claudetools/temp/bardach_dedup_step4_delete_batch.py

#!/usr/bin/env python3
"""Step 4: Delete duplicate contacts in batches.
Usage: python bardach_dedup_step4_delete_batch.py [start_offset]
Processes 500 deletes per run. Saves progress."""

import json
import subprocess
import sys
import time
import os

TENANT_ID = "dd4a82e8-85a3-44ac-8800-07945ab4d95f"
CLIENT_ID = "fabb3421-8b34-484b-bc17-e46de9703418"
CLIENT_SECRET = "~QJ8Q~NyQSs4OcGqHZyPrA2CVnq9KBfKiimntbMO"
SCOPE = "https://graph.microsoft.com/.default"
USER = "barbara@bardach.net"
PLAN_FILE = "D:/ClaudeTools/temp/bardach_dedup_plan.json"
PROGRESS_FILE = "D:/ClaudeTools/temp/bardach_dedup_delete_progress.json"
RESULTS_FILE = "D:/ClaudeTools/temp/bardach_dedup_delete_results.json"
BATCH_SIZE = 500


def get_token():
    url = f"https://login.microsoftonline.com/{TENANT_ID}/oauth2/v2.0/token"
    result = subprocess.run(
        ["curl", "-s", "-X", "POST", url,
         "-H", "Content-Type: application/x-www-form-urlencoded",
         "-d", f"client_id={CLIENT_ID}&scope={SCOPE}&client_secret={CLIENT_SECRET}&grant_type=client_credentials"],
        capture_output=True, text=True
    )
    data = json.loads(result.stdout)
    if "access_token" not in data:
        print(f"[ERROR] Failed to get token: {data}", flush=True)
        sys.exit(1)
    return data["access_token"]


def delete_contact(token, contact_id):
    url = f"https://graph.microsoft.com/v1.0/users/{USER}/contacts/{contact_id}"
    result = subprocess.run(
        ["curl", "-s", "-X", "DELETE", url,
         "-H", f"Authorization: Bearer {token}",
         "-w", "%{http_code}"],
        capture_output=True, text=True
    )
    output = result.stdout.strip()
    try:
        status_code = int(output[-3:])
    except (ValueError, IndexError):
        status_code = 0
    return status_code


def load_progress():
    if os.path.exists(PROGRESS_FILE):
        with open(PROGRESS_FILE, "r", encoding="utf-8") as f:
            return json.load(f)
    return {"completed_index": 0, "successes": 0, "failures": []}


def save_progress(progress):
    with open(PROGRESS_FILE, "w", encoding="utf-8") as f:
        json.dump(progress, f, indent=2, ensure_ascii=False)


def main():
    start_offset = int(sys.argv[1]) if len(sys.argv) > 1 else None

    with open(PLAN_FILE, "r", encoding="utf-8") as f:
        plan_data = json.load(f)

    all_deletes = []
    for entry in plan_data["plan"]:
        for did in entry["delete_ids"]:
            all_deletes.append({"id": did, "display_name": entry["display_name"]})

    total = len(all_deletes)

    progress = load_progress()
    start = start_offset if start_offset is not None else progress["completed_index"]
    end = min(start + BATCH_SIZE, total)

    print(f"DELETE BATCH: {start}-{end} of {total} (batch size {BATCH_SIZE})", flush=True)

    if start >= total:
        print(f"[OK] All {total} deletes already processed!", flush=True)
        print(f"  Successes: {progress['successes']}", flush=True)
        print(f"  Failures: {len(progress['failures'])}", flush=True)
        # Save final results
        with open(RESULTS_FILE, "w", encoding="utf-8") as f:
            json.dump({
                "total_attempted": total,
                "successes": progress["successes"],
                "failures": len(progress["failures"]),
                "failure_details": progress["failures"]
            }, f, indent=2, ensure_ascii=False)
        return

    token = get_token()
    print("[OK] Token acquired", flush=True)

    successes = progress["successes"]
    failures = list(progress["failures"])
    batch_start_time = time.time()

    for i in range(start, end):
        item = all_deletes[i]
        status_code = delete_contact(token, item["id"])

        if status_code == 204 or status_code == 200:
            successes += 1
        elif status_code == 404:
            # Already deleted (maybe from previous run)
            successes += 1
        else:
            failures.append({"display_name": item["display_name"], "id": item["id"], "status": status_code})

        if (i - start + 1) % 100 == 0:
            elapsed = time.time() - batch_start_time
            rate = (i - start + 1) / elapsed if elapsed > 0 else 0
            print(f"  {i + 1}/{total} | Batch: {i - start + 1}/{end - start} | OK: {successes} | Fail: {len(failures)} | {rate:.1f}/sec", flush=True)

        if (i - start + 1) % 50 == 0:
            time.sleep(0.3)

    # Save progress
    progress = {"completed_index": end, "successes": successes, "failures": failures}
    save_progress(progress)

    elapsed = time.time() - batch_start_time
    print(f"\nBatch complete: {start}-{end} in {elapsed:.0f}s", flush=True)
    print(f"  Total successes so far: {successes}", flush=True)
    print(f"  Total failures so far: {len(failures)}", flush=True)
    print(f"  Next batch starts at: {end}", flush=True)

    if end < total:
        print(f"  Remaining: {total - end}", flush=True)
    else:
        print(f"[OK] ALL DELETES COMPLETE!", flush=True)
        with open(RESULTS_FILE, "w", encoding="utf-8") as f:
            json.dump({
                "total_attempted": total,
                "successes": successes,
                "failures": len(failures),
                "failure_details": failures
            }, f, indent=2, ensure_ascii=False)
        print(f"[OK] Results saved to {RESULTS_FILE}", flush=True)


if __name__ == "__main__":
    main()