Odoo18-Base/setup/delete_records.py

#!/usr/bin/env python
"""
Version :
This script deletes records from an Odoo database based on a given model and domain filter.

Usage:
    delete_records.py <db_name> <base_model>

Configuration:
Create a .env file with the following variables to configure the Odoo connection:
    HOST = localhost
    PORT = 8069
    DOMAIN = '[('employee', '=', False), ('customer_rank', '>', 0)]'
    USERNAME = admin
    PASSWORD = admin
    PROCESS_SIZE = 30

Example:
    delete_records.py enterprise-ambio res.partner
    delete_records.py enterprise-ambio res.partner --domain "[('active', '=', False)]" --process_size 10
"""
__copyright__ = "Copyright 2025, NextZenOS"
__email__ = "techgroup@nextzenos.com"
__license__ = "GPLv3"
__maintainer__ = "hoangvv"
__status__ = "Development"
__version__ = "0.1.0"
import odoorpc
import json
import os
import multiprocessing as mp
import ast
import argparse
from dotenv import load_dotenv
import color_log
from collections import deque

# Load environment variables
load_dotenv()
HOST = os.getenv("HOST", "localhost")
PORT = int(os.getenv("PORT", "8069"))
USERNAME = os.getenv("USERNAME", "admin")
PASSWORD = os.getenv("PASSWORD", "admin")
DEFAULT_DOMAIN = os.getenv("DOMAIN", "[]")
DEFAULT_PROCESS_SIZE = int(os.getenv("PROCESS_SIZE", "30"))

# Parse command-line arguments
parser = argparse.ArgumentParser(
    description="Delete records from an Odoo model based on a domain filter."
)
parser.add_argument("db_name", help="Database name")
parser.add_argument("base_model", help="Odoo model to delete records from")
parser.add_argument(
    "--domain",
    type=str,
    default=DEFAULT_DOMAIN,
    help="Domain filter (default from .env)",
)
parser.add_argument(
    "--process_size",
    type=int,
    default=DEFAULT_PROCESS_SIZE,
    help="Number of parallel processes (default from .env)",
)
parser.add_argument(
    "--force",
    action="store_true",
    help="Force delete cascade all records instead of archiving",
)
args = parser.parse_args()

db_name = args.db_name
base_model = args.base_model
domain_str = args.domain
process_size = args.process_size
OK = 0
FAIL = 1
INFO = 2
NOTICE = 3
# Odoo connection setup
odoo = odoorpc.ODOO(HOST, port=PORT)
color_log.Show(INFO, ("Available databases:", odoo.db.list()))

# Login to Odoo
try:
    odoo.login(db_name, USERNAME, PASSWORD)
    color_log.Show(
        0,
        f"Connected to Odoo at {HOST}:{PORT}, Database: {db_name}, Model: {base_model}",
    )
except Exception as e:
    color_log.Show(FAIL, f"Fail to Connect to Odoo Server {e}")
    exit(1)

# Convert DOMAIN from string to list
try:
    domain_filter = ast.literal_eval(domain_str)
except Exception as e:
    color_log.Show(FAIL, f"Invalid DOMAIN format: {e}")
    exit(1)


skipped_models_cache = set()  # Cache to store skipped models


def fetch_related_fields_for_model(model, base_model):
    """Fetch related fields for a single model with caching for skipped models."""
    if model in skipped_models_cache:
        color_log.Show(NOTICE, f"Skipping cached model {model}")
        return {}

    related_fields = {}
    try:
        fields = odoo.env[model].fields_get()
        for field_name, field_info in fields.items():
            if (
                field_info.get("type") in ["many2one", "many2many", "one2many"]
                and field_info.get("relation") == base_model
            ):
                related_fields.setdefault(model, []).append(field_name)
    except odoorpc.error.RPCError as e:
        color_log.Show(FAIL, f"Access denied for model {model}: {e}")
        skipped_models_cache.add(model)  # Add to cache if access is denied
        return {}
    except Exception as e:
        color_log.Show(NOTICE, f"Skipping model {model}: {e}")
        skipped_models_cache.add(model)  # Add to cache for any other errors
    return related_fields


# Function to fetch related models
def get_related_fields(db_name, base_model):
    """Fetch related fields for a given model using multiprocessing."""
    cache_file = f"cache/{db_name}/{base_model}.cache.json"

    if os.path.exists(cache_file):
        with open(cache_file, "r") as f:
            related_models = json.load(f)
        return related_models

    color_log.Show(INFO, f"Fetching related models for {base_model} from Odoo...")
    related_models = {}

    try:
        all_model_ids = odoo.env["ir.model"].search([])
        all_models = odoo.env["ir.model"].read(all_model_ids, ["model"])
        model_names = [model["model"] for model in all_models]

        # Use multiprocessing to fetch related fields
        with mp.Pool(processes=process_size) as pool:
            results = pool.starmap(
                fetch_related_fields_for_model,
                [(model_name, base_model) for model_name in model_names],
            )

        # Combine results from all processes
        for result in results:
            for model, fields in result.items():
                related_models.setdefault(model, []).extend(fields)

        # Save to cache
        os.makedirs(os.path.dirname(cache_file), exist_ok=True)
        with open(cache_file, "w") as f:
            json.dump(related_models, f, indent=4)

        return related_models
    except Exception as e:
        color_log.Show(FAIL, f"Error fetching related models: {e}")
        return {}


def delete_records_cascade(
    record_ids, model_name, process_count, related_models, db_name
):
    """Iteratively delete related records before deleting the base records."""
    # Stack for traversal: list of (model_name, record_ids) tuples
    stack = [(model_name, record_ids)]
    # Track visited records to avoid cycles: set of (model_name, record_id) tuples
    visited = set()
    # Deletion order: list of (model_name, record_ids) in post-order
    deletion_order = []

    # Step 1: Collect all records to delete in post-order
    while stack:
        current_model, current_ids = stack.pop()
        current_pairs = {(current_model, rid) for rid in current_ids}

        if not current_pairs.issubset(visited):
            visited.update(current_pairs)
            has_dependents = False

            for related_model, fields in get_related_fields(
                db_name, current_model
            ).items():
                if related_model in skipped_models_cache:
                    color_log.Show(NOTICE, f"Skipping cached model {related_model}")
                    continue

                related_model_obj = odoo.env[related_model]
                for field in fields:
                    try:
                        related_ids = related_model_obj.search(
                            [(field, "in", current_ids)]
                        )
                        if related_ids:
                            stack.append((related_model, related_ids))
                            has_dependents = True
                            color_log.Show(
                                NOTICE,
                                f"{process_count}: Found {len(related_ids)} related records in {related_model} ({field})",
                            )
                            process_count += 1
                    except odoorpc.error.RPCError as e:
                        color_log.Show(
                            FAIL, f"Access denied for model {related_model}: {e}"
                        )
                        skipped_models_cache.add(related_model)  # Add to cache
                        break  # Skip further processing for this model
                    except Exception as e:
                        color_log.Show(
                            NOTICE, f"Skipping model {related_model} due to error: {e}"
                        )
                        skipped_models_cache.add(related_model)  # Add to cache
                        break  # Skip further processing for this model

            if not has_dependents:
                deletion_order.append((current_model, current_ids))
        else:
            deletion_order.append((current_model, current_ids))

    # Step 2: Delete records in the collected order
    for del_model, del_ids in deletion_order:
        try:
            if del_ids:
                odoo.env[del_model].browse(del_ids).unlink()
                color_log.Show(OK, f"Deleted {len(del_ids)} records from {del_model}")
        except Exception as e:
            color_log.Show(FAIL, f"Error deleting {del_model} records: {e}")

    # Now delete the main records after all dependencies are gone
    try:
        if record_ids:
            odoo.env[model_name].browse(record_ids).unlink()
            color_log.Show(OK, f"Deleted {len(record_ids)} records from {model_name}")
    except Exception as e:
        color_log.Show(FAIL, f"Error deleting {model_name} records: {e}")


# Function to delete records in parallel
def process_batch(batch, model_name, process_count, related_models):
    """Process a batch of records - archive or delete based on references."""
    model = odoo.env[model_name]
    archived_count = 0
    deleted_count = 0
    skipped_count = 0

    for record_id in batch:
        is_referenced = any(
            odoo.env[related_model].search_count([(field, "=", record_id)]) > 0
            for related_model, fields in related_models.items()
            for field in fields
        )

        try:
            if is_referenced:
                model.browse(record_id).write({"active": False})
                archived_count += 1
                color_log.Show(
                    INFO, f"{process_count}: Archived {model_name} ID {record_id}"
                )
            else:
                model.unlink([record_id])
                deleted_count += 1
                color_log.Show(
                    INFO, f"{process_count}: Deleted {model_name} ID {record_id}"
                )
        except Exception as e:
            skipped_count += 1
            color_log.Show(
                INFO, f"{process_count}: Skipped {model_name} ID {record_id}: {e}"
            )

    color_log.Show(
        OK,
        f"{process_count}: {model_name} - Deleted: {deleted_count}, Archived: {archived_count}, Skipped: {skipped_count}.",
    )


# Main function to execute deletion
def main():
    """Main function to fetch records and process them in parallel."""
    model = odoo.env[base_model]
    color_log.Show(INFO, f"{domain_filter}")
    record_ids = model.search(domain_filter)

    if not record_ids:
        color_log.Show(
            FAIL, f"No records found for model {base_model} with the given domain."
        )
        return

    related_models = get_related_fields(db_name, base_model)

    batch_list = [
        record_ids[i : i + process_size]
        for i in range(0, len(record_ids), process_size)
    ]
    processes = []
    if args.force:
        delete_records_cascade(record_ids, base_model, 0, related_models, db_name)
        # for i, batch in enumerate(batch_list, start=1):
        #     process = mp.Process(
        #         target=delete_records_cascade,
        #         args=(batch, base_model, f"Process-{i}", related_models,db_name),
        #     )
        #     processes.append(process)
        #     process.start()

        # for process in processes:
        #     process.join()
    else:
        for i, batch in enumerate(batch_list, start=1):
            process = mp.Process(
                target=process_batch,
                args=(batch, base_model, f"Process-{i}", related_models),
            )
            processes.append(process)
            process.start()

        for process in processes:
            process.join()

    color_log.Show(OK, "Record deletion process completed.")


if __name__ == "__main__":
    main()