增强风险管理师分析功能
This commit is contained in:
oficcejo
@@ -0,0 +1,469 @@
|
||||
"""
|
||||
风险数据获取模块
|
||||
使用pywencai获取股票风险相关信息：
|
||||
1. 限售解禁数据
|
||||
2. 大股东减持公告
|
||||
3. 近期重要事件
|
||||
"""
|
||||
|
||||
import pywencai
|
||||
import pandas as pd
|
||||
from typing import Dict, Any
|
||||
import time
|
||||
import warnings
|
||||
import os
|
||||
|
||||
# 屏蔽pywencai的Node.js警告信息（不影响功能）
|
||||
warnings.filterwarnings('ignore', category=DeprecationWarning)
|
||||
os.environ['PYTHONWARNINGS'] = 'ignore::DeprecationWarning'
|
||||
os.environ['NODE_NO_WARNINGS'] = '1' # 屏蔽Node.js警告
|
||||
|
||||
|
||||
class RiskDataFetcher:
|
||||
"""风险数据获取类"""
|
||||
|
||||
def __init__(self):
|
||||
"""初始化"""
|
||||
pass
|
||||
|
||||
def get_risk_data(self, symbol: str) -> Dict[str, Any]:
|
||||
"""
|
||||
获取股票风险相关数据
|
||||
|
||||
Args:
|
||||
symbol: 股票代码（如：600000）
|
||||
|
||||
Returns:
|
||||
包含风险数据的字典
|
||||
"""
|
||||
print(f"\n正在获取 {symbol} 的风险数据...")
|
||||
|
||||
risk_data = {
|
||||
'symbol': symbol,
|
||||
'data_success': False,
|
||||
'lifting_ban': None, # 限售解禁数据
|
||||
'shareholder_reduction': None, # 大股东减持数据
|
||||
'important_events': None, # 重要事件数据
|
||||
'error': None
|
||||
}
|
||||
|
||||
try:
|
||||
# 1. 获取限售解禁数据
|
||||
print(" 查询限售解禁数据...")
|
||||
lifting_ban = self._get_lifting_ban_data(symbol)
|
||||
risk_data['lifting_ban'] = lifting_ban
|
||||
if lifting_ban and lifting_ban.get('has_data'):
|
||||
print(f" 获取到限售解禁数据")
|
||||
else:
|
||||
print(f" 暂无限售解禁数据")
|
||||
|
||||
time.sleep(1) # 避免请求过快
|
||||
|
||||
# 2. 获取大股东减持公告
|
||||
print(" 查询大股东减持公告...")
|
||||
reduction = self._get_shareholder_reduction_data(symbol)
|
||||
risk_data['shareholder_reduction'] = reduction
|
||||
if reduction and reduction.get('has_data'):
|
||||
print(f" 获取到大股东减持数据")
|
||||
else:
|
||||
print(f" 暂无大股东减持数据")
|
||||
|
||||
time.sleep(1) # 避免请求过快
|
||||
|
||||
# 3. 获取近期重要事件
|
||||
print(" 查询近期重要事件...")
|
||||
events = self._get_important_events_data(symbol)
|
||||
risk_data['important_events'] = events
|
||||
if events and events.get('has_data'):
|
||||
print(f" 获取到重要事件数据")
|
||||
else:
|
||||
print(f" 暂无重要事件数据")
|
||||
|
||||
# 如果至少有一个数据源成功，则认为获取成功
|
||||
if (lifting_ban and lifting_ban.get('has_data')) or \
|
||||
(reduction and reduction.get('has_data')) or \
|
||||
(events and events.get('has_data')):
|
||||
risk_data['data_success'] = True
|
||||
print(f"风险数据获取完成")
|
||||
else:
|
||||
print(f"未获取到风险相关数据")
|
||||
|
||||
except Exception as e:
|
||||
print(f"风险数据获取失败: {str(e)}")
|
||||
risk_data['error'] = str(e)
|
||||
|
||||
return risk_data
|
||||
|
||||
def _get_lifting_ban_data(self, symbol: str) -> Dict[str, Any]:
|
||||
"""获取限售解禁数据"""
|
||||
result = {
|
||||
'has_data': False,
|
||||
'query': f"{symbol}限售解禁",
|
||||
'data': None,
|
||||
'summary': None
|
||||
}
|
||||
|
||||
try:
|
||||
# 构建问句
|
||||
query = f"{symbol}限售解禁"
|
||||
|
||||
# 使用pywencai查询
|
||||
response = pywencai.get(query=query, loop=True)
|
||||
|
||||
if response is None:
|
||||
return result
|
||||
|
||||
# 处理返回结果
|
||||
df_result = self._convert_to_dataframe(response)
|
||||
|
||||
if df_result is None or df_result.empty:
|
||||
return result
|
||||
|
||||
# 提取有用的信息
|
||||
result['has_data'] = True
|
||||
result['data'] = df_result
|
||||
|
||||
# 生成摘要
|
||||
summary = []
|
||||
|
||||
# 尝试提取关键字段
|
||||
if '解禁时间' in df_result.columns or '限售解禁日' in df_result.columns:
|
||||
time_col = '解禁时间' if '解禁时间' in df_result.columns else '限售解禁日'
|
||||
summary.append(f"发现 {len(df_result)} 条解禁记录")
|
||||
|
||||
# 提取最近的解禁记录
|
||||
recent_records = df_result.head(5)
|
||||
for idx, row in recent_records.iterrows():
|
||||
record_info = []
|
||||
if time_col in row.index:
|
||||
record_info.append(f"日期: {row[time_col]}")
|
||||
if '解禁股数' in row.index:
|
||||
record_info.append(f"解禁股数: {row['解禁股数']}")
|
||||
if '解禁市值' in row.index:
|
||||
record_info.append(f"解禁市值: {row['解禁市值']}")
|
||||
if '股东名称' in row.index:
|
||||
record_info.append(f"股东: {row['股东名称']}")
|
||||
|
||||
if record_info:
|
||||
summary.append(" | ".join(record_info))
|
||||
else:
|
||||
# 如果没有标准字段，只记录有数据
|
||||
summary.append(f"获取到 {len(df_result)} 条相关记录")
|
||||
|
||||
result['summary'] = "\n".join(summary) if summary else "有限售解禁数据"
|
||||
|
||||
except Exception as e:
|
||||
result['error'] = str(e)
|
||||
|
||||
return result
|
||||
|
||||
def _get_shareholder_reduction_data(self, symbol: str) -> Dict[str, Any]:
|
||||
"""获取大股东减持公告数据"""
|
||||
result = {
|
||||
'has_data': False,
|
||||
'query': f"{symbol}大股东减持公告",
|
||||
'data': None,
|
||||
'summary': None
|
||||
}
|
||||
|
||||
try:
|
||||
# 构建问句
|
||||
query = f"{symbol}大股东减持公告"
|
||||
|
||||
# 使用pywencai查询
|
||||
response = pywencai.get(query=query, loop=True)
|
||||
|
||||
if response is None:
|
||||
return result
|
||||
|
||||
# 处理返回结果
|
||||
df_result = self._convert_to_dataframe(response)
|
||||
|
||||
if df_result is None or df_result.empty:
|
||||
return result
|
||||
|
||||
# 提取有用的信息
|
||||
result['has_data'] = True
|
||||
result['data'] = df_result
|
||||
|
||||
# 生成摘要
|
||||
summary = []
|
||||
|
||||
# 尝试提取关键字段
|
||||
if '公告日期' in df_result.columns or '减持日期' in df_result.columns:
|
||||
date_col = '公告日期' if '公告日期' in df_result.columns else '减持日期'
|
||||
summary.append(f"发现 {len(df_result)} 条减持公告")
|
||||
|
||||
# 提取最近的减持记录
|
||||
recent_records = df_result.head(5)
|
||||
for idx, row in recent_records.iterrows():
|
||||
record_info = []
|
||||
if date_col in row.index:
|
||||
record_info.append(f"日期: {row[date_col]}")
|
||||
if '股东名称' in row.index:
|
||||
record_info.append(f"股东: {row['股东名称']}")
|
||||
if '减持股数' in row.index:
|
||||
record_info.append(f"减持股数: {row['减持股数']}")
|
||||
if '减持比例' in row.index:
|
||||
record_info.append(f"减持比例: {row['减持比例']}")
|
||||
|
||||
if record_info:
|
||||
summary.append(" | ".join(record_info))
|
||||
else:
|
||||
# 如果没有标准字段，只记录有数据
|
||||
summary.append(f"获取到 {len(df_result)} 条相关记录")
|
||||
|
||||
result['summary'] = "\n".join(summary) if summary else "有大股东减持数据"
|
||||
|
||||
except Exception as e:
|
||||
result['error'] = str(e)
|
||||
|
||||
return result
|
||||
|
||||
def _get_important_events_data(self, symbol: str) -> Dict[str, Any]:
|
||||
"""获取近期重要事件数据"""
|
||||
result = {
|
||||
'has_data': False,
|
||||
'query': f"{symbol}近期重要事件",
|
||||
'data': None,
|
||||
'summary': None
|
||||
}
|
||||
|
||||
try:
|
||||
# 构建问句
|
||||
query = f"{symbol}近期重要事件"
|
||||
|
||||
# 使用pywencai查询
|
||||
response = pywencai.get(query=query, loop=True)
|
||||
|
||||
if response is None:
|
||||
return result
|
||||
|
||||
# 处理返回结果
|
||||
df_result = self._convert_to_dataframe(response)
|
||||
|
||||
if df_result is None or df_result.empty:
|
||||
return result
|
||||
|
||||
# 提取有用的信息
|
||||
result['has_data'] = True
|
||||
result['data'] = df_result
|
||||
|
||||
# 生成摘要
|
||||
summary = []
|
||||
|
||||
# 尝试提取关键字段
|
||||
if '事件时间' in df_result.columns or '公告日期' in df_result.columns:
|
||||
time_col = '事件时间' if '事件时间' in df_result.columns else '公告日期'
|
||||
summary.append(f"发现 {len(df_result)} 条重要事件")
|
||||
|
||||
# 提取最近的事件
|
||||
recent_events = df_result.head(10)
|
||||
for idx, row in recent_events.iterrows():
|
||||
event_info = []
|
||||
if time_col in row.index:
|
||||
event_info.append(f"时间: {row[time_col]}")
|
||||
if '事件类型' in row.index:
|
||||
event_info.append(f"类型: {row['事件类型']}")
|
||||
if '事件内容' in row.index:
|
||||
content = str(row['事件内容'])[:100] # 限制长度
|
||||
event_info.append(f"内容: {content}")
|
||||
elif '标题' in row.index:
|
||||
title = str(row['标题'])[:100]
|
||||
event_info.append(f"标题: {title}")
|
||||
|
||||
if event_info:
|
||||
summary.append(" | ".join(event_info))
|
||||
else:
|
||||
# 如果没有标准字段，只记录有数据
|
||||
summary.append(f"获取到 {len(df_result)} 条相关记录")
|
||||
|
||||
result['summary'] = "\n".join(summary) if summary else "有重要事件数据"
|
||||
|
||||
except Exception as e:
|
||||
result['error'] = str(e)
|
||||
|
||||
return result
|
||||
|
||||
def _convert_to_dataframe(self, result) -> pd.DataFrame:
|
||||
"""将pywencai返回结果转换为DataFrame"""
|
||||
try:
|
||||
if result is None:
|
||||
return None
|
||||
|
||||
df_result = None
|
||||
|
||||
if isinstance(result, dict):
|
||||
try:
|
||||
df_result = pd.DataFrame([result])
|
||||
except Exception:
|
||||
return None
|
||||
elif isinstance(result, pd.DataFrame):
|
||||
df_result = result
|
||||
else:
|
||||
return None
|
||||
|
||||
if df_result is None or df_result.empty:
|
||||
return None
|
||||
|
||||
# 处理嵌套结构（tableV1）
|
||||
if 'tableV1' in df_result.columns and len(df_result.columns) == 1:
|
||||
table_v1_data = df_result.iloc[0]['tableV1']
|
||||
if isinstance(table_v1_data, pd.DataFrame):
|
||||
df_result = table_v1_data
|
||||
elif isinstance(table_v1_data, list) and len(table_v1_data) > 0:
|
||||
df_result = pd.DataFrame(table_v1_data)
|
||||
else:
|
||||
return None
|
||||
|
||||
# 处理嵌套结构（title_content等单列嵌套）
|
||||
# 如果只有一列，且该列的值是DataFrame，则展开
|
||||
if len(df_result.columns) == 1:
|
||||
col_name = df_result.columns[0]
|
||||
first_value = df_result.iloc[0][col_name]
|
||||
if isinstance(first_value, pd.DataFrame):
|
||||
print(f" 检测到嵌套DataFrame（列名: {col_name}），正在展开...")
|
||||
df_result = first_value
|
||||
|
||||
return df_result if not df_result.empty else None
|
||||
|
||||
except Exception as e:
|
||||
print(f" 转换DataFrame时出错: {str(e)}")
|
||||
return None
|
||||
|
||||
def format_risk_data_for_ai(self, risk_data: Dict[str, Any]) -> str:
|
||||
"""格式化风险数据供AI分析使用 - 直接转换DataFrame为字符串"""
|
||||
if not risk_data or not risk_data.get('data_success'):
|
||||
return "未获取到风险数据"
|
||||
|
||||
formatted_text = []
|
||||
|
||||
try:
|
||||
# 1. 限售解禁数据
|
||||
lifting_ban = risk_data.get('lifting_ban')
|
||||
if lifting_ban and lifting_ban.get('has_data') and lifting_ban.get('data') is not None:
|
||||
formatted_text.append("=" * 80)
|
||||
formatted_text.append("【限售解禁数据】")
|
||||
formatted_text.append("=" * 80)
|
||||
formatted_text.append(f"查询语句: {lifting_ban.get('query', '')}")
|
||||
formatted_text.append("")
|
||||
|
||||
# 直接将DataFrame转换为字符串（最多50行）
|
||||
df = lifting_ban.get('data')
|
||||
try:
|
||||
df_str = df.head(50).to_string(index=False, max_rows=50, max_cols=20)
|
||||
formatted_text.append(f"共 {len(df)} 条记录，显示前50条：")
|
||||
formatted_text.append(df_str)
|
||||
except Exception as e:
|
||||
formatted_text.append(f"数据转换失败: {str(e)}")
|
||||
formatted_text.append("")
|
||||
|
||||
# 2. 大股东减持数据
|
||||
reduction = risk_data.get('shareholder_reduction')
|
||||
if reduction and reduction.get('has_data') and reduction.get('data') is not None:
|
||||
formatted_text.append("=" * 80)
|
||||
formatted_text.append("【大股东减持数据】")
|
||||
formatted_text.append("=" * 80)
|
||||
formatted_text.append(f"查询语句: {reduction.get('query', '')}")
|
||||
formatted_text.append("")
|
||||
|
||||
# 直接将DataFrame转换为字符串（最多50行）
|
||||
df = reduction.get('data')
|
||||
try:
|
||||
df_str = df.head(50).to_string(index=False, max_rows=50, max_cols=20)
|
||||
formatted_text.append(f"共 {len(df)} 条记录，显示前50条：")
|
||||
formatted_text.append(df_str)
|
||||
except Exception as e:
|
||||
formatted_text.append(f"数据转换失败: {str(e)}")
|
||||
formatted_text.append("")
|
||||
|
||||
# 3. 重要事件数据
|
||||
events = risk_data.get('important_events')
|
||||
if events and events.get('has_data') and events.get('data') is not None:
|
||||
formatted_text.append("=" * 80)
|
||||
formatted_text.append("【重要事件数据】")
|
||||
formatted_text.append("=" * 80)
|
||||
formatted_text.append(f"查询语句: {events.get('query', '')}")
|
||||
formatted_text.append("")
|
||||
|
||||
# 直接将DataFrame转换为字符串（最多50行）
|
||||
df = events.get('data')
|
||||
try:
|
||||
df_str = df.head(50).to_string(index=False, max_rows=50, max_cols=20)
|
||||
formatted_text.append(f"共 {len(df)} 条记录，显示前50条：")
|
||||
formatted_text.append(df_str)
|
||||
except Exception as e:
|
||||
formatted_text.append(f"数据转换失败: {str(e)}")
|
||||
formatted_text.append("")
|
||||
|
||||
return "\n".join(formatted_text) if formatted_text else "暂无风险数据"
|
||||
|
||||
except Exception as e:
|
||||
print(f"格式化风险数据时出错: {str(e)}")
|
||||
import traceback
|
||||
traceback.print_exc()
|
||||
return f"格式化风险数据时出错: {str(e)}"
|
||||
|
||||
def _format_dataframe_for_ai(self, df: pd.DataFrame, data_type: str) -> str:
|
||||
"""将DataFrame格式化为AI易读的文本格式"""
|
||||
lines = []
|
||||
|
||||
# 显示数据总数
|
||||
lines.append(f"共 {len(df)} 条{data_type}记录")
|
||||
lines.append("")
|
||||
|
||||
# 显示列名
|
||||
lines.append(f"数据字段：{', '.join(df.columns.tolist())}")
|
||||
lines.append("")
|
||||
|
||||
# 逐行显示数据（最多显示50条，避免数据过大）
|
||||
max_rows = min(50, len(df))
|
||||
|
||||
for idx, row in df.head(max_rows).iterrows():
|
||||
lines.append(f"【记录 {idx + 1}】")
|
||||
|
||||
# 显示每个字段的值
|
||||
for col in df.columns:
|
||||
value = row[col]
|
||||
|
||||
# 处理不同类型的值
|
||||
if pd.isna(value):
|
||||
value_str = "无数据"
|
||||
elif isinstance(value, (int, float)):
|
||||
value_str = str(value)
|
||||
else:
|
||||
value_str = str(value)
|
||||
# 限制过长的字符串
|
||||
if len(value_str) > 200:
|
||||
value_str = value_str[:200] + "..."
|
||||
|
||||
lines.append(f" {col}: {value_str}")
|
||||
|
||||
lines.append("")
|
||||
|
||||
if len(df) > max_rows:
|
||||
lines.append(f"... 还有 {len(df) - max_rows} 条记录（已省略）")
|
||||
lines.append("")
|
||||
|
||||
return "\n".join(lines)
|
||||
|
||||
|
||||
# 测试代码
|
||||
if __name__ == "__main__":
|
||||
fetcher = RiskDataFetcher()
|
||||
|
||||
# 测试获取风险数据
|
||||
test_symbol = "600000"
|
||||
print(f"测试获取 {test_symbol} 的风险数据...")
|
||||
|
||||
risk_data = fetcher.get_risk_data(test_symbol)
|
||||
|
||||
print("\n" + "=" * 60)
|
||||
print("获取结果:")
|
||||
print("=" * 60)
|
||||
print(f"数据获取成功: {risk_data['data_success']}")
|
||||
|
||||
if risk_data['data_success']:
|
||||
print("\n格式化的风险数据:")
|
||||
print(fetcher.format_risk_data_for_ai(risk_data))
|
||||
|
||||
Reference in New Issue
Block a user