Files
GX.O3.WMS/WebApi/docs/compare_bom.py
T

58 lines
2.4 KiB
Python
Executable File

import pandas as pd
# 读取金蝶导出
kd = pd.read_excel('物料清单正查_.xlsx', sheet_name='树形单据体')
kd = kd[kd['子项物料编码'].notna()].copy()
kd['code'] = kd['子项物料编码'].astype(str).str.strip()
kd['level'] = kd['BOM层级'].astype(str).str.strip()
kd['usage'] = pd.to_numeric(kd['标准用量'], errors='coerce').fillna(1.0)
print(f"金蝶展开条数: {len(kd)}")
# 读取数据库展开
db = pd.read_csv('db_bom.csv', dtype={'ChildMaterialId': str, 'ParentMaterialId': str})
db['code'] = db['ChildMaterialId'].astype(str).str.strip()
db['level'] = db['Level'].astype(str).str.strip()
db['usage'] = pd.to_numeric(db['UsageQty'], errors='coerce').fillna(1.0)
print(f"数据库展开条数: {len(db)}")
kd_codes = kd['code'].tolist()
db_codes = db['code'].tolist()
from collections import Counter
kd_cnt = Counter(kd_codes)
db_cnt = Counter(db_codes)
# 编码集合差异
kd_set = set(kd_codes)
db_set = set(db_codes)
only_kd = kd_set - db_set
only_db = db_set - kd_set
print(f"\n=== 编码集合对比 ===")
print(f"金蝶有但数据库没有的编码数: {len(only_kd)}")
for c in sorted(only_kd):
print(f" 金蝶独有: {c} (金蝶出现{kd_cnt[c]}次)")
print(f"数据库有但金蝶没有的编码数: {len(only_db)}")
for c in sorted(only_db):
print(f" 数据库独有: {c} (数据库出现{db_cnt[c]}次)")
# 出现次数差异(同一编码在两边出现次数不同)
print(f"\n=== 出现次数差异(同一编码两边次数不同) ===")
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")
# 用量差异(同编码同层级)
print(f"\n=== 用量差异(同编码出现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()
diff_usage = 0
for c in sorted(set(kd_grp.keys()) & set(db_grp.keys())):
if kd_grp[c] != db_grp[c]:
diff_usage += 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: {diff_usage}")
# 总条数差异说明
print(f"\n=== 汇总 ===")
print(f"金蝶总条数: {len(kd)}, 数据库总条数: {len(db)}, 差值: {len(db)-len(kd)}")