# -*- coding: utf-8 -*- """ log_parser.py — 公共日志解析模块 所有可视化工具共用同一套正则和解析函数，避免重复维护。 """ import re, math, os _ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__))) DEFAULT_LOG = os.path.join(_ROOT, "my_trader.log") FK = ['TR', 'OB', 'TK', 'OI', 'FR', 'MP', 'VD', 'BTC', 'GM', 'IV', 'EX', 'LC', 'MR', 'SM', 'CT'] FN = { 'TR': '趋势', 'OB': '挂单', 'TK': '成交', 'OI': '持仓量', 'FR': '费率', 'MP': '痛点', 'VD': '量变', 'BTC': 'BTC', 'GM': 'Gamma', 'IV': '波动率', 'EX': '衰竭', 'LC': '清算', 'MR': '回归', 'SM': '聪明钱', 'CT': '跟单', } FC = { 'TR': '#ff6b6b', 'OB': '#ffa502', 'TK': '#2ed573', 'OI': '#1e90ff', 'FR': '#a55eea', 'MP': '#ff4757', 'VD': '#7bed9f', 'BTC': '#eccc68', 'GM': '#e056fd', 'IV': '#686de0', 'EX': '#ff6348', 'LC': '#3ae374', 'MR': '#18dcff', 'SM': '#7d5fff', 'CT': '#ffd700', } # ── 正则 ────────────────────────────────────────────────────── RE_PRICE = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}),\d+ \[MyTrader\] INFO: \[ETH\] \$([0-9,.]+) \|' ) RE_FACTOR = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}),\d+ \[MyTrader\] INFO: \[ETH\] ' r'TR:([+-]?\d+\.?\d*) OB:([+-]?\d+\.?\d*) FR:([+-]?\d+\.?\d*) ' r'TK:([+-]?\d+\.?\d*) OI:([+-]?\d+\.?\d*) MP:([+-]?\d+\.?\d*) ' r'VD:([+-]?\d+\.?\d*) BTC:([+-]?\d+\.?\d*) GM:([+-]?\d+\.?\d*) ' r'IV:([+-]?\d+\.?\d*) EX:([+-]?\d+\.?\d*) LC:([+-]?\d+\.?\d*) ' r'MR:([+-]?\d+\.?\d*) SM:([+-]?\d+\.?\d*) CT:([+-]?\d+\.?\d*) ' r'mom:[+-]?\d+\.?\d* flip:[+-]?\d+\.?\d* => ([+-]?\d+\.?\d*) -> (\w+)' ) RE_NEWS_DS = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}),\d+ \[MyTrader\] INFO: \[News\] DeepSeek: (.+?) ~' ) RE_NEWS_FULL = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}),\d+ \[MyTrader\] INFO: \[News\] DeepSeek: ' r'AI判断:([+-]?\d+\.?\d*)\(把握(\d+)%[^)]*\) 恐贪:(\d+) (.+?) \[(.+?)\]' ) RE_NEWS_SCORE = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}),\d+ \[MyTrader\] INFO: \[News\] (?!DeepSeek).+=>\s*([+-]?\d+)' ) RE_FLASH = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}),\d+ \[MyTrader\] INFO: \[FLASH\] ★(\d+) (.+)' ) RE_TRADE_OPEN = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}).*ETH (LONG|SHORT) [\d.]+\S* @ \$([\d.]+) TP:\$([\d.]+) SL:\$([\d.]+)' ) RE_TRADE_TP = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}).*\[止盈\].*\$([-\d.]+)' ) RE_TRADE_SL = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}).*\[止损\].*\$([-\d.]+)' ) RE_TRADE_CLOSE = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}).*(锁利|因子反向.*锁|强制平仓|决策树.*平)' ) def _parse_factor_row(m): return { 'TR': float(m.group(2)), 'OB': float(m.group(3)), 'FR': float(m.group(4)), 'TK': float(m.group(5)), 'OI': float(m.group(6)), 'MP': float(m.group(7)), 'VD': float(m.group(8)), 'BTC': float(m.group(9)), 'GM': float(m.group(10)), 'IV': float(m.group(11)), 'EX': float(m.group(12)), 'LC': float(m.group(13)), 'MR': float(m.group(14)), 'SM': float(m.group(15)), 'CT': float(m.group(16)), 'total': float(m.group(17)), 'dir': m.group(18), } def parse_rows(log_path=None, max_rows=6000): """ 解析日志，返回因子行列表（含价格填充）和事件列表。 每行格式: {'ts', 'price', 'TR'..'SM', 'total', 'dir', 'news': {...}|None} 事件格式: {'ts', 'type': open/tp/sl/close, 'price', 'label', 'color', 'dir', 'tp', 'sl'} """ path = log_path or DEFAULT_LOG try: lines = open(path, encoding='utf-8', errors='replace').readlines() except Exception: return [], [] prices_map = {} rows = [] events = [] last_news = None for line in lines: m = RE_PRICE.search(line) if m: prices_map[m.group(1)] = float(m.group(2).replace(',', '')) m = RE_NEWS_FULL.search(line) if m: last_news = { 'ai': float(m.group(2)), 'conf': int(m.group(3)), 'fg': int(m.group(4)), 'summary': m.group(5).strip()[:80], 'headline':m.group(6).strip()[:100], } m = RE_FACTOR.search(line) if m: ts = m.group(1) row = {'ts': ts, 'price': prices_map.get(ts), 'news': last_news} row.update(_parse_factor_row(m)) rows.append(row) m = RE_TRADE_OPEN.search(line) if m: d = m.group(2) events.append({ 'ts': m.group(1), 'type': 'open', 'price': float(m.group(3)), 'label': f"{'▲' if d == 'LONG' else '▼'}{d[0]} @{float(m.group(3)):.0f}", 'color': '#2ed573' if d == 'LONG' else '#ff4757', 'dir': d, 'tp': float(m.group(4)), 'sl': float(m.group(5)), }) m = RE_TRADE_TP.search(line) if m: events.append({'ts': m.group(1), 'type': 'tp', 'price': 0, 'label': f"止盈 ${float(m.group(2)):+.4f}", 'color': '#f0c040', 'dir': ''}) m = RE_TRADE_SL.search(line) if m: events.append({'ts': m.group(1), 'type': 'sl', 'price': 0, 'label': f"止损 ${float(m.group(2)):+.4f}", 'color': '#ff6348', 'dir': ''}) m = RE_TRADE_CLOSE.search(line) if m: events.append({'ts': m.group(1), 'type': 'close', 'price': 0, 'label': f"平仓:{m.group(2)[:6]}", 'color': '#aaa', 'dir': ''}) # 填充价格（前向填充） last = None for r in rows: if r['price'] is None: r['price'] = last else: last = r['price'] rows = [r for r in rows if r['price'] is not None][-max_rows:] # 填充事件价格（取最近一条 row 的价格） if rows: ts_map = {r['ts']: r['price'] for r in rows} all_ts = sorted(ts_map.keys()) t0, t1 = rows[0]['ts'], rows[-1]['ts'] for ev in events: if not ev['price']: best = next((t for t in reversed(all_ts) if t <= ev['ts']), None) ev['price'] = ts_map.get(best or all_ts[0], 0) events = [e for e in events if t0 <= e['ts'] <= t1] return rows, events def parse_rows_simple(log_path=None, max_rows=2000): """轻量版：只返回 rows，不解析事件（供 factor_correlation 使用）""" rows, _ = parse_rows(log_path, max_rows) return rows def parse_news_events(log_path=None): """解析新闻事件（供 factor_chart 使用）""" path = log_path or DEFAULT_LOG try: lines = open(path, encoding='utf-8', errors='replace').readlines() except Exception: return [], [] news_events, flash_events = [], [] ds_cache = {} for line in lines: m = RE_NEWS_DS.search(line) if m: ds_cache[m.group(1)] = m.group(2)[:120] m = RE_NEWS_SCORE.search(line) if m: ts = m.group(1) news_events.append({'ts': ts, 'score': int(m.group(2)), 'summary': ds_cache.get(ts, '')}) m = RE_FLASH.search(line) if m: flash_events.append({'ts': m.group(1), 'score': int(m.group(2)), 'text': m.group(3)[:100]}) return news_events, flash_events def pearson(x, y): n = len(x) if n < 3: return 0.0 mx, my = sum(x) / n, sum(y) / n num = sum((a - mx) * (b - my) for a, b in zip(x, y)) dx = math.sqrt(sum((a - mx) ** 2 for a in x)) dy = math.sqrt(sum((b - my) ** 2 for b in y)) return round(num / (dx * dy), 4) if dx > 1e-10 and dy > 1e-10 else 0.0 def normalize_prices(rows): """返回价格归一化序列 [-1, 1]""" prices = [r['price'] for r in rows] px_min, px_max = min(prices), max(prices) px_range = px_max - px_min if px_max != px_min else 1 px_mid = (px_max + px_min) / 2 return [(p - px_mid) / (px_range / 2) for p in prices] # ═══════════════════════════════════════════════════════════════ # BTC 日志解析器 — 策略A历史数据提取 # 格式: [BTC] $79,246.00 | ... (价格行) # [BTC] TR:-1 OB:-0.10 ... (因子行, 紧接价格行) # [News] DeepSeek: ... (新闻行) # ═══════════════════════════════════════════════════════════════ RE_BTC_PRICE = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}),\d+ \[MyTrader\].*\[BTC\] \$([0-9,.]+)' ) RE_BTC_FACTOR = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}),\d+ \[MyTrader\].*\[BTC\] ' r'TR:([+-]?\d+\.?\d*)\s+OB:([+-]?\d+\.?\d*)\s+FR:([+-]?\d+\.?\d*)\s+' r'TK:([+-]?\d+\.?\d*)\s+OI:([+-]?\d+\.?\d*)\s+MP:([+-]?\d+\.?\d*)\s+' r'VD:([+-]?\d+\.?\d*)\s+BTC:([+-]?\d+\.?\d*)\s+GM:([+-]?\d+\.?\d*)\s+' r'IV:([+-]?\d+\.?\d*)\s+EX:([+-]?\d+\.?\d*)\s+LC:([+-]?\d+\.?\d*)\s+' r'MR:([+-]?\d+\.?\d*)\s+SM:([+-]?\d+\.?\d*)\s+CT:([+-]?\d+\.?\d*)\s+' r'mom:[+-]?\d+\.?\d*\s+flip:[+-]?\d+\.?\d*\s*=>\s*([+-]?\d+\.?\d*)\s*->\s*(\w+)' ) RE_BTC_NEWS = re.compile( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}),\d+ \[MyTrader\].*\[News\]\s*' r'(?:DeepSeek:\s*)?(.*?)(?:=>\s*([+-]?\d+))?\s*$' ) def parse_rows_btc(log_path=None, max_rows=1000): """ 解析BTC格式日志, 返回 (rows, news_events) rows: [{'ts', 'price', 'TR','OB',...'CT', 'total', 'dir'}, ...] news_events: [{'ts', 'summary', 'score', 'headline'}, ...] """ path = log_path or DEFAULT_LOG try: lines = open(path, encoding='utf-8', errors='replace').readlines() except Exception: return [], [] # 第一遍：收集价格时间戳映射 price_map = {} # ts -> price for line in lines: m = RE_BTC_PRICE.search(line) if m: ts = m.group(1) try: price_map[ts] = float(m.group(2).replace(',', '')) except ValueError: pass # 第二遍：解析因子行 + 新闻行 rows = [] news_events = [] last_price_ts = None for line in lines: # --- 因子行 --- m = RE_BTC_FACTOR.search(line) if m: ts = m.group(1) # 取最接近的价格（按时间排序选上一个） available_ts = sorted(price_map.keys()) best_ts = None for t in available_ts: if t <= ts: best_ts = t price = price_map.get(best_ts) if best_ts else None row = { 'ts': ts, 'price': price, 'TR': float(m.group(2)), 'OB': float(m.group(3)), 'FR': float(m.group(4)), 'TK': float(m.group(5)), 'OI': float(m.group(6)), 'MP': float(m.group(7)), 'VD': float(m.group(8)), 'BTC': float(m.group(9)), 'GM': float(m.group(10)), 'IV': float(m.group(11)), 'EX': float(m.group(12)), 'LC': float(m.group(13)), 'MR': float(m.group(14)), 'SM': float(m.group(15)), 'CT': float(m.group(16)), 'total': float(m.group(17)), 'dir': m.group(18), } rows.append(row) # --- 新闻行 --- if '[News]' in line and '[MyTrader]' in line: ts_match = re.match( r'(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2})', line ) if not ts_match: continue ts = ts_match.group(1) # DeepSeek 格式 ds = re.search( r'\[News\]\s*DeepSeek:\s*AI判断:([+-]?\d+\.?\d*)\(.*?\)\s+恐贪:(\d+)\s+(.*?)\s+\[(.+?)\]', line ) if ds: news_events.append({ 'ts': ts, 'ai_score': float(ds.group(1)), 'fear_greed': int(ds.group(2)), 'summary': ds.group(3).strip()[:120], 'headline': ds.group(4).strip()[:100], 'score': 0, # 后续填充 }) else: # 快讯格式 nx = re.search(r'\[News\]\s*(.+?)(?:=>\s*([+-]?\d+))?\s*$', line) if nx: score_str = nx.group(2) news_events.append({ 'ts': ts, 'ai_score': 0, 'fear_greed': 0, 'summary': nx.group(1).strip()[:120], 'headline': '', 'score': int(score_str) if score_str else 0, }) # 前向填充价格（没有精确匹配时用最近价格） last_px = None for r in rows: if r['price'] is not None: last_px = r['price'] else: r['price'] = last_px # 过滤无价格的行 rows = [r for r in rows if r['price'] is not None] # 限制行数 rows = rows[-max_rows:] # 过滤新闻到有效时间范围 if rows: t_min, t_max = rows[0]['ts'], rows[-1]['ts'] news_events = [ n for n in news_events if t_min <= n['ts'] <= t_max ] return rows, news_events