#!/usr/bin/env python3
"""检查 user_dashou 中有 user_id 的记录"""
import re

with open('xaio_cheng_xu_backup.sql', 'r', encoding='utf-8') as f:
    content = f.read()

pattern = re.compile(r"INSERT\s+INTO\s+`user_dashou`\s+VALUES\s*\((.+?)\)\s*;", re.IGNORECASE | re.DOTALL)
matches = list(pattern.finditer(content))

# 找有 user_id 的记录
has_uid = 0
no_uid = 0
uid_values = []
for m in matches:
    v = m.group(1).rstrip()
    if v.endswith('NULL'):
        no_uid += 1
    else:
        has_uid += 1
        # 提取最后一个值（user_id）
        last_comma = v.rfind(',')
        uid_val = v[last_comma+1:].strip()
        uid_values.append(uid_val)

print(f"有 user_id: {has_uid}, 无 user_id: {no_uid}")
print(f"user_id 值样本: {uid_values[:10]}")

# 检查 user_main 中这些 user_id 是否存在
# user_main 的 id 字段是 vals[1]
main_pattern = re.compile(r"INSERT\s+INTO\s+`user_main`\s+VALUES\s*\((.+?)\)\s*;", re.IGNORECASE | re.DOTALL)
main_ids = set()
for m in main_pattern.finditer(content):
    # 简单提取第二个字段 (id)
    v = m.group(1)
    # 找前几个逗号
    in_q = False
    commas = []
    for i, ch in enumerate(v):
        if ch == "'": in_q = not in_q
        elif ch == ',' and not in_q:
            commas.append(i)
            if len(commas) >= 2: break
    if len(commas) >= 2:
        id_val = v[commas[0]+1:commas[1]].strip()
        main_ids.add(id_val)

print(f"\nuser_main id 范围: {len(main_ids)} 个")
matched = sum(1 for uid in uid_values if uid in main_ids)
print(f"user_dashou user_id 匹配 user_main: {matched}/{len(uid_values)}")

# 也检查 user_type 分布
type_pattern = re.compile(r"INSERT\s+INTO\s+`user_main`\s+VALUES\s*\((.+?)\)\s*;", re.IGNORECASE | re.DOTALL)
type_counts = {}
for m in type_pattern.finditer(content):
    v = m.group(1)
    # 找第12个字段 (user_type, index 11)
    in_q = False
    commas = []
    for i, ch in enumerate(v):
        if ch == "'": in_q = not in_q
        elif ch == ',' and not in_q:
            commas.append(i)
            if len(commas) >= 12: break
    if len(commas) >= 12:
        type_val = v[commas[10]+1:commas[11]].strip().strip("'")
        type_counts[type_val] = type_counts.get(type_val, 0) + 1

print(f"\nuser_type 分布: {type_counts}")