Model/orchestration/sharepoint_renamer_orchestrator.py

import csv
import os
from typing import Optional

from domain.pashub_fetcher.sharepoint_subfolders import SharepointSubfolders
from utilities.logger import setup_logger
from utils.sharepoint.domna_sharepoint_client import DomnaSharepointClient

BASE_PATH = (
    "Osmosis-ACD Projects/Sero-Clarion Housing/"
    "Sero Project Documents/Property Folders"
)
ASSESSMENT_SUBFOLDER = "A. Assessment"
BATCH_FOLDER_PREFIX = "_Sero Batch"

logger = setup_logger()


def build_canonical_filename(
    uprn: str, address: str, postcode: str, original_name: str
) -> Optional[str]:
    """
    Returns the canonical filename, or None if the file is already renamed.

    Already-renamed: name starts with "{uprn}_".
    Strips any existing address prefix (address+postcode first, then address alone)
    before inserting the canonical prefix.
    """
    if original_name.startswith(f"{uprn}_"):
        return None

    stem, ext = os.path.splitext(original_name)
    stem_lower = stem.lower()

    street = address.split(",")[0].strip()
    prefixes = [
        f"{address} {postcode}",
        address,
        f"{street} {postcode}",
        street,
    ]

    doc_name = stem
    for prefix in prefixes:
        if stem_lower.startswith(prefix.lower()):
            doc_name = stem[len(prefix) :]
            break

    if doc_name.startswith(" - "):
        doc_name = doc_name[3:]
    elif doc_name.startswith(" _ "):
        doc_name = doc_name[3:]
    doc_name = doc_name.strip()

    street_post = f"{street} {postcode}"
    if doc_name:
        return f"{uprn}_{street_post}_{doc_name}{ext}"
    return f"{uprn}_{street_post}{ext}"


class SharepointRenamerOrchestrator:
    def __init__(
        self, sp_client: DomnaSharepointClient, csv_path: str, dry_run: bool = False
    ) -> None:
        self._sp_client = sp_client
        self._csv_path = csv_path
        self._dry_run = dry_run

    def _discover_roots(self) -> list[str]:
        contents = self._sp_client.get_folders_in_path(BASE_PATH)
        batch_roots = sorted(
            f"{BASE_PATH}/{item['name']}"
            for item in contents.get("value", [])
            if "folder" in item
            and item["name"].lower().startswith(BATCH_FOLDER_PREFIX.lower())
        )
        return [BASE_PATH, *batch_roots]

    def run(self) -> None:
        roots = self._discover_roots()
        with open(self._csv_path, newline="", encoding="utf-8-sig") as f:
            reader = csv.DictReader(f)
            required = {"UPRN", "Address", "Postcode"}
            if not reader.fieldnames or not required.issubset(set(reader.fieldnames)):
                raise ValueError(
                    f"CSV missing required columns. Expected {required}, got {reader.fieldnames}"
                )

            for row in reader:
                self._process_row(
                    roots,
                    uprn=row["UPRN"].strip(),
                    address=row["Address"].strip(),
                    postcode=row["Postcode"].strip(),
                )

    def _process_row(
        self, roots: list[str], uprn: str, address: str, postcode: str
    ) -> None:
        # Batch folders name properties "{uprn}_{address}"; filenames must not
        # repeat the UPRN, so the folder lookup and the canonical name diverge.
        display_address = address.removeprefix(f"{uprn}_")
        for root in roots:
            folder_path = (
                f"{root}/{address}, {postcode}"
                f"/{SharepointSubfolders.ASSESSMENT.value}/{ASSESSMENT_SUBFOLDER}"
            )
            if self._process_folder(folder_path, uprn, display_address, postcode):
                return
        logger.warning(
            f"Missing folder for UPRN {uprn} in any root: {address}, {postcode}"
        )

    def _process_folder(
        self, folder_path: str, uprn: str, address: str, postcode: str
    ) -> bool:
        try:
            contents = self._sp_client.get_folders_in_path(folder_path)
        except ValueError:
            return False

        for item in contents.get("value", []):
            if "folder" in item:
                self._process_folder(
                    f"{folder_path}/{item['name']}", uprn, address, postcode
                )
            elif "file" in item:
                original_name: str = item["name"]
                if os.path.splitext(original_name)[1].lower() in {".jpg", ".heic"}:
                    continue
                new_name = build_canonical_filename(
                    uprn, address, postcode, original_name
                )

                if new_name is None:
                    continue

                if self._dry_run:
                    logger.info(
                        f'Would rename: "{original_name}" → "{new_name}" (UPRN: {uprn})'
                    )
                else:
                    try:
                        self._sp_client.rename_file(item["id"], new_name)
                        logger.info(
                            f'Renamed: "{original_name}" → "{new_name}" (UPRN: {uprn})'
                        )
                    except Exception as e:
                        logger.error(
                            f'Failed to rename "{original_name}" → "{new_name}" (UPRN: {uprn}): {e}'
                        )
        return True