Chan/research/live/probe_inner.py

"""把 `inner_ms` 拆成和本地一致的分档，用来定位两边测不一致的那部分。

起因：本地量到的构成是 TF_DF 占 ~80%、信号链 ~20%，服务器报的是 chan 构建
22ms、信号链 86ms。按机器差（×1.36）也解释不了四倍差距，说明两边测的不是
同一件事，或者有个环节只在服务器上贵。

用同一批窗口跑，比较分档而不是总数。两边都跑一遍再对表：

    python research/live/probe_inner.py --syms BTC,ETH,SOL --repeat 5

分档口径（与 shadow_signal.compute 的调用顺序一致）：
    rebuild      payload → DataFrame
    ind_ltf/htf  两条腿各自的 add_indicators（TF_DF 内部会做，这里单独计时）
    chan_ltf/htf TF_DF 构建（lean）
    zones        build_htf_zones
    ladder       add_zone_ladder
    bsp          find_fast_bsp3
    timeline     htf_fx_timeline（5m 分型时间线）
    attach       attach_htf_agree + attach_zone_ladder

注意 `ind_*` 与 `chan_*` 在真实路径里是合一的（TF_DF 内部调 add_indicators），
这里拆开只为定位。总和会略大于实际 inner_ms。
"""
from __future__ import annotations

import argparse
import os
import sys
import time
import warnings
from pathlib import Path

import numpy as np
import pandas as pd

warnings.filterwarnings("ignore")
for _v in ("OMP_NUM_THREADS", "OPENBLAS_NUM_THREADS", "MKL_NUM_THREADS"):
    os.environ.setdefault(_v, "1")

HERE = Path(__file__).resolve()
sys.path.insert(0, str(HERE.parents[1]))
sys.path.insert(0, str(HERE.parents[2]))
sys.path.insert(0, str(HERE.parent))

LTF_BARS, HTF_BARS = 2001, 801


def med(fn, n: int):
    ts = []
    out = None
    for _ in range(n):
        t = time.perf_counter()
        out = fn()
        ts.append((time.perf_counter() - t) * 1000)
    return float(np.median(ts)), out


def probe(sym: str, repeat: int) -> dict:
    from chanlun import TF_DF
    from chanlun.analysis.fast_bsp import (
        add_zone_ladder, attach_htf_agree, attach_zone_ladder,
        build_htf_zones, find_fast_bsp3, htf_fx_timeline,
    )
    from lib.data import fetch_ohlcv

    dl = fetch_ohlcv(f"{sym}/USDT:USDT", "1m", LTF_BARS * 3).tail(LTF_BARS).reset_index(drop=True)
    dh = fetch_ohlcv(f"{sym}/USDT:USDT", "5m", HTF_BARS * 3).tail(HTF_BARS).reset_index(drop=True)

    probe_tf = TF_DF(lean=True)
    r = {"sym": sym}
    r["ind_ltf"], _ = med(lambda: probe_tf.add_indicators(dl.copy()), repeat)
    r["ind_htf"], _ = med(lambda: probe_tf.add_indicators(dh.copy()), repeat)
    r["chan_ltf"], cl = med(lambda: TF_DF(dl, 1, "1m", lean=True), repeat)
    r["chan_htf"], ch = med(lambda: TF_DF(dh, 1, "5m", lean=True), repeat)
    cdf = cl.dataframe
    r["zones"], z = med(lambda: build_htf_zones(cdf, "1m", chan=cl), repeat)
    if z is None or z.empty:
        r["n_zones"] = 0
        return r
    z0 = z.reset_index(drop=True)
    r["ladder"], zl = med(lambda: add_zone_ladder(z0), repeat)
    r["bsp"], sg = med(lambda: find_fast_bsp3(cdf, zl), repeat)
    r["timeline"], tl = med(lambda: htf_fx_timeline(ch, ch.dataframe), repeat)
    r["attach"], _ = med(
        lambda: attach_zone_ladder(attach_htf_agree(sg, cdf, tl), zl), repeat)
    r["n_zones"], r["n_sig"] = len(z0), len(sg)

    # 增量口径：init 一次后追加，看稳态单根成本
    c = TF_DF(lean=True)
    c.init_stream(dl.iloc[:-60].reset_index(drop=True), 1, "1m")
    ts = []
    for k in range(len(dl) - 60, len(dl)):
        t = time.perf_counter()
        c.append_bar(dl.iloc[k])
        ts.append((time.perf_counter() - t) * 1000)
    r["append_bar"] = float(np.median(ts[20:]))
    return r


def main() -> None:
    ap = argparse.ArgumentParser()
    ap.add_argument("--syms", default="BTC,ETH,SOL")
    ap.add_argument("--repeat", type=int, default=5)
    args = ap.parse_args()

    rows = [probe(s.strip(), args.repeat) for s in args.syms.split(",") if s.strip()]
    d = pd.DataFrame(rows).set_index("sym")
    parts = [c for c in ("ind_ltf", "ind_htf", "chan_ltf", "chan_htf", "zones",
                         "ladder", "bsp", "timeline", "attach") if c in d]
    d["合计"] = d[parts].sum(axis=1)
    pd.set_option("display.width", 220)
    print("\n分档耗时（ms，中位）")
    print(d[parts + ["合计", "append_bar"]].round(2).to_string())
    print("\n占比（%）")
    print((d[parts].div(d["合计"], axis=0) * 100).round(1).to_string())
    print("\n规模")
    print(d[[c for c in ("n_zones", "n_sig") if c in d]].to_string())
    print("\n注：ind_* 与 chan_* 在真实路径里合一（TF_DF 内部调 add_indicators），"
          "拆开只为定位，合计会略大于实际 inner_ms。")


if __name__ == "__main__":
    main()