Gaokao Eol Cn

Gaokao Data Extraction Skill

This skill extracts structured table data from the Chinese College Entrance Examination (高考) portal at gaokao.eol.cn.

Features

Extracted Data Types

1. Admission Lines (投档线)

   • University admission scores by major group
   • Data includes: School ID, School Name, Group ID, Subject Requirements, Cut-off Score
   • Example: Beijing 2024 undergraduate admission lines

2. Score Distribution (一分一段表)

   • Score distribution tables showing student count per score segment
   • Data includes: Score segment, Count in segment, Cumulative count
   • Example: Beijing 2023 score distribution

Supported URLs

The skill can fetch any page from gaokao.eol.cn that contains tabular data, including:

• Province-specific pages (e.g., /bei_jing/dongtai/)
• Historical data (multiple years)
• Admission announcements with score tables

Functions

get_admission_lines

Extracts admission line data from a gaokao.eol.cn URL.

Parameters:

• url (required): Full URL to the page containing admission line tables

Returns:

• Success indicator
• Metadata (title, province, year)
• Array of records with: school_id, school_name, group_id, group_subject, score

Example:

result = await execute({
    'function': 'get_admission_lines',
    'url': 'https://gaokao.eol.cn/bei_jing/dongtai/202407/t20240720_2625168.shtml'
})

get_score_distribution

Extracts score distribution data from a gaokao.eol.cn URL.

Parameters:

• url (required): Full URL to the page containing score distribution tables

Returns:

• Success indicator
• Metadata (title, province, year)
• Array of records with: score, original_score, count, cumulative_count

Example:

result = await execute({
    'function': 'get_score_distribution',
    'url': 'https://gaokao.eol.cn/bei_jing/dongtai/202306/t20230625_2446872.shtml'
})

fetch_page

Generic function to fetch any gaokao.eol.cn page and extract all table data.

Parameters:

• url (required): Full URL to the page

Returns:

• Success indicator
• Page metadata (title, province, year, data_type)
• All tables found on the page with raw data
• Auto-detected data type based on page content
• Parsed records if data type is recognized

Example:

result = await execute({
    'function': 'fetch_page',
    'url': 'https://gaokao.eol.cn/bei_jing/dongtai/202407/t20240720_2625168.shtml'
})

Implementation Details

• Uses aiohttp for async HTTP requests
• BeautifulSoup for HTML parsing
• Automatically detects data type from page title
• Extracts metadata (year, province) from URL and title
• Handles Chinese text encoding properly
• No JavaScript execution needed (server-side rendered HTML)

Error Handling

Returns structured error messages for:

• Invalid URLs
• Missing required parameters
• HTTP errors
• Timeout errors
• Missing or invalid table data
• Data type mismatches

Response Format

All functions return a dictionary with:

{
    'success': bool,           # Whether the operation succeeded
    'url': str,                # The fetched URL
    'title': str,              # Page title
    'province': str,           # Extracted province code
    'year': str,               # Extracted year
    'data_type': str,          # 'admission_lines' or 'score_distribution'
    'total_records': int,      # Number of extracted records
    'records': list,           # Array of structured records
    'error': str               # Error message if success=False
}