#!/usr/bin/env python3 """2026-09-08 钉盘数据增量导入：日出库补录(09-04/05/06店铺+09-07旺店通) + 库存09-08""" import csv, os, json, subprocess, pymysql, importlib.util, sys from datetime import datetime from collections import defaultdict sys.path.insert(0, os.path.dirname(os.path.abspath(__file__))) s = importlib.util.spec_from_file_location('config', 'config.py') m = importlib.util.module_from_spec(s); s.loader.exec_module(m) DB = m.DB_CONFIG SPACE_ID = '28814565790' OB_FOLDER = '3NwLYZXWynyBjRB4tZjwNxDvVkyEqBQm' IV_FOLDER = 'NkDwLng8ZLGrXmrnFxYq3vRqVKMEvZBY' TMP = 'tmp_csv/update_20260908' os.makedirs(TMP, exist_ok=True) def si(v): try: return int(float(str(v or 0).replace('"','').replace('=','').strip())) except: return 0 def sf(v): try: return float(str(v or 0).replace('"','').replace('=','').strip()) except: return 0.0 def clean(v): return str(v or '').replace('"','').replace('=','').strip() def bad(c): n = clean(c).upper().rstrip(':') return not n or n in ('NA','N/A','合计') SPECIALS = {'POZ-FM-0184-TM-SM':.055,'POZ-XFS-9014-TM-MD':.062,'POZ-FM-9742-TM-MD':.181, 'POZ-MM-0580-TM-MD':.22,'POZ-SKS-3374-LD':.1,'POZ-SKS-3367-LD':.1, 'POZ-SKS-3350-LD':.1,'POZ-SKS-3343-LD':.1} def dws_list(folder): out=[]; cursor=None while True: args=['dws','drive','list','--space-id',SPACE_ID,'--folder',folder,'--limit','50','--format','json'] if cursor: args+=['--cursor',cursor] r=subprocess.run(args,capture_output=True,text=True) if r.returncode!=0: print(' [list err]',r.stderr[:150]); break d=json.loads(r.stdout).get('result',{}) out+=d.get('items',[]) cursor=d.get('nextToken') if not cursor: break return out def dl(name, fileId): dest=os.path.join(TMP,name) if os.path.exists(dest) and os.path.getsize(dest)>10: return dest r=subprocess.run(['dws','drive','download','--node',fileId,'--space-id',SPACE_ID,'--output',dest,'--format','json'],capture_output=True,text=True) if r.returncode!=0 or not os.path.exists(dest) or os.path.getsize(dest)<10: print(' [dl err]',name,r.stderr[:120]); return None return dest def read_rows_auto(path): """尝试多种编码，返回 (header_list, data_rows)""" for enc in ['utf-8-sig','gbk','gb18030']: try: with open(path, encoding=enc) as f: rr=list(csv.reader(f)) if rr: return rr[0], rr[1:] except: continue return None, None def import_outbound(conn, date, src, path): """wdt 全格式 或 店铺2列简版；返回插入条数""" with conn.cursor() as x: x.execute('DELETE FROM daily_outbound WHERE 日期=%s AND 数据来源=%s',(date,src)) header, rows = read_rows_auto(path) if rows is None: print(' [read err]',path); return 0 inserted=0 with conn.cursor() as x: if src == 'wdt': # 全格式 DictReader（wdt 文件为 GBK） with open(path, encoding='gb18030') as f: for row in csv.DictReader(f): code=clean(row.get('货品编号')) if bad(code): continue x.execute('INSERT INTO daily_outbound (货品编号,货品名称,品牌,仓库,店铺,一级类目,二级类目,三级类目,货品数量,货品成本,日期,数据来源,数据类型) VALUES (%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s)', (code,clean(row.get('货品名称')),clean(row.get('品牌')),clean(row.get('仓库')),clean(row.get('店铺')), clean(row.get('一级类目')),clean(row.get('二级类目')),clean(row.get('三级类目')), si(row.get('货品数量')),sf(row.get('货品成本')),date,src,'出库数据')) inserted+=1 else: # 店铺简版：跳过表头，col0=编码 col1=数量 for raw in rows: if not raw or len(raw)<2: continue code=clean(raw[0]); qty=si(raw[1]) if bad(code): continue x.execute('SELECT product_name, brand, category1, category2, category3, cost_price FROM products WHERE product_code=%s LIMIT 1',(code,)) p=x.fetchone() if p: pname=p['product_name'] or '' if clean(pname) in ('-','--'): continue brand=p['brand'] or ''; c1=p['category1'] or ''; c2=p['category2'] or ''; c3=p['category3'] or '' cost=float(p['cost_price']) if p['cost_price'] else 0 else: pname=brand=c1=c2=c3=''; cost=0 x.execute('INSERT INTO daily_outbound (货品编号,货品名称,品牌,仓库,店铺,一级类目,二级类目,三级类目,货品数量,货品成本,日期,数据来源,数据类型) VALUES (%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s)', (code,pname,brand,'','',c1,c2,c3,qty,cost,date,src,'出库数据')) inserted+=1 return inserted def import_inventory(conn, date, path, wh): with conn.cursor() as x: x.execute('DELETE FROM inventory_detail WHERE data_date=%s AND warehouse_code=%s',(date,wh)) inserted=0 with conn.cursor() as x: with open(path, encoding='utf-8-sig') as f: for row in csv.DictReader(f): code=clean(row.get('货品编号')) if bad(code) or clean(row.get('货品名称')) in ('-','--'): continue cost=SPECIALS.get(code, sf(row.get('成本价'))) ct=clean(row.get('单品创建时间')); dt=None for fmt in ('%Y-%m-%d %H:%M:%S','%Y-%m-%d'): try: dt=datetime.strptime(ct,fmt); break except: pass x.execute('INSERT INTO inventory_detail (data_date,source,warehouse_code,goods_code,product_code,product_name,brand,cost_price,available_qty,production_in_transit,purchase_in_transit,category1,category2,category3,created_time) VALUES (%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s,%s)', (date,'wdt',wh,clean(row.get('商家编码')),code,clean(row.get('货品名称')),clean(row.get('品牌')),cost, si(row.get('库存')),si(row.get('生产在途量')),si(row.get('采购在途')), clean(row.get('一级类目')),clean(row.get('二级类目')),clean(row.get('三级类目')),dt)) inserted+=1 return inserted def main(): print('扫描钉盘...') ob = dws_list(OB_FOLDER) iv = dws_list(IV_FOLDER) # 组织文件 ob_files = {} # name -> fileId for x in ob: if x['type']=='FILE' and x['name'].endswith('.csv'): ob_files[x['name']]=x['fileId'] iv_files = {} for x in iv: if x['type']=='FILE' and x['name'].endswith('.csv'): iv_files[x['name']]=x['fileId'] # 现有出库组合 conn = pymysql.connect(**DB, cursorclass=pymysql.cursors.DictCursor) existing_ob = set() with conn.cursor() as x: x.execute("SELECT 日期,数据来源 FROM daily_outbound") for r in x.fetchall(): d = r['日期'].strftime('%Y-%m-%d') if hasattr(r['日期'],'strftime') else str(r['日期'])[:10] existing_ob.add((d, r['数据来源'])) x.execute("SELECT DISTINCT data_date FROM inventory_detail") existing_iv = set(str(r['data_date'])[:10] for r in x.fetchall()) # 目标：日出库 09-04 ~ 09-07 所有来源 need_ob = [] for name, fid in sorted(ob_files.items()): p = name.split('-') if len(p) < 3 or not p[0].isdigit(): continue date = f'{p[0][:4]}-{p[0][4:6]}-{p[0][6:8]}' if not (date >= '2026-09-04' and date <= '2026-09-07'): continue src = p[1] if (date, src) in existing_ob: continue need_ob.append((name, fid, date, src)) # 目标：库存 09-08 need_iv = [] for name, fid in sorted(iv_files.items()): p = name.split('-') if not (p and p[0]=='20260908' and name.endswith('.csv')): continue wh = name.rsplit('-',1)[1][:-4] if '2026-09-08' in existing_iv and wh in [n.rsplit('-',1)[1][:-4] for n in iv_files if n.startswith('20260908') and '2026-09-08' in existing_iv]: pass need_iv.append((name, fid, '2026-09-08', wh)) print(f'需导入日出库 {len(need_ob)} 个文件, 库存 {len(need_iv)} 个文件') # 下载并导入出库 ob_ins = 0 for name, fid, date, src in need_ob: p = dl(name, fid) if not p: continue n = import_outbound(conn, date, src, p) ob_ins += n print(f' 出库 {name} ({date} {src}) 导入 {n} 条') # 下载并导入库存 09-08 iv_ins = 0 for name, fid, date, wh in need_iv: p = dl(name, fid) if not p: continue n = import_inventory(conn, date, p, wh) iv_ins += n print(f' 库存 {name} (仓库{wh}) 导入 {n} 条') conn.commit() conn.close() print(f'\n完成: 出库新增 {ob_ins} 条, 库存新增 {iv_ins} 条') if __name__ == '__main__': main()