kabigon 0.8.0__py3-none-any.whl → 0.8.2__py3-none-any.whl

kabigon/compose.py

@@ -1,6 +1,5 @@
 from loguru import logger
 
-from .errors import KabigonError
 from .loader import Loader
 
 
@@ -23,7 +22,7 @@ class Compose(Loader):
             except Exception as e:
                 logger.info("[{}] Failed to load URL: {}, got error: {}", loader.__class__.__name__, url, e)
 
-        raise KabigonError(f"Failed to load URL: {url}")
+        raise Exception(f"Failed to load URL: {url}")
 
     async def async_load(self, url: str) -> str:
         for loader in self.loaders:
@@ -40,4 +39,4 @@ class Compose(Loader):
             except Exception as e:
                 logger.info("[{}] Failed to load URL: {}, got error: {}", loader.__class__.__name__, url, e)
 
-        raise KabigonError(f"Failed to load URL: {url}")
+        raise Exception(f"Failed to load URL: {url}")

kabigon/firecrawl.py

@@ -2,8 +2,6 @@ import os
 
 from firecrawl import FirecrawlApp
 
-from .errors import FirecrawlError
-from .errors import FirecrawlKeyError
 from .loader import Loader
 
 
@@ -13,7 +11,7 @@ class FirecrawlLoader(Loader):
 
         api_key = os.getenv("FIRECRAWL_API_KEY")
         if not api_key:
-            raise FirecrawlKeyError()
+            raise ValueError("FIRECRAWL_API_KEY is not set.")
 
         self.app = FirecrawlApp(api_key=api_key)
 
@@ -25,7 +23,7 @@ class FirecrawlLoader(Loader):
         )
 
         if not result.success:
-            raise FirecrawlError(url, result.error)
+            raise Exception(f"Failed to load URL: {url}, got: {result.error}")
 
         return result.markdown
 

kabigon/pdf.py

@@ -6,7 +6,6 @@ from typing import Any
 import httpx
 from pypdf import PdfReader
 
-from .errors import NotPDFError
 from .loader import Loader
 
 DEFAULT_HEADERS = {
@@ -15,6 +14,11 @@ DEFAULT_HEADERS = {
 }
 
 
+class NotPDFError(Exception):
+    def __init__(self, url: str) -> None:
+        super().__init__(f"URL is not a PDF: {url}")
+
+
 class PDFLoader(Loader):
     def load(self, url_or_file: str) -> str:
         if not url_or_file.startswith("http"):

kabigon/ptt.py

@@ -1,12 +1,12 @@
 from urllib.parse import urlparse
 
-from .errors import NotTwitterURLError
 from .httpx import HttpxLoader
 from .loader import Loader
 
 
-def is_ptt_url(url: str) -> bool:
-    return urlparse(url).netloc == "www.ptt.cc"
+def check_ptt_url(url: str) -> None:
+    if urlparse(url).netloc != "www.ptt.cc":
+        raise ValueError(f"URL must be from ptt.cc, got {url}")
 
 
 class PttLoader(Loader):
@@ -20,13 +20,11 @@ class PttLoader(Loader):
         )
 
     def load(self, url: str) -> str:
-        if not is_ptt_url(url):
-            raise NotTwitterURLError(url)
+        check_ptt_url(url)
 
         return self.httpx_loader.load(url)
 
     async def async_load(self, url: str):
-        if not is_ptt_url(url):
-            raise NotTwitterURLError(url)
+        check_ptt_url(url)
 
         return await self.httpx_loader.async_load(url)

kabigon/reel.py

@@ -1,11 +1,11 @@
-from .errors import NotReelURLError
 from .httpx import HttpxLoader
 from .loader import Loader
 from .ytdlp import YtdlpLoader
 
 
-def is_reel_url(url: str) -> bool:
-    return url.startswith("https://www.instagram.com/reel")
+def check_reel_url(url: str) -> None:
+    if not url.startswith("https://www.instagram.com/reel"):
+        raise ValueError(f"URL is not an Instagram Reel: {url}")
 
 
 class ReelLoader(Loader):
@@ -14,8 +14,7 @@ class ReelLoader(Loader):
         self.ytdlp_loader = YtdlpLoader()
 
     def load(self, url: str) -> str:
-        if not is_reel_url(url):
-            raise NotReelURLError(url)
+        check_reel_url(url)
 
         audio_content = self.ytdlp_loader.load(url)
         html_content = self.httpx_loader.load(url)
@@ -23,8 +22,7 @@ class ReelLoader(Loader):
         return f"{audio_content}\n\n{html_content}"
 
     async def async_load(self, url: str):
-        if not is_reel_url(url):
-            raise NotReelURLError(url)
+        check_reel_url(url)
 
         audio_content = await self.ytdlp_loader.async_load(url)
         html_content = await self.httpx_loader.async_load(url)

kabigon/twitter.py

@@ -1,7 +1,6 @@
 from urllib.parse import urlparse
 from urllib.parse import urlunparse
 
-from .errors import NotTwitterURLError
 from .loader import Loader
 from .playwright import PlaywrightLoader
 
@@ -21,8 +20,9 @@ def replace_domain(url: str, new_domain: str = "x.com") -> str:
     return urlunparse(urlparse(url)._replace(netloc=new_domain))
 
 
-def is_x_url(url: str) -> bool:
-    return urlparse(url).netloc in TWITTER_DOMAINS
+def check_x_url(url: str) -> None:
+    if urlparse(url).netloc not in TWITTER_DOMAINS:
+        raise ValueError(f"URL is not a Twitter URL: {url}")
 
 
 class TwitterLoader(Loader):
@@ -30,16 +30,14 @@ class TwitterLoader(Loader):
         self.playwright_loader = PlaywrightLoader(wait_until="networkidle")
 
     def load(self, url: str) -> str:
-        if not is_x_url(url):
-            raise NotTwitterURLError(url)
+        check_x_url(url)
 
         url = replace_domain(url)
 
         return self.playwright_loader.load(url)
 
     async def async_load(self, url: str):
-        if not is_x_url(url):
-            raise NotTwitterURLError(url)
+        check_x_url(url)
 
         url = replace_domain(url)
 

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

@@ -1,25 +1,21 @@
-import functools
+# import functools
 import hashlib
 import os
-import subprocess
+
+# import subprocess
 import tempfile
 from typing import Final
 
-import numpy as np
-import timeout_decorator
-import whisper
+# import numpy as np
 import yt_dlp
 from loguru import logger
 
 from .loader import Loader
 
 try:
-    import mlx_whisper  # noqa: F401 # type: ignore
-
-    _mlx_whisper_installed = True
+    import whisper
 except ImportError:
-    _mlx_whisper_installed = False
-
+    logger.info("OpenAI Whisper not installed. Please install it with `pip install openai-whisper`.")
 
 DEFAULT_FFMPEG_PATH: Final[str] = "ffmpeg"
 
@@ -65,68 +61,58 @@ 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)
+# 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
 
 
 class YtdlpLoader(Loader):
-    @timeout_decorator.timeout(300)
+    def __init__(self, model: str = "tiny") -> None:
+        self.model = whisper.load_model(model)
+
     def load(self, url: str) -> str:
         audio_file = download_audio(url)
-        audio = load_audio(audio_file)
+        # audio = load_audio(audio_file)
+        audio = whisper.load_audio(audio_file)
 
         # Clean up the audio file
         os.remove(audio_file)
 
-        result = _transcribe(audio)
+        result = self.model.transcribe(audio)
         return result.get("text", "")

{kabigon-0.8.0.dist-info → kabigon-0.8.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kabigon
-Version: 0.8.0
+Version: 0.8.2
 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

kabigon-0.8.2.dist-info/RECORD

@@ -0,0 +1,20 @@
+kabigon/__init__.py,sha256=MUfTFUe5ezA249L2yuU5_2FiewLu86H3VsIpJSne2vQ,560
+kabigon/cli.py,sha256=PJ0wnwp_AgHA54YxGr1jNJ_q3ls7fEymgTJaJxCVU7M,650
+kabigon/compose.py,sha256=Kb6_-SNeh08QELMF-r3mWxasDTxJBuJJQamFPLcwQ1I,1463
+kabigon/firecrawl.py,sha256=-5AI9tla_684dtpubY_BRudqLgw28158WdwA1RjJvAA,778
+kabigon/httpx.py,sha256=Zup9DURyWLqoWzaxBbCYAaV-5LSlHUuAcNyyUsZTVag,696
+kabigon/loader.py,sha256=KhOJvlzLDM0o05o4VqMRgiyRVjofkzuGEcihQagj-8M,452
+kabigon/pdf.py,sha256=PNOx-Dz_VpN-EVnVar_wJJZdxWrDZrAFE-gSuUR9q7o,1870
+kabigon/playwright.py,sha256=MZ-r0Ej2wWAOJkDLwYRvO77wcDvh38KXz2wgDsCTgm0,1358
+kabigon/ptt.py,sha256=Gyp2nJrjptkjbwZJ9VEQHX0DEgKBe5QRQOmGVHUUgNA,896
+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=F9GpLa0iUy03wYU94RrrnrXa6ExqbG6CZpqx5bPENWE,1106
+kabigon/ytdlp.py,sha256=oeFoE7nWZWaT0dR9nwt_SZh_FE0gJ6Gulh6QzGCB6xo,2956
+kabigon-0.8.2.dist-info/METADATA,sha256=UEq0yUqP0dgzyqMhxiHgLvP0wT_nf0PlQLy8VkufrEg,1287
+kabigon-0.8.2.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+kabigon-0.8.2.dist-info/entry_points.txt,sha256=O3FYAO9w-NQvlGMJrBvtrnGHSK2QkUnQBTa30YXRbVE,45
+kabigon-0.8.2.dist-info/licenses/LICENSE,sha256=H2T3_RTgmcngMeC7p_SXT3GwBLkd2DaNgAZuxulcfiA,1066
+kabigon-0.8.2.dist-info/RECORD,,

kabigon/errors.py

@@ -1,28 +0,0 @@
-class KabigonError(Exception):
-    pass
-
-
-class FirecrawlKeyError(KabigonError):
-    def __init__(self) -> None:
-        super().__init__("FIRECRAWL_API_KEY is not set.")
-
-
-class FirecrawlError(KabigonError):
-    def __init__(self, url: str, error: str) -> None:
-        msg = f"Failed to load URL: {url}, got: {error}"
-        super().__init__(msg)
-
-
-class NotPDFError(KabigonError):
-    def __init__(self, url: str) -> None:
-        super().__init__(f"URL is not a PDF: {url}")
-
-
-class NotReelURLError(KabigonError):
-    def __init__(self, url: str):
-        super().__init__(f"URL is not an Instagram Reel: {url}")
-
-
-class NotTwitterURLError(KabigonError):
-    def __init__(self, url: str):
-        super().__init__(f"URL is not a Twitter URL: {url}")

kabigon-0.8.0.dist-info/RECORD

@@ -1,21 +0,0 @@
-kabigon/__init__.py,sha256=MUfTFUe5ezA249L2yuU5_2FiewLu86H3VsIpJSne2vQ,560
-kabigon/cli.py,sha256=PJ0wnwp_AgHA54YxGr1jNJ_q3ls7fEymgTJaJxCVU7M,650
-kabigon/compose.py,sha256=l2D5OK91VcN2a6DbjMdwBk3YSqzVV7fOVX0TqNm2gJo,1502
-kabigon/errors.py,sha256=iri_YS71UsOHwaVtfy5IA6iUfq30DCsptZsChmZaZic,755
-kabigon/firecrawl.py,sha256=Xnrlhco_R58x5kwGy9ZCKTnVqS4Pp-D3G0u-qnuGEsU,800
-kabigon/httpx.py,sha256=Zup9DURyWLqoWzaxBbCYAaV-5LSlHUuAcNyyUsZTVag,696
-kabigon/loader.py,sha256=KhOJvlzLDM0o05o4VqMRgiyRVjofkzuGEcihQagj-8M,452
-kabigon/pdf.py,sha256=yJcgkdMMF52baFsFxJp9Jn89KsTKBboLTLwD3gs6U5U,1775
-kabigon/playwright.py,sha256=MZ-r0Ej2wWAOJkDLwYRvO77wcDvh38KXz2wgDsCTgm0,1358
-kabigon/ptt.py,sha256=S2d6SeFGxM4E4kg-n5blN2BX56CWj_eOhapJxzUnxu8,965
-kabigon/py.typed,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
-kabigon/reel.py,sha256=J2QOxGMYi_HaEscQPIipPEoHGN_iksGxR6pV_XvryME,929
-kabigon/twitter.py,sha256=U07pa8xA0nHAaaDPeUelQRvXR5ZnUvYJZW35xRAvHA8,1114
-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-0.8.0.dist-info/METADATA,sha256=SAazEEnqklluyiH2psGQwUI40ee400GELlcxv0OyvuU,1079
-kabigon-0.8.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-kabigon-0.8.0.dist-info/entry_points.txt,sha256=O3FYAO9w-NQvlGMJrBvtrnGHSK2QkUnQBTa30YXRbVE,45
-kabigon-0.8.0.dist-info/licenses/LICENSE,sha256=H2T3_RTgmcngMeC7p_SXT3GwBLkd2DaNgAZuxulcfiA,1066
-kabigon-0.8.0.dist-info/RECORD,,