import pandas as pd from collections import Counter # 金蝶正查 CG0S175E0000120 kd = pd.read_excel('物料清单正查_1.xlsx', sheet_name='树形单据体') kd = kd[kd['子项物料编码'].notna()].copy() kd = kd[kd['BOM层级'].astype(str).str.strip() != '0'] # 去掉根 kd['code'] = kd['子项物料编码'].astype(str).str.strip() kd['level'] = kd['BOM层级'].astype(str).str.strip() kd['name'] = kd['物料名称'].astype(str).str.strip() kd['usage'] = pd.to_numeric(kd['标准用量'], errors='coerce').fillna(1.0) # 数据库展开 db = pd.read_csv('db_bom.csv', dtype={'ChildMaterialId': str, 'ParentMaterialId': str}) db['code'] = db['ChildMaterialId'].astype(str).str.strip() db['usage'] = pd.to_numeric(db['UsageQty'], errors='coerce').fillna(1.0) print(f"金蝶子项条数: {len(kd)}") print(f"数据库条数: {len(db)}") print(f"差异: {len(db) - len(kd)}") kd_cnt = Counter(kd['code']) db_cnt = Counter(db['code']) kd_set = set(kd['code']) db_set = set(db['code']) only_kd = kd_set - db_set only_db = db_set - kd_set print(f"\n=== 编码集合差异 ===") print(f"金蝶独有编码: {len(only_kd)} 个") for c in sorted(only_kd): rows = kd[kd['code']==c] names = set(rows['name']) print(f" {c} (金蝶{kd_cnt[c]}次) 名称={names}") print(f"\n数据库独有编码: {len(only_db)} 个") for c in sorted(only_db): print(f" {c} (数据库{db_cnt[c]}次)") print(f"\n=== 出现次数差异(同编码两边次数不同) ===") for c in sorted(set(kd_cnt.keys()) & set(db_cnt.keys())): if kd_cnt[c] != db_cnt[c]: print(f" {c}: 金蝶{kd_cnt[c]}次 vs 数据库{db_cnt[c]}次") print(f"\n=== 用量差异(同编码次数相同但用量不同) ===") kd_grp = kd.groupby('code')['usage'].apply(lambda x: sorted(x.tolist())).to_dict() db_grp = db.groupby('code')['usage'].apply(lambda x: sorted(x.tolist())).to_dict() cnt = 0 for c in sorted(set(kd_grp.keys()) & set(db_grp.keys())): if kd_grp[c] != db_grp[c]: cnt += 1 print(f" {c}: 金蝶用量{kd_grp[c]} vs 数据库用量{db_grp[c]}") print(f"用量不同的编码数: {cnt}")