kabigon 0.8.1__py3-none-any.whl → 0.8.3__py3-none-any.whl

kabigon/youtube.py

@@ -1,6 +1,5 @@
 import aioytt
 import aioytt.video_id
-import timeout_decorator
 from youtube_transcript_api import YouTubeTranscriptApi
 
 from .loader import Loader
@@ -12,7 +11,6 @@ class YoutubeLoader(Loader):
     def __init__(self, languages: list[str] | None = None) -> None:
         self.languages = languages or DEFAULT_LANGUAGES
 
-    @timeout_decorator.timeout(20)
     def load(self, url: str) -> str:
         video_id = aioytt.video_id.parse_video_id(url)
 

kabigon/ytdlp.py

@@ -1,49 +1,15 @@
-import functools
 import hashlib
 import os
-import subprocess
 import tempfile
-from typing import Final
 
-import numpy as np
-import timeout_decorator
-import whisper
 import yt_dlp
 from loguru import logger
 
 from .loader import Loader
 
-try:
-    import mlx_whisper  # noqa: F401 # type: ignore
-
-    _mlx_whisper_installed = True
-except ImportError:
-    _mlx_whisper_installed = False
-
-
-DEFAULT_FFMPEG_PATH: Final[str] = "ffmpeg"
-
-
-def hash_url(url: str) -> str:
-    return hashlib.sha512(url.encode("utf-8")).hexdigest()
-
-
-def get_ffmpeg_path() -> str:
-    path = os.getenv("FFMPEG_PATH")
-    if not path:
-        path = DEFAULT_FFMPEG_PATH
-        logger.warning("FFMPEG_PATH not set, using default: {}", DEFAULT_FFMPEG_PATH)
-
-    return path
-
 
 def download_audio(url: str) -> str:
-    ffmpeg_path = get_ffmpeg_path()
-
-    filename = os.path.join(
-        tempfile.gettempdir(),
-        hash_url(url),
-    )
+    filename = os.path.join(tempfile.gettempdir(), hashlib.sha512(url.encode("utf-8")).hexdigest())
 
     ydl_opts = {
         "format": "bestaudio/best",
@@ -55,7 +21,7 @@ def download_audio(url: str) -> str:
             }
         ],
         "outtmpl": filename,
-        "ffmpeg_location": ffmpeg_path,
+        "ffmpeg_location": os.getenv("FFMPEG_PATH", "ffmpeg"),
         "match_filter": yt_dlp.match_filter_func(["!is_live"]),
     }
 
@@ -65,68 +31,25 @@ def download_audio(url: str) -> str:
     return filename + ".mp3"
 
 
-def load_audio(file: str, sr: int = 16000):
-    """
-    Open an audio file and read as mono waveform, resampling as necessary
-
-    Parameters
-    ----------
-    file: str
-        The audio file to open
-
-    sr: int
-        The sample rate to resample the audio if necessary
-
-    Returns
-    -------
-    A NumPy array containing the audio waveform, in float32 dtype.
-    """
-    ffmpeg_path = get_ffmpeg_path()
-
-    # This launches a subprocess to decode audio while down-mixing
-    # and resampling as necessary.  Requires the ffmpeg CLI in PATH.
-    # fmt: off
-    cmd = [
-        ffmpeg_path,
-        "-nostdin",
-        "-threads", "0",
-        "-i", file,
-        "-f", "s16le",
-        "-ac", "1",
-        "-acodec", "pcm_s16le",
-        "-ar", str(sr),
-        "-"
-    ]
-    # fmt: on
-    try:
-        out = subprocess.run(cmd, capture_output=True, check=True).stdout
-    except subprocess.CalledProcessError as e:
-        raise RuntimeError(f"Failed to load audio: {e.stderr.decode()}") from e
-
-    return np.frombuffer(out, np.int16).flatten().astype(np.float32) / 32768.0
-
-
-@functools.cache
-def _load_whisper_model() -> whisper.Whisper:
-    return whisper.load_model("tiny")
-
-
-def _transcribe(audio: np.ndarray) -> dict:
-    if _mlx_whisper_installed:
-        return mlx_whisper.transcribe(audio, path_or_hf_repo="mlx-community/whisper-tiny")
-
-    model = _load_whisper_model()
-    return model.transcribe(audio)
+class YtdlpLoader(Loader):
+    def __init__(self, model: str = "tiny") -> None:
+        try:
+            import whisper
+        except ImportError as e:
+            raise ImportError(
+                "OpenAI Whisper not installed. Please install it with `pip install openai-whisper`."
+            ) from e
 
+        self.model = whisper.load_model(model)
+        self.load_audio = whisper.load_audio
 
-class YtdlpLoader(Loader):
-    @timeout_decorator.timeout(300)
     def load(self, url: str) -> str:
         audio_file = download_audio(url)
-        audio = load_audio(audio_file)
+        audio = self.load_audio(audio_file)
 
         # Clean up the audio file
         os.remove(audio_file)
 
-        result = _transcribe(audio)
+        logger.info("Transcribing audio file: {}", audio_file)
+        result = self.model.transcribe(audio)
         return result.get("text", "")

{kabigon-0.8.1.dist-info → kabigon-0.8.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kabigon
-Version: 0.8.1
+Version: 0.8.3
 Author-email: narumi <toucans-cutouts0f@icloud.com>
 License-File: LICENSE
 Requires-Python: >=3.10
@@ -9,14 +9,19 @@ Requires-Dist: firecrawl-py>=2.4.1
 Requires-Dist: httpx>=0.28.1
 Requires-Dist: loguru>=0.7.3
 Requires-Dist: markdownify>=0.14.1
-Requires-Dist: openai-whisper>=20240930
-Requires-Dist: playwright>=1.52.0
+Requires-Dist: numpy>=2.2.5
 Requires-Dist: pypdf>=5.3.0
 Requires-Dist: rich>=13.9.4
-Requires-Dist: timeout-decorator>=0.5.0
 Requires-Dist: typer>=0.15.3
 Requires-Dist: youtube-transcript-api>=0.6.3
 Requires-Dist: yt-dlp>=2025.4.30
+Provides-Extra: all
+Requires-Dist: openai-whisper>=20240930; extra == 'all'
+Requires-Dist: playwright>=1.52.0; extra == 'all'
+Provides-Extra: playwright
+Requires-Dist: playwright>=1.52.0; extra == 'playwright'
+Provides-Extra: whisper
+Requires-Dist: openai-whisper>=20240930; extra == 'whisper'
 Description-Content-Type: text/markdown
 
 # kabigon
@@ -24,7 +29,7 @@ Description-Content-Type: text/markdown
 ## Installation
 
 ```shell
-pip install kabigon
+pip install kabigon[all]
 playwright install chromium
 ```
 

{kabigon-0.8.1.dist-info → kabigon-0.8.3.dist-info}/RECORD

@@ -11,10 +11,10 @@ kabigon/py.typed,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
 kabigon/reel.py,sha256=qOwWCvcp7xNKg0JDunq_Bsl8yqqMzrnAOI9k5mSqrOU,874
 kabigon/twitter.py,sha256=aRqAiFxIwln6lteWdoF6SmvbzO62yBTQRzcB7UcVJwk,1046
 kabigon/utils.py,sha256=eNTLtHLSB2erDac2HH3jWemgfr8Ou_ozwVb8h9BD-4g,922
-kabigon/youtube.py,sha256=HoiFNq0ookPL7_rO_wloBaY8yTIX6xP8A77F7y02q64,1166
-kabigon/ytdlp.py,sha256=_QRcyFx9s7NnI1MvcWdKKxlX-hHLnqtduCSL5_UH6dU,3140
-kabigon-0.8.1.dist-info/METADATA,sha256=8nZX3ukADj-qndGHMMuZluG_HdI8-wqNu8QJBoplPRc,1079
-kabigon-0.8.1.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-kabigon-0.8.1.dist-info/entry_points.txt,sha256=O3FYAO9w-NQvlGMJrBvtrnGHSK2QkUnQBTa30YXRbVE,45
-kabigon-0.8.1.dist-info/licenses/LICENSE,sha256=H2T3_RTgmcngMeC7p_SXT3GwBLkd2DaNgAZuxulcfiA,1066
-kabigon-0.8.1.dist-info/RECORD,,
+kabigon/youtube.py,sha256=F9GpLa0iUy03wYU94RrrnrXa6ExqbG6CZpqx5bPENWE,1106
+kabigon/ytdlp.py,sha256=OGVrKRGXY_eHePgDmN52Kkui12NN44LsIPonD_61QBM,1501
+kabigon-0.8.3.dist-info/METADATA,sha256=uidye_pVnhpAivxY26MlZMSascT1MmuMjBuCAw84Ics,1292
+kabigon-0.8.3.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+kabigon-0.8.3.dist-info/entry_points.txt,sha256=O3FYAO9w-NQvlGMJrBvtrnGHSK2QkUnQBTa30YXRbVE,45
+kabigon-0.8.3.dist-info/licenses/LICENSE,sha256=H2T3_RTgmcngMeC7p_SXT3GwBLkd2DaNgAZuxulcfiA,1066
+kabigon-0.8.3.dist-info/RECORD,,