This commit is contained in:
nose
2025-12-14 00:53:52 -08:00
parent 52a79b0086
commit a03eb0d1be
24 changed files with 2785 additions and 1868 deletions

View File

@@ -3,7 +3,6 @@ from __future__ import annotations
from typing import Any, Dict, Sequence, List, Optional, Tuple
from pathlib import Path
from dataclasses import dataclass, field
from collections import OrderedDict
import re
import json
@@ -11,57 +10,9 @@ import sys
from SYS.logger import log, debug
from ._shared import Cmdlet, CmdletArg, get_field, should_show_help
from ._shared import Cmdlet, CmdletArg, get_field, should_show_help, normalize_hash, first_title_tag
import pipeline as ctx
# Optional dependencies
try:
import mutagen # type: ignore
except ImportError: # pragma: no cover
mutagen = None # type: ignore
try:
from config import get_hydrus_url, resolve_output_dir
except Exception: # pragma: no cover
get_hydrus_url = None # type: ignore
resolve_output_dir = None # type: ignore
try:
from API.HydrusNetwork import HydrusNetwork, HydrusRequestError
except ImportError: # pragma: no cover
HydrusNetwork = None # type: ignore
HydrusRequestError = RuntimeError # type: ignore
try:
from SYS.utils import sha256_file
except ImportError: # pragma: no cover
sha256_file = None # type: ignore
try:
from SYS.utils_constant import mime_maps
except ImportError: # pragma: no cover
mime_maps = {} # type: ignore
@dataclass(slots=True)
class SearchRecord:
path: str
size_bytes: int | None = None
duration_seconds: str | None = None
tag: str | None = None
hash: str | None = None
def as_dict(self) -> dict[str, str]:
payload: dict[str, str] = {"path": self.path}
if self.size_bytes is not None:
payload["size"] = str(self.size_bytes)
if self.duration_seconds:
payload["duration"] = self.duration_seconds
if self.tag:
payload["tag"] = self.tag
if self.hash:
payload["hash"] = self.hash
return payload
STORAGE_ORIGINS = {"local", "hydrus", "folder"}
@@ -86,12 +37,15 @@ class Search_Store(Cmdlet):
detail=[
"Search across storage backends: Folder stores and Hydrus instances",
"Use -store to search a specific backend by name",
"URL search: url:* (any URL) or url:<value> (URL substring)",
"Filter results by: tag, size, type, duration",
"Results include hash for downstream commands (get-file, add-tag, etc.)",
"Examples:",
"search-store foo # Search all storage backends",
"search-store -store home '*' # Search 'home' Hydrus instance",
"search-store -store test 'video' # Search 'test' folder store",
"search-store 'url:*' # Files that have any URL",
"search-store 'url:youtube.com' # Files whose URL contains substring",
"search-store song -type audio # Search for audio files",
"search-store movie -tag action # Search with tag filter",
],
@@ -100,6 +54,40 @@ class Search_Store(Cmdlet):
self.register()
# --- Helper methods -------------------------------------------------
@staticmethod
def _parse_hash_query(query: str) -> List[str]:
"""Parse a `hash:` query into a list of normalized 64-hex SHA256 hashes.
Supported examples:
- hash:<h1>,<h2>,<h3>
- Hash: <h1> <h2> <h3>
- hash:{<h1>, <h2>}
"""
q = str(query or "").strip()
if not q:
return []
m = re.match(r"^hash(?:es)?\s*:\s*(.+)$", q, flags=re.IGNORECASE)
if not m:
return []
rest = (m.group(1) or "").strip()
if rest.startswith("{") and rest.endswith("}"):
rest = rest[1:-1].strip()
if rest.startswith("[") and rest.endswith("]"):
rest = rest[1:-1].strip()
# Split on commas and whitespace.
raw_parts = [p.strip() for p in re.split(r"[\s,]+", rest) if p.strip()]
out: List[str] = []
for part in raw_parts:
h = normalize_hash(part)
if not h:
continue
if h not in out:
out.append(h)
return out
@staticmethod
def _normalize_extension(ext_value: Any) -> str:
"""Sanitize extension strings to alphanumerics and cap at 5 chars."""
@@ -150,10 +138,10 @@ class Search_Store(Cmdlet):
# Parse arguments
query = ""
tag_filters: List[str] = []
size_filter: Optional[Tuple[str, int]] = None
duration_filter: Optional[Tuple[str, float]] = None
type_filter: Optional[str] = None
_tag_filters: List[str] = []
_size_filter: Optional[Tuple[str, int]] = None
_duration_filter: Optional[Tuple[str, float]] = None
_type_filter: Optional[str] = None
storage_backend: Optional[str] = None
limit = 100
searched_backends: List[str] = []
@@ -166,7 +154,7 @@ class Search_Store(Cmdlet):
storage_backend = args_list[i + 1]
i += 2
elif low in {"-tag", "--tag"} and i + 1 < len(args_list):
tag_filters.append(args_list[i + 1])
_tag_filters.append(args_list[i + 1])
i += 2
elif low in {"-limit", "--limit"} and i + 1 < len(args_list):
try:
@@ -175,7 +163,7 @@ class Search_Store(Cmdlet):
limit = 100
i += 2
elif low in {"-type", "--type"} and i + 1 < len(args_list):
type_filter = args_list[i + 1].lower()
_type_filter = args_list[i + 1].lower()
i += 2
elif not arg.startswith("-"):
query = f"{query} {arg}".strip() if query else arg
@@ -195,6 +183,8 @@ class Search_Store(Cmdlet):
if store_filter and not storage_backend:
storage_backend = store_filter
hash_query = self._parse_hash_query(query)
if not query:
log("Provide a search query", file=sys.stderr)
return 1
@@ -230,12 +220,136 @@ class Search_Store(Cmdlet):
table_title += f" [{storage_backend}]"
table = ResultTable(table_title)
if hash_query:
try:
table.set_preserve_order(True)
except Exception:
pass
from Store import Store
storage = Store(config=config or {})
from Store._base import Store as BaseStore
backend_to_search = storage_backend or None
if hash_query:
# Explicit hash list search: build rows from backend metadata.
backends_to_try: List[str] = []
if backend_to_search:
backends_to_try = [backend_to_search]
else:
backends_to_try = list(storage.list_backends())
found_any = False
for h in hash_query:
resolved_backend_name: Optional[str] = None
resolved_backend = None
for backend_name in backends_to_try:
try:
backend = storage[backend_name]
except Exception:
continue
try:
# If get_metadata works, consider it a hit; get_file can be optional (e.g. remote URL).
meta = backend.get_metadata(h)
if meta is None:
continue
resolved_backend_name = backend_name
resolved_backend = backend
break
except Exception:
continue
if resolved_backend_name is None or resolved_backend is None:
continue
found_any = True
searched_backends.append(resolved_backend_name)
# Resolve a path/URL string if possible
path_str: Optional[str] = None
try:
maybe_path = resolved_backend.get_file(h)
if isinstance(maybe_path, Path):
path_str = str(maybe_path)
elif isinstance(maybe_path, str) and maybe_path:
path_str = maybe_path
except Exception:
path_str = None
meta_obj: Dict[str, Any] = {}
try:
meta_obj = resolved_backend.get_metadata(h) or {}
except Exception:
meta_obj = {}
tags_list: List[str] = []
try:
tag_result = resolved_backend.get_tag(h)
if isinstance(tag_result, tuple) and tag_result:
maybe_tags = tag_result[0]
else:
maybe_tags = tag_result
if isinstance(maybe_tags, list):
tags_list = [str(t).strip() for t in maybe_tags if isinstance(t, str) and str(t).strip()]
except Exception:
tags_list = []
title_from_tag: Optional[str] = None
try:
title_tag = first_title_tag(tags_list)
if title_tag and ":" in title_tag:
title_from_tag = title_tag.split(":", 1)[1].strip()
except Exception:
title_from_tag = None
title = title_from_tag or meta_obj.get("title") or meta_obj.get("name")
if not title and path_str:
try:
title = Path(path_str).stem
except Exception:
title = path_str
ext_val = meta_obj.get("ext") or meta_obj.get("extension")
if not ext_val and path_str:
try:
ext_val = Path(path_str).suffix
except Exception:
ext_val = None
size_bytes = meta_obj.get("size")
if size_bytes is None:
size_bytes = meta_obj.get("size_bytes")
try:
size_bytes_int: Optional[int] = int(size_bytes) if size_bytes is not None else None
except Exception:
size_bytes_int = None
payload: Dict[str, Any] = {
"title": str(title or h),
"hash": h,
"store": resolved_backend_name,
"path": path_str,
"ext": self._normalize_extension(ext_val),
"size_bytes": size_bytes_int,
"tag": tags_list,
}
table.add_result(payload)
results_list.append(payload)
ctx.emit(payload)
if found_any:
ctx.set_last_result_table(table, results_list)
db.append_worker_stdout(worker_id, json.dumps(results_list, indent=2))
db.update_worker_status(worker_id, 'completed')
return 0
log("No results found", file=sys.stderr)
db.append_worker_stdout(worker_id, json.dumps([], indent=2))
db.update_worker_status(worker_id, 'completed')
return 0
if backend_to_search:
searched_backends.append(backend_to_search)
target_backend = storage[backend_to_search]
@@ -243,7 +357,9 @@ class Search_Store(Cmdlet):
log(f"Backend '{backend_to_search}' does not support searching", file=sys.stderr)
db.update_worker_status(worker_id, 'error')
return 1
debug(f"[search-store] Searching '{backend_to_search}'")
results = target_backend.search(query, limit=limit)
debug(f"[search-store] '{backend_to_search}' -> {len(results or [])} result(s)")
else:
from API.HydrusNetwork import is_hydrus_available
hydrus_available = is_hydrus_available(config or {})
@@ -257,7 +373,9 @@ class Search_Store(Cmdlet):
continue
searched_backends.append(backend_name)
debug(f"[search-store] Searching '{backend_name}'")
backend_results = backend.search(query, limit=limit - len(all_results))
debug(f"[search-store] '{backend_name}' -> {len(backend_results or [])} result(s)")
if backend_results:
all_results.extend(backend_results)
if len(all_results) >= limit:
@@ -317,11 +435,6 @@ class Search_Store(Cmdlet):
results_list.append(normalized)
ctx.emit(normalized)
# Debug: Verify table rows match items list
debug(f"[search-store] Added {len(table.rows)} rows to table, {len(results_list)} items to results_list")
if len(table.rows) != len(results_list):
debug(f"[search-store] WARNING: Table/items mismatch! rows={len(table.rows)} items={len(results_list)}", file=sys.stderr)
ctx.set_last_result_table(table, results_list)
db.append_worker_stdout(worker_id, json.dumps(results_list, indent=2))
else: