KudoBot / bot /parser.py
saske560's picture
Upload 9 files
337aa19 verified
Raw History Blame Contribute Delete
6.57 kB
from __future__ import annotations
import re
from typing import Any
from .utils import compact_title, normalize_text
QUALITY_RE = re.compile(r"\b(2160p|1440p|1080p|720p|480p|360p|4k|uhd|hd)\b", re.I)
YEAR_RE = re.compile(r"\b(19\d{2}|20\d{2}|2100)\b")
PART_RE = re.compile(r"\b(?:parte|part|pt)\s*([0-9]{1,3})\b", re.I)
SEASON_EPISODE_PATTERNS = [
re.compile(r"\bS(\d{1,2})\s*E(\d{1,3})\b", re.I),
re.compile(r"\b(\d{1,2})x(\d{1,3})\b", re.I),
re.compile(r"\b(?:temporada|season|t)\s*(\d{1,2})\D{0,6}(?:capitulo|capítulo|episodio|episode|ep|e|c)\s*(\d{1,3})\b", re.I),
]
EPISODE_ONLY_PATTERNS = [
re.compile(r"\b(?:capitulo|capítulo|episodio|episode|ep|e|c)\s*(\d{1,3})\b", re.I),
]
SEASON_ONLY_PATTERNS = [
re.compile(r"\b(?:temporada|season|t)\s*(\d{1,2})\b", re.I),
]
LANG_PATTERNS = [
(re.compile(r"espa(?:n|ñ)ol\s+latino", re.I), ("Español Latino", "🇲🇽")),
(re.compile(r"latino", re.I), ("Español Latino", "🇲🇽")),
(re.compile(r"castellano", re.I), ("Español Castellano", "🇪🇸")),
(re.compile(r"sub(?:titulado)?\s+espa(?:n|ñ)ol", re.I), ("Sub Español", "🇪🇸")),
(re.compile(r"dual\s+audio", re.I), ("Dual Audio", "🌐")),
(re.compile(r"english|ingl[eé]s", re.I), ("Inglés", "🇺🇸")),
(re.compile(r"japon[eé]s|japanese", re.I), ("Japonés", "🇯🇵")),
]
EXPLICIT_MOVIE = re.compile(r"#movie|#pelicula|\bpel[ií]cula\b|\bfilm\b", re.I)
EXPLICIT_SERIES = re.compile(r"#serie|#series|#anime|\banime\b|\bova\b|\btemporada\b|\bseason\b|\bcap[ií]tulo\b|\bepisodio\b", re.I)
FINAL_SEASON_RE = re.compile(r"final\s+de\s+temporada|season\s+finale", re.I)
TITLE_CLEAN_RE = re.compile(
r"\b(?:2160p|1440p|1080p|720p|480p|360p|4k|uhd|hd|espa(?:n|ñ)ol|latino|castellano|sub|dual|audio|cap[ií]tulo|episodio|episode|season|temporada|parte|part|final|de|la|pel[ií]cula|movie|mkv|mp4)\b",
re.I,
)
TRAILING_NUMBER_RE = re.compile(r"(.+?)\s+([0-9]{1,3})$")
def _first_text(*values: str | None) -> str:
for value in values:
if value and value.strip():
return value.strip()
return ""
def _detect_language(text: str) -> tuple[str | None, str | None]:
for pattern, payload in LANG_PATTERNS:
if pattern.search(text):
return payload
return None, None
def _clean_title(text: str) -> str:
lines = [line.strip() for line in text.splitlines() if line.strip()]
if not lines:
return "Sin título"
base = lines[0]
base = re.sub(r"\[[^\]]+\]", " ", base)
base = re.sub(r"\([^)]*\)", " ", base)
base = QUALITY_RE.sub(" ", base)
base = YEAR_RE.sub(" ", base)
base = PART_RE.sub(" ", base)
for pattern in SEASON_EPISODE_PATTERNS + EPISODE_ONLY_PATTERNS + SEASON_ONLY_PATTERNS:
base = pattern.sub(" ", base)
base = TITLE_CLEAN_RE.sub(" ", base)
base = re.sub(r"\.(mkv|mp4|avi|mov|m4v|webm)$", " ", base, flags=re.I)
base = re.sub(r"[_\-.]+", " ", base)
base = re.sub(r"\s+", " ", base).strip(" -")
return compact_title(base or lines[0])
def _detect_episode(text: str) -> tuple[int | None, int | None]:
for pattern in SEASON_EPISODE_PATTERNS:
match = pattern.search(text)
if match:
return int(match.group(1)), int(match.group(2))
season = None
episode = None
for pattern in SEASON_ONLY_PATTERNS:
match = pattern.search(text)
if match:
season = int(match.group(1))
break
for pattern in EPISODE_ONLY_PATTERNS:
match = pattern.search(text)
if match:
episode = int(match.group(1))
break
return season, episode
def parse_media_message(message) -> dict[str, Any] | None:
media = message.video or message.document
if not media:
return None
file_name = getattr(media, "file_name", None) or ""
raw_caption = message.caption or ""
raw_text = _first_text(raw_caption, file_name, f"media-{message.id}")
source_text = f"{raw_caption}\n{file_name}".strip()
quality_match = QUALITY_RE.search(source_text)
quality = quality_match.group(1).upper() if quality_match else None
year_match = YEAR_RE.search(source_text)
year = int(year_match.group(1)) if year_match else None
season, episode = _detect_episode(source_text)
part_match = PART_RE.search(source_text)
part = int(part_match.group(1)) if part_match else None
duration = getattr(media, "duration", None) or None
language, language_flag = _detect_language(source_text)
is_finale = bool(FINAL_SEASON_RE.search(source_text))
title = _clean_title(raw_text)
lowered = source_text.lower()
content_type = "movie"
if EXPLICIT_SERIES.search(source_text) or season or episode:
content_type = "series"
elif EXPLICIT_MOVIE.search(source_text):
content_type = "movie"
elif duration and duration < 55 * 60:
content_type = "series"
elif duration and duration >= 55 * 60:
content_type = "movie"
if content_type == "movie" and not part:
trailing = TRAILING_NUMBER_RE.match(_clean_title(raw_text))
if trailing:
maybe_number = int(trailing.group(2))
if maybe_number < 100 and maybe_number != year:
title = compact_title(trailing.group(1))
part = maybe_number
if content_type == "series" and season is None:
season = 1
category = "anime" if "anime" in lowered or "ova" in lowered else content_type
search_blob = " ".join(
part for part in [title, file_name, raw_caption, language or "", quality or "", str(year or "")] if part
)
return {
"item_id": None,
"chat_id": message.chat.id,
"message_id": message.id,
"file_id": getattr(media, "file_id", None),
"file_unique_id": getattr(media, "file_unique_id", None),
"file_name": file_name,
"file_size": getattr(media, "file_size", None),
"mime_type": getattr(media, "mime_type", None),
"duration": duration,
"media_kind": "video" if message.video else "document",
"content_type": content_type,
"category": category,
"title": title,
"normalized_title": normalize_text(title),
"year": year,
"season": season,
"episode": episode,
"part": part,
"quality": quality,
"language": language,
"language_flag": language_flag,
"is_finale": is_finale,
"source_caption": raw_caption,
"search_blob": search_blob,
"notes": "",
}