Www Mi Com

Mi.com Product Spec Extractor

Extract detailed product specifications from Xiaomi's official website (mi.com).

Overview

This skill accesses Xiaomi's official product pages to extract structured specification data. It supports both Hong Kong and Mainland China product pages.

Key Features

• Structured Spec Extraction: Parses specification tables into organized sections
• Multi-Region Support: Works with both HK (/hk/) and Mainland China URLs
• No JavaScript Required: HK specs pages can be parsed directly with HTML parsing
• Batch Processing: Extract specs for multiple products in one call

Supported URL Patterns

Hong Kong Specs Pages (Recommended)

https://www.mi.com/hk/product/{product}/specs/

Example: https://www.mi.com/hk/product/redmi-note-12-5g/specs/

These URLs contain fully structured specification tables that can be parsed directly without JavaScript execution. This is the most reliable method.

Hong Kong Product Pages

https://www.mi.com/hk/product/{product}/

Example: https://www.mi.com/hk/product/redmi-note-12-5g/

Basic product info available; for full specs, use the /specs/ URL.

Mainland China Product Pages

https://www.mi.com/{product}
https://www.mi.com/prod/{product}

Examples:

• https://www.mi.com/redmi-note-13-5g
• https://www.mi.com/prod/redmi-note-14

These pages are Vue.js applications and require Playwright for full spec extraction.

Usage

Basic Usage

from executor import execute

# Extract specs from a Hong Kong specs page
result = await execute({
    'url': 'https://www.mi.com/hk/product/redmi-note-12-5g/specs/'
})

print(result['product']['name'])  # "Redmi Note 12 5G"
print(result['product']['specs'].keys())  # ['螢幕', '設計', '後置鏡頭', ...]

Extracting Multiple Products

result = await execute({
    'urls': [
        'https://www.mi.com/hk/product/redmi-note-12-5g/specs/',
        'https://www.mi.com/hk/product/redmi-note-13-5g/specs/',
    ]
})

for product in result['products']:
    print(f"{product['name']}: {len(product['specs'])} sections")

Using Playwright for Mainland China Pages

result = await execute({
    'url': 'https://www.mi.com/redmi-note-13-5g',
    'use_playwright': True
})

Output Structure

{
  "success": true,
  "product": {
    "name": "Redmi Note 12 5G",
    "url": "https://www.mi.com/hk/product/redmi-note-12-5g/specs/",
    "title": "Redmi Note 12 5G 規格與功能 | Xiaomi 香港",
    "description": "Redmi Note 12 5G 規格...",
    "image": "//i02.appmifile.com/.../product.png",
    "region": "Hong Kong",
    "specs": {
      "螢幕": [
        {"text": "6.67\" AMOLED DotDisplay 螢幕", "data_key": "spec_3"},
        {"text": "螢幕更新頻率：高達 120Hz", "data_key": "spec_5"},
        {"text": "450 尼特（典型值），HBM 700 尼特，1200 尼特峰值亮度", "data_key": "spec_7"}
      ],
      "設計": [
        {"text": "165.88*76.21*7.98mm", "data_key": "spec_16"},
        {"text": "189g", "data_key": "spec_17"},
        {"text": "瑪瑙灰、森林綠、冰河藍", "data_key": "spec_18"}
      ],
      "後置鏡頭": [
        {"text": "4800 萬像素主鏡頭", "data_key": "spec_20"},
        {"text": "f/1.8", "data_key": "spec_23"},
        ...
      ],
      ...
    },
    "raw_specs": [
      {"section": "螢幕", "text": "6.67\" AMOLED DotDisplay 螢幕", "data_key": "spec_3"},
      ...
    ]
  }
}

Spec Sections

Hong Kong specs pages typically include the following sections (in Traditional Chinese):

Parameters

Technical Details

Parsing Method

1. Hong Kong Specs Pages: Direct HTML parsing using BeautifulSoup

   • Identifies spec sections by font-size styling (headers use 35px font)
   • Extracts spec items from <span class="xm-text"> elements
   • Each item has a data-key attribute for cross-referencing

2. Mainland China Pages: Playwright browser rendering

   • Waits for Vue.js app to fully render
   • Extracts data from DOM after JavaScript execution
   • Parses tables, definition lists, and spec items

Data Key System

Each spec item has a data-key attribute like spec_3, spec_5, etc. This can be used to:

• Cross-reference specs between different language versions
• Track which spec is which when product information is updated

Dependencies

Required:

• aiohttp - HTTP client for async requests
• beautifulsoup4 - HTML parsing

Optional:

• playwright - For Mainland China pages (install with: pip install playwright && playwright install chromium)

Examples

Get display specifications

result = await execute({'url': 'https://www.mi.com/hk/product/redmi-note-12-5g/specs/'})

display_specs = result['product']['specs'].get('螢幕', [])
for spec in display_specs:
    print(f"- {spec['text']}")

Output:

- 6.67" AMOLED DotDisplay 螢幕
- 螢幕更新頻率：高達 120Hz
- 450 尼特（典型值），HBM 700 尼特，1200 尼特峰值亮度
- 對比度：4500000:1
- DCI-P3 寬廣色域
- 解析度：2400 x 1080
- 陽光螢幕
- 閱讀模式

Search for specific spec

result = await execute({
    'url': 'https://www.mi.com/hk/product/redmi-note-12-5g/specs/',
    'include_raw_specs': True
})

# Find battery capacity
for spec in result['product']['raw_specs']:
    if '電池' in spec['section'] or 'Battery' in spec['section']:
        print(f"{spec['section']}: {spec['text']}")

Limitations

1. Mainland China pages require Playwright and are slower to process
2. Spec section names are in Chinese and may vary by region
3. Some specs may be listed under different sections for different products
4. The parser relies on HTML structure which may change if Xiaomi updates their site

Error Handling

result = await execute({'url': 'https://www.mi.com/hk/product/invalid-product/specs/'})

if not result['success']:
    print(f"Error: {result.get('error')}")