整理项目
This commit is contained in:
321
tmp/etl_billiards_misc/backups/manual_ingest_task.py
Normal file
321
tmp/etl_billiards_misc/backups/manual_ingest_task.py
Normal file
@@ -0,0 +1,321 @@
|
||||
# -*- coding: utf-8 -*-
|
||||
"""鎵嬪伐绀轰緥鏁版嵁鐏屽叆锛氭寜 schema_ODS_doc.sql 涓婚敭/鍞竴閿壒閲忓啓鍏?ODS銆?""
|
||||
from __future__ import annotations
|
||||
|
||||
import json
|
||||
import os
|
||||
from datetime import datetime
|
||||
from typing import Any, Iterable
|
||||
|
||||
from psycopg2.extras import Json
|
||||
|
||||
from .base_task import BaseTask
|
||||
|
||||
|
||||
class ManualIngestTask(BaseTask):
|
||||
"""鏈湴绀轰緥 JSON 鐏屽叆 ODS锛岀‘淇濊〃鍚嶃€佷富閿€佹彃鍏ュ垪涓?schema_ODS_doc.sql 瀵归綈銆?""
|
||||
|
||||
def __init__(self, config, db_connection, api_client, logger):
|
||||
"""鍒濆鍖栫紦瀛橈紝閬垮厤閲嶅鏌ヨ琛ㄧ粨鏋勩€?""
|
||||
super().__init__(config, db_connection, api_client, logger)
|
||||
self._table_columns_cache: dict[str, list[str]] = {}
|
||||
|
||||
# 鏂囦欢鍏抽敭璇?-> 鐩爣琛紙鍖归厤 C:\dev\LLTQ\export\temp\source-data-doc 涓嬬ず鑼?JSON 鍚嶇О锛? FILE_MAPPING: list[tuple[tuple[str, ...], str]] = [
|
||||
(("浼氬憳妗ｆ", "member_profiles"), "billiards_ods.member_profiles"),
|
||||
(("浣欓鍙樻洿璁板綍", "member_balance_changes"), "billiards_ods.member_balance_changes"),
|
||||
(("鍌ㄥ€煎崱鍒楄〃", "member_stored_value_cards"), "billiards_ods.member_stored_value_cards"),
|
||||
(("鍏呭€艰褰?, "recharge_settlements"), "billiards_ods.recharge_settlements"),
|
||||
(("缁撹处璁板綍", "settlement_records"), "billiards_ods.settlement_records"),
|
||||
(("鍔╂暀搴熼櫎", "assistant_cancellation_records"), "billiards_ods.assistant_cancellation_records"),
|
||||
(("鍔╂暀璐﹀彿", "assistant_accounts_master"), "billiards_ods.assistant_accounts_master"),
|
||||
(("鍔╂暀娴佹按", "assistant_service_records"), "billiards_ods.assistant_service_records"),
|
||||
(("鍙版鍒楄〃", "site_tables_master"), "billiards_ods.site_tables_master"),
|
||||
(("鍙拌垂鎵撴姌", "table_fee_discount_records"), "billiards_ods.table_fee_discount_records"),
|
||||
(("鍙拌垂娴佹按", "table_fee_transactions"), "billiards_ods.table_fee_transactions"),
|
||||
(("搴撳瓨鍙樺寲璁板綍1", "goods_stock_movements"), "billiards_ods.goods_stock_movements"),
|
||||
(("搴撳瓨鍙樺寲璁板綍2", "stock_goods_category_tree"), "billiards_ods.stock_goods_category_tree"),
|
||||
(("搴撳瓨姹囨€?, "goods_stock_summary"), "billiards_ods.goods_stock_summary"),
|
||||
(("鏀粯璁板綍", "payment_transactions"), "billiards_ods.payment_transactions"),
|
||||
(("閫€娆捐褰?, "refund_transactions"), "billiards_ods.refund_transactions"),
|
||||
(("骞冲彴楠屽埜璁板綍", "platform_coupon_redemption_records"), "billiards_ods.platform_coupon_redemption_records"),
|
||||
(("鍥㈣喘濂楅娴佹按", "group_buy_redemption_records"), "billiards_ods.group_buy_packages_ledger"),
|
||||
(("鍥㈣喘濂楅", "group_buy_packages"), "billiards_ods.group_buy_packages"),
|
||||
(("灏忕エ璇︽儏", "settlement_ticket_details"), "billiards_ods.settlement_ticket_details"),
|
||||
(("闂ㄥ簵鍟嗗搧妗ｆ", "store_goods_master"), "billiards_ods.store_goods_master"),
|
||||
(("鍟嗗搧妗ｆ", "tenant_goods_master"), "billiards_ods.tenant_goods_master"),
|
||||
(("闂ㄥ簵鍟嗗搧閿€鍞褰?, "store_goods_sales_records"), "billiards_ods.store_goods_sales_records"),
|
||||
]
|
||||
|
||||
# 琛ㄧ粨鏋勮鏄庯細pk=涓婚敭鍒?None 琛ㄧず鏃犲啿绐佹洿鏂?锛宩son_cols=闇€瑕佸崟鍒楀瓨 JSONB 鐨勫瓧娈? TABLE_SPECS: dict[str, dict[str, Any]] = {
|
||||
"billiards_ods.member_profiles": {"pk": "id"},
|
||||
"billiards_ods.member_balance_changes": {"pk": "id"},
|
||||
"billiards_ods.member_stored_value_cards": {"pk": "id"},
|
||||
"billiards_ods.recharge_settlements": {"pk": None, "json_cols": ["settleList", "siteProfile"]},
|
||||
"billiards_ods.settlement_records": {"pk": None, "json_cols": ["settleList", "siteProfile"]},
|
||||
"billiards_ods.assistant_cancellation_records": {"pk": "id", "json_cols": ["siteProfile"]},
|
||||
"billiards_ods.assistant_accounts_master": {"pk": "id"},
|
||||
"billiards_ods.assistant_service_records": {"pk": "id", "json_cols": ["siteProfile"]},
|
||||
"billiards_ods.site_tables_master": {"pk": "id"},
|
||||
"billiards_ods.table_fee_discount_records": {"pk": "id", "json_cols": ["siteProfile", "tableProfile"]},
|
||||
"billiards_ods.table_fee_transactions": {"pk": "id", "json_cols": ["siteProfile"]},
|
||||
"billiards_ods.goods_stock_movements": {"pk": "siteGoodsStockId"},
|
||||
"billiards_ods.stock_goods_category_tree": {"pk": "id", "json_cols": ["categoryBoxes"]},
|
||||
"billiards_ods.goods_stock_summary": {"pk": "siteGoodsId"},
|
||||
"billiards_ods.payment_transactions": {"pk": "id", "json_cols": ["siteProfile"]},
|
||||
"billiards_ods.refund_transactions": {"pk": "id", "json_cols": ["siteProfile"]},
|
||||
"billiards_ods.platform_coupon_redemption_records": {"pk": "id"},
|
||||
"billiards_ods.tenant_goods_master": {"pk": "id"},
|
||||
"billiards_ods.group_buy_packages": {"pk": "id"},
|
||||
"billiards_ods.group_buy_packages_ledger": {"pk": "id"},
|
||||
"billiards_ods.settlement_ticket_details": {
|
||||
"pk": "orderSettleId",
|
||||
"json_cols": ["memberProfile", "orderItem", "tenantMemberCardLogs"],
|
||||
},
|
||||
"billiards_ods.store_goods_master": {"pk": "id"},
|
||||
"billiards_ods.store_goods_sales_records": {"pk": "id"},
|
||||
}
|
||||
|
||||
def get_task_code(self) -> str:
|
||||
"""杩斿洖浠诲姟缂栫爜銆?""
|
||||
return "MANUAL_INGEST"
|
||||
|
||||
def execute(self, cursor_data: dict | None = None) -> dict:
|
||||
"""浠庣ず鑼冪洰褰曡鍙?JSON锛屾寜琛?涓婚敭鎵归噺鍏ュ簱銆?""
|
||||
data_dir = (
|
||||
self.config.get("manual.data_dir")
|
||||
or self.config.get("pipeline.ingest_source_dir")
|
||||
or r"c:\dev\LLTQ\ETL\feiqiu-ETL\etl_billiards\tests\testdata_json"
|
||||
)
|
||||
if not os.path.exists(data_dir):
|
||||
self.logger.error("Data directory not found: %s", data_dir)
|
||||
return {"status": "error", "message": "Directory not found"}
|
||||
|
||||
counts = {"fetched": 0, "inserted": 0, "updated": 0, "skipped": 0, "errors": 0}
|
||||
|
||||
for filename in sorted(os.listdir(data_dir)):
|
||||
if not filename.endswith(".json"):
|
||||
continue
|
||||
filepath = os.path.join(data_dir, filename)
|
||||
try:
|
||||
with open(filepath, "r", encoding="utf-8") as fh:
|
||||
raw_entries = json.load(fh)
|
||||
except Exception:
|
||||
counts["errors"] += 1
|
||||
self.logger.exception("Failed to read %s", filename)
|
||||
continue
|
||||
|
||||
if not isinstance(raw_entries, list):
|
||||
raw_entries = [raw_entries]
|
||||
|
||||
records = self._extract_records(raw_entries)
|
||||
if not records:
|
||||
counts["skipped"] += 1
|
||||
continue
|
||||
|
||||
target_table = self._match_by_filename(filename)
|
||||
if not target_table:
|
||||
self.logger.warning("No mapping found for file: %s", filename)
|
||||
counts["skipped"] += 1
|
||||
continue
|
||||
|
||||
self.logger.info("Ingesting %s into %s", filename, target_table)
|
||||
try:
|
||||
inserted, updated = self._ingest_table(target_table, records, filename)
|
||||
counts["inserted"] += inserted
|
||||
counts["updated"] += updated
|
||||
counts["fetched"] += len(records)
|
||||
except Exception:
|
||||
counts["errors"] += 1
|
||||
self.logger.exception("Error processing %s", filename)
|
||||
self.db.rollback()
|
||||
continue
|
||||
|
||||
try:
|
||||
self.db.commit()
|
||||
except Exception:
|
||||
self.db.rollback()
|
||||
raise
|
||||
|
||||
return {"status": "SUCCESS", "counts": counts}
|
||||
|
||||
# ------------------------------------------------------------------ helpers
|
||||
def _match_by_filename(self, filename: str) -> str | None:
|
||||
"""鏍规嵁鏂囦欢鍚嶅叧閿瘝鎵惧埌鐩爣琛ㄣ€?""
|
||||
for keywords, table in self.FILE_MAPPING:
|
||||
if any(keyword and keyword in filename for keyword in keywords):
|
||||
return table
|
||||
return None
|
||||
|
||||
def _extract_records(self, raw_entries: Iterable[Any]) -> list[dict]:
|
||||
"""鍏煎澶氱 JSON 缁撴瀯锛屾彁鍙栨垚璁板綍鍒楄〃銆?""
|
||||
records: list[dict] = []
|
||||
for entry in raw_entries:
|
||||
if isinstance(entry, dict):
|
||||
# 濡傛灉鍚?data 涓旇繕鍖呭惈鍏朵粬閿紙濡?orderSettleId锛夛紝浼樺厛淇濈暀澶栧眰浠ュ厤涓㈠け涓婚敭
|
||||
preferred = entry
|
||||
if "data" in entry and not any(k not in {"data", "code"} for k in entry.keys()):
|
||||
preferred = entry["data"]
|
||||
data = preferred
|
||||
if isinstance(data, dict):
|
||||
list_used = False
|
||||
for v in data.values():
|
||||
if isinstance(v, list) and v and isinstance(v[0], dict):
|
||||
records.extend(v)
|
||||
list_used = True
|
||||
break
|
||||
if list_used:
|
||||
continue
|
||||
if isinstance(data, list) and data and isinstance(data[0], dict):
|
||||
records.extend(data)
|
||||
elif isinstance(data, dict):
|
||||
records.append(data)
|
||||
elif isinstance(entry, list):
|
||||
records.extend([item for item in entry if isinstance(item, dict)])
|
||||
return records
|
||||
|
||||
def _get_table_columns(self, table: str) -> list[str]:
|
||||
"""鏌ヨ淇℃伅_schema锛岃幏鍙栫洰鏍囪〃鐨勫叏閮ㄥ垪鍚嶏紙鎸夐『搴忥級銆?""
|
||||
if table in self._table_columns_cache:
|
||||
return self._table_columns_cache[table]
|
||||
if "." in table:
|
||||
schema, name = table.split(".", 1)
|
||||
else:
|
||||
schema, name = "public", table
|
||||
sql = """
|
||||
SELECT column_name, data_type, udt_name
|
||||
FROM information_schema.columns
|
||||
WHERE table_schema = %s AND table_name = %s
|
||||
ORDER BY ordinal_position
|
||||
"""
|
||||
with self.db.conn.cursor() as cur:
|
||||
cur.execute(sql, (schema, name))
|
||||
cols = [(r[0], (r[1] or "").lower(), (r[2] or "").lower()) for r in cur.fetchall()]
|
||||
self._table_columns_cache[table] = cols
|
||||
return cols
|
||||
|
||||
def _ingest_table(self, table: str, records: list[dict], source_file: str) -> tuple[int, int]:
|
||||
"""鏋勯€?INSERT/ON CONFLICT 璇彞骞舵壒閲忔墽琛屻€?""
|
||||
spec = self.TABLE_SPECS.get(table)
|
||||
if not spec:
|
||||
raise ValueError(f"No table spec for {table}")
|
||||
|
||||
pk_col = spec.get("pk")
|
||||
json_cols = set(spec.get("json_cols", []))
|
||||
json_cols_lower = {c.lower() for c in json_cols}
|
||||
|
||||
columns_info = self._get_table_columns(table)
|
||||
columns = [c[0] for c in columns_info]
|
||||
db_json_cols_lower = {
|
||||
c[0].lower() for c in columns_info if c[1] in ("json", "jsonb") or c[2] in ("json", "jsonb")
|
||||
}
|
||||
pk_col_db = None
|
||||
if pk_col:
|
||||
pk_col_db = next((c for c in columns if c.lower() == pk_col.lower()), pk_col)
|
||||
|
||||
placeholders = ", ".join(["%s"] * len(columns))
|
||||
col_list = ", ".join(f'"{c}"' for c in columns)
|
||||
sql = f'INSERT INTO {table} ({col_list}) VALUES ({placeholders})'
|
||||
if pk_col_db:
|
||||
update_cols = [c for c in columns if c != pk_col_db]
|
||||
set_clause = ", ".join(f'"{c}"=EXCLUDED."{c}"' for c in update_cols)
|
||||
sql += f' ON CONFLICT ("{pk_col_db}") DO UPDATE SET {set_clause}'
|
||||
sql += " RETURNING (xmax = 0) AS inserted"
|
||||
|
||||
params = []
|
||||
now = datetime.now()
|
||||
json_dump = lambda v: json.dumps(v, ensure_ascii=False) # noqa: E731
|
||||
for rec in records:
|
||||
merged_rec = rec if isinstance(rec, dict) else {}
|
||||
# 閫愬眰灞曞紑 data -> data.data 缁撴瀯锛屽～鍏呯己澶卞瓧娈? data_part = merged_rec.get("data")
|
||||
while isinstance(data_part, dict):
|
||||
merged_rec = {**data_part, **merged_rec}
|
||||
data_part = data_part.get("data")
|
||||
|
||||
pk_val = self._get_value_case_insensitive(merged_rec, pk_col) if pk_col else None
|
||||
if pk_col and (pk_val is None or pk_val == ""):
|
||||
continue
|
||||
|
||||
row_vals = []
|
||||
for col_name, data_type, udt in columns_info:
|
||||
col_lower = col_name.lower()
|
||||
if col_lower == "payload":
|
||||
row_vals.append(Json(rec, dumps=json_dump))
|
||||
continue
|
||||
if col_lower == "source_file":
|
||||
row_vals.append(source_file)
|
||||
continue
|
||||
if col_lower == "fetched_at":
|
||||
row_vals.append(merged_rec.get(col_name, now))
|
||||
continue
|
||||
|
||||
value = self._normalize_scalar(self._get_value_case_insensitive(merged_rec, col_name))
|
||||
|
||||
if col_lower in json_cols_lower or col_lower in db_json_cols_lower:
|
||||
row_vals.append(Json(value, dumps=json_dump) if value is not None else None)
|
||||
continue
|
||||
|
||||
casted = self._cast_value(value, data_type)
|
||||
row_vals.append(casted)
|
||||
params.append(tuple(row_vals))
|
||||
|
||||
if not params:
|
||||
return 0, 0
|
||||
|
||||
inserted = 0
|
||||
updated = 0
|
||||
with self.db.conn.cursor() as cur:
|
||||
for row in params:
|
||||
cur.execute(sql, row)
|
||||
try:
|
||||
flag = cur.fetchone()[0]
|
||||
except Exception:
|
||||
flag = None
|
||||
if flag:
|
||||
inserted += 1
|
||||
else:
|
||||
updated += 1
|
||||
return inserted, updated
|
||||
|
||||
def _get_value_case_insensitive(self, record: dict, col: str):
|
||||
"""蹇界暐澶у皬鍐欒幏鍙栧€硷紝鍏煎 information_schema 灏忓啓鍒楀悕涓?JSON 鍘熷澶у皬鍐欍€?""
|
||||
if record is None:
|
||||
return None
|
||||
if col is None:
|
||||
return None
|
||||
if col in record:
|
||||
return record.get(col)
|
||||
col_lower = col.lower()
|
||||
for k, v in record.items():
|
||||
if isinstance(k, str) and k.lower() == col_lower:
|
||||
return v
|
||||
return None
|
||||
|
||||
def _normalize_scalar(self, value):
|
||||
"""灏嗙┖瀛楃涓叉爣鍑嗗寲涓?None锛岄伩鍏嶆暟鍊?鏃堕棿瀛楁绫诲瀷閿欒銆?""
|
||||
if value == "" or value == "{}" or value == "[]":
|
||||
return None
|
||||
return value
|
||||
|
||||
def _cast_value(self, value, data_type: str):
|
||||
"""鏍规嵁鍒楃被鍨嬪仛杞婚噺杞崲锛岄伩鍏嶇被鍨嬩笉鍖归厤銆?""
|
||||
if value is None:
|
||||
return None
|
||||
dt = (data_type or "").lower()
|
||||
if dt in ("integer", "bigint", "smallint"):
|
||||
if isinstance(value, bool):
|
||||
return int(value)
|
||||
try:
|
||||
return int(value)
|
||||
except Exception:
|
||||
return None
|
||||
if dt in ("numeric", "double precision", "real", "decimal"):
|
||||
if isinstance(value, bool):
|
||||
return int(value)
|
||||
try:
|
||||
return float(value)
|
||||
except Exception:
|
||||
return None
|
||||
if dt.startswith("timestamp") or dt in ("date", "time", "interval"):
|
||||
# 浠呮帴鍙楀瓧绗︿覆/鏃ユ湡锛屾暟鍊肩瓑涓€寰嬬疆绌? return value if isinstance(value, str) else None
|
||||
return value
|
||||
|
||||
Reference in New Issue
Block a user