"""Step 4: independent photometry for the whole ALFALFA gold+silver sample via the CDS XMatch service.

  * 2MASS Extended Source Catalogue (VII/233/xsc): total K-band magnitude and 2MASS axis ratio
    -> independent check on SDSS/unWISE catalogue photometry (shredding of large galaxies)
  * AllWISE (II/328/allwise): W1-W4 profile-fit magnitudes -> mid-infrared (warm waste heat) test
Outputs: RAW/xm_2mass_xsc.csv, RAW/xm_allwise.csv ; data/alfalfa_phot.csv (merged, one row per AGC)
"""
import io
import os
import subprocess
import numpy as np
import pandas as pd
from common import RAW, DATA

XMATCH = "https://cdsxmatch.u-strasbg.fr/xmatch/api/v1/sync"


def xmatch(df, cat2, radius_arcsec, outfile, force=False):
    path = os.path.join(RAW, outfile)
    if os.path.exists(path) and os.path.getsize(path) > 100 and not force:
        return pd.read_csv(path)
    up = os.path.join(RAW, "_upload.csv")
    df[["AGC", "ra", "dec"]].to_csv(up, index=False)
    # multipart upload is simplest with curl (uses the pre-configured proxy/CA bundle)
    cmd = ["curl", "-sS", "--max-time", "600", "-X", "POST",
           "-F", "request=xmatch", "-F", f"distMaxArcsec={radius_arcsec}", "-F", "RESPONSEFORMAT=csv",
           "-F", f"cat1=@{up}", "-F", "colRA1=ra", "-F", "colDec1=dec", "-F", f"cat2={cat2}", XMATCH]
    out = subprocess.run(cmd, capture_output=True, text=True, check=True).stdout
    if not out.startswith("angDist"):
        raise RuntimeError(f"xmatch failed for {cat2}: {out[:400]}")
    with open(path, "w") as f:
        f.write(out)
    return pd.read_csv(io.StringIO(out))


def main():
    s = pd.read_csv(os.path.join(DATA, "alfalfa_resid.csv"), low_memory=False)
    s["ra"], s["dec"] = s.RAO, s.DEO
    x = xmatch(s, "vizier:VII/233/xsc", 15, "xm_2mass_xsc.csv")
    x = x.sort_values("angDist").drop_duplicates("AGC")
    x = x.rename(columns={"K.ext": "Kext", "e_K.ext": "e_Kext", "Sb/a": "ba_2mass", "angDist": "sep_2mass",
                          "J.ext": "Jext", "H.ext": "Hext"})[
        ["AGC", "sep_2mass", "Jext", "Hext", "Kext", "e_Kext", "ba_2mass", "r3sig"]]
    w = xmatch(s, "vizier:II/328/allwise", 6, "xm_allwise.csv")
    w = w.sort_values("angDist").drop_duplicates("AGC")
    keep = ["AGC", "angDist", "AllWISE", "W1mag", "e_W1mag", "W2mag", "e_W2mag", "W3mag", "e_W3mag",
            "W4mag", "e_W4mag", "ccf", "ex", "qph"]
    w = w[[c for c in keep if c in w.columns]].rename(columns={"angDist": "sep_wise"})
    m = s[["AGC"]].merge(x, on="AGC", how="left").merge(w, on="AGC", how="left")
    m.to_csv(os.path.join(DATA, "alfalfa_phot.csv"), index=False)
    print(f"  sample rows {len(s)}; 2MASS XSC matches {m.Kext.notna().sum()}; AllWISE matches {m.W1mag.notna().sum()}"
          f"; W3 with error {m.e_W3mag.notna().sum()}; W4 with error {m.e_W4mag.notna().sum()}")


if __name__ == "__main__":
    main()