Medios-Macina/Store/Folder.py

from __future__ import annotations

import json
import re
import shutil
import sys
from fnmatch import translate
from pathlib import Path
from typing import Any, Dict, List, Optional, Tuple

from SYS.logger import debug, log
from SYS.utils import sha256_file

from Store._base import Store


def _normalize_hash(value: Any) -> Optional[str]:
    candidate = str(value or "").strip().lower()
    if len(candidate) != 64:
        return None
    if any(ch not in "0123456789abcdef" for ch in candidate):
        return None
    return candidate


def _resolve_file_hash(db_hash: Optional[str], file_path: Path) -> Optional[str]:
    normalized = _normalize_hash(db_hash) if db_hash else None
    if normalized:
        return normalized
    return _normalize_hash(file_path.stem)


class Folder(Store):
    """"""

    # Track which locations have already been migrated to avoid repeated migrations
    _migrated_locations = set()
    # Cache scan results to avoid repeated full scans across repeated instantiations
    _scan_cache: Dict[str,
                      Tuple[bool,
                            str,
                            Dict[str,
                                 int]]] = {}

    def __new__(cls, *args: Any, **kwargs: Any) -> "Folder":
        return super().__new__(cls)

    setattr(__new__, "keys", ("NAME", "PATH"))

    def __init__(
        self,
        location: Optional[str] = None,
        name: Optional[str] = None,
        *,
        NAME: Optional[str] = None,
        PATH: Optional[str] = None,
    ) -> None:
        if name is None and NAME is not None:
            name = str(NAME)
        if location is None and PATH is not None:
            location = str(PATH)

        self._location = str(location) if location is not None else ""
        self._name = name

        # Scan status (set during init)
        self.scan_ok: bool = True
        self.scan_detail: str = ""
        self.scan_stats: Dict[str,
                              int] = {}

        if self._location:
            try:
                from API.folder import API_folder_store
                from API.folder import LocalLibraryInitializer
                from pathlib import Path

                location_path = Path(self._location).expanduser()

                # Use context manager to ensure connection is properly closed
                with API_folder_store(location_path) as db:
                    if db.connection:
                        db.connection.commit()

                # Call migration and discovery at startup
                Folder.migrate_location(self._location)

                # Local library scan/index (one-time per location per process)
                location_key = str(location_path)
                cached = Folder._scan_cache.get(location_key)
                if cached is None:
                    try:
                        initializer = LocalLibraryInitializer(location_path)
                        stats = initializer.scan_and_index() or {}
                        files_new = int(stats.get("files_new", 0) or 0)
                        sidecars = int(stats.get("sidecars_imported", 0) or 0)
                        total_db = int(stats.get("files_total_db", 0) or 0)
                        if files_new > 0 or sidecars > 0:
                            detail = f"New: {files_new}, Sidecars: {sidecars}" + (
                                f" (Total: {total_db})" if total_db else ""
                            )
                        else:
                            detail = "Up to date" + (
                                f" (Total: {total_db})" if total_db else ""
                            )
                        Folder._scan_cache[location_key] = (True, detail, dict(stats))
                    except Exception as exc:
                        Folder._scan_cache[location_key] = (
                            False,
                            f"Scan failed: {exc}",
                            {}
                        )

                ok, detail, stats = Folder._scan_cache.get(location_key, (True, "", {}))
                self.scan_ok = bool(ok)
                self.scan_detail = str(detail or "")
                self.scan_stats = dict(stats or {})
            except Exception as exc:
                debug(f"Failed to initialize database for '{name}': {exc}")

    @classmethod
    def migrate_location(cls, location: Optional[str]) -> None:
        """Migrate a location to hash-based storage (one-time operation, call explicitly at startup)."""
        if not location:
            return

        from pathlib import Path

        location_path = Path(location).expanduser()
        location_str = str(location_path)

        # Only migrate once per location
        if location_str in cls._migrated_locations:
            return

        cls._migrated_locations.add(location_str)

        cls._migrate_to_hash_storage(location_path)

    @classmethod
    def _migrate_to_hash_storage(cls, location_path: Path) -> None:
        """Migrate existing files from filename-based to hash-based storage.

        Checks for sidecars (.metadata, .tag) and imports them before renaming.
        Also ensures all files have a title: tag.
        """
        from API.folder import API_folder_store, read_sidecar, write_sidecar, find_sidecar

        try:
            with API_folder_store(location_path) as db:
                cursor = db.connection.cursor()

                # First pass: migrate filename-based files and add title tags
                # Scan all files in the storage directory
                for file_path in sorted(location_path.iterdir()):
                    if not file_path.is_file():
                        continue

                    # Skip database files and sidecars
                    if file_path.suffix in (".db", ".metadata", ".tag", "-shm", "-wal"):
                        continue
                    # Also skip if the file ends with -shm or -wal (SQLite journal files)
                    if file_path.name.endswith(("-shm", "-wal")):
                        continue

                    # Check if filename is already a hash (without extension)
                    if len(file_path.stem) == 64 and all(
                            c in "0123456789abcdef" for c in file_path.stem.lower()):
                        continue  # Already migrated, will process in second pass

                    try:
                        # Compute file hash
                        file_hash = sha256_file(file_path)
                        # Preserve extension in the hash-based filename
                        file_ext = file_path.suffix  # e.g., '.mp4'
                        hash_filename = file_hash + file_ext if file_ext else file_hash
                        hash_path = location_path / hash_filename

                        # Check for sidecars and import them
                        sidecar_path = find_sidecar(file_path)
                        tags_to_add = []
                        url_to_add = []
                        has_title_tag = False

                        if sidecar_path and sidecar_path.exists():
                            try:
                                _, tags, url = read_sidecar(sidecar_path)
                                if tags:
                                    tags_to_add = list(tags)
                                    # Check if title tag exists
                                    has_title_tag = any(
                                        t.lower().startswith("title:")
                                        for t in tags_to_add
                                    )
                                if url:
                                    url_to_add = list(url)
                                debug(
                                    f"Found sidecar for {file_path.name}: {len(tags_to_add)} tags, {len(url_to_add)} url",
                                    file=sys.stderr,
                                )
                                # Delete the sidecar after importing
                                sidecar_path.unlink()
                            except Exception as exc:
                                debug(
                                    f"Failed to read sidecar for {file_path.name}: {exc}",
                                    file=sys.stderr,
                                )

                        # Ensure there's a title tag (use original filename if not present)
                        if not has_title_tag:
                            tags_to_add.append(f"title:{file_path.name}")

                        # Rename file to hash if needed
                        if hash_path != file_path and not hash_path.exists():
                            debug(
                                f"Migrating: {file_path.name} -> {hash_filename}",
                                file=sys.stderr
                            )
                            file_path.rename(hash_path)

                            # Ensure DB points to the renamed path (update by hash).
                            try:
                                cursor.execute(
                                    "UPDATE file SET file_path = ?, updated_at = CURRENT_TIMESTAMP WHERE hash = ?",
                                    (db._to_db_file_path(hash_path),
                                     file_hash),
                                )
                            except Exception:
                                pass

                            # Create or update database entry
                            db.get_or_create_file_entry(hash_path)

                            # Save extension metadata
                            ext_clean = file_ext.lstrip(".") if file_ext else ""
                            db.save_metadata(
                                hash_path,
                                {
                                    "hash": file_hash,
                                    "ext": ext_clean,
                                    "size": hash_path.stat().st_size,
                                },
                            )

                            # Add all tags (including title tag)
                            if tags_to_add:
                                db.save_tags(hash_path, tags_to_add)
                                debug(
                                    f"Added {len(tags_to_add)} tags to {file_hash}",
                                    file=sys.stderr
                                )

                            # Note: url would need a separate table if you want to store them
                            # For now, we're just noting them in debug
                            if url_to_add:
                                debug(
                                    f"Imported {len(url_to_add)} url for {file_hash}: {url_to_add}",
                                    file=sys.stderr,
                                )

                    except Exception as exc:
                        debug(
                            f"Failed to migrate file {file_path.name}: {exc}",
                            file=sys.stderr
                        )

                # Second pass: ensure all files in database have a title: tag
                db.connection.commit()
                cursor.execute(
                    """
                    SELECT f.hash, f.file_path 
                    FROM file f 
                    WHERE NOT EXISTS (
                        SELECT 1 FROM tag t WHERE t.hash = f.hash AND LOWER(t.tag) LIKE 'title:%'
                    )
                """
                )
                files_without_title = cursor.fetchall()

                for file_hash, file_path_str in files_without_title:
                    try:
                        file_path = location_path / str(file_path_str)
                        if file_path.exists():
                            # Use the filename as the title
                            title_tag = f"title:{file_path.name}"
                            db.save_tags(file_path, [title_tag])
                            debug(
                                f"Added title tag to {file_path.name}",
                                file=sys.stderr
                            )
                    except Exception as exc:
                        debug(
                            f"Failed to add title tag to file {file_path_str}: {exc}",
                            file=sys.stderr,
                        )

                db.connection.commit()

                # Third pass: discover files on disk that aren't in the database yet
                # These are hash-named files that were added after initial indexing
                cursor.execute("SELECT LOWER(hash) FROM file")
                db_hashes = {row[0]
                             for row in cursor.fetchall()}

                discovered = 0
                for file_path in sorted(location_path.rglob("*")):
                    if file_path.is_file():
                        # Check if file name (without extension) is a 64-char hex hash
                        name_without_ext = file_path.stem
                        if len(name_without_ext) == 64 and all(
                                c in "0123456789abcdef"
                                for c in name_without_ext.lower()):
                            file_hash = name_without_ext.lower()

                            # Skip if already in DB
                            if file_hash in db_hashes:
                                continue

                            try:
                                # Add file to DB (creates entry and auto-adds title: tag)
                                db.get_or_create_file_entry(file_path)

                                # Save extension metadata
                                file_ext = file_path.suffix
                                ext_clean = file_ext.lstrip(".") if file_ext else ""
                                db.save_metadata(
                                    file_path,
                                    {
                                        "hash": file_hash,
                                        "ext": ext_clean,
                                        "size": file_path.stat().st_size,
                                    },
                                )

                                discovered += 1
                            except Exception as e:
                                debug(
                                    f"Failed to discover file {file_path.name}: {e}",
                                    file=sys.stderr,
                                )

                if discovered > 0:
                    debug(
                        f"Discovered and indexed {discovered} undiscovered files in {location_path.name}",
                        file=sys.stderr,
                    )
                    db.connection.commit()
        except Exception as exc:
            debug(f"Migration to hash storage failed: {exc}", file=sys.stderr)

    def location(self) -> str:
        return self._location

    def name(self) -> str:
        return self._name

    def add_file(self, file_path: Path, **kwargs: Any) -> str:
        """Add file to local folder storage with full metadata support.

        Args:
            file_path: Path to the file to add
            move: If True, move file instead of copy (default: False)
            tag: Optional list of tag values to add
            url: Optional list of url to associate with the file
            title: Optional title (will be added as 'title:value' tag)

        Returns:
            File hash (SHA256 hex string) as identifier
        """
        move_file = bool(kwargs.get("move"))
        tag_list = kwargs.get("tag", [])
        url = kwargs.get("url", [])
        title = kwargs.get("title")

        # Extract title from tags if not explicitly provided
        if not title:
            for candidate in tag_list:
                if isinstance(candidate,
                              str) and candidate.lower().startswith("title:"):
                    title = candidate.split(":", 1)[1].strip()
                    break

        # Fallback to filename if no title
        if not title:
            title = file_path.name

        # Ensure title is in tags
        title_tag = f"title:{title}"
        if not any(str(candidate).lower().startswith("title:")
                   for candidate in tag_list):
            tag_list = [title_tag] + list(tag_list)

        try:
            file_hash = sha256_file(file_path)
            debug(f"File hash: {file_hash}", file=sys.stderr)

            # Preserve extension in the stored filename
            file_ext = file_path.suffix  # e.g., '.mp4'
            save_filename = file_hash + file_ext if file_ext else file_hash
            save_file = Path(self._location) / save_filename

            # Check if file already exists
            from API.folder import API_folder_store

            with API_folder_store(Path(self._location)) as db:
                existing_path = db.search_hash(file_hash)
                if existing_path and existing_path.exists():
                    log(
                        f"✓ File already in local storage: {existing_path}",
                        file=sys.stderr,
                    )
                    # Still add tags and url if provided
                    if tag_list:
                        self.add_tag(file_hash, tag_list)
                    if url:
                        self.add_url(file_hash, url)
                    return file_hash

            # Move or copy file (with progress bar on actual byte transfer).
            # Note: a same-volume move may be a fast rename and won't show progress.
            def _copy_with_progress(src: Path, dst: Path, *, label: str) -> None:
                from SYS.models import ProgressFileReader

                total_bytes = None
                try:
                    total_bytes = int(src.stat().st_size)
                except Exception:
                    total_bytes = None

                with src.open("rb") as r, dst.open("wb") as w:
                    reader = ProgressFileReader(r, total_bytes=total_bytes, label=label)
                    while True:
                        chunk = reader.read(1024 * 1024)
                        if not chunk:
                            break
                        w.write(chunk)

                # Preserve file metadata similar to shutil.copy2
                try:
                    shutil.copystat(str(src), str(dst))
                except Exception:
                    pass

            if move_file:
                # Prefer native move; fall back to copy+delete with progress on failure.
                try:
                    shutil.move(str(file_path), str(save_file))
                    debug(f"Local move: {save_file}", file=sys.stderr)
                    # After a move, the original path no longer exists; use destination for subsequent ops.
                    file_path = save_file
                except Exception:
                    _copy_with_progress(
                        file_path,
                        save_file,
                        label=f"folder:{self._name} move"
                    )
                    try:
                        file_path.unlink(missing_ok=True)  # type: ignore[arg-type]
                    except Exception:
                        try:
                            if file_path.exists():
                                file_path.unlink()
                        except Exception:
                            pass
                    debug(f"Local move (copy+delete): {save_file}", file=sys.stderr)
                    file_path = save_file
            else:
                _copy_with_progress(
                    file_path,
                    save_file,
                    label=f"folder:{self._name} copy"
                )
                debug(f"Local copy: {save_file}", file=sys.stderr)

            # Best-effort: capture duration for media
            duration_value: float | None = None
            try:
                from SYS.utils import ffprobe

                probe = ffprobe(str(save_file))
                duration = probe.get("duration")
                if isinstance(duration, (int, float)) and duration > 0:
                    duration_value = float(duration)
            except Exception:
                duration_value = None

            # Save to database (metadata + tag/url updates share one connection)
            with API_folder_store(Path(self._location)) as db:
                conn = getattr(db, "connection", None)
                if conn is None:
                    raise RuntimeError("Folder store DB connection unavailable")
                cursor = conn.cursor()

                debug(
                    f"[Folder.add_file] saving metadata for hash {file_hash}",
                    file=sys.stderr,
                )
                ext_clean = file_ext.lstrip(".") if file_ext else ""
                db.save_metadata(
                    save_file,
                    {
                        "hash": file_hash,
                        "ext": ext_clean,
                        "size": save_file.stat().st_size,
                        "duration": duration_value,
                    },
                )
                debug(
                    f"[Folder.add_file] metadata stored for hash {file_hash}",
                    file=sys.stderr,
                )

                if tag_list:
                    try:
                        debug(
                            f"[Folder.add_file] merging {len(tag_list)} tags for {file_hash}",
                            file=sys.stderr,
                        )
                        from SYS.metadata import compute_namespaced_tag_overwrite

                        existing_tags = [
                            t for t in (db.get_tags(file_hash) or [])
                            if isinstance(t, str) and t.strip()
                        ]
                        _to_remove, _to_add, merged = compute_namespaced_tag_overwrite(
                            existing_tags, tag_list or []
                        )
                        if _to_remove or _to_add:
                            cursor.execute("DELETE FROM tag WHERE hash = ?",
                                           (file_hash,))
                            for t in merged:
                                tag_val = str(t).strip().lower()
                                if tag_val:
                                    cursor.execute(
                                        "INSERT OR IGNORE INTO tag (hash, tag) VALUES (?, ?)",
                                        (file_hash, tag_val),
                                    )
                            conn.commit()
                            debug(
                                f"[Folder.add_file] tags rewritten for {file_hash}",
                                file=sys.stderr,
                            )
                            try:
                                db._update_metadata_modified_time(file_hash)
                            except Exception:
                                pass
                    except Exception as exc:
                        debug(f"Local DB tag merge failed: {exc}", file=sys.stderr)

                if url:
                    try:
                        debug(
                            f"[Folder.add_file] merging {len(url)} URLs for {file_hash}",
                            file=sys.stderr,
                        )
                        from SYS.metadata import normalize_urls

                        existing_meta = db.get_metadata(file_hash) or {}
                        existing_urls = normalize_urls(existing_meta.get("url"))
                        incoming_urls = normalize_urls(url)
                        changed = False
                        for entry in list(incoming_urls or []):
                            if not entry:
                                continue
                            if entry not in existing_urls:
                                existing_urls.append(entry)
                                changed = True
                        if changed:
                            db.update_metadata_by_hash(
                                file_hash,
                                {"url": existing_urls},
                            )
                            debug(
                                f"[Folder.add_file] URLs merged for {file_hash}",
                                file=sys.stderr,
                            )
                    except Exception as exc:
                        debug(f"Local DB URL merge failed: {exc}", file=sys.stderr)

            ##log(f"✓ Added to local storage: {save_file.name}", file=sys.stderr)
            return file_hash

        except Exception as exc:
            log(f"❌ Local storage failed: {exc}", file=sys.stderr)
            raise

    def search(self, query: str, **kwargs: Any) -> list[Dict[str, Any]]:
        """Search local database for files by title tag or filename."""
        from fnmatch import fnmatch
        from API.folder import DatabaseAPI
        import unicodedata

        limit = kwargs.get("limit")
        try:
            limit = int(limit) if limit is not None else None
        except (TypeError, ValueError):
            limit = None
        if isinstance(limit, int) and limit <= 0:
            limit = None

        query = query.lower()
        query_lower = query  # Ensure query_lower is defined for all code paths

        def _normalize_namespace_text(text: str, *, allow_wildcards: bool) -> str:
            """Normalize tag namespace values for consistent matching.

            Removes control/format chars (e.g. zero-width spaces) that frequently appear in scraped tags,
            collapses whitespace, and lowercases.
            """
            s = str(text or "")
            # Normalize newlines/tabs/etc to spaces early.
            s = s.replace("\r", " ").replace("\n", " ").replace("\t", " ")
            # Drop control / format chars (Cc/Cf) while preserving wildcard tokens when requested.
            cleaned_chars: list[str] = []
            for ch in s:
                if allow_wildcards and ch in {"*",
                                              "?"}:
                    cleaned_chars.append(ch)
                    continue
                cat = unicodedata.category(ch)
                if cat in {"Cc",
                           "Cf"}:
                    continue
                cleaned_chars.append(ch)
            s = "".join(cleaned_chars)
            # Collapse any remaining unicode whitespace runs.
            s = " ".join(s.split())
            return s.strip().lower()

        def _normalize_ext_filter(value: str) -> str:
            v = str(value or "").strip().lower().lstrip(".")
            v = "".join(ch for ch in v if ch.isalnum())
            return v

        def _extract_system_filetype_ext(text: str) -> Optional[str]:
            # Match: system:filetype = png  (allow optional '=' and flexible spaces)
            m = re.search(r"\bsystem:filetype\s*(?:=\s*)?([^\s,]+)", text)
            if not m:
                m = re.search(r"\bsystem:filetype\s*=\s*([^\s,]+)", text)
            if not m:
                return None
            return _normalize_ext_filter(m.group(1)) or None

        # Support `ext:<value>` and Hydrus-style `system:filetype = <value>` anywhere
        # in the query (space or comma separated).
        ext_filter: Optional[str] = None
        try:
            sys_ext = _extract_system_filetype_ext(query_lower)
            if sys_ext:
                ext_filter = sys_ext
                query_lower = re.sub(
                    r"\s*\bsystem:filetype\s*(?:=\s*)?[^\s,]+",
                    " ",
                    query_lower
                )
                query_lower = re.sub(r"\s{2,}", " ", query_lower).strip().strip(",")
                query = query_lower

            m = re.search(r"\bext:([^\s,]+)", query_lower)
            if not m:
                m = re.search(r"\bextension:([^\s,]+)", query_lower)
            if m:
                ext_filter = _normalize_ext_filter(m.group(1)) or None
                query_lower = re.sub(
                    r"\s*\b(?:ext|extension):[^\s,]+",
                    " ",
                    query_lower
                )
                query_lower = re.sub(r"\s{2,}", " ", query_lower).strip().strip(",")
                query = query_lower
        except Exception:
            ext_filter = None

        match_all = query == "*" or (not query and bool(ext_filter))
        results = []
        search_dir = Path(self._location).expanduser()

        def _url_like_pattern(value: str) -> str:
            # Interpret user patterns as substring matches (with optional glob wildcards).
            v = (value or "").strip().lower()
            if not v or v == "*":
                return "%"
            v = v.replace("%", "\\%").replace("_", "\\_")
            v = v.replace("*", "%").replace("?", "_")
            if "%" not in v and "_" not in v:
                return f"%{v}%"
            if not v.startswith("%"):
                v = "%" + v
            if not v.endswith("%"):
                v = v + "%"
            return v

        def _like_pattern(term: str) -> str:
            # Convert glob-like tokens to SQL LIKE wildcards.
            return str(term or "").replace("*", "%").replace("?", "_")

        tokens = [t.strip() for t in query.split(",") if t.strip()]

        if not match_all and len(tokens) == 1 and _normalize_hash(query):
            debug("Hash queries require 'hash:' prefix for local search")
            return results

        if not match_all and _normalize_hash(query):
            debug("Hash queries require 'hash:' prefix for local search")
            return results

        def _create_entry(
            file_path: Path,
            tags: list[str],
            size_bytes: int | None,
            db_hash: Optional[str]
        ) -> dict[str,
                  Any]:
            path_str = str(file_path)
            # Get title from tags if available, otherwise use hash as fallback
            title = next(
                (t.split(":",
                         1)[1] for t in tags if t.lower().startswith("title:")),
                None
            )
            if not title:
                # Fallback to hash if no title tag exists
                hash_value = _resolve_file_hash(db_hash, file_path)
                title = hash_value if hash_value else file_path.stem

            # Extract extension from file path
            ext = file_path.suffix.lstrip(".")
            if not ext:
                # Fallback: try to extract from title (original filename might be in title)
                title_path = Path(title)
                ext = title_path.suffix.lstrip(".")

            # Build clean entry with only necessary fields
            hash_value = _resolve_file_hash(db_hash, file_path)
            entry = {
                "title": title,
                "ext": ext,
                "path": path_str,
                "target": path_str,
                "store": self._name,
                "size": size_bytes,
                "hash": hash_value,
                "tag": tags,
            }
            return entry

        try:
            if not search_dir.exists():
                debug(f"Search directory does not exist: {search_dir}")
                return results

            try:
                with DatabaseAPI(search_dir) as api:
                    ext_hashes: set[str] | None = None
                    if ext_filter:
                        # Fetch a bounded set of hashes to intersect with other filters.
                        ext_fetch_limit = (limit or 45) * 50
                        ext_hashes = api.get_file_hashes_by_ext(
                            ext_filter,
                            limit=ext_fetch_limit
                        )

                    # ext-only search: query is empty (or coerced to match_all above).
                    if ext_filter and (not query_lower or query_lower == "*"):
                        rows = api.get_files_by_ext(ext_filter, limit)
                        for file_hash, file_path_str, size_bytes, ext in rows:
                            if not file_path_str:
                                continue
                            file_path = search_dir / str(file_path_str)
                            if not file_path.exists():
                                continue
                            if size_bytes is None:
                                try:
                                    size_bytes = file_path.stat().st_size
                                except OSError:
                                    size_bytes = None
                            tags = api.get_tags_for_file(file_hash)
                            entry = _create_entry(
                                file_path,
                                tags,
                                size_bytes,
                                file_hash
                            )
                            try:
                                db_ext = str(ext or "").strip().lstrip(".")
                                if db_ext:
                                    entry["ext"] = db_ext
                            except Exception:
                                pass
                            results.append(entry)
                            if limit is not None and len(results) >= limit:
                                return results
                        backend_label = str(
                            getattr(self,
                                    "_name",
                                    "") or getattr(self,
                                                   "NAME",
                                                   "") or "folder"
                        )
                        debug(f"[folder:{backend_label}] {len(results)} result(s)")
                        return results

                    if tokens and len(tokens) > 1:
                        url_fetch_limit = (limit or 45) * 50

                        def _ids_for_token(token: str) -> set[int]:
                            token = token.strip()
                            if not token:
                                return set()

                            if ":" in token and not token.startswith(":"):
                                namespace, pattern = token.split(":", 1)
                                namespace = namespace.strip().lower()
                                pattern = pattern.strip().lower()

                                if namespace == "hash":
                                    normalized_hash = _normalize_hash(pattern)
                                    if not normalized_hash:
                                        return set()
                                    h = api.get_file_hash_by_hash(normalized_hash)
                                    return {h} if h else set()

                                if namespace == "url":
                                    if not pattern or pattern == "*":
                                        return api.get_file_hashes_with_any_url(
                                            limit=url_fetch_limit
                                        )
                                    return api.get_file_hashes_by_url_like(
                                        _url_like_pattern(pattern),
                                        limit=url_fetch_limit
                                    )

                                if namespace == "system":
                                    # Hydrus-compatible query: system:filetype = png
                                    m_ft = re.match(
                                        r"^filetype\s*(?:=\s*)?(.+)$",
                                        pattern
                                    )
                                    if m_ft:
                                        normalized_ext = _normalize_ext_filter(
                                            m_ft.group(1)
                                        )
                                        if not normalized_ext:
                                            return set()
                                        return api.get_file_hashes_by_ext(
                                            normalized_ext,
                                            limit=url_fetch_limit
                                        )
                                    return set()

                                if namespace in {"ext",
                                                 "extension"}:
                                    normalized_ext = _normalize_ext_filter(pattern)
                                    if not normalized_ext:
                                        return set()
                                    return api.get_file_hashes_by_ext(
                                        normalized_ext,
                                        limit=url_fetch_limit
                                    )

                                if namespace == "store":
                                    if pattern not in {"local",
                                                       "file",
                                                       "filesystem"}:
                                        return set()
                                    return api.get_all_file_hashes()

                                query_pattern = f"{namespace}:%"
                                tag_rows = api.get_file_hashes_by_tag_pattern(
                                    query_pattern
                                )
                                matched: set[str] = set()
                                for file_hash, tag_val in tag_rows:
                                    if not tag_val:
                                        continue
                                    tag_lower = str(tag_val).lower()
                                    if not tag_lower.startswith(f"{namespace}:"):
                                        continue
                                    value = _normalize_namespace_text(
                                        tag_lower[len(namespace) + 1:],
                                        allow_wildcards=False
                                    )
                                    pat = _normalize_namespace_text(
                                        pattern,
                                        allow_wildcards=True
                                    )
                                    if fnmatch(value, pat):
                                        matched.add(file_hash)
                                return matched

                            term = token.lower()
                            like_pattern = f"%{_like_pattern(term)}%"
                            # Unqualified token: match file path, title: tags, and non-namespaced tags.
                            # Do NOT match other namespaces by default (e.g., artist:men at work).
                            hashes = set(
                                api.get_file_hashes_by_path_pattern(like_pattern)
                                or set()
                            )

                            try:
                                title_rows = api.get_files_by_namespace_pattern(
                                    f"title:{like_pattern}",
                                    url_fetch_limit
                                )
                                hashes.update(
                                    {
                                        row[0]
                                        for row in (title_rows or []) if row and row[0]
                                    }
                                )
                            except Exception:
                                pass

                            try:
                                simple_rows = api.get_files_by_simple_tag_pattern(
                                    like_pattern,
                                    url_fetch_limit
                                )
                                hashes.update(
                                    {
                                        row[0]
                                        for row in (simple_rows or []) if row and row[0]
                                    }
                                )
                            except Exception:
                                pass

                            return hashes

                        try:
                            matching_hashes: set[str] | None = None
                            for token in tokens:
                                hashes = _ids_for_token(token)
                                matching_hashes = (
                                    hashes if matching_hashes is None else
                                    matching_hashes & hashes
                                )
                                if not matching_hashes:
                                    return results

                            if ext_hashes is not None:
                                matching_hashes = (
                                    matching_hashes or set()
                                ) & ext_hashes
                                if not matching_hashes:
                                    return results

                            if not matching_hashes:
                                return results

                            rows = api.get_file_metadata(matching_hashes, limit)
                            for file_hash, file_path_str, size_bytes, ext in rows:
                                if not file_path_str:
                                    continue
                                file_path = search_dir / str(file_path_str)
                                if not file_path.exists():
                                    continue
                                if size_bytes is None:
                                    try:
                                        size_bytes = file_path.stat().st_size
                                    except OSError:
                                        size_bytes = None
                                tags = api.get_tags_for_file(file_hash)
                                entry = _create_entry(
                                    file_path,
                                    tags,
                                    size_bytes,
                                    file_hash
                                )
                                try:
                                    db_ext = str(ext or "").strip().lstrip(".")
                                    if db_ext:
                                        entry["ext"] = db_ext
                                except Exception:
                                    pass
                                results.append(entry)
                                if limit is not None and len(results) >= limit:
                                    return results
                            return results
                        except Exception as exc:
                            log(f"⚠️  AND search failed: {exc}", file=sys.stderr)
                            debug(f"AND search exception details: {exc}")
                            return []

                    if ":" in query and not query.startswith(":"):
                        namespace, pattern = query.split(":", 1)
                        namespace = namespace.strip().lower()
                        pattern = pattern.strip().lower()
                        debug(f"Performing namespace search: {namespace}:{pattern}")

                        if namespace == "hash":
                            normalized_hash = _normalize_hash(pattern)
                            if not normalized_hash:
                                return results
                            h = api.get_file_hash_by_hash(normalized_hash)
                            hashes = {h} if h else set()
                            rows = api.get_file_metadata(hashes, limit)
                            for file_hash, file_path_str, size_bytes, ext in rows:
                                if not file_path_str:
                                    continue
                                file_path = search_dir / str(file_path_str)
                                if not file_path.exists():
                                    continue
                                if size_bytes is None:
                                    try:
                                        size_bytes = file_path.stat().st_size
                                    except OSError:
                                        size_bytes = None
                                tags = api.get_tags_for_file(file_hash)
                                entry = _create_entry(
                                    file_path,
                                    tags,
                                    size_bytes,
                                    file_hash
                                )
                                try:
                                    db_ext = str(ext or "").strip().lstrip(".")
                                    if db_ext:
                                        entry["ext"] = db_ext
                                except Exception:
                                    pass
                                results.append(entry)
                                if limit is not None and len(results) >= limit:
                                    return results
                            return results

                        if namespace == "url":
                            if not pattern or pattern == "*":
                                rows = api.get_files_with_any_url(limit)
                            else:
                                rows = api.get_files_by_url_like(
                                    _url_like_pattern(pattern),
                                    limit
                                )
                            for file_hash, file_path_str, size_bytes, ext in rows:
                                if not file_path_str:
                                    continue
                                file_path = search_dir / str(file_path_str)
                                if not file_path.exists():
                                    continue
                                if size_bytes is None:
                                    try:
                                        size_bytes = file_path.stat().st_size
                                    except OSError:
                                        size_bytes = None
                                tags = api.get_tags_for_file(file_hash)
                                entry = _create_entry(
                                    file_path,
                                    tags,
                                    size_bytes,
                                    file_hash
                                )
                                results.append(entry)
                                if limit is not None and len(results) >= limit:
                                    return results
                            return results

                        if namespace == "system":
                            # Hydrus-compatible query: system:filetype = png
                            m_ft = re.match(r"^filetype\s*(?:=\s*)?(.+)$", pattern)
                            if m_ft:
                                normalized_ext = _normalize_ext_filter(m_ft.group(1))
                                if not normalized_ext:
                                    return results
                                rows = api.get_files_by_ext(normalized_ext, limit)
                                for file_hash, file_path_str, size_bytes, ext in rows:
                                    if not file_path_str:
                                        continue
                                    file_path = search_dir / str(file_path_str)
                                    if not file_path.exists():
                                        continue
                                    if size_bytes is None:
                                        try:
                                            size_bytes = file_path.stat().st_size
                                        except OSError:
                                            size_bytes = None
                                    tags = api.get_tags_for_file(file_hash)
                                    entry = _create_entry(
                                        file_path,
                                        tags,
                                        size_bytes,
                                        file_hash
                                    )
                                    try:
                                        db_ext = str(ext or "").strip().lstrip(".")
                                        if db_ext:
                                            entry["ext"] = db_ext
                                    except Exception:
                                        pass
                                    results.append(entry)
                                    if limit is not None and len(results) >= limit:
                                        return results
                                return results

                        if namespace in {"ext",
                                         "extension"}:
                            normalized_ext = _normalize_ext_filter(pattern)
                            if not normalized_ext:
                                return results
                            rows = api.get_files_by_ext(normalized_ext, limit)
                            for file_hash, file_path_str, size_bytes, ext in rows:
                                if not file_path_str:
                                    continue
                                file_path = search_dir / str(file_path_str)
                                if not file_path.exists():
                                    continue
                                if size_bytes is None:
                                    try:
                                        size_bytes = file_path.stat().st_size
                                    except OSError:
                                        size_bytes = None
                                tags = api.get_tags_for_file(file_hash)
                                entry = _create_entry(
                                    file_path,
                                    tags,
                                    size_bytes,
                                    file_hash
                                )
                                try:
                                    db_ext = str(ext or "").strip().lstrip(".")
                                    if db_ext:
                                        entry["ext"] = db_ext
                                except Exception:
                                    pass
                                results.append(entry)
                                if limit is not None and len(results) >= limit:
                                    return results
                            return results

                        query_pattern = f"{namespace}:%"
                        rows = api.get_files_by_namespace_pattern(query_pattern, limit)
                        debug(f"Found {len(rows)} potential matches in DB")

                        for file_hash, file_path_str, size_bytes, ext in rows:
                            if not file_path_str:
                                continue

                            tags = api.get_tags_by_namespace_and_file(
                                file_hash,
                                query_pattern
                            )

                            for tag in tags:
                                tag_lower = tag.lower()
                                if tag_lower.startswith(f"{namespace}:"):
                                    value = _normalize_namespace_text(
                                        tag_lower[len(namespace) + 1:],
                                        allow_wildcards=False
                                    )
                                    pat = _normalize_namespace_text(
                                        pattern,
                                        allow_wildcards=True
                                    )
                                    if fnmatch(value, pat):
                                        if ext_hashes is not None and file_hash not in ext_hashes:
                                            break
                                        file_path = search_dir / str(file_path_str)
                                        if file_path.exists():
                                            if size_bytes is None:
                                                size_bytes = file_path.stat().st_size
                                            all_tags = api.get_tags_for_file(file_hash)
                                            entry = _create_entry(
                                                file_path,
                                                all_tags,
                                                size_bytes,
                                                file_hash
                                            )
                                            try:
                                                db_ext = str(ext
                                                             or "").strip().lstrip(".")
                                                if db_ext:
                                                    entry["ext"] = db_ext
                                            except Exception:
                                                pass
                                            results.append(entry)
                                        else:
                                            debug(f"File missing on disk: {file_path}")
                                        break

                            if limit is not None and len(results) >= limit:
                                return results
                    elif not match_all:
                        # Default (unqualified) search: AND semantics across terms.
                        # Each term must match at least one of:
                        # - file path (filename)
                        # - title: namespace tag
                        # - non-namespaced tag
                        # Other namespaces (artist:, series:, etc.) are excluded unless explicitly queried.
                        terms = [
                            t.strip() for t in query_lower.replace(",", " ").split()
                            if t.strip()
                        ]
                        if not terms:
                            terms = [query_lower]

                        fetch_limit = (limit or 45) * 50

                        matching_hashes: Optional[set[str]] = None
                        for term in terms:
                            if not term:
                                continue
                            like_term = _like_pattern(term)
                            like_pattern = f"%{like_term}%"

                            term_hashes: set[str] = set()
                            try:
                                term_hashes.update(
                                    api.get_file_hashes_by_path_pattern(like_pattern)
                                )
                            except Exception:
                                pass

                            try:
                                title_rows = api.get_files_by_namespace_pattern(
                                    f"title:{like_pattern}",
                                    fetch_limit
                                )
                                term_hashes.update(
                                    {
                                        row[0]
                                        for row in (title_rows or []) if row and row[0]
                                    }
                                )
                            except Exception:
                                pass

                            try:
                                simple_rows = api.get_files_by_simple_tag_pattern(
                                    like_pattern,
                                    fetch_limit
                                )
                                term_hashes.update(
                                    {
                                        row[0]
                                        for row in (simple_rows or []) if row and row[0]
                                    }
                                )
                            except Exception:
                                pass

                            if ext_hashes is not None:
                                term_hashes &= ext_hashes

                            matching_hashes = (
                                term_hashes if matching_hashes is None else
                                (matching_hashes & term_hashes)
                            )
                            if not matching_hashes:
                                return results

                        if not matching_hashes:
                            return results

                        rows = api.get_file_metadata(set(matching_hashes), limit)
                        for file_hash, file_path_str, size_bytes, ext in rows:
                            if not file_path_str:
                                continue
                            file_path = search_dir / str(file_path_str)
                            if not file_path.exists():
                                continue
                            if size_bytes is None:
                                try:
                                    size_bytes = file_path.stat().st_size
                                except OSError:
                                    size_bytes = None
                            tags = api.get_tags_for_file(file_hash)
                            entry_obj = _create_entry(
                                file_path,
                                tags,
                                size_bytes,
                                file_hash
                            )
                            try:
                                db_ext = str(ext or "").strip().lstrip(".")
                                if db_ext:
                                    entry_obj["ext"] = db_ext
                            except Exception:
                                pass
                            results.append(entry_obj)
                            if limit is not None and len(results) >= limit:
                                break

                    else:
                        rows = api.get_all_files(limit)
                        for file_hash, file_path_str, size_bytes, ext in rows:
                            if file_path_str:
                                if ext_hashes is not None and file_hash not in ext_hashes:
                                    continue
                                file_path = search_dir / str(file_path_str)
                                if file_path.exists():
                                    if size_bytes is None:
                                        size_bytes = file_path.stat().st_size

                                    tags = api.get_tags_for_file(file_hash)
                                    entry = _create_entry(
                                        file_path,
                                        tags,
                                        size_bytes,
                                        file_hash
                                    )
                                    try:
                                        db_ext = str(ext or "").strip().lstrip(".")
                                        if db_ext:
                                            entry["ext"] = db_ext
                                    except Exception:
                                        pass
                                    results.append(entry)

                    backend_label = str(
                        getattr(self,
                                "_name",
                                "") or getattr(self,
                                               "NAME",
                                               "") or "folder"
                    )
                    debug(f"[folder:{backend_label}] {len(results)} result(s)")
                    return results

            except Exception as e:
                log(f"⚠️  Database search failed: {e}", file=sys.stderr)
                debug(f"DB search exception details: {e}")
                return []

        except Exception as exc:
            log(f"❌ Local search failed: {exc}", file=sys.stderr)
            raise

    def _resolve_library_root(self,
                              file_path: Path,
                              config: Dict[str,
                                           Any]) -> Optional[Path]:
        """Return the library root containing medios-macina.db.

        Prefer the store's configured location, then config override, then walk parents
        of the file path to find a directory with medios-macina.db."""
        candidates: list[Path] = []
        if self._location:
            candidates.append(Path(self._location).expanduser())
        cfg_root = get_local_storage_path(config) if config else None
        if cfg_root:
            candidates.append(Path(cfg_root).expanduser())

        for root in candidates:
            db_path = root / "medios-macina.db"
            if db_path.exists():
                return root

        try:
            for parent in [file_path] + list(file_path.parents):
                db_path = parent / "medios-macina.db"
                if db_path.exists():
                    return parent
        except Exception:
            pass
        return None

    def get_file(self, file_hash: str, **kwargs: Any) -> Optional[Path]:
        """Retrieve file by hash, returning path to the file.

        Args:
            file_hash: SHA256 hash of the file (64-char hex string)

        Returns:
            Path to the file or None if not found
        """
        try:
            # Normalize the hash
            normalized_hash = _normalize_hash(file_hash)
            if not normalized_hash:
                return None

            search_dir = Path(self._location).expanduser()
            from API.folder import API_folder_store

            with API_folder_store(search_dir) as db:
                # Search for file by hash
                file_path = db.search_hash(normalized_hash)

                if file_path and file_path.exists():
                    return file_path

                return None

        except Exception as exc:
            debug(f"Failed to get file for hash {file_hash}: {exc}")
            return None

    def get_metadata(self, file_hash: str) -> Optional[Dict[str, Any]]:
        """Get metadata for a file from the database by hash.

        Args:
            file_hash: SHA256 hash of the file (64-char hex string)

        Returns:
            Dict with metadata fields (ext, size, hash, duration, etc.) or None if not found
        """
        try:
            # Normalize the hash
            normalized_hash = _normalize_hash(file_hash)
            if not normalized_hash:
                return None

            search_dir = Path(self._location).expanduser()
            from API.folder import DatabaseAPI

            with DatabaseAPI(search_dir) as api:
                # Get file hash
                file_hash_result = api.get_file_hash_by_hash(normalized_hash)
                if not file_hash_result:
                    return None

                # Query metadata directly from database
                cursor = api.get_cursor()
                cursor.execute(
                    """
                    SELECT * FROM metadata WHERE hash = ?
                """,
                    (file_hash_result,
                     ),
                )

                row = cursor.fetchone()
                if not row:
                    return None

                metadata = dict(row)

                # Canonicalize metadata keys (no legacy aliases)
                if "file_path" in metadata and "path" not in metadata:
                    metadata["path"] = metadata.get("file_path")
                metadata.pop("file_path", None)

                # Parse JSON fields
                for field in ["url", "relationships"]:
                    if metadata.get(field):
                        try:
                            metadata[field] = json.loads(metadata[field])
                        except (json.JSONDecodeError, TypeError):
                            metadata[field] = [] if field == "url" else []

                return metadata
        except Exception as exc:
            debug(f"Failed to get metadata for hash {file_hash}: {exc}")
            return None

    def set_relationship(self, alt_hash: str, king_hash: str, kind: str = "alt") -> bool:
        """Persist a relationship in the folder store DB.

        This is a thin wrapper around the folder DB API so cmdlets can avoid
        backend-specific branching.
        """
        try:
            if not self._location:
                return False

            alt_norm = _normalize_hash(alt_hash)
            king_norm = _normalize_hash(king_hash)
            if not alt_norm or not king_norm or alt_norm == king_norm:
                return False

            from API.folder import API_folder_store

            with API_folder_store(Path(self._location).expanduser()) as db:
                db.set_relationship_by_hash(
                    alt_norm,
                    king_norm,
                    str(kind or "alt"),
                    bidirectional=False,
                )
            return True
        except Exception:
            return False

    def get_tag(self, file_identifier: str, **kwargs: Any) -> Tuple[List[str], str]:
        """Get tags for a local file by hash.

        Returns:
            Tuple of (tags_list, store_name) where store_name is the actual store name
        """
        from API.folder import API_folder_store

        try:
            file_hash = file_identifier
            if self._location:
                try:
                    with API_folder_store(Path(self._location)) as db:
                        db_tags = db.get_tags(file_hash)
                        if db_tags:
                            # Return actual store name instead of generic "local_db"
                            store_name = self._name if self._name else "local"
                            return [
                                str(t).strip().lower()
                                for t in db_tags
                                if isinstance(t, str) and t.strip()
                            ], store_name
                except Exception as exc:
                    debug(f"Local DB lookup failed: {exc}")
            return [], "unknown"
        except Exception as exc:
            debug(f"get_tags failed for local file: {exc}")
            return [], "unknown"

    def add_tag(self, hash: str, tag: List[str], **kwargs: Any) -> bool:
        """Add tags to a local file by hash (via API_folder_store).

        Handles namespace collapsing: when adding namespace:value, removes existing namespace:* tags.
        Returns True if tags were successfully added.
        """
        from API.folder import API_folder_store

        try:
            if not self._location:
                return False

            try:
                with API_folder_store(Path(self._location)) as db:
                    existing_tags = [
                        t for t in (db.get_tags(hash) or [])
                        if isinstance(t, str) and t.strip()
                    ]

                    from SYS.metadata import compute_namespaced_tag_overwrite

                    _to_remove, _to_add, merged = compute_namespaced_tag_overwrite(
                        existing_tags, tag or []
                    )
                    if not _to_remove and not _to_add:
                        return True

                    # Folder DB tag table is case-sensitive and add_tags_to_hash() is additive.
                    # To enforce lowercase-only tags and namespace overwrites, rewrite the full tag set.
                    cursor = db.connection.cursor()
                    cursor.execute("DELETE FROM tag WHERE hash = ?",
                                   (hash,
                                    ))
                    for t in merged:
                        t = str(t).strip().lower()
                        if t:
                            cursor.execute(
                                "INSERT OR IGNORE INTO tag (hash, tag) VALUES (?, ?)",
                                (hash,
                                 t),
                            )
                    db.connection.commit()
                    try:
                        db._update_metadata_modified_time(hash)
                    except Exception:
                        pass
                    return True
            except Exception as exc:
                debug(f"Local DB add_tags failed: {exc}")
                return False
        except Exception as exc:
            debug(f"add_tag failed for local file: {exc}")
            return False

    def delete_tag(self, file_identifier: str, tags: List[str], **kwargs: Any) -> bool:
        """Remove tags from a local file by hash."""
        from API.folder import API_folder_store

        try:
            file_hash = file_identifier
            if self._location:
                try:
                    with API_folder_store(Path(self._location)) as db:
                        tag_list = [
                            str(t).strip().lower() for t in (tags or [])
                            if isinstance(t, str) and str(t).strip()
                        ]
                        if not tag_list:
                            return True
                        db.remove_tags_from_hash(file_hash, tag_list)
                        return True
                except Exception as exc:
                    debug(f"Local DB remove_tags failed: {exc}")
            return False
        except Exception as exc:
            debug(f"delete_tag failed for local file: {exc}")
            return False

    def get_url(self, file_identifier: str, **kwargs: Any) -> List[str]:
        """Get known url for a local file by hash."""
        from API.folder import API_folder_store

        try:
            file_hash = file_identifier
            if self._location:
                try:
                    from SYS.metadata import normalize_urls

                    with API_folder_store(Path(self._location)) as db:
                        meta = db.get_metadata(file_hash) or {}
                        urls = normalize_urls(meta.get("url"))
                        return urls
                except Exception as exc:
                    debug(f"Local DB get_metadata failed: {exc}")
            return []
        except Exception as exc:
            debug(f"get_url failed for local file: {exc}")
            return []

    def add_url(self, file_identifier: str, url: List[str], **kwargs: Any) -> bool:
        """Add known url to a local file by hash."""
        from API.folder import API_folder_store

        try:
            file_hash = file_identifier
            if self._location:
                try:
                    from SYS.metadata import normalize_urls

                    with API_folder_store(Path(self._location)) as db:
                        meta = db.get_metadata(file_hash) or {}
                        existing_urls = normalize_urls(meta.get("url"))
                        incoming_urls = normalize_urls(url)
                        changed = False
                        for u in list(incoming_urls or []):
                            if not u:
                                continue
                            if u not in existing_urls:
                                existing_urls.append(u)
                                changed = True
                        if changed:
                            db.update_metadata_by_hash(
                                file_hash,
                                {
                                    "url": existing_urls
                                }
                            )
                            return True
                except Exception as exc:
                    debug(f"Local DB add_url failed: {exc}")
            return False
        except Exception as exc:
            debug(f"add_url failed for local file: {exc}")
            return False

    def add_url_bulk(self, items: List[tuple[str, List[str]]], **kwargs: Any) -> bool:
        """Add known urls to many local files in one DB session.

        This is a performance optimization used by cmdlets that receive many PipeObjects.
        """
        from API.folder import API_folder_store

        try:
            if not self._location:
                return False

            # Normalize + coalesce duplicates per hash.
            try:
                from SYS.metadata import normalize_urls
            except Exception:
                normalize_urls = None  # type: ignore

            merged_by_hash: Dict[str,
                                 List[str]] = {}
            for file_identifier, url_list in items or []:
                file_hash = str(file_identifier or "").strip().lower()
                if not file_hash:
                    continue

                incoming: List[str]
                if normalize_urls is not None:
                    try:
                        incoming = normalize_urls(url_list)
                    except Exception:
                        incoming = [
                            str(u).strip() for u in (url_list or []) if str(u).strip()
                        ]
                else:
                    incoming = [
                        str(u).strip() for u in (url_list or []) if str(u).strip()
                    ]

                if not incoming:
                    continue

                existing = merged_by_hash.get(file_hash) or []
                for u in incoming:
                    if u and u not in existing:
                        existing.append(u)
                merged_by_hash[file_hash] = existing

            if not merged_by_hash:
                return True

            import json

            with API_folder_store(Path(self._location)) as db:
                conn = getattr(db, "connection", None)
                if conn is None:
                    return False
                cursor = conn.cursor()

                # Ensure metadata rows exist (may be needed for older entries).
                for file_hash in merged_by_hash.keys():
                    try:
                        cursor.execute(
                            "INSERT OR IGNORE INTO metadata (hash) VALUES (?)",
                            (file_hash,
                             )
                        )
                    except Exception:
                        continue

                # Load existing urls for all hashes in chunks.
                existing_urls_by_hash: Dict[str,
                                            List[str]] = {
                                                h: []
                                                for h in merged_by_hash.keys()
                                            }
                hashes = list(merged_by_hash.keys())
                chunk_size = 400
                for i in range(0, len(hashes), chunk_size):
                    chunk = hashes[i:i + chunk_size]
                    if not chunk:
                        continue
                    placeholders = ",".join(["?"] * len(chunk))
                    try:
                        cursor.execute(
                            f"SELECT hash, url FROM metadata WHERE hash IN ({placeholders})",
                            chunk
                        )
                        rows = cursor.fetchall() or []
                    except Exception:
                        rows = []

                    for row in rows:
                        try:
                            row_hash = str(row[0]).strip().lower()
                        except Exception:
                            continue
                        raw_urls = None
                        try:
                            raw_urls = row[1]
                        except Exception:
                            raw_urls = None

                        parsed_urls: List[str] = []
                        if raw_urls:
                            try:
                                parsed = json.loads(raw_urls)
                                if normalize_urls is not None:
                                    parsed_urls = normalize_urls(parsed)
                                else:
                                    if isinstance(parsed, list):
                                        parsed_urls = [
                                            str(u).strip() for u in parsed
                                            if str(u).strip()
                                        ]
                            except Exception:
                                parsed_urls = []

                        existing_urls_by_hash[row_hash] = parsed_urls

                # Compute updates and write in one commit.
                updates: List[tuple[str, str]] = []
                for file_hash, incoming_urls in merged_by_hash.items():
                    existing_urls = existing_urls_by_hash.get(file_hash) or []
                    final = list(existing_urls)
                    for u in incoming_urls:
                        if u and u not in final:
                            final.append(u)
                    if final != existing_urls:
                        try:
                            updates.append((json.dumps(final), file_hash))
                        except Exception:
                            continue

                if updates:
                    cursor.executemany(
                        "UPDATE metadata SET url = ?, time_modified = CURRENT_TIMESTAMP, updated_at = CURRENT_TIMESTAMP WHERE hash = ?",
                        updates,
                    )

                conn.commit()
                return True
        except Exception as exc:
            debug(f"add_url_bulk failed for local file: {exc}")
            return False

    def delete_url(self, file_identifier: str, url: List[str], **kwargs: Any) -> bool:
        """Delete known url from a local file by hash."""
        from API.folder import API_folder_store

        try:
            file_hash = file_identifier
            if self._location:
                try:
                    from SYS.metadata import normalize_urls

                    with API_folder_store(Path(self._location)) as db:
                        meta = db.get_metadata(file_hash) or {}
                        existing_urls = normalize_urls(meta.get("url"))
                        remove_set = {u
                                      for u in normalize_urls(url) if u}
                        if not remove_set:
                            return False
                        new_urls = [u for u in existing_urls if u not in remove_set]
                        if new_urls != existing_urls:
                            db.update_metadata_by_hash(file_hash,
                                                       {
                                                           "url": new_urls
                                                       })
                            return True
                except Exception as exc:
                    debug(f"Local DB delete_url failed: {exc}")
            return False
        except Exception as exc:
            debug(f"delete_url failed for local file: {exc}")
            return False

    def delete_url_bulk(
        self,
        items: List[tuple[str,
                          List[str]]],
        **kwargs: Any
    ) -> bool:
        """Delete known urls from many local files in one DB session."""
        from API.folder import API_folder_store

        try:
            if not self._location:
                return False

            try:
                from SYS.metadata import normalize_urls
            except Exception:
                normalize_urls = None  # type: ignore

            remove_by_hash: Dict[str,
                                 set[str]] = {}
            for file_identifier, url_list in items or []:
                file_hash = str(file_identifier or "").strip().lower()
                if not file_hash:
                    continue

                incoming: List[str]
                if normalize_urls is not None:
                    try:
                        incoming = normalize_urls(url_list)
                    except Exception:
                        incoming = [
                            str(u).strip() for u in (url_list or []) if str(u).strip()
                        ]
                else:
                    incoming = [
                        str(u).strip() for u in (url_list or []) if str(u).strip()
                    ]

                remove = {u
                          for u in incoming if u}
                if not remove:
                    continue
                remove_by_hash.setdefault(file_hash, set()).update(remove)

            if not remove_by_hash:
                return True

            import json

            with API_folder_store(Path(self._location)) as db:
                conn = getattr(db, "connection", None)
                if conn is None:
                    return False
                cursor = conn.cursor()

                # Ensure metadata rows exist.
                for file_hash in remove_by_hash.keys():
                    try:
                        cursor.execute(
                            "INSERT OR IGNORE INTO metadata (hash) VALUES (?)",
                            (file_hash,
                             )
                        )
                    except Exception:
                        continue

                # Load existing urls for hashes in chunks.
                existing_urls_by_hash: Dict[str,
                                            List[str]] = {
                                                h: []
                                                for h in remove_by_hash.keys()
                                            }
                hashes = list(remove_by_hash.keys())
                chunk_size = 400
                for i in range(0, len(hashes), chunk_size):
                    chunk = hashes[i:i + chunk_size]
                    if not chunk:
                        continue
                    placeholders = ",".join(["?"] * len(chunk))
                    try:
                        cursor.execute(
                            f"SELECT hash, url FROM metadata WHERE hash IN ({placeholders})",
                            chunk
                        )
                        rows = cursor.fetchall() or []
                    except Exception:
                        rows = []

                    for row in rows:
                        try:
                            row_hash = str(row[0]).strip().lower()
                        except Exception:
                            continue
                        raw_urls = None
                        try:
                            raw_urls = row[1]
                        except Exception:
                            raw_urls = None

                        parsed_urls: List[str] = []
                        if raw_urls:
                            try:
                                parsed = json.loads(raw_urls)
                                if normalize_urls is not None:
                                    parsed_urls = normalize_urls(parsed)
                                else:
                                    if isinstance(parsed, list):
                                        parsed_urls = [
                                            str(u).strip() for u in parsed
                                            if str(u).strip()
                                        ]
                            except Exception:
                                parsed_urls = []

                        existing_urls_by_hash[row_hash] = parsed_urls

                # Apply removals + write updates.
                updates: List[tuple[str, str]] = []
                for file_hash, remove_set in remove_by_hash.items():
                    existing_urls = existing_urls_by_hash.get(file_hash) or []
                    new_urls = [u for u in existing_urls if u not in remove_set]
                    if new_urls != existing_urls:
                        try:
                            updates.append((json.dumps(new_urls), file_hash))
                        except Exception:
                            continue

                if updates:
                    cursor.executemany(
                        "UPDATE metadata SET url = ?, time_modified = CURRENT_TIMESTAMP, updated_at = CURRENT_TIMESTAMP WHERE hash = ?",
                        updates,
                    )

                conn.commit()
                return True
        except Exception as exc:
            debug(f"delete_url_bulk failed for local file: {exc}")
            return False

    def get_note(self, file_identifier: str, **kwargs: Any) -> Dict[str, str]:
        """Get notes for a local file by hash."""
        from API.folder import API_folder_store

        try:
            if not self._location:
                return {}
            file_hash = str(file_identifier or "").strip().lower()
            if not _normalize_hash(file_hash):
                return {}
            with API_folder_store(Path(self._location)) as db:
                getter = getattr(db, "get_notes", None)
                if callable(getter):
                    notes = getter(file_hash)
                    return notes if isinstance(notes,
                                               dict) else {}
                # Fallback: default-only
                note = db.get_note(file_hash)
                return {
                    "default": str(note or "")
                } if note else {}
        except Exception as exc:
            debug(f"get_note failed for local file: {exc}")
            return {}

    def set_note(
        self,
        file_identifier: str,
        name: str,
        text: str,
        **kwargs: Any
    ) -> bool:
        """Set a named note for a local file by hash."""
        from API.folder import API_folder_store

        try:
            if not self._location:
                return False
            file_hash = str(file_identifier or "").strip().lower()
            if not _normalize_hash(file_hash):
                return False

            file_path = self.get_file(file_hash, **kwargs)
            if not file_path or not isinstance(file_path,
                                               Path) or not file_path.exists():
                return False

            with API_folder_store(Path(self._location)) as db:
                setter = getattr(db, "set_note", None)
                if callable(setter):
                    setter(file_path, str(name), str(text))
                    return True
                db.save_note(file_path, str(text))
                return True
        except Exception as exc:
            debug(f"set_note failed for local file: {exc}")
            return False

    def set_note_bulk(self, items: List[tuple[str, str, str]], **kwargs: Any) -> bool:
        """Set notes for many local files in one DB session.

        Preserves existing semantics by only setting notes for hashes that still
        map to a file path that exists on disk.
        """
        from API.folder import API_folder_store

        try:
            if not self._location:
                return False

            # Normalize input.
            normalized: List[tuple[str, str, str]] = []
            for file_identifier, name, text in items or []:
                file_hash = str(file_identifier or "").strip().lower()
                note_name = str(name or "").strip()
                note_text = str(text or "")
                if not file_hash or not _normalize_hash(file_hash) or not note_name:
                    continue
                normalized.append((file_hash, note_name, note_text))

            if not normalized:
                return True

            with API_folder_store(Path(self._location)) as db:
                conn = getattr(db, "connection", None)
                if conn is None:
                    return False
                cursor = conn.cursor()

                # Look up file paths for hashes in chunks (to verify existence).
                wanted_hashes = sorted({h
                                        for (h, _n, _t) in normalized})
                hash_to_path: Dict[str,
                                   str] = {}
                chunk_size = 400
                for i in range(0, len(wanted_hashes), chunk_size):
                    chunk = wanted_hashes[i:i + chunk_size]
                    if not chunk:
                        continue
                    placeholders = ",".join(["?"] * len(chunk))
                    try:
                        cursor.execute(
                            f"SELECT hash, file_path FROM file WHERE hash IN ({placeholders})",
                            chunk,
                        )
                        rows = cursor.fetchall() or []
                    except Exception:
                        rows = []
                    for row in rows:
                        try:
                            h = str(row[0]).strip().lower()
                            p = str(row[1]).strip()
                        except Exception:
                            continue
                        if h and p:
                            hash_to_path[h] = p

                # Ensure notes rows exist and only write for existing files.
                inserts: List[tuple[str, str, str]] = []
                for h, note_name, note_text in normalized:
                    p = hash_to_path.get(h)
                    if not p:
                        continue
                    try:
                        if not (Path(self._location) / p).exists():
                            continue
                    except Exception:
                        continue
                    inserts.append((h, note_name, note_text))

                if not inserts:
                    return False

                # Prefer upsert when supported, else fall back to INSERT OR REPLACE.
                try:
                    cursor.executemany(
                        "INSERT INTO note (hash, name, note) VALUES (?, ?, ?) "
                        "ON CONFLICT(hash, name) DO UPDATE SET note = excluded.note, updated_at = CURRENT_TIMESTAMP",
                        inserts,
                    )
                except Exception:
                    cursor.executemany(
                        "INSERT OR REPLACE INTO note (hash, name, note) VALUES (?, ?, ?)",
                        inserts,
                    )

                conn.commit()
                return True
        except Exception as exc:
            debug(f"set_note_bulk failed for local file: {exc}")
            return False

    def delete_note(self, file_identifier: str, name: str, **kwargs: Any) -> bool:
        """Delete a named note for a local file by hash."""
        from API.folder import API_folder_store

        try:
            if not self._location:
                return False
            file_hash = str(file_identifier or "").strip().lower()
            if not _normalize_hash(file_hash):
                return False
            with API_folder_store(Path(self._location)) as db:
                deleter = getattr(db, "delete_note", None)
                if callable(deleter):
                    deleter(file_hash, str(name))
                    return True
                # Default-only fallback
                if str(name).strip().lower() == "default":
                    deleter2 = getattr(db, "save_note", None)
                    if callable(deleter2):
                        file_path = self.get_file(file_hash, **kwargs)
                        if file_path and isinstance(file_path,
                                                    Path) and file_path.exists():
                            deleter2(file_path, "")
                            return True
            return False
        except Exception as exc:
            debug(f"delete_note failed for local file: {exc}")
            return False

    def delete_file(self, file_identifier: str, **kwargs: Any) -> bool:
        """Delete a file from the folder store.

        Args:
            file_identifier: The file path (as string) or hash of the file to delete
            **kwargs: Optional parameters

        Returns:
            True if deletion succeeded, False otherwise
        """
        from API.folder import API_folder_store

        try:
            if not self._location:
                return False

            raw = str(file_identifier or "").strip()
            if not raw:
                return False

            store_root = Path(self._location).expanduser()

            # Support deletion by hash (common for store items where `path` is the hash).
            file_hash = _normalize_hash(raw)
            resolved_path: Optional[Path] = None
            with API_folder_store(store_root) as db:
                if file_hash:
                    resolved_path = db.search_hash(file_hash)
                else:
                    p = Path(raw)
                    resolved_path = p if p.is_absolute() else (store_root / p)

                if resolved_path is None:
                    debug(f"delete_file: could not resolve identifier: {raw}")
                    return False

                # Delete from database (also cleans up relationship backlinks).
                db.delete_file(resolved_path)

            # Delete the actual file from disk (best-effort).
            try:
                if resolved_path.exists():
                    resolved_path.unlink()
                    debug(f"Deleted file: {resolved_path}")
                else:
                    debug(f"File not found on disk: {resolved_path}")
            except Exception:
                pass

            return True
        except Exception as exc:
            debug(f"delete_file failed: {exc}")
            return False