#!/usr/bin/env python3
"""Verify njghes FTS and content quality"""
import sqlite3

DB_PATH = "/root/search.db"
conn = sqlite3.connect(DB_PATH)
conn.row_factory = sqlite3.Row
c = conn.cursor()

# FTS search
search_term = "土壤污染调查"
c.execute(
    "SELECT COUNT(*) FROM gov_search WHERE summary MATCH ?",
    ("\"" + search_term + "\"",)
)
fts_hits = c.fetchone()[0]
print(f'FTS搜索 "{search_term}": {fts_hits}条')

# Count
c.execute("SELECT COUNT(*) FROM gov_raw WHERE script_name='crawl_njghes_tzgg.py'")
total = c.fetchone()[0]
print(f'总记录数: {total}条')

# Show samples with content quality
c.execute("""
    SELECT id, title, length(content), 
           CASE WHEN content LIKE '%\n\n%' THEN '分段OK' ELSE '未分段' END as quality
    FROM gov_raw 
    WHERE script_name='crawl_njghes_tzgg.py' 
    ORDER BY id DESC LIMIT 5
""")
for row in c.fetchall():
    print(f'#{row["id"]} {row["title"][:50]} ({row["length(content)"]}chars) [{row["quality"]}]')

# Show one full content
c.execute("""
    SELECT title, content FROM gov_raw 
    WHERE script_name='crawl_njghes_tzgg.py' 
    ORDER BY id DESC LIMIT 1
""")
row = c.fetchone()
if row:
    print(f'\n最新条目标题: {row["title"]}')
    print(f'正文({len(row["content"])}字符):')
    for i, p in enumerate(row["content"].split("\n\n")):
        print(f'  [段落{i+1}] {p[:100]}')

conn.close()