aishare-txt 1.0.0__py3-none-any.whl

AIShareTxt/indicators/technical_indicators.py

@@ -0,0 +1,507 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+技术指标计算模块
+负责计算各种技术指标
+"""
+
+import talib
+import pandas as pd
+import numpy as np
+from ..core.config import IndicatorConfig as Config
+from ..utils.utils import LoggerManager
+import warnings
+warnings.filterwarnings('ignore')
+
+
+class TechnicalIndicators:
+    """技术指标计算器"""
+    
+    def __init__(self):
+        self.config = Config()
+        self.logger = LoggerManager.get_logger('technical_indicators')
+    
+    def calculate_all_indicators(self, data):
+        """
+        计算所有技术指标
+        
+        Args:
+            data (pd.DataFrame): 股票数据
+            
+        Returns:
+            dict: 包含所有指标的字典
+        """
+        if data is None or len(data) == 0:
+            return None
+            
+        # 获取价格数据
+        close = data['close'].values
+        high = data['high'].values
+        low = data['low'].values
+        volume = data['volume'].values
+        open_price = data['open'].values
+        
+        indicators = {}
+        
+        # 计算各类指标
+        indicators.update(self._calculate_moving_averages(close))
+        indicators.update(self._calculate_ma_derived_indicators(close))
+        indicators.update(self._calculate_volume_price_indicators(high, low, close, volume))
+        indicators.update(self._calculate_trend_strength_indicators(high, low, close))
+        indicators.update(self._calculate_momentum_oscillators(high, low, close))
+        indicators.update(self._calculate_volatility_indicators(high, low, close))
+        indicators.update(self._calculate_volume_indicators(volume))
+        
+        # 添加基础数据
+        indicators['current_price'] = close[-1]
+        # 将日期转换为字符串格式以便JSON序列化
+        date_value = data['date'].iloc[-1]
+        if hasattr(date_value, 'strftime'):
+            indicators['date'] = date_value.strftime('%Y-%m-%d')
+        else:
+            indicators['date'] = str(date_value)
+        
+        return indicators
+    
+    def _calculate_moving_averages(self, close):
+        """计算移动平均线指标"""
+        indicators = {}
+        
+        try:
+            # MA（移动平均线）
+            all_periods = (self.config.MA_PERIODS['short'] + 
+                          self.config.MA_PERIODS['medium'] + 
+                          self.config.MA_PERIODS['long'])
+            
+            for period in all_periods:
+                ma = talib.SMA(close, timeperiod=period)
+                indicators[f'MA_{period}'] = ma[-1]
+                indicators[f'SMA_{period}'] = ma[-1]  # SMA与MA相同，保留用于兼容
+            
+            # EMA（指数移动平均线）
+            for period in self.config.EMA_PERIODS:
+                ema = talib.EMA(close, timeperiod=period)
+                indicators[f'EMA_{period}'] = ema[-1]
+            
+            # WMA（加权移动平均线）
+            for period in self.config.WMA_PERIODS:
+                wma = talib.WMA(close, timeperiod=period)
+                indicators[f'WMA_{period}'] = wma[-1]
+            
+            # 均线形态分析
+            ma_patterns = self.analyze_ma_patterns(close)
+            indicators.update(ma_patterns)
+            
+        except Exception as e:
+            self.logger.warning(f"均线指标计算失败：{str(e)}")
+        
+        return indicators
+    
+    def _calculate_ma_derived_indicators(self, close):
+        """计算均线衍生指标"""
+        indicators = {}
+        
+        try:
+            # BIAS 乖离率
+            for period in self.config.BIAS_PERIODS:
+                bias = self.calculate_bias(close, timeperiod=period)
+                if bias is not None:
+                    indicators[f'BIAS_{period}'] = bias[-1]
+            
+            # MACD
+            macd_config = self.config.MACD_CONFIG
+            macd, macdsignal, macdhist = talib.MACD(
+                close, 
+                fastperiod=macd_config['fastperiod'],
+                slowperiod=macd_config['slowperiod'], 
+                signalperiod=macd_config['signalperiod']
+            )
+            indicators['MACD_DIF'] = macd[-1]
+            indicators['MACD_DEA'] = macdsignal[-1]
+            indicators['MACD_HIST'] = macdhist[-1]
+            
+            # 布林带
+            bb_config = self.config.BOLLINGER_BANDS_CONFIG
+            upperband, middleband, lowerband = talib.BBANDS(
+                close, 
+                timeperiod=bb_config['timeperiod'],
+                nbdevup=bb_config['nbdevup'], 
+                nbdevdn=bb_config['nbdevdn'], 
+                matype=int(bb_config['matype'])  # type: ignore
+            )
+            indicators['BB_UPPER'] = upperband[-1]
+            indicators['BB_MIDDLE'] = middleband[-1]
+            indicators['BB_LOWER'] = lowerband[-1]
+            indicators['BB_WIDTH'] = (upperband[-1] - lowerband[-1]) / middleband[-1] * 100
+            
+        except Exception as e:
+            self.logger.warning(f"均线衍生指标计算失败：{str(e)}")
+        
+        return indicators
+    
+    def _calculate_volume_price_indicators(self, high, low, close, volume):
+        """计算量价指标"""
+        indicators = {}
+        
+        try:
+            # VWAP (Volume Weighted Average Price)
+            typical_price = (high + low + close) / 3
+            vwap_period = self.config.VWAP_PERIOD
+            if len(typical_price) >= vwap_period:
+                price_volume = typical_price[-vwap_period:] * volume[-vwap_period:]
+                indicators['VWAP_14'] = np.sum(price_volume) / np.sum(volume[-vwap_period:])
+            
+            # OBV (On Balance Volume)
+            close_float = close.astype(np.float64)
+            volume_float = volume.astype(np.float64)
+            obv = talib.OBV(close_float, volume_float)
+            
+            indicators.update(self._analyze_obv(obv, close))
+            
+        except Exception as e:
+            self.logger.warning(f"量价指标计算失败：{str(e)}")
+        
+        return indicators
+    
+    def _calculate_trend_strength_indicators(self, high, low, close):
+        """计算趋势强度指标"""
+        indicators = {}
+        
+        try:
+            # ADX (Average Directional Index) 和 DMI 指标
+            adx_period = self.config.ADX_PERIOD
+            adx = talib.ADX(high, low, close, timeperiod=adx_period)
+            plus_di = talib.PLUS_DI(high, low, close, timeperiod=adx_period)
+            minus_di = talib.MINUS_DI(high, low, close, timeperiod=adx_period)
+            
+            indicators['ADX'] = adx[-1]
+            indicators['DI_PLUS'] = plus_di[-1]
+            indicators['DI_MINUS'] = minus_di[-1]
+            indicators['DI_DIFF'] = plus_di[-1] - minus_di[-1]
+            
+            # ADX趋势强度跟踪
+            if len(adx) >= 5:
+                adx_recent = adx[-5:]
+                adx_slope = np.polyfit(range(len(adx_recent)), adx_recent, 1)[0]
+                indicators['ADX_TREND'] = adx_slope
+                indicators['ADX_5D_CHANGE'] = ((adx[-1] - adx[-5]) / adx[-5] * 100) if adx[-5] != 0 else 0
+            
+        except Exception as e:
+            self.logger.warning(f"趋势强度指标计算失败：{str(e)}")
+        
+        return indicators
+    
+    def _calculate_momentum_oscillators(self, high, low, close):
+        """计算动量振荡指标"""
+        indicators = {}
+        
+        try:
+            # RSI (Relative Strength Index)
+            for period in self.config.RSI_PERIODS:
+                rsi = talib.RSI(close, timeperiod=period)
+                indicators[f'RSI_{period}'] = rsi[-1]
+            
+            # Stochastic Oscillator (KD指标)
+            stoch_config = self.config.STOCH_CONFIG
+            slowk, slowd = talib.STOCH(
+                high, low, close,
+                fastk_period=stoch_config['fastk_period'],
+                slowk_period=stoch_config['slowk_period'],
+                slowk_matype=int(stoch_config['slowk_matype']),  # type: ignore
+                slowd_period=stoch_config['slowd_period'],
+                slowd_matype=int(stoch_config['slowd_matype'])  # type: ignore
+            )
+            indicators['STOCH_K'] = slowk[-1]
+            indicators['STOCH_D'] = slowd[-1]
+            indicators['STOCH_KD_DIFF'] = slowk[-1] - slowd[-1]
+            
+        except Exception as e:
+            self.logger.warning(f"动量振荡指标计算失败：{str(e)}")
+        
+        return indicators
+    
+    def _calculate_volatility_indicators(self, high, low, close):
+        """计算波动率指标"""
+        indicators = {}
+        
+        try:
+            # ATR (Average True Range)
+            atr_period = self.config.ATR_PERIOD
+            atr = talib.ATR(high, low, close, timeperiod=atr_period)
+            indicators['ATR_14'] = atr[-1]
+            indicators['ATR_RATIO'] = (atr[-1] / close[-1] * 100) if close[-1] != 0 else 0
+            
+        except Exception as e:
+            self.logger.warning(f"波动率指标计算失败：{str(e)}")
+        
+        return indicators
+    
+    def _calculate_volume_indicators(self, volume):
+        """计算成交量指标"""
+        indicators = {}
+        
+        try:
+            # 量比指标
+            if len(volume) >= 6:
+                current_volume = volume[-1]
+                avg_volume_5d = np.mean(volume[-6:-1])  # 过去5日平均
+                indicators['VOLUME_RATIO'] = current_volume / avg_volume_5d if avg_volume_5d != 0 else 0
+            
+            # 成交量趋势
+            if len(volume) >= 5:
+                volume_recent = volume[-5:]
+                volume_slope = np.polyfit(range(len(volume_recent)), volume_recent, 1)[0]
+                indicators['VOLUME_TREND'] = volume_slope
+            
+        except Exception as e:
+            self.logger.warning(f"成交量指标计算失败：{str(e)}")
+        
+        return indicators
+    
+    def calculate_bias(self, close, timeperiod=20):
+        """
+        计算乖离率 (BIAS)
+        BIAS = (收盘价 - N日SMA) / N日SMA * 100
+        """
+        sma = talib.SMA(close, timeperiod=timeperiod)
+        bias = ((close - sma) / sma) * 100
+        return bias
+    
+    def analyze_ma_patterns(self, close):
+        """分析均线形态"""
+        patterns = {
+            'trend_pattern': '',
+            'cross_pattern': '',
+            'position_pattern': '',
+            'arrangement_pattern': '',
+            'support_resistance': ''
+        }
+        
+        # 计算均线
+        ma5 = talib.SMA(close, timeperiod=5)
+        ma10 = talib.SMA(close, timeperiod=10)
+        ma20 = talib.SMA(close, timeperiod=20)
+        ma60 = talib.SMA(close, timeperiod=60)
+        
+        if len(ma5) < 2 or len(ma10) < 2 or len(ma20) < 2:
+            return patterns
+            
+        current_price = close[-1]
+        current_ma5 = ma5[-1]
+        current_ma10 = ma10[-1]
+        current_ma20 = ma20[-1]
+        current_ma60 = ma60[-1] if len(ma60) > 0 else 0
+        
+        prev_ma5 = ma5[-2]
+        prev_ma10 = ma10[-2]
+        prev_ma20 = ma20[-2]
+        
+        # 使用配置中的阈值
+        thresholds = self.config.MA_THRESHOLDS
+        
+        # 1. 趋势形态分析
+        patterns['trend_pattern'] = self._analyze_trend_pattern(
+            current_ma5, current_ma10, current_ma20, thresholds
+        )
+        
+        # 2. 交叉形态分析
+        patterns['cross_pattern'] = self._analyze_cross_pattern(
+            current_ma5, current_ma10, current_ma20,
+            prev_ma5, prev_ma10, prev_ma20
+        )
+        
+        # 3. 位置形态分析
+        patterns['position_pattern'] = self._analyze_position_pattern(
+            current_price, current_ma5, current_ma10, current_ma20, current_ma60, thresholds
+        )
+        
+        # 4. 排列形态分析
+        patterns['arrangement_pattern'] = self._analyze_arrangement_pattern(
+            current_ma5, current_ma10, current_ma20, current_ma60
+        )
+        
+        # 5. 支撑阻力分析
+        patterns['support_resistance'] = self._analyze_support_resistance(
+            current_price, current_ma5, current_ma10, current_ma20, thresholds
+        )
+        
+        return patterns
+    
+    def _analyze_trend_pattern(self, ma5, ma10, ma20, thresholds):
+        """分析趋势形态"""
+        if ma5 > ma10 > ma20:
+            # 检查发散程度
+            ma_diff_5_10 = (ma5 - ma10) / ma10 * 100
+            ma_diff_10_20 = (ma10 - ma20) / ma20 * 100
+            
+            if ma_diff_5_10 > thresholds['divergence_strong'] * 100 and ma_diff_10_20 > thresholds['divergence_strong'] * 100:
+                return "多头发散"
+            elif ma_diff_5_10 < thresholds['divergence_weak'] * 100 and ma_diff_10_20 < thresholds['divergence_weak'] * 100:
+                return "多头收敛"
+            else:
+                return "多头排列"
+                
+        elif ma5 < ma10 < ma20:
+            # 检查发散程度
+            ma_diff_5_10 = abs(ma5 - ma10) / ma10 * 100
+            ma_diff_10_20 = abs(ma10 - ma20) / ma20 * 100
+            
+            if ma_diff_5_10 > thresholds['divergence_strong'] * 100 and ma_diff_10_20 > thresholds['divergence_strong'] * 100:
+                return "空头发散"
+            elif ma_diff_5_10 < thresholds['divergence_weak'] * 100 and ma_diff_10_20 < thresholds['divergence_weak'] * 100:
+                return "空头收敛"
+            else:
+                return "空头排列"
+        else:
+            # 检查是否为盘整形态
+            ma_diff_5_10 = abs(ma5 - ma10) / ma10 * 100
+            ma_diff_10_20 = abs(ma10 - ma20) / ma20 * 100
+            ma_diff_5_20 = abs(ma5 - ma20) / ma20 * 100
+            
+            if (ma_diff_5_10 < thresholds['adhesion_loose'] * 100 and 
+                ma_diff_10_20 < thresholds['adhesion_loose'] * 100 and 
+                ma_diff_5_20 < thresholds['adhesion_range'] * 100):
+                return "三均线粘合"
+            elif ma_diff_5_10 < thresholds['adhesion_tight'] * 100:
+                return "MA5与MA10粘合"
+            elif ma_diff_10_20 < thresholds['adhesion_tight'] * 100:
+                return "MA10与MA20粘合"
+            else:
+                return "均线缠绕"
+    
+    def _analyze_cross_pattern(self, ma5, ma10, ma20, prev_ma5, prev_ma10, prev_ma20):
+        """分析交叉形态"""
+        cross_signals = []
+        
+        # 均线交叉检测
+        if ma5 > ma10 and prev_ma5 <= prev_ma10:
+            cross_signals.append("MA5上穿MA10")
+        if ma10 > ma20 and prev_ma10 <= prev_ma20:
+            cross_signals.append("MA10上穿MA20")
+            
+        if ma5 < ma10 and prev_ma5 >= prev_ma10:
+            cross_signals.append("MA5下穿MA10")
+        if ma10 < ma20 and prev_ma10 >= prev_ma20:
+            cross_signals.append("MA10下穿MA20")
+            
+        return "，".join(cross_signals) if cross_signals else "无交叉信号"
+    
+    def _analyze_position_pattern(self, price, ma5, ma10, ma20, ma60, thresholds):
+        """分析位置形态"""
+        position_signals = []
+        support_threshold = thresholds['support_resistance']
+        
+        # 价格与均线关系
+        for ma_value, ma_name in [(ma5, "MA5"), (ma10, "MA10"), (ma20, "MA20")]:
+            if price > ma_value:
+                position_signals.append(f"站上{ma_name}")
+            elif abs(price - ma_value) / ma_value < support_threshold:
+                position_signals.append(f"接近{ma_name}")
+        
+        if ma60 > 0:
+            if price > ma60:
+                position_signals.append("站上MA60")
+            elif abs(price - ma60) / ma60 < support_threshold:
+                position_signals.append("接近MA60")
+        
+        return "，".join(position_signals) if position_signals else "均线下方运行"
+    
+    def _analyze_arrangement_pattern(self, ma5, ma10, ma20, ma60):
+        """分析排列形态"""
+        if ma5 > ma10 > ma20:
+            if ma60 > 0 and ma20 > ma60:
+                return "MA5>MA10>MA20>MA60"
+            else:
+                return "MA5>MA10>MA20"
+        elif ma5 < ma10 < ma20:
+            if ma60 > 0 and ma20 < ma60:
+                return "MA5<MA10<MA20<MA60"
+            else:
+                return "MA5<MA10<MA20"
+        else:
+            return "均线无规律排列"
+    
+    def _analyze_support_resistance(self, price, ma5, ma10, ma20, thresholds):
+        """分析支撑阻力"""
+        support_resistance = []
+        support_threshold = thresholds['support_resistance']
+        
+        for ma_value, ma_name in [(ma5, "MA5"), (ma10, "MA10"), (ma20, "MA20")]:
+            if abs(price - ma_value) / ma_value < support_threshold:
+                support_resistance.append(f"接近{ma_name}")
+        
+        return "，".join(support_resistance) if support_resistance else "无接近均线"
+    
+    def _analyze_obv(self, obv, close):
+        """分析OBV指标"""
+        indicators = {}
+        obv_config = self.config.OBV_CONFIG
+        
+        # 计算OBV基础指标
+        indicators['OBV_current'] = obv[-1]
+        indicators['OBV_5d_ago'] = obv[-6] if len(obv) >= 6 else obv[0]
+        indicators['OBV_20d_ago'] = obv[-21] if len(obv) >= 21 else obv[0]
+        
+        # 计算OBV变化率
+        obv_5d_change = ((obv[-1] - obv[-6]) / abs(obv[-6]) * 100) if len(obv) >= 6 and obv[-6] != 0 else 0
+        obv_20d_change = ((obv[-1] - obv[-21]) / abs(obv[-21]) * 100) if len(obv) >= 21 and obv[-21] != 0 else 0
+        
+        indicators['OBV_5d_change'] = obv_5d_change
+        indicators['OBV_20d_change'] = obv_20d_change
+        
+        # 计算OBV趋势（最近5日斜率）
+        if len(obv) >= 5:
+            recent_obv = obv[-5:]
+            x = np.arange(len(recent_obv))
+            slope = np.polyfit(x, recent_obv, 1)[0]
+            indicators['OBV_trend'] = slope
+        
+        # OBV背离验证
+        if len(obv) >= 20 and len(close) >= 20:
+            try:
+                from scipy.signal import argrelextrema
+                indicators['OBV_DIVERGENCE'] = self._detect_obv_divergence(obv, close, obv_config)
+            except ImportError:
+                indicators['OBV_DIVERGENCE'] = "需要scipy库进行背离分析"
+        
+        return indicators
+    
+    def _detect_obv_divergence(self, obv, close, config):
+        """检测OBV背离"""
+        try:
+            from scipy.signal import argrelextrema
+            
+            recent_close = close[-20:]
+            recent_obv = obv[-20:]
+            
+            # 找局部高点和低点
+            order = config['extrema_order']
+            price_highs = argrelextrema(recent_close, np.greater, order=order)[0]
+            price_lows = argrelextrema(recent_close, np.less, order=order)[0]
+            obv_highs = argrelextrema(recent_obv, np.greater, order=order)[0]
+            obv_lows = argrelextrema(recent_obv, np.less, order=order)[0]
+            
+            # 顶背离：价格创新高，OBV未创新高
+            if len(price_highs) >= 2 and len(obv_highs) >= 2:
+                latest_price_high_idx = price_highs[-1]
+                prev_price_high_idx = price_highs[-2]
+                
+                if latest_price_high_idx < len(obv_highs) and prev_price_high_idx < len(obv_highs):
+                    if (recent_close[latest_price_high_idx] > recent_close[prev_price_high_idx] and
+                        recent_obv[latest_price_high_idx] < recent_obv[prev_price_high_idx]):
+                        return "顶背离"
+            
+            # 底背离：价格创新低，OBV未创新低
+            if len(price_lows) >= 2 and len(obv_lows) >= 2:
+                latest_price_low_idx = price_lows[-1]
+                prev_price_low_idx = price_lows[-2]
+                
+                if latest_price_low_idx < len(obv_lows) and prev_price_low_idx < len(obv_lows):
+                    if (recent_close[latest_price_low_idx] < recent_close[prev_price_low_idx] and
+                        recent_obv[latest_price_low_idx] > recent_obv[prev_price_low_idx]):
+                        return "底背离"
+            
+            return "无明显背离"
+            
+        except Exception:
+            return "背离检测失败"

AIShareTxt/tests/__init__.py

@@ -0,0 +1,11 @@
+"""
+AIShareTxt测试模块
+
+包含单元测试和集成测试。
+"""
+
+# 这里可以添加测试相关的导入
+# from .test_analyzer import TestStockAnalyzer
+# from .test_indicators import TestTechnicalIndicators
+
+__all__ = []

AIShareTxt/utils/__init__.py

@@ -0,0 +1,17 @@
+"""
+AIShareTxt工具模块
+
+包含日志管理、股票列表获取等辅助功能。
+"""
+
+from .utils import Logger
+from .stock_list import get_stock_list
+
+__all__ = [
+    "Logger",
+    "get_stock_list",
+]
+
+# 这里可以添加更多工具函数
+# from .data_validator import DataValidator
+# __all__.append("DataValidator")