# Data Cross-Reference: Employment Report × Admission Data × School Map

## When to use
Building articles that combine external data (employment blue book, policy) with the site's own admission database — notably "春考生能报的高薪专业清单" type articles.

## Data sources (gdcjgk.net)

| Source | Table | Key fields |
|--------|-------|------------|
| Admission data | `2025` (diyform) | `nianfen`, `glxy` (school FK), `zhaoshengpici` (批次), `title` (专业名), `zhaoshengjihua`, `luquzuidifen`, `luquzuidipaiwei`, `zhengshuyaoqiu`, `xuefei` |
| School names | `fa_cms_archives` (model_id=2) | `id`, `title` (school name) |
| Employment data | External — 麦可思就业蓝皮书 | 高职月薪, 专业对口率 by 大类 |

## Cross-reference query

```python
import sqlite3, re

# Step 1: Build school name map from DB dump
schools = {}
for m in re.finditer(
    r"INSERT INTO `fa_cms_archives` VALUES \((\d+), \d+, \d+, '[^']*', 2, '[^']*', \d+, '([^']+)'",
    dump_content
):
    schools[m.group(1)] = m.group(2)

# Step 2: Parse 2025 table (字段位置固定)
# (NULL, year, glxy, '批次', '专业', zyz, code, plan, 'cert', 'features', 'fee', 'dorm', 'location', memo, admit, min_rank, min_score, name_id)
pattern = r"NULL, (\d+), (\d+), '([^']+)', '([^']+)', (\d+), (\d+), (\d+), '([^']*?)', '([^']*?)', '([^']+)', '([^']*?)', '([^']*?)', (NULL|\d+), (\d+), (\d+), ([\d.]+), (\d+)\);"

for m in re.finditer(pattern, dump_content):
    year, glxy, batch, title, zyz, code, plan, cert, features, fee, dorm, location, memo, admit, min_rank, min_score, name_id = m.groups()
    if batch != '专科-中职生':
        continue
    school_name = schools.get(glxy, f'ID{glxy}')
    # ... match against high-demand keyword list

# Step 3: Categorize by employment report fields
high_kw_map = {
    '医药卫生': ['护理', '药学', '康复治疗', '医学检验', '医学影像', '口腔医学'],
    '能源制造': ['电气自动化技术', '新能源', '机电一体化', '数控技术', '工业机器人'],
    '生物化工': ['食品检验', '环境监测', '生物制药', '应用化工'],
    'IT数字': ['计算机应用', '软件技术', '大数据', '人工智能', '云计算', '物联网'],
}
```

## Verified data (2025 intake)

- 专科-中职生: 2,731 majors across 106 schools
- 高需求专业 matches: 750 (178 医药 + 192 能源 + 45 生物 + 335 IT)
- Top schools mapped: 深圳职业技术大学, 顺德职业技术大学, 广东轻工职业技术大学, 东莞职业技术学院, etc.

## DB dump location

`/home/agentuser/.hermes/cache/documents/doc_1777df84589d_网站数据库备份（5.9）.sql`

The live DB is on `gdcjgk.net` server — the dump is a snapshot, adequate for content planning but not real-time.