This commit is contained in:
2026-01-05 07:51:19 -08:00
parent 8545367e28
commit 1f765cffda
32 changed files with 3447 additions and 3250 deletions

View File

@@ -244,7 +244,7 @@ class SharedArgs:
description="Destination location",
)
DELETE_FLAG = CmdletArg(
DELETE = CmdletArg(
"delete",
type="flag",
description="Delete the file and its .tag after successful operation.",
@@ -2081,6 +2081,12 @@ def extract_url_from_result(result: Any) -> list[str]:
_extend(result.metadata.get("url"))
_extend(result.metadata.get("url"))
_extend(result.metadata.get("url"))
if isinstance(getattr(result, "full_metadata", None), dict):
fm = getattr(result, "full_metadata", None)
if isinstance(fm, dict):
_extend(fm.get("url"))
_extend(fm.get("url"))
_extend(fm.get("url"))
elif hasattr(result, "url") or hasattr(result, "url"):
# Handle objects with url/url attribute
_extend(getattr(result, "url", None))
@@ -2090,6 +2096,11 @@ def extract_url_from_result(result: Any) -> list[str]:
_extend(result.get("url"))
_extend(result.get("url"))
_extend(result.get("url"))
fm = result.get("full_metadata")
if isinstance(fm, dict):
_extend(fm.get("url"))
_extend(fm.get("url"))
_extend(fm.get("url"))
extra = result.get("extra")
if isinstance(extra, dict):
_extend(extra.get("url"))
@@ -2531,6 +2542,30 @@ def resolve_tidal_manifest_path(item: Any) -> Optional[str]:
metadata["_tidal_track_details_fetched"] = True
except Exception:
pass
if not metadata.get("url"):
try:
resp_info = httpx.get(
"https://tidal-api.binimum.org/info/",
params={"id": str(track_int)},
timeout=10.0,
)
resp_info.raise_for_status()
info_payload = resp_info.json()
info_data = info_payload.get("data") if isinstance(info_payload, dict) else None
if isinstance(info_data, dict) and info_data:
try:
for k, v in info_data.items():
if k not in metadata:
metadata[k] = v
except Exception:
pass
try:
if info_data.get("url"):
metadata["url"] = info_data.get("url")
except Exception:
pass
except Exception:
pass
except Exception:
pass

View File

@@ -345,6 +345,14 @@ class Add_File(Cmdlet):
else:
items_to_process = [result]
total_items = len(items_to_process) if isinstance(items_to_process, list) else 0
processed_items = 0
try:
if total_items:
progress.set_percent(0)
except Exception:
pass
# Minimal step-based progress for single-item runs.
# Many add-file flows don't emit intermediate items, so without steps the pipe can look "stuck".
use_steps = False
@@ -496,9 +504,25 @@ class Add_File(Cmdlet):
and len(items_to_process) > 1
)
for item in items_to_process:
for idx, item in enumerate(items_to_process, 1):
pipe_obj = coerce_to_pipe_object(item, path_arg)
try:
label = pipe_obj.title or pipe_obj.name
if not label and pipe_obj.path:
try:
label = Path(str(pipe_obj.path)).name
except Exception:
label = pipe_obj.path
if not label:
label = "file"
if total_items:
pending_pct = int(round(((idx - 1) / max(1, total_items)) * 100))
progress.set_percent(pending_pct)
progress.set_status(f"adding {idx}/{total_items}: {label}")
except Exception:
pass
temp_dir_to_cleanup: Optional[Path] = None
delete_after_item = delete_after
try:
@@ -597,6 +621,14 @@ class Add_File(Cmdlet):
shutil.rmtree(temp_dir_to_cleanup, ignore_errors=True)
except Exception:
pass
processed_items += 1
try:
pct = int(round((processed_items / max(1, total_items)) * 100))
progress.set_percent(pct)
if processed_items >= total_items:
progress.clear_status()
except Exception:
pass
# Apply deferred url associations (bulk) before showing the final store table.
if pending_url_associations:

289
cmdlet/convert_file.py Normal file
View File

@@ -0,0 +1,289 @@
from __future__ import annotations
from typing import Any, Dict, Sequence, Optional
from pathlib import Path
import sys
import shutil
import subprocess
from SYS.logger import log, debug
from SYS.utils import sha256_file
from . import _shared as sh
from SYS import pipeline as ctx
Cmdlet = sh.Cmdlet
CmdletArg = sh.CmdletArg
QueryArg = sh.QueryArg
SharedArgs = sh.SharedArgs
parse_cmdlet_args = sh.parse_cmdlet_args
normalize_result_input = sh.normalize_result_input
extract_title_from_result = sh.extract_title_from_result
VIDEO_EXTS = {
"mp4",
"mkv",
"webm",
"mov",
"avi",
"flv",
"mpeg",
"mpg",
"m4v",
}
AUDIO_EXTS = {
"mp3",
"m4a",
"m4b",
"aac",
"flac",
"wav",
"ogg",
"opus",
"mka",
}
IMAGE_EXTS = {
"png",
"jpg",
"jpeg",
"webp",
"bmp",
"tif",
"tiff",
"gif",
}
DOC_EXTS = {
"pdf",
"mobi",
"epub",
"azw3",
"txt",
"rtf",
"html",
"htm",
"md",
"doc",
"docx",
}
def _detect_kind(ext: str) -> str:
e = ext.lower().lstrip(".")
if e in VIDEO_EXTS:
return "video"
if e in AUDIO_EXTS:
return "audio"
if e in IMAGE_EXTS:
return "image"
if e in DOC_EXTS:
return "doc"
return "unknown"
def _allowed(source_kind: str, target_kind: str) -> bool:
if source_kind == target_kind:
return True
if source_kind == "video" and target_kind == "audio":
return True
return False
def _ffmpeg_convert(
input_path: Path,
output_path: Path,
target_kind: str,
copy_metadata: bool,
) -> bool:
ffmpeg_path = shutil.which("ffmpeg")
if not ffmpeg_path:
log("ffmpeg not found in PATH", file=sys.stderr)
return False
cmd = [ffmpeg_path, "-y", "-i", str(input_path)]
if target_kind == "audio":
cmd.extend(["-vn"])
if copy_metadata:
cmd.extend(["-map_metadata", "0"])
cmd.append(str(output_path))
debug(f"[convert-file] Running ffmpeg: {' '.join(cmd)}")
proc = subprocess.run(cmd, capture_output=True, text=True)
if proc.returncode != 0:
log(f"ffmpeg error: {proc.stderr}", file=sys.stderr)
return False
return True
def _doc_convert(input_path: Path, output_path: Path) -> bool:
try:
import pypandoc # type: ignore
except Exception:
log("pypandoc is required for document conversion; install pypandoc-binary", file=sys.stderr)
return False
target_fmt = output_path.suffix.lstrip(".").lower() or "pdf"
try:
pypandoc.convert_file(
str(input_path),
to=target_fmt,
outputfile=str(output_path),
)
except OSError as exc:
log(f"pandoc is missing or failed to run: {exc}", file=sys.stderr)
return False
except Exception as exc:
log(f"pypandoc conversion failed: {exc}", file=sys.stderr)
return False
if not output_path.exists():
log("pypandoc conversion did not produce an output file", file=sys.stderr)
return False
return True
CMDLET = Cmdlet(
name="convert-file",
summary="Convert files between media/container formats (video, audio, image, documents).",
usage="convert-file -to <format> [-path <file|dir>] [-delete] [-query format:<fmt>]",
arg=[
QueryArg("to", key="format", query_only=False, required=True,
description="Target format/extension (e.g., mp4, mp3, wav, jpg, pdf)."),
SharedArgs.PATH,
SharedArgs.QUERY,
SharedArgs.DELETE,
],
detail=[
"Allows video↔video, audio↔audio, image↔image, doc↔doc, and video→audio conversions.",
"Disallows incompatible conversions (e.g., video→pdf).",
"Uses ffmpeg for media and pypandoc-binary (bundled pandoc) for document formats (mobi/epub→pdf/txt/etc).",
],
)
def _resolve_output_path(input_path: Path, outdir: Optional[Path], target_ext: str) -> Path:
base = input_path.stem
directory = outdir if outdir is not None else input_path.parent
directory.mkdir(parents=True, exist_ok=True)
candidate = directory / f"{base}.{target_ext}"
if candidate.exists():
for i in range(1, 1000):
alt = directory / f"{base}_{i}.{target_ext}"
if not alt.exists():
candidate = alt
break
return candidate
def _run(result: Any, args: Sequence[str], config: Dict[str, Any]) -> int:
parsed = parse_cmdlet_args(args, CMDLET)
target_fmt_raw = parsed.get("to") or parsed.get("format")
if not target_fmt_raw:
log("-to <format> is required", file=sys.stderr)
return 1
target_fmt = str(target_fmt_raw).lower().lstrip(".")
target_kind = _detect_kind(target_fmt)
if target_kind == "unknown":
log(f"Unsupported target format: {target_fmt}", file=sys.stderr)
return 1
delete_src = bool(parsed.get("delete", False))
inputs = normalize_result_input(result)
path_arg = parsed.get("path")
outdir_override: Optional[Path] = None
if path_arg:
try:
p = Path(str(path_arg)).expanduser()
if p.exists() and p.is_dir():
outdir_override = p
else:
inputs.append({"path": p})
except Exception:
inputs.append({"path": path_arg})
if not inputs:
log("No input provided to convert-file", file=sys.stderr)
return 1
success = 0
for item in inputs:
input_path: Optional[Path] = None
if isinstance(item, dict):
p = item.get("path") or item.get("target")
elif hasattr(item, "path"):
p = getattr(item, "path")
else:
p = item
try:
input_path = Path(str(p)) if p else None
except Exception:
input_path = None
if not input_path or not input_path.exists() or not input_path.is_file():
log("convert-file: input path missing or not found", file=sys.stderr)
continue
source_ext = input_path.suffix.lower().lstrip(".")
source_kind = _detect_kind(source_ext)
if not _allowed(source_kind, target_kind):
log(
f"Conversion from {source_kind or 'unknown'} to {target_kind} is not allowed",
file=sys.stderr,
)
continue
output_path = _resolve_output_path(input_path, outdir_override, target_fmt)
converted = False
if target_kind in {"video", "audio", "image"}:
converted = _ffmpeg_convert(input_path, output_path, target_kind, copy_metadata=True)
elif target_kind == "doc":
converted = _doc_convert(input_path, output_path)
else:
log(f"No converter for target kind {target_kind}", file=sys.stderr)
if not converted:
continue
try:
out_hash = sha256_file(output_path)
except Exception:
out_hash = None
title = extract_title_from_result(item) or output_path.stem
ctx.emit({
"path": str(output_path),
"title": title,
"hash": out_hash,
"media_kind": target_kind,
"source_path": str(input_path),
})
if delete_src:
try:
input_path.unlink()
log(f"Deleted source file: {input_path}", file=sys.stderr)
except Exception as exc:
log(f"Failed to delete source {input_path}: {exc}", file=sys.stderr)
success += 1
return 0 if success else 1
CMDLET.exec = _run
CMDLET.register()

View File

@@ -22,8 +22,8 @@ from Provider import internetarchive as ia_provider
from Provider import alldebrid as ad_provider
from Provider import openlibrary as ol_provider
from SYS.download import DownloadError, _download_direct_file
from SYS.models import DownloadOptions, DownloadMediaResult
from API.HTTP import _download_direct_file
from SYS.models import DownloadError, DownloadOptions, DownloadMediaResult
from SYS.logger import log, debug
from SYS.pipeline_progress import PipelineProgress
from SYS.result_table import ResultTable
@@ -890,7 +890,6 @@ class Download_File(Cmdlet):
return expanded_items
def _process_provider_items(
self,
*,
piped_items: Sequence[Any],
final_output_dir: Path,
@@ -900,8 +899,9 @@ class Download_File(Cmdlet):
registry: Dict[str,
Any],
progress: PipelineProgress,
) -> int:
) -> tuple[int, int]:
downloaded_count = 0
queued_magnet_submissions = 0
get_search_provider = registry.get("get_search_provider")
SearchResult = registry.get("SearchResult")
@@ -911,8 +911,17 @@ class Download_File(Cmdlet):
config=config
)
total_items = len(expanded_items)
processed_items = 0
try:
if total_items:
progress.set_percent(0)
except Exception:
pass
for item in expanded_items:
try:
label = "item"
table = get_field(item, "table")
title = get_field(item, "title")
target = get_field(item, "path") or get_field(item, "url")
@@ -933,6 +942,25 @@ class Download_File(Cmdlet):
if isinstance(extra_md, dict):
full_metadata = extra_md
try:
label = title or target
label = str(label or "item").strip()
if total_items:
pct = int(round((processed_items / max(1, total_items)) * 100))
progress.set_percent(pct)
progress.set_status(
f"downloading {processed_items + 1}/{total_items}: {label}"
)
except Exception:
pass
transfer_label = label
if str(table or "").lower() == "hifi":
try:
progress.begin_transfer(label=transfer_label, total=None)
except Exception:
pass
# If this looks like a provider item and providers are available, prefer provider.download()
downloaded_path: Optional[Path] = None
attempted_provider_download = False
@@ -1065,6 +1093,45 @@ class Download_File(Cmdlet):
continue
# Magnet targets (e.g., torrent provider results) -> submit/download via AllDebrid
if downloaded_path is None and isinstance(target, str) and is_magnet_link(str(target)):
magnet_spec = ad_provider.resolve_magnet_spec(str(target))
if magnet_spec:
def _on_emit(path: Path, file_url: str, relpath: str, metadata: Dict[str, Any]) -> None:
title_hint = metadata.get("name") or relpath or title
self._emit_local_file(
downloaded_path=path,
source=file_url or target,
title_hint=title_hint,
tags_hint=None,
media_kind_hint="file",
full_metadata=metadata,
progress=progress,
config=config,
provider_hint="alldebrid",
)
downloaded, magnet_id = ad_provider.download_magnet(
magnet_spec,
str(target),
final_output_dir,
config,
progress,
quiet_mode,
self._path_from_download_result,
_on_emit,
)
if downloaded > 0:
downloaded_count += downloaded
continue
# If queued but not yet ready, skip the generic unsupported-target error.
if magnet_id is not None:
queued_magnet_submissions += 1
continue
# Fallback: if we have a direct HTTP URL, download it directly
if (downloaded_path is None and isinstance(target,
str)
@@ -1080,6 +1147,7 @@ class Download_File(Cmdlet):
file=sys.stderr,
)
continue
debug(
f"[download-file] Provider item looks like direct URL, downloading: {target}"
)
@@ -1150,8 +1218,22 @@ class Download_File(Cmdlet):
log(f"Download failed: {e}", file=sys.stderr)
except Exception as e:
log(f"Error downloading item: {e}", file=sys.stderr)
finally:
if str(table or "").lower() == "hifi":
try:
progress.finish_transfer(label=transfer_label)
except Exception:
pass
processed_items += 1
try:
pct = int(round((processed_items / max(1, total_items)) * 100))
progress.set_percent(pct)
if processed_items >= total_items:
progress.clear_status()
except Exception:
pass
return downloaded_count
return downloaded_count, queued_magnet_submissions
# === Streaming helpers (yt-dlp) ===
@@ -2687,6 +2769,15 @@ class Download_File(Cmdlet):
debug(f"Output directory: {final_output_dir}")
try:
PipelineProgress(pipeline_context).ensure_local_ui(
label="download-file",
total_items=len(supported_url),
items_preview=supported_url,
)
except Exception:
pass
clip_spec = parsed.get("clip")
query_spec = parsed.get("query")
@@ -3572,7 +3663,7 @@ class Download_File(Cmdlet):
if early_exit is not None:
return int(early_exit)
downloaded_count += self._process_provider_items(
provider_downloaded, magnet_submissions = self._process_provider_items(
piped_items=piped_items,
final_output_dir=final_output_dir,
config=config,
@@ -3580,9 +3671,13 @@ class Download_File(Cmdlet):
registry=registry,
progress=progress,
)
downloaded_count += provider_downloaded
if downloaded_count > 0 or streaming_downloaded > 0:
debug(f"✓ Successfully processed {downloaded_count} file(s)")
if downloaded_count > 0 or streaming_downloaded > 0 or magnet_submissions > 0:
msg = f"✓ Successfully processed {downloaded_count} file(s)"
if magnet_submissions:
msg += f" and queued {magnet_submissions} magnet(s)"
debug(msg)
return 0
if streaming_exit_code is not None:

View File

@@ -255,7 +255,7 @@ def _pick_supported_ytdlp_url(urls: List[str]) -> Optional[str]:
# Prefer a true support check when the Python module is available.
try:
from SYS.download import is_url_supported_by_ytdlp
from tool.ytdlp import is_url_supported_by_ytdlp
for text in candidates:
try:

View File

@@ -246,7 +246,7 @@ class search_file(Cmdlet):
else:
table_title = f"{provider_label}: {query}".strip().rstrip(":")
preserve_order = provider_lower in {"youtube", "openlibrary", "loc"}
preserve_order = provider_lower in {"youtube", "openlibrary", "loc", "torrent"}
table_type = provider_name
table_meta: Dict[str, Any] = {"provider": provider_name}
if provider_lower == "hifi":