#!/usr/bin/env python3
"""Step 2: Build dedup plan from backup contacts."""

import json
import os
from collections import defaultdict
from datetime import datetime

BACKUP_FILE = "D:/ClaudeTools/temp/bardach_temp_backup_prededup.json"
PLAN_FILE = "D:/ClaudeTools/temp/bardach_dedup_plan.json"


def load_backup():
    with open(BACKUP_FILE, "r", encoding="utf-8") as f:
        data = json.load(f)
    return data["contacts"]


def normalize_name(name):
    """Normalize display name for grouping."""
    if not name:
        return ""
    return name.strip().lower()


def get_emails(contact):
    """Extract email addresses as lowercase set."""
    emails = set()
    for e in (contact.get("emailAddresses") or []):
        addr = (e.get("address") or "").strip().lower()
        if addr:
            emails.add(addr)
    return emails


def get_phones(contact, field):
    """Extract phone numbers as set."""
    phones = set()
    for p in (contact.get(field) or []):
        cleaned = p.strip()
        if cleaned:
            phones.add(cleaned)
    return phones


def is_address_empty(addr):
    """Check if an address object is empty."""
    if not addr:
        return True
    for key in ["street", "city", "state", "postalCode", "countryOrRegion"]:
        if (addr.get(key) or "").strip():
            return False
    return True


def score_contact(contact):
    """Score a contact by richness of data."""
    score = 0

    # Email addresses (2 pts each)
    emails = get_emails(contact)
    score += len(emails) * 2

    # Phone numbers (2 pts each)
    for field in ["homePhones", "businessPhones"]:
        score += len(get_phones(contact, field)) * 2

    # Text fields (1 pt each if non-empty)
    for field in ["companyName", "jobTitle", "nickName", "birthday"]:
        if (contact.get(field) or "").strip():
            score += 1

    # Personal notes (2 pts if non-empty, more for longer)
    notes = (contact.get("personalNotes") or "").strip()
    if notes:
        score += 2
        if len(notes) > 50:
            score += 1

    # Addresses (2 pts each if non-empty)
    for field in ["homeAddress", "businessAddress", "otherAddress"]:
        if not is_address_empty(contact.get(field)):
            score += 2

    # Categories (1 pt if has any)
    if contact.get("categories"):
        score += 1

    # Given/surname (1 pt each)
    if (contact.get("givenName") or "").strip():
        score += 1
    if (contact.get("surname") or "").strip():
        score += 1

    # Recency bonus: slight preference for more recently modified
    lm = contact.get("lastModifiedDateTime")
    if lm:
        try:
            dt = datetime.fromisoformat(lm.replace("Z", "+00:00"))
            # Give up to 2 bonus points for recency (within last year = 2, older = less)
            days_ago = (datetime.now(dt.tzinfo) - dt).days
            if days_ago < 365:
                score += 2
            elif days_ago < 730:
                score += 1
        except Exception:
            pass

    return score


def build_merge_updates(keeper, duplicates):
    """Determine what unique data from duplicates should be merged into keeper."""
    updates = {}

    # Merge emails
    keeper_emails = get_emails(keeper)
    new_emails = set()
    for dup in duplicates:
        new_emails |= get_emails(dup)
    new_emails -= keeper_emails
    if new_emails:
        # Build new emailAddresses list: keeper's existing + new ones
        existing = list(keeper.get("emailAddresses") or [])
        for addr in new_emails:
            existing.append({"address": addr, "name": ""})
        updates["emailAddresses"] = existing

    # Merge phones
    for field in ["homePhones", "businessPhones"]:
        keeper_phones = get_phones(keeper, field)
        new_phones = set()
        for dup in duplicates:
            new_phones |= get_phones(dup, field)
        new_phones -= keeper_phones
        if new_phones:
            existing = list(keeper.get(field) or [])
            existing.extend(list(new_phones))
            updates[field] = existing

    # Merge notes (append unique notes)
    keeper_notes = (keeper.get("personalNotes") or "").strip()
    for dup in duplicates:
        dup_notes = (dup.get("personalNotes") or "").strip()
        if dup_notes and dup_notes != keeper_notes and dup_notes not in keeper_notes:
            if keeper_notes:
                keeper_notes += "\n---\n" + dup_notes
            else:
                keeper_notes = dup_notes
    if keeper_notes != (keeper.get("personalNotes") or "").strip():
        updates["personalNotes"] = keeper_notes

    # Fill blank fields from duplicates
    for field in ["companyName", "jobTitle", "nickName", "birthday"]:
        if not (keeper.get(field) or "").strip():
            for dup in duplicates:
                val = (dup.get(field) or "").strip()
                if val:
                    updates[field] = val
                    break

    # Fill blank addresses
    for field in ["homeAddress", "businessAddress", "otherAddress"]:
        if is_address_empty(keeper.get(field)):
            for dup in duplicates:
                if not is_address_empty(dup.get(field)):
                    updates[field] = dup[field]
                    break

    # Fill given/surname if blank
    for field in ["givenName", "surname"]:
        if not (keeper.get(field) or "").strip():
            for dup in duplicates:
                val = (dup.get(field) or "").strip()
                if val:
                    updates[field] = val
                    break

    # Merge categories
    keeper_cats = set(keeper.get("categories") or [])
    new_cats = set()
    for dup in duplicates:
        new_cats |= set(dup.get("categories") or [])
    new_cats -= keeper_cats
    if new_cats:
        updates["categories"] = list(keeper_cats | new_cats)

    return updates


def main():
    print("=" * 60)
    print("STEP 2: Build dedup plan")
    print("=" * 60)

    contacts = load_backup()
    print(f"[OK] Loaded {len(contacts)} contacts from backup")

    # Group by normalized displayName
    groups = defaultdict(list)
    no_name_count = 0
    for c in contacts:
        name = normalize_name(c.get("displayName"))
        if not name:
            no_name_count += 1
            continue
        groups[name].append(c)

    print(f"[INFO] Unique names: {len(groups)}")
    print(f"[INFO] Contacts without displayName: {no_name_count}")

    # Find duplicate groups (2+ contacts with same name)
    dup_groups = {name: clist for name, clist in groups.items() if len(clist) >= 2}
    print(f"[INFO] Duplicate groups (2+ contacts with same name): {len(dup_groups)}")

    total_dupes = sum(len(v) for v in dup_groups.values())
    total_to_delete = total_dupes - len(dup_groups)  # keep one per group
    print(f"[INFO] Total contacts in duplicate groups: {total_dupes}")
    print(f"[INFO] Contacts to delete (extras): {total_to_delete}")

    # Build merge plan
    plan = []
    keepers_needing_updates = 0

    for name, clist in sorted(dup_groups.items()):
        # Score each contact
        scored = [(score_contact(c), c) for c in clist]
        scored.sort(key=lambda x: x[0], reverse=True)

        keeper = scored[0][1]
        duplicates = [s[1] for s in scored[1:]]

        # Build updates
        updates = build_merge_updates(keeper, duplicates)

        entry = {
            "display_name": name,
            "group_size": len(clist),
            "keeper_id": keeper["id"],
            "keeper_score": scored[0][0],
            "updates_to_apply": updates,
            "delete_ids": [d["id"] for d in duplicates],
            "delete_count": len(duplicates)
        }
        plan.append(entry)

        if updates:
            keepers_needing_updates += 1

    # Save plan
    with open(PLAN_FILE, "w", encoding="utf-8") as f:
        json.dump({"total_groups": len(plan), "plan": plan}, f, indent=2, ensure_ascii=False)

    # Summary
    total_deletes = sum(e["delete_count"] for e in plan)
    print(f"\n{'=' * 60}")
    print(f"DEDUP PLAN SUMMARY")
    print(f"{'=' * 60}")
    print(f"  Duplicate groups:         {len(plan)}")
    print(f"  Keepers needing updates:  {keepers_needing_updates}")
    print(f"  Contacts to delete:       {total_deletes}")
    print(f"  Contacts to keep (dupes): {len(plan)}")
    print(f"  Unique contacts (no dup): {len(groups) - len(dup_groups)}")
    print(f"  Final expected count:     {len(groups) - len(dup_groups) + len(plan) + no_name_count}")
    print(f"[OK] Plan saved to {PLAN_FILE}")

    # Show top 10 largest duplicate groups
    by_size = sorted(plan, key=lambda x: x["group_size"], reverse=True)[:10]
    print(f"\nTop 10 largest duplicate groups:")
    for e in by_size:
        print(f"  {e['display_name']}: {e['group_size']} copies (delete {e['delete_count']})")


if __name__ == "__main__":
    main()