#!/usr/bin/env python3
"""Backfill Asterisk Master.csv CDR rows into the MySQL `cdr` table.

CSV column mapping for Asterisk 16 default cdr_csv.conf:
  calldate=9, clid=4, src=1, dst=2, dcontext=3,
  channel=5, dstchannel=6, lastapp=7, lastdata=8,
  duration=12, billsec=13, disposition=14, amaflags=15,
  accountcode=0, uniqueid=16, userfield=17
  peeraccount/linkedid/sequence are not in the CSV -> NULL

Idempotent: skips rows whose uniqueid already exists; for rows with no
uniqueid (early Asterisk) dedups on a row-content hash.

Run as asterisk:
  sudo -u asterisk /opt/vm-transcribe/venv/bin/python3 /opt/vm-transcribe/vm_backfill_cdr.py
"""
import csv
import hashlib
import os
import sys

sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
import vm_store

CSV_PATH = "/var/log/asterisk/cdr-csv/Master.csv"
# CSV positions (0-based) as parsed from a real Master.csv row
POS = {
    "accountcode": 0, "src": 1, "dst": 2, "dcontext": 3, "clid": 4,
    "channel": 5, "dstchannel": 6, "lastapp": 7, "lastdata": 8,
    "calldate": 9, "duration": 12, "billsec": 13, "disposition": 14,
    "amaflags": 15, "uniqueid": 16, "userfield": 17,
}
# Column order matching the MySQL cdr table
COLS = [
    "calldate", "clid", "src", "dst", "dcontext",
    "channel", "dstchannel", "lastapp", "lastdata",
    "duration", "billsec", "disposition", "amaflags",
    "accountcode", "uniqueid", "peeraccount", "linkedid",
    "sequence", "userfield",
]
# Which of those are NULL because not in the CSV
NULL_AFTER = {"peeraccount", "linkedid", "sequence"}


def parse(v):
    if v is None or v == "":
        return None
    try:
        return int(v)
    except ValueError:
        return v


def main():
    if not os.path.exists(CSV_PATH):
        print("csv not found:", CSV_PATH)
        return
    con = vm_store.connect()
    seen = set()
    total = 0
    skip = 0
    with open(CSV_PATH, newline="", encoding="utf-8", errors="replace") as f:
        reader = csv.reader(f, quotechar='"', escapechar="\\", doublequote=False)
        for row in reader:
            if not row or len(row) < 17:
                continue
            uid = (row[POS["uniqueid"]] if len(row) > POS["uniqueid"] else "") or ""
            key = uid
            if not key:
                key = hashlib.sha1(("\x00".join(row)).encode("utf-8", errors="replace")).hexdigest()[:32]
            if key in seen:
                skip += 1
                continue
            seen.add(key)
            total += 1
            vals = []
            for c in COLS:
                if c in NULL_AFTER:
                    vals.append(None)
                    continue
                v = row[POS[c]] if POS[c] < len(row) else None
                vals.append(parse(v))
            ph = ",".join(["%s"] * len(COLS))
            try:
                con.execute(
                    "INSERT IGNORE INTO cdr (%s) VALUES (%s)"
                    % (",".join(COLS), ph),
                    vals,
                )
            except Exception as e:
                print("  skip key=%s: %s" % (key, e))
    con.commit()
    mysql_total = con.execute("SELECT COUNT(*) c FROM cdr").fetchone()["c"]
    print("csv scanned: %d  skipped: %d  -> MySQL cdr total: %d" % (total, skip, mysql_total))


if __name__ == "__main__":
    main()