update 26.7.25

4600Pro/tgd/module/nfo.py

@@ -0,0 +1,161 @@
+"""Kodi/Jellyfin/Emby compatible NFO generation."""
+
+import math
+import os
+import re
+import xml.etree.ElementTree as ET
+from typing import List, Optional
+
+from module.download_record import DownloadRecord
+
+HASHTAG_RE = re.compile(r"(?<!\w)#\S+")
+PROMOTION_KEYWORDS = (
+    "点击下方",
+    "评论区",
+    "置顶",
+    "预览视频",
+    "支持一下",
+    "谢谢大家",
+)
+
+
+def get_nfo_path(video_path: str) -> str:
+    """Return the sidecar NFO path for a video path."""
+    return os.path.splitext(video_path)[0] + ".nfo"
+
+
+def clean_caption_line(line: str) -> str:
+    """Remove hashtags and surrounding whitespace from one caption line."""
+    return HASHTAG_RE.sub("", line).strip()
+
+
+def extract_genres(caption: Optional[str]) -> List[str]:
+    """Extract ordered unique genres from caption hashtags."""
+    if not caption:
+        return []
+
+    genres: List[str] = []
+    seen = set()
+    for tag in HASHTAG_RE.findall(caption):
+        genre = tag.lstrip("#").strip()
+        if genre and genre not in seen:
+            genres.append(genre)
+            seen.add(genre)
+    return genres
+
+
+def is_promotional_line(line: str) -> bool:
+    """Return True when a caption line is likely channel promotion text."""
+    return any(keyword in line for keyword in PROMOTION_KEYWORDS)
+
+
+def clean_caption(caption: Optional[str]) -> str:
+    """Return media-library friendly caption text without tags or promo lines."""
+    if not caption:
+        return ""
+
+    lines: List[str] = []
+    for raw_line in caption.splitlines():
+        line = clean_caption_line(raw_line)
+        if not line or is_promotional_line(line):
+            continue
+        lines.append(line)
+
+    paragraphs: List[str] = []
+    current: List[str] = []
+    for line in lines:
+        if line:
+            current.append(line)
+        elif current:
+            paragraphs.append("\n".join(current))
+            current = []
+    if current:
+        paragraphs.append("\n".join(current))
+
+    return "\n\n".join(paragraphs).strip()
+
+
+def title_from_record(record: DownloadRecord) -> str:
+    """Choose a human-readable title for the media library."""
+    cleaned_caption = clean_caption(record.caption)
+    if cleaned_caption:
+        return cleaned_caption.splitlines()[0]
+    if record.original_file_name:
+        return os.path.splitext(record.original_file_name)[0]
+    if record.chat_title:
+        return f"{record.chat_title} {record.message_id}"
+    return f"{record.chat_id} {record.message_id}"
+
+
+def date_part(value: Optional[str]) -> Optional[str]:
+    """Return YYYY-MM-DD from an ISO-like timestamp."""
+    if not value:
+        return None
+    return value[:10]
+
+
+def year_part(value: Optional[str]) -> Optional[str]:
+    """Return YYYY from an ISO-like timestamp."""
+    if not value:
+        return None
+    return value[:4]
+
+
+def runtime_minutes(duration: Optional[int]) -> Optional[str]:
+    """Convert seconds to rounded-up minutes for NFO runtime."""
+    if not duration:
+        return None
+    return str(int(math.ceil(duration / 60.0)))
+
+
+def add_text(parent, tag: str, text: Optional[str]):
+    """Add an XML child only when text is meaningful."""
+    if text is None or text == "":
+        return
+    child = ET.SubElement(parent, tag)
+    child.text = str(text)
+
+
+def build_video_nfo_xml(record: DownloadRecord) -> str:
+    """Build Kodi/Jellyfin/Emby-compatible movie XML."""
+    root = ET.Element("movie")
+    title = title_from_record(record)
+    plot = clean_caption(record.caption)
+    add_text(root, "title", title)
+    add_text(root, "plot", plot)
+
+    unique_id = ET.SubElement(
+        root,
+        "uniqueid",
+        {"type": "telegram", "default": "true"},
+    )
+    unique_id.text = f"{record.chat_id}_{record.message_id}"
+
+    add_text(root, "premiered", date_part(record.message_date))
+    add_text(root, "year", year_part(record.message_date))
+    add_text(root, "runtime", runtime_minutes(record.duration))
+    add_text(root, "studio", record.chat_title)
+    for genre in extract_genres(record.caption):
+        add_text(root, "genre", genre)
+
+    return ET.tostring(root, encoding="unicode")
+
+
+def write_video_nfo(record: DownloadRecord) -> str:
+    """Write a sidecar NFO file for successful video records."""
+    if (
+        record.media_type != "video"
+        or record.download_status != "success"
+        or not record.local_path
+    ):
+        return ""
+
+    nfo_path = get_nfo_path(record.local_path)
+    nfo_dir = os.path.dirname(nfo_path)
+    if nfo_dir:
+        os.makedirs(nfo_dir, exist_ok=True)
+
+    root = ET.fromstring(build_video_nfo_xml(record))
+    tree = ET.ElementTree(root)
+    tree.write(nfo_path, encoding="utf-8", xml_declaration=True)
+    return nfo_path