import pandas as pd 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() db = pd.read_csv('db_bom.csv', dtype={'ChildMaterialId': str, 'ParentMaterialId': str}) db['code'] = db['ChildMaterialId'].astype(str).str.strip() db['parent'] = db['ParentMaterialId'].astype(str).str.strip() kd_codes = set(kd['code']) db_codes = set(db['code']) only_db = sorted(db_codes - kd_codes) print("=== 数据库独有编码的上下文(父级/层级/名称/规格) ===") for c in only_db: rows = db[db['code'] == c] for _, r in rows.iterrows(): print(f" {c} | 父级={r['parent']} | 层级={r['Level']} | 用量={r['UsageQty']} | 名称={r['MaterialName']} | 规格={r['Spec']}") # 尝试匹配:数据库独有编码 vs 金蝶编码(去掉版本后缀比较) import re def strip_suffix(code): # 去掉末尾的 -XX 版本后缀 return re.sub(r'-[A-Z]\d+$', '', code) print("\n=== 数据库独有编码 与 金蝶编码 去后缀后的匹配 ===") kd_stripped = {} for c in kd_codes: s = strip_suffix(c) kd_stripped.setdefault(s, []).append(c) for c in only_db: s = strip_suffix(c) if s in kd_stripped: print(f" {c} (数据库) 去后缀={s} -> 金蝶有: {kd_stripped[s]}")