
|
| 199 |
query = f"""
|
| 200 |
WITH cohort AS (
|
| 201 |
-
SELECT u.
|
| 202 |
-
CAST(u.
|
| 203 |
FROM {self.users_table} u
|
| 204 |
-
WHERE u.
|
| 205 |
),
|
| 206 |
activity AS (
|
| 207 |
-
SELECT t.
|
| 208 |
-
CAST(t.
|
| 209 |
FROM {self.trips_table} t
|
| 210 |
-
WHERE t.
|
| 211 |
-
AND t.
|
| 212 |
)
|
| 213 |
SELECT c.cohort_date,
|
| 214 |
COUNT(DISTINCT c.user_id) AS cohort_size,
|
|
@@ -241,29 +224,26 @@ class QueryBuilder:
|
|
| 241 |
start_date: datetime,
|
| 242 |
end_date: datetime,
|
| 243 |
granularity: str = "day",
|
| 244 |
-
driver_type: Optional[
|
| 245 |
is_solo: Optional[bool] = None
|
| 246 |
) -> tuple[str, Dict[str, Any]]:
|
| 247 |
"""Query for trip volume over time with optional filters."""
|
| 248 |
date_trunc = {
|
| 249 |
-
"day": "CAST(
|
| 250 |
-
"week": "DATEADD(week, DATEDIFF(week, 0,
|
| 251 |
-
"month": "DATEADD(month, DATEDIFF(month, 0,
|
| 252 |
}
|
| 253 |
|
| 254 |
-
|
| 255 |
-
trip_id_col = get_column_name(self.history_table, "trip_id")
|
| 256 |
-
|
| 257 |
-
trunc_expr = date_trunc[granularity].format(col=f"thd.{completed_col}")
|
| 258 |
|
| 259 |
trip_filters, filter_params = self._build_trip_filters(driver_type, is_solo)
|
| 260 |
|
| 261 |
query = f"""
|
| 262 |
SELECT {trunc_expr} AS period,
|
| 263 |
-
COUNT(DISTINCT thd.
|
| 264 |
FROM {self.history_table} thd
|
| 265 |
-
JOIN {self.trips_table} t ON t.
|
| 266 |
-
WHERE thd.
|
| 267 |
{trip_filters}
|
| 268 |
GROUP BY {trunc_expr}
|
| 269 |
ORDER BY period
|
|
@@ -277,35 +257,31 @@ class QueryBuilder:
|
|
| 277 |
self,
|
| 278 |
start_date: datetime,
|
| 279 |
end_date: datetime,
|
| 280 |
-
driver_type: Optional[
|
| 281 |
is_solo: Optional[bool] = None
|
| 282 |
) -> tuple[str, Dict[str, Any]]:
|
| 283 |
"""Query for aggregate trip metrics (distance, impact, etc.)."""
|
| 284 |
-
completed_col = get_column_name(self.history_table, "completed_on")
|
| 285 |
-
distance_col = get_column_name(self.history_table, "distance_miles")
|
| 286 |
-
shared_col = get_column_name(self.history_table, "shared_miles")
|
| 287 |
-
co2_col = get_column_name(self.history_table, "co2_reduced")
|
| 288 |
-
nox_col = get_column_name(self.history_table, "nox_reduced")
|
| 289 |
-
pm25_col = get_column_name(self.history_table, "pm25_reduced")
|
| 290 |
-
points_col = get_column_name(self.history_table, "points")
|
| 291 |
-
trip_id_col = get_column_name(self.history_table, "trip_id")
|
| 292 |
-
|
| 293 |
trip_filters, filter_params = self._build_trip_filters(driver_type, is_solo)
|
| 294 |
|
| 295 |
query = f"""
|
| 296 |
SELECT
|
| 297 |
-
COUNT(DISTINCT thd.
|
| 298 |
-
AVG(thd.
|
| 299 |
-
SUM(thd.
|
| 300 |
-
SUM(thd.
|
| 301 |
-
SUM(thd.
|
| 302 |
-
AVG(thd.
|
| 303 |
-
SUM(thd.
|
| 304 |
-
SUM(thd.
|
| 305 |
-
SUM(thd.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 306 |
FROM {self.history_table} thd
|
| 307 |
-
JOIN {self.trips_table} t ON t.
|
| 308 |
-
WHERE thd.
|
| 309 |
{trip_filters}
|
| 310 |
"""
|
| 311 |
|
|
@@ -319,18 +295,20 @@ class QueryBuilder:
|
|
| 319 |
end_date: datetime
|
| 320 |
) -> tuple[str, Dict[str, Any]]:
|
| 321 |
"""Query for distribution of trips by driver type."""
|
| 322 |
-
completed_col = get_column_name(self.history_table, "completed_on")
|
| 323 |
-
trip_id_col = get_column_name(self.history_table, "trip_id")
|
| 324 |
-
driver_type_col = get_column_name(self.trips_table, "driver_type")
|
| 325 |
-
|
| 326 |
query = f"""
|
| 327 |
-
SELECT
|
| 328 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 329 |
FROM {self.history_table} thd
|
| 330 |
-
JOIN {self.trips_table} t ON t.
|
| 331 |
-
WHERE thd.
|
| 332 |
-
AND t.
|
| 333 |
-
GROUP BY t.
|
| 334 |
ORDER BY trip_count DESC
|
| 335 |
"""
|
| 336 |
|
|
@@ -343,18 +321,14 @@ class QueryBuilder:
|
|
| 343 |
end_date: datetime
|
| 344 |
) -> tuple[str, Dict[str, Any]]:
|
| 345 |
"""Query for solo vs shared trip distribution."""
|
| 346 |
-
completed_col = get_column_name(self.history_table, "completed_on")
|
| 347 |
-
trip_id_col = get_column_name(self.history_table, "trip_id")
|
| 348 |
-
is_solo_col = get_column_name(self.trips_table, "is_solo")
|
| 349 |
-
|
| 350 |
query = f"""
|
| 351 |
SELECT
|
| 352 |
-
CASE WHEN t.
|
| 353 |
-
COUNT(DISTINCT thd.
|
| 354 |
FROM {self.history_table} thd
|
| 355 |
-
JOIN {self.trips_table} t ON t.
|
| 356 |
-
WHERE thd.
|
| 357 |
-
GROUP BY t.
|
| 358 |
"""
|
| 359 |
|
| 360 |
params = {"start_date": start_date, "end_date": end_date}
|
|
@@ -366,19 +340,15 @@ class QueryBuilder:
|
|
| 366 |
end_date: datetime
|
| 367 |
) -> tuple[str, Dict[str, Any]]:
|
| 368 |
"""Query for trip completion rate."""
|
| 369 |
-
started_col = get_column_name(self.trips_table, "started_on")
|
| 370 |
-
status_col = get_column_name(self.trips_table, "status")
|
| 371 |
-
id_col = get_column_name(self.trips_table, "id")
|
| 372 |
-
|
| 373 |
query = f"""
|
| 374 |
SELECT
|
| 375 |
-
COUNT(DISTINCT t.
|
| 376 |
-
COUNT(DISTINCT CASE WHEN t.
|
| 377 |
-
CAST(COUNT(DISTINCT CASE WHEN t.
|
| 378 |
-
/ NULLIF(COUNT(DISTINCT t.
|
| 379 |
FROM {self.trips_table} t
|
| 380 |
-
WHERE t.
|
| 381 |
-
AND t.
|
| 382 |
"""
|
| 383 |
|
| 384 |
params = {
|
|
@@ -388,31 +358,45 @@ class QueryBuilder:
|
|
| 388 |
}
|
| 389 |
return query, params
|
| 390 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 391 |
# =========================================================================
|
| 392 |
# GEOGRAPHY QUERIES
|
| 393 |
# =========================================================================
|
| 394 |
|
| 395 |
def get_user_locations_query(self) -> tuple[str, Dict[str, Any]]:
|
| 396 |
"""Query for user home locations (for heat map)."""
|
| 397 |
-
user_id_col = get_column_name(self.address_table, "user_id")
|
| 398 |
-
type_col = get_column_name(self.address_table, "address_type")
|
| 399 |
-
lat_col = get_column_name(self.address_table, "latitude")
|
| 400 |
-
lon_col = get_column_name(self.address_table, "longitude")
|
| 401 |
-
city_col = get_column_name(self.address_table, "city")
|
| 402 |
-
state_col = get_column_name(self.address_table, "state")
|
| 403 |
-
|
| 404 |
query = f"""
|
| 405 |
SELECT
|
| 406 |
-
ua.
|
| 407 |
-
ua.
|
| 408 |
-
ua.
|
| 409 |
-
ua.
|
| 410 |
-
|
|
|
|
| 411 |
FROM {self.address_table} ua
|
| 412 |
-
WHERE ua.
|
| 413 |
-
AND ua.
|
| 414 |
-
AND ua.
|
| 415 |
-
GROUP BY ua.
|
| 416 |
"""
|
| 417 |
|
| 418 |
return query, {}
|
|
@@ -424,34 +408,29 @@ class QueryBuilder:
|
|
| 424 |
level: str = "state" # state, city, or zip
|
| 425 |
) -> tuple[str, Dict[str, Any]]:
|
| 426 |
"""Query for geographic distribution of users or trips."""
|
| 427 |
-
|
| 428 |
-
|
| 429 |
-
|
| 430 |
-
|
| 431 |
-
|
| 432 |
-
addr_user_col = get_column_name(self.address_table, "user_id")
|
| 433 |
-
created_col = get_column_name(self.users_table, "created_at")
|
| 434 |
-
|
| 435 |
-
group_col = {
|
| 436 |
-
"state": state_col,
|
| 437 |
-
"city": f"ua.{city_col}, ua.{state_col}",
|
| 438 |
-
"zip": zip_col
|
| 439 |
-
}.get(level, state_col)
|
| 440 |
|
| 441 |
select_cols = {
|
| 442 |
-
"state":
|
| 443 |
-
"city":
|
| 444 |
-
"zip":
|
| 445 |
-
}
|
|
|
|
|
|
|
|
|
|
| 446 |
|
| 447 |
query = f"""
|
| 448 |
SELECT
|
| 449 |
-
{
|
| 450 |
-
COUNT(DISTINCT u.
|
| 451 |
FROM {self.users_table} u
|
| 452 |
-
JOIN {self.address_table} ua ON ua.
|
| 453 |
-
WHERE u.
|
| 454 |
-
AND ua.
|
| 455 |
GROUP BY {group_col}
|
| 456 |
ORDER BY user_count DESC
|
| 457 |
"""
|
|
@@ -459,6 +438,28 @@ class QueryBuilder:
|
|
| 459 |
params = {"start_date": start_date, "end_date": end_date}
|
| 460 |
return query, params
|
| 461 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 462 |
# =========================================================================
|
| 463 |
# TRANSACTION/REWARD QUERIES
|
| 464 |
# =========================================================================
|
|
@@ -468,37 +469,32 @@ class QueryBuilder:
|
|
| 468 |
start_date: datetime,
|
| 469 |
end_date: datetime,
|
| 470 |
granularity: str = "day",
|
| 471 |
-
transaction_type: Optional[
|
| 472 |
) -> tuple[str, Dict[str, Any]]:
|
| 473 |
"""Query for transaction volume and amounts over time."""
|
| 474 |
date_trunc = {
|
| 475 |
-
"day": "CAST(
|
| 476 |
-
"week": "DATEADD(week, DATEDIFF(week, 0,
|
| 477 |
-
"month": "DATEADD(month, DATEDIFF(month, 0,
|
| 478 |
}
|
| 479 |
|
| 480 |
-
|
| 481 |
-
amount_col = get_column_name(self.transaction_table, "amount")
|
| 482 |
-
type_col = get_column_name(self.transaction_table, "type")
|
| 483 |
-
id_col = get_column_name(self.transaction_table, "id")
|
| 484 |
-
|
| 485 |
-
trunc_expr = date_trunc[granularity].format(col=f"tr.{date_col}")
|
| 486 |
|
| 487 |
-
type_filter =
|
| 488 |
|
| 489 |
query = f"""
|
| 490 |
SELECT {trunc_expr} AS period,
|
| 491 |
-
COUNT(DISTINCT tr.
|
| 492 |
-
SUM(tr.
|
| 493 |
FROM {self.transaction_table} tr
|
| 494 |
-
WHERE tr.
|
| 495 |
{type_filter}
|
| 496 |
GROUP BY {trunc_expr}
|
| 497 |
ORDER BY period
|
| 498 |
"""
|
| 499 |
|
| 500 |
params = {"start_date": start_date, "end_date": end_date}
|
| 501 |
-
if transaction_type:
|
| 502 |
params["transaction_type"] = transaction_type
|
| 503 |
|
| 504 |
return query, params
|
|
@@ -508,22 +504,48 @@ class QueryBuilder:
|
|
| 508 |
start_date: datetime,
|
| 509 |
end_date: datetime
|
| 510 |
) -> tuple[str, Dict[str, Any]]:
|
| 511 |
-
"""Query for sponsor-level
|
| 512 |
-
|
| 513 |
-
|
| 514 |
-
|
| 515 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 516 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 517 |
query = f"""
|
| 518 |
SELECT
|
| 519 |
-
|
| 520 |
-
COUNT(DISTINCT
|
| 521 |
-
|
| 522 |
-
|
| 523 |
-
|
| 524 |
-
|
| 525 |
-
|
| 526 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 527 |
"""
|
| 528 |
|
| 529 |
params = {"start_date": start_date, "end_date": end_date}
|
|
@@ -531,4 +553,4 @@ class QueryBuilder:
|
|
| 531 |
|
| 532 |
|
| 533 |
# Global query builder instance
|
| 534 |
-
query_builder = QueryBuilder()
|
|
|
|
| 1 |
"""
|
| 2 |
Parameterized query builders for all metrics.
|
| 3 |
Uses config.py mapping layer for table/column names.
|
| 4 |
+
Updated for Hytch database schema.
|
| 5 |
"""
|
| 6 |
|
| 7 |
from typing import Optional, Dict, Any, List
|
|
|
|
| 25 |
self.history_table = get_table_name("TripHistoryDetail")
|
| 26 |
self.address_table = get_table_name("HYUserAddress")
|
| 27 |
self.transaction_table = get_table_name("HYTransaction")
|
| 28 |
+
self.participants_table = get_table_name("HYTripParticipant")
|
| 29 |
|
| 30 |
def _build_trip_filters(
|
| 31 |
self,
|
| 32 |
+
driver_type: Optional[int] = None,
|
| 33 |
is_solo: Optional[bool] = None,
|
| 34 |
+
status: Optional[int] = None
|
|
|
|
| 35 |
) -> tuple[str, Dict[str, Any]]:
|
| 36 |
"""Build WHERE clause and params for trip filters."""
|
| 37 |
conditions = []
|
| 38 |
params = {}
|
| 39 |
|
| 40 |
+
if driver_type is not None:
|
| 41 |
+
conditions.append("t.driver_type = :driver_type")
|
| 42 |
params['driver_type'] = driver_type
|
| 43 |
|
| 44 |
if is_solo is not None:
|
| 45 |
+
conditions.append("t.is_solo_hytch = :is_solo")
|
| 46 |
params['is_solo'] = is_solo
|
| 47 |
|
| 48 |
+
if status is not None:
|
| 49 |
+
conditions.append("t.status = :status")
|
| 50 |
params['status'] = status
|
| 51 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
where_clause = " AND " + " AND ".join(conditions) if conditions else ""
|
| 53 |
return where_clause, params
|
| 54 |
|
|
|
|
| 64 |
) -> tuple[str, Dict[str, Any]]:
|
| 65 |
"""Query for new registered users over time."""
|
| 66 |
date_trunc = {
|
| 67 |
+
"day": "CAST(u.created_at AS DATE)",
|
| 68 |
+
"week": "DATEADD(week, DATEDIFF(week, 0, u.created_at), 0)",
|
| 69 |
+
"month": "DATEADD(month, DATEDIFF(month, 0, u.created_at), 0)"
|
| 70 |
}
|
| 71 |
|
| 72 |
+
trunc_expr = date_trunc.get(granularity, date_trunc["day"])
|
|
|
|
| 73 |
|
| 74 |
query = f"""
|
| 75 |
SELECT {trunc_expr} AS period,
|
| 76 |
COUNT(*) AS new_users
|
| 77 |
FROM {self.users_table} u
|
| 78 |
+
WHERE u.created_at BETWEEN :start_date AND :end_date
|
| 79 |
GROUP BY {trunc_expr}
|
| 80 |
ORDER BY period
|
| 81 |
"""
|
|
|
|
| 89 |
end_date: datetime,
|
| 90 |
granularity: str = "day"
|
| 91 |
) -> tuple[str, Dict[str, Any]]:
|
| 92 |
+
"""Query for users with completed profiles (using complete_profile_code)."""
|
| 93 |
date_trunc = {
|
| 94 |
+
"day": "CAST(u.updated_at AS DATE)",
|
| 95 |
+
"week": "DATEADD(week, DATEDIFF(week, 0, u.updated_at), 0)",
|
| 96 |
+
"month": "DATEADD(month, DATEDIFF(month, 0, u.updated_at), 0)"
|
| 97 |
}
|
| 98 |
|
| 99 |
+
trunc_expr = date_trunc.get(granularity, date_trunc["day"])
|
|
|
|
| 100 |
|
| 101 |
+
# Users with complete profiles (complete_profile_code is set)
|
| 102 |
query = f"""
|
| 103 |
SELECT {trunc_expr} AS period,
|
| 104 |
COUNT(*) AS verified_users
|
| 105 |
FROM {self.users_table} u
|
| 106 |
+
WHERE u.updated_at BETWEEN :start_date AND :end_date
|
| 107 |
+
AND u.complete_profile_code IS NOT NULL
|
| 108 |
+
AND u.complete_profile_code != ''
|
| 109 |
GROUP BY {trunc_expr}
|
| 110 |
ORDER BY period
|
| 111 |
"""
|
|
|
|
| 121 |
) -> tuple[str, Dict[str, Any]]:
|
| 122 |
"""Query for users activated by their first completed trip."""
|
| 123 |
date_trunc = {
|
| 124 |
+
"day": "CAST(f.first_trip_at AS DATE)",
|
| 125 |
+
"week": "DATEADD(week, DATEDIFF(week, 0, f.first_trip_at), 0)",
|
| 126 |
+
"month": "DATEADD(month, DATEDIFF(month, 0, f.first_trip_at), 0)"
|
| 127 |
}
|
| 128 |
|
| 129 |
+
trunc_expr = date_trunc.get(granularity, date_trunc["day"])
|
|
|
|
|
|
|
|
|
|
|
|
|
| 130 |
|
| 131 |
query = f"""
|
| 132 |
WITH first_trip AS (
|
| 133 |
+
SELECT t.owner_user_id AS user_id,
|
| 134 |
+
MIN(t.finished_at) AS first_trip_at
|
| 135 |
FROM {self.trips_table} t
|
| 136 |
+
WHERE t.status = :completed_status
|
| 137 |
+
AND t.finished_at IS NOT NULL
|
| 138 |
+
GROUP BY t.owner_user_id
|
| 139 |
)
|
| 140 |
SELECT {trunc_expr} AS period,
|
| 141 |
COUNT(*) AS activated_users
|
|
|
|
| 158 |
window_days: int = 30
|
| 159 |
) -> tuple[str, Dict[str, Any]]:
|
| 160 |
"""Query for rolling active users (users with trips in last N days)."""
|
|
|
|
|
|
|
|
|
|
|
|
|
| 161 |
query = f"""
|
| 162 |
+
SELECT COUNT(DISTINCT t.owner_user_id) AS active_users
|
| 163 |
FROM {self.trips_table} t
|
| 164 |
+
WHERE t.finished_at BETWEEN DATEADD(day, -:window_days, :as_of_date) AND :as_of_date
|
| 165 |
+
AND t.status = :completed_status
|
| 166 |
"""
|
| 167 |
|
| 168 |
params = {
|
|
|
|
| 179 |
retention_days: int = 7
|
| 180 |
) -> tuple[str, Dict[str, Any]]:
|
| 181 |
"""Query for cohort retention analysis."""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 182 |
query = f"""
|
| 183 |
WITH cohort AS (
|
| 184 |
+
SELECT u.user_id,
|
| 185 |
+
CAST(u.created_at AS DATE) AS cohort_date
|
| 186 |
FROM {self.users_table} u
|
| 187 |
+
WHERE u.created_at BETWEEN :cohort_start AND :cohort_end
|
| 188 |
),
|
| 189 |
activity AS (
|
| 190 |
+
SELECT t.owner_user_id AS user_id,
|
| 191 |
+
CAST(t.finished_at AS DATE) AS active_date
|
| 192 |
FROM {self.trips_table} t
|
| 193 |
+
WHERE t.status = :completed_status
|
| 194 |
+
AND t.finished_at IS NOT NULL
|
| 195 |
)
|
| 196 |
SELECT c.cohort_date,
|
| 197 |
COUNT(DISTINCT c.user_id) AS cohort_size,
|
|
|
|
| 224 |
start_date: datetime,
|
| 225 |
end_date: datetime,
|
| 226 |
granularity: str = "day",
|
| 227 |
+
driver_type: Optional[int] = None,
|
| 228 |
is_solo: Optional[bool] = None
|
| 229 |
) -> tuple[str, Dict[str, Any]]:
|
| 230 |
"""Query for trip volume over time with optional filters."""
|
| 231 |
date_trunc = {
|
| 232 |
+
"day": "CAST(thd.completed_on AS DATE)",
|
| 233 |
+
"week": "DATEADD(week, DATEDIFF(week, 0, thd.completed_on), 0)",
|
| 234 |
+
"month": "DATEADD(month, DATEDIFF(month, 0, thd.completed_on), 0)"
|
| 235 |
}
|
| 236 |
|
| 237 |
+
trunc_expr = date_trunc.get(granularity, date_trunc["day"])
|
|
|
|
|
|
|
|
|
|
| 238 |
|
| 239 |
trip_filters, filter_params = self._build_trip_filters(driver_type, is_solo)
|
| 240 |
|
| 241 |
query = f"""
|
| 242 |
SELECT {trunc_expr} AS period,
|
| 243 |
+
COUNT(DISTINCT thd.trip_id) AS trip_count
|
| 244 |
FROM {self.history_table} thd
|
| 245 |
+
JOIN {self.trips_table} t ON t.trip_id = thd.trip_id
|
| 246 |
+
WHERE thd.completed_on BETWEEN :start_date AND :end_date
|
| 247 |
{trip_filters}
|
| 248 |
GROUP BY {trunc_expr}
|
| 249 |
ORDER BY period
|
|
|
|
| 257 |
self,
|
| 258 |
start_date: datetime,
|
| 259 |
end_date: datetime,
|
| 260 |
+
driver_type: Optional[int] = None,
|
| 261 |
is_solo: Optional[bool] = None
|
| 262 |
) -> tuple[str, Dict[str, Any]]:
|
| 263 |
"""Query for aggregate trip metrics (distance, impact, etc.)."""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 264 |
trip_filters, filter_params = self._build_trip_filters(driver_type, is_solo)
|
| 265 |
|
| 266 |
query = f"""
|
| 267 |
SELECT
|
| 268 |
+
COUNT(DISTINCT thd.trip_id) AS total_trips,
|
| 269 |
+
AVG(thd.distance_miles) AS avg_distance_miles,
|
| 270 |
+
SUM(thd.distance_miles) AS total_distance_miles,
|
| 271 |
+
SUM(thd.shared_miles) AS total_shared_miles,
|
| 272 |
+
SUM(thd.co2_reduced) AS total_co2_reduced,
|
| 273 |
+
AVG(thd.co2_reduced) AS avg_co2_per_trip,
|
| 274 |
+
SUM(thd.nox_reduced) AS total_nox_reduced,
|
| 275 |
+
SUM(thd.pm25_reduced) AS total_pm25_reduced,
|
| 276 |
+
SUM(thd.pm10_reduced) AS total_pm10_reduced,
|
| 277 |
+
SUM(thd.voc_reduced) AS total_voc_reduced,
|
| 278 |
+
SUM(thd.co_reduced) AS total_co_reduced,
|
| 279 |
+
SUM(thd.trees_saved) AS total_trees_saved,
|
| 280 |
+
SUM(thd.points) AS total_points,
|
| 281 |
+
SUM(thd.gas_savings) AS total_gas_savings
|
| 282 |
FROM {self.history_table} thd
|
| 283 |
+
JOIN {self.trips_table} t ON t.trip_id = thd.trip_id
|
| 284 |
+
WHERE thd.completed_on BETWEEN :start_date AND :end_date
|
| 285 |
{trip_filters}
|
| 286 |
"""
|
| 287 |
|
|
|
|
| 295 |
end_date: datetime
|
| 296 |
) -> tuple[str, Dict[str, Any]]:
|
| 297 |
"""Query for distribution of trips by driver type."""
|
|
|
|
|
|
|
|
|
|
|
|
|
| 298 |
query = f"""
|
| 299 |
+
SELECT
|
| 300 |
+
CASE t.driver_type
|
| 301 |
+
WHEN 0 THEN 'Owner'
|
| 302 |
+
WHEN 1 THEN 'Participant'
|
| 303 |
+
WHEN 2 THEN 'External'
|
| 304 |
+
ELSE 'Unknown'
|
| 305 |
+
END AS driver_type,
|
| 306 |
+
COUNT(DISTINCT thd.trip_id) AS trip_count
|
| 307 |
FROM {self.history_table} thd
|
| 308 |
+
JOIN {self.trips_table} t ON t.trip_id = thd.trip_id
|
| 309 |
+
WHERE thd.completed_on BETWEEN :start_date AND :end_date
|
| 310 |
+
AND t.driver_type IS NOT NULL
|
| 311 |
+
GROUP BY t.driver_type
|
| 312 |
ORDER BY trip_count DESC
|
| 313 |
"""
|
| 314 |
|
|
|
|
| 321 |
end_date: datetime
|
| 322 |
) -> tuple[str, Dict[str, Any]]:
|
| 323 |
"""Query for solo vs shared trip distribution."""
|
|
|
|
|
|
|
|
|
|
|
|
|
| 324 |
query = f"""
|
| 325 |
SELECT
|
| 326 |
+
CASE WHEN t.is_solo_hytch = 1 THEN 'Solo' ELSE 'Shared' END AS trip_type,
|
| 327 |
+
COUNT(DISTINCT thd.trip_id) AS trip_count
|
| 328 |
FROM {self.history_table} thd
|
| 329 |
+
JOIN {self.trips_table} t ON t.trip_id = thd.trip_id
|
| 330 |
+
WHERE thd.completed_on BETWEEN :start_date AND :end_date
|
| 331 |
+
GROUP BY t.is_solo_hytch
|
| 332 |
"""
|
| 333 |
|
| 334 |
params = {"start_date": start_date, "end_date": end_date}
|
|
|
|
| 340 |
end_date: datetime
|
| 341 |
) -> tuple[str, Dict[str, Any]]:
|
| 342 |
"""Query for trip completion rate."""
|
|
|
|
|
|
|
|
|
|
|
|
|
| 343 |
query = f"""
|
| 344 |
SELECT
|
| 345 |
+
COUNT(DISTINCT t.trip_id) AS total_started,
|
| 346 |
+
COUNT(DISTINCT CASE WHEN t.status = :completed_status THEN t.trip_id END) AS completed,
|
| 347 |
+
CAST(COUNT(DISTINCT CASE WHEN t.status = :completed_status THEN t.trip_id END) AS FLOAT) * 100.0
|
| 348 |
+
/ NULLIF(COUNT(DISTINCT t.trip_id), 0) AS completion_rate
|
| 349 |
FROM {self.trips_table} t
|
| 350 |
+
WHERE t.started_at BETWEEN :start_date AND :end_date
|
| 351 |
+
AND t.started_at IS NOT NULL
|
| 352 |
"""
|
| 353 |
|
| 354 |
params = {
|
|
|
|
| 358 |
}
|
| 359 |
return query, params
|
| 360 |
|
| 361 |
+
def get_trip_impact_grade_distribution_query(
|
| 362 |
+
self,
|
| 363 |
+
start_date: datetime,
|
| 364 |
+
end_date: datetime
|
| 365 |
+
) -> tuple[str, Dict[str, Any]]:
|
| 366 |
+
"""Query for distribution of trips by environmental impact grade."""
|
| 367 |
+
query = f"""
|
| 368 |
+
SELECT
|
| 369 |
+
thd.trip_impact_grade AS grade,
|
| 370 |
+
COUNT(*) AS trip_count
|
| 371 |
+
FROM {self.history_table} thd
|
| 372 |
+
WHERE thd.completed_on BETWEEN :start_date AND :end_date
|
| 373 |
+
AND thd.trip_impact_grade IS NOT NULL
|
| 374 |
+
GROUP BY thd.trip_impact_grade
|
| 375 |
+
ORDER BY thd.trip_impact_grade
|
| 376 |
+
"""
|
| 377 |
+
|
| 378 |
+
params = {"start_date": start_date, "end_date": end_date}
|
| 379 |
+
return query, params
|
| 380 |
+
|
| 381 |
# =========================================================================
|
| 382 |
# GEOGRAPHY QUERIES
|
| 383 |
# =========================================================================
|
| 384 |
|
| 385 |
def get_user_locations_query(self) -> tuple[str, Dict[str, Any]]:
|
| 386 |
"""Query for user home locations (for heat map)."""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 387 |
query = f"""
|
| 388 |
SELECT
|
| 389 |
+
ua.latitude,
|
| 390 |
+
ua.longitude,
|
| 391 |
+
ua.city,
|
| 392 |
+
ua.state,
|
| 393 |
+
ua.zip_code,
|
| 394 |
+
COUNT(DISTINCT ua.user_id) AS user_count
|
| 395 |
FROM {self.address_table} ua
|
| 396 |
+
WHERE ua.address_type = 'home'
|
| 397 |
+
AND ua.latitude IS NOT NULL
|
| 398 |
+
AND ua.longitude IS NOT NULL
|
| 399 |
+
GROUP BY ua.latitude, ua.longitude, ua.city, ua.state, ua.zip_code
|
| 400 |
"""
|
| 401 |
|
| 402 |
return query, {}
|
|
|
|
| 408 |
level: str = "state" # state, city, or zip
|
| 409 |
) -> tuple[str, Dict[str, Any]]:
|
| 410 |
"""Query for geographic distribution of users or trips."""
|
| 411 |
+
group_cols = {
|
| 412 |
+
"state": "ua.state",
|
| 413 |
+
"city": "ua.city, ua.state",
|
| 414 |
+
"zip": "ua.zip_code"
|
| 415 |
+
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 416 |
|
| 417 |
select_cols = {
|
| 418 |
+
"state": "ua.state AS region",
|
| 419 |
+
"city": "ua.city AS city, ua.state AS state",
|
| 420 |
+
"zip": "ua.zip_code AS zip_code"
|
| 421 |
+
}
|
| 422 |
+
|
| 423 |
+
group_col = group_cols.get(level, group_cols["state"])
|
| 424 |
+
select_col = select_cols.get(level, select_cols["state"])
|
| 425 |
|
| 426 |
query = f"""
|
| 427 |
SELECT
|
| 428 |
+
{select_col},
|
| 429 |
+
COUNT(DISTINCT u.user_id) AS user_count
|
| 430 |
FROM {self.users_table} u
|
| 431 |
+
JOIN {self.address_table} ua ON ua.user_id = u.user_id
|
| 432 |
+
WHERE u.created_at BETWEEN :start_date AND :end_date
|
| 433 |
+
AND ua.address_type = 'home'
|
| 434 |
GROUP BY {group_col}
|
| 435 |
ORDER BY user_count DESC
|
| 436 |
"""
|
|
|
|
| 438 |
params = {"start_date": start_date, "end_date": end_date}
|
| 439 |
return query, params
|
| 440 |
|
| 441 |
+
def get_trip_locations_query(
|
| 442 |
+
self,
|
| 443 |
+
start_date: datetime,
|
| 444 |
+
end_date: datetime
|
| 445 |
+
) -> tuple[str, Dict[str, Any]]:
|
| 446 |
+
"""Query for trip start locations from location_history."""
|
| 447 |
+
query = f"""
|
| 448 |
+
SELECT
|
| 449 |
+
lh.latitude,
|
| 450 |
+
lh.longitude,
|
| 451 |
+
COUNT(DISTINCT lh.trip_id) AS trip_count
|
| 452 |
+
FROM location_history lh
|
| 453 |
+
JOIN {self.trips_table} t ON t.trip_id = lh.trip_id
|
| 454 |
+
WHERE t.started_at BETWEEN :start_date AND :end_date
|
| 455 |
+
AND lh.latitude IS NOT NULL
|
| 456 |
+
AND lh.longitude IS NOT NULL
|
| 457 |
+
GROUP BY lh.latitude, lh.longitude
|
| 458 |
+
"""
|
| 459 |
+
|
| 460 |
+
params = {"start_date": start_date, "end_date": end_date}
|
| 461 |
+
return query, params
|
| 462 |
+
|
| 463 |
# =========================================================================
|
| 464 |
# TRANSACTION/REWARD QUERIES
|
| 465 |
# =========================================================================
|
|
|
|
| 469 |
start_date: datetime,
|
| 470 |
end_date: datetime,
|
| 471 |
granularity: str = "day",
|
| 472 |
+
transaction_type: Optional[int] = None
|
| 473 |
) -> tuple[str, Dict[str, Any]]:
|
| 474 |
"""Query for transaction volume and amounts over time."""
|
| 475 |
date_trunc = {
|
| 476 |
+
"day": "CAST(tr.created_at AS DATE)",
|
| 477 |
+
"week": "DATEADD(week, DATEDIFF(week, 0, tr.created_at), 0)",
|
| 478 |
+
"month": "DATEADD(month, DATEDIFF(month, 0, tr.created_at), 0)"
|
| 479 |
}
|
| 480 |
|
| 481 |
+
trunc_expr = date_trunc.get(granularity, date_trunc["day"])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 482 |
|
| 483 |
+
type_filter = "AND tr.type = :transaction_type" if transaction_type is not None else ""
|
| 484 |
|
| 485 |
query = f"""
|
| 486 |
SELECT {trunc_expr} AS period,
|
| 487 |
+
COUNT(DISTINCT tr.transaction_id) AS transaction_count,
|
| 488 |
+
SUM(tr.amount) AS total_amount
|
| 489 |
FROM {self.transaction_table} tr
|
| 490 |
+
WHERE tr.created_at BETWEEN :start_date AND :end_date
|
| 491 |
{type_filter}
|
| 492 |
GROUP BY {trunc_expr}
|
| 493 |
ORDER BY period
|
| 494 |
"""
|
| 495 |
|
| 496 |
params = {"start_date": start_date, "end_date": end_date}
|
| 497 |
+
if transaction_type is not None:
|
| 498 |
params["transaction_type"] = transaction_type
|
| 499 |
|
| 500 |
return query, params
|
|
|
|
| 504 |
start_date: datetime,
|
| 505 |
end_date: datetime
|
| 506 |
) -> tuple[str, Dict[str, Any]]:
|
| 507 |
+
"""Query for sponsor-level point activity."""
|
| 508 |
+
query = f"""
|
| 509 |
+
SELECT
|
| 510 |
+
ms.sponsor_id,
|
| 511 |
+
ms.name AS sponsor_name,
|
| 512 |
+
COUNT(DISTINCT usa.user_id) AS user_count,
|
| 513 |
+
SUM(usa.points_earned) AS total_points
|
| 514 |
+
FROM market_sponsors ms
|
| 515 |
+
JOIN user_sponsor_associations usa ON usa.sponsor_id = ms.sponsor_id
|
| 516 |
+
WHERE usa.created_at BETWEEN :start_date AND :end_date
|
| 517 |
+
AND ms.is_active = 1
|
| 518 |
+
GROUP BY ms.sponsor_id, ms.name
|
| 519 |
+
ORDER BY total_points DESC
|
| 520 |
+
"""
|
| 521 |
|
| 522 |
+
params = {"start_date": start_date, "end_date": end_date}
|
| 523 |
+
return query, params
|
| 524 |
+
|
| 525 |
+
# =========================================================================
|
| 526 |
+
# PARTICIPANT QUERIES
|
| 527 |
+
# =========================================================================
|
| 528 |
+
|
| 529 |
+
def get_participant_stats_query(
|
| 530 |
+
self,
|
| 531 |
+
start_date: datetime,
|
| 532 |
+
end_date: datetime
|
| 533 |
+
) -> tuple[str, Dict[str, Any]]:
|
| 534 |
+
"""Query for trip participant statistics."""
|
| 535 |
query = f"""
|
| 536 |
SELECT
|
| 537 |
+
COUNT(DISTINCT tp.participant_id) AS total_participations,
|
| 538 |
+
COUNT(DISTINCT tp.user_id) AS unique_participants,
|
| 539 |
+
COUNT(DISTINCT tp.trip_id) AS trips_with_participants,
|
| 540 |
+
AVG(participants_per_trip.participant_count) AS avg_participants_per_trip
|
| 541 |
+
FROM {self.participants_table} tp
|
| 542 |
+
JOIN {self.trips_table} t ON t.trip_id = tp.trip_id
|
| 543 |
+
JOIN (
|
| 544 |
+
SELECT trip_id, COUNT(*) AS participant_count
|
| 545 |
+
FROM {self.participants_table}
|
| 546 |
+
GROUP BY trip_id
|
| 547 |
+
) participants_per_trip ON participants_per_trip.trip_id = tp.trip_id
|
| 548 |
+
WHERE t.started_at BETWEEN :start_date AND :end_date
|
| 549 |
"""
|
| 550 |
|
| 551 |
params = {"start_date": start_date, "end_date": end_date}
|
|
|
|
| 553 |
|
| 554 |
|
| 555 |
# Global query builder instance
|
| 556 |
+
query_builder = QueryBuilder()
|