{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 属性统计\n",
    "def parse_attr_set(attr_set_str):\n",
    "    \"\"\"解析属性集，返回 {属性值: 属性名} 的字典\"\"\"\n",
    "    mapping = {}\n",
    "    for item in attr_set_str.split(\";\"):\n",
    "        item = item.strip()\n",
    "        if not item:\n",
    "            continue\n",
    "        try:\n",
    "            left, value = item.split(\"~\", 1)   # \"231012:大小~23206184:140*80*75\"\n",
    "            attr_name = left.split(\":\")[1]\n",
    "            value = value.split(\":\")[1]\n",
    "            mapping[value] = attr_name\n",
    "        except Exception:\n",
    "            continue\n",
    "    return mapping\n",
    "def map_spec_to_attrs(spec, attr_mapping):\n",
    "    \"\"\"把规格里的值映射为 {属性名: 规格值}\"\"\"\n",
    "    results = {}\n",
    "    if not isinstance(attr_mapping, dict):  # 如果不是字典，直接返回 None\n",
    "        return None\n",
    "\n",
    "    for val in str(spec).split():  # spec 也转成 str，避免 NaN\n",
    "        if val in attr_mapping:\n",
    "            attr_name = attr_mapping[val]\n",
    "            results[attr_name] = val\n",
    "    return results if results else None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from utils.gtools import MySQLconnect\n",
    "\n",
    "batch_size = 1000000\n",
    "offset = 0\n",
    "results = []\n",
    "leimu = \"furniture\"\n",
    "with MySQLconnect('ods') as ods:\n",
    "    while True:\n",
    "        sql = f\"\"\"\n",
    "        SELECT * FROM erp_{leimu}_sku\n",
    "        LIMIT {batch_size} OFFSET {offset}\n",
    "        \"\"\"\n",
    "        chunk = pd.read_sql(sql, ods.con)\n",
    "        if chunk.empty:\n",
    "            break\n",
    "        chunk[\"规格属性映射\"] = chunk.apply(\n",
    "            lambda row: map_spec_to_attrs(row[\"规格\"], parse_attr_set(row[\"标准/预设属性集\"])),\n",
    "            axis=1\n",
    "        )\n",
    "        print(f\"处理了 {offset} 到 {offset+batch_size} 条数据\")\n",
    "        results.append(chunk)\n",
    "        offset += batch_size\n",
    "df = pd.concat(results, ignore_index=True)\n",
    "all_attrs = {}\n",
    "for mapping in df[\"规格属性映射\"].dropna():\n",
    "    for attr_name, val in mapping.items():\n",
    "        if attr_name not in all_attrs:  # 只保留一个样本\n",
    "            all_attrs[attr_name] = val\n",
    "\n",
    "print(all_attrs)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 保存\n",
    "out_df = pd.DataFrame(list(all_attrs.items()), columns=[\"规格属性映射\", \"属性值\"])\n",
    "out_df.to_excel(f\"D:/test/logistics/test_excel/{leimu}-规格属性映射.xlsx\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}