osekit 0.4.3__py3-none-any.whl

osekit/__init__.py

@@ -0,0 +1,51 @@
+from __future__ import annotations
+
+import logging.config
+import os.path
+from pathlib import Path
+
+import yaml
+
+from osekit import utils
+
+__all__ = [
+    "setup_logging",
+    "utils",
+]
+
+
+def setup_logging(
+    config_file: str | Path = "logging_config.yaml",
+    default_level: int = logging.INFO,
+) -> None:
+    """Configure logger using a configuration yaml file.
+
+    Parameters
+    ----------
+    config_file: str | Path
+        Path to a logging configuration file
+    default_level: int
+        Logging level to use
+        Default value, `logging.INFO`
+
+    """
+    user_config_file_path = Path(os.getenv("OSMOSE_USER_CONFIG", ".")) / config_file
+    default_config_file_path = Path(__file__).parent / config_file
+
+    config_file_path = next(
+        (
+            file
+            for file in (user_config_file_path, default_config_file_path)
+            if file.exists()
+        ),
+        None,
+    )
+
+    if config_file_path:
+        with Path.open(config_file_path) as configuration:
+            logging_config = yaml.safe_load(configuration)
+        logging.config.dictConfig(logging_config)
+    else:
+        logging.basicConfig(level=default_level)
+        msg = "Configuration file not found, using default configuration."
+        logging.getLogger(__name__).warning(msg)

osekit/audio_backend/audio_backend.py

@@ -0,0 +1,49 @@
+from os import PathLike
+from typing import Protocol
+
+import numpy as np
+
+
+class AudioBackend(Protocol):
+    def info(self, path: PathLike | str) -> tuple[int, int, int]:
+        """Return the sample rate, number of frames and channels of the audio file.
+
+        Parameters
+        ----------
+        path: PathLike | str
+            Path to the audio file.
+
+        Returns
+        -------
+        tuple[int,int,int]:
+            Sample rate, number of frames and channels of the audio file.
+
+        """
+        ...
+
+    def read(self, path: PathLike | str, start: int, stop: int) -> np.ndarray:
+        """Read the content of an audio file.
+
+        If the audio file is not the current opened file,
+        the current opened file is switched.
+
+        Parameters
+        ----------
+        path: PathLike | str
+            Path to the audio file.
+        start: int
+            First frame to read.
+        stop: int
+            Frame after the last frame to read.
+
+        Returns
+        -------
+        np.ndarray:
+            A ``(channel * frames)`` array containing the audio data.
+
+        """
+        ...
+
+    def close(self) -> None:
+        """Close the currently opened file."""
+        ...

osekit/audio_backend/audio_file_manager.py

@@ -0,0 +1,104 @@
+"""Audio File Manager which keeps an audio file open until a request in another file is made.
+
+This workflow avoids closing/opening a same file repeatedly.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import TYPE_CHECKING
+
+from osekit.audio_backend.mseed_backend import MSeedBackend
+from osekit.audio_backend.soundfile_backend import SoundFileBackend
+
+if TYPE_CHECKING:
+    from os import PathLike
+
+    import numpy as np
+
+
+class AudioFileManager:
+    """Audio File Manager which keeps an audio file open until a request in another file is made."""
+
+    def __init__(self) -> None:
+        """Initialize an audio file manager."""
+        self._soundfile = SoundFileBackend()
+        self._mseed: MSeedBackend | None = None
+
+    def close(self) -> None:
+        """Close the currently opened file."""
+        self._soundfile.close()
+        if self._mseed:
+            self._mseed.close()
+
+    def _backend(self, path: PathLike | str) -> SoundFileBackend | MSeedBackend:
+        suffix = Path(path).suffix.lower()
+
+        if suffix == ".mseed":
+            if self._mseed is None:
+                self._mseed = MSeedBackend()
+            return self._mseed
+
+        return self._soundfile
+
+    def info(self, path: PathLike | str) -> tuple[int, int, int]:
+        """Return the sample rate, number of frames and channels of the audio file.
+
+        Parameters
+        ----------
+        path: PathLike | str
+            Path to the audio file.
+
+        Returns
+        -------
+        tuple[int,int,int]:
+            Sample rate, number of frames and channels of the audio file.
+
+        """
+        return self._backend(path).info(path)
+
+    def read(
+        self,
+        path: PathLike | str,
+        start: int = 0,
+        stop: int | None = None,
+    ) -> np.ndarray:
+        """Read the content of an audio file.
+
+        If the audio file is not the current opened file,
+        the current opened file is switched.
+
+        Parameters
+        ----------
+        path: PathLike | str
+            Path to the audio file.
+        start: int
+            First frame to read.
+        stop: int | None
+            Frame after the last frame to read.
+
+        Returns
+        -------
+        np.ndarray:
+            A ``(channel * frames)`` array containing the audio data.
+
+        """
+        _, frames, _ = self.info(path)
+
+        if stop is None:
+            stop = frames
+
+        if stop is None:
+            stop = frames
+
+        if not 0 <= start < frames:
+            msg = "Start should be between 0 and the last frame of the audio file."
+            raise ValueError(msg)
+        if not 0 <= stop <= frames:
+            msg = "Stop should be between 0 and the last frame of the audio file."
+            raise ValueError(msg)
+        if start > stop:
+            msg = "Start should be inferior to Stop."
+            raise ValueError(msg)
+
+        return self._backend(path).read(path=path, start=start, stop=stop)

osekit/audio_backend/mseed_backend.py

@@ -0,0 +1,85 @@
+from os import PathLike
+
+import numpy as np
+
+
+def _require_obspy() -> None:
+    try:
+        import obspy  # noqa: PLC0415, F401
+    except ImportError as e:
+        msg = "MSEED support requires the optional dependency 'obspy' "
+        "Install with: ``pip install osekit[mseed]``. "
+        "If you're on windows and don't use conda, may the force be with you."
+        raise ImportError(msg) from e
+
+
+class MSeedBackend:
+    """Backend for reading sismology MSEED files."""
+
+    def __init__(self) -> None:
+        """Initialize the MSEED backend."""
+        _require_obspy()
+
+    def close(self) -> None:
+        """Close the currently opened file. No use in MSEED files."""
+
+    def info(self, path: PathLike | str) -> tuple[int, int, int]:
+        """Return the sample rate, number of frames and channels of the MSEED file.
+
+        Parameters
+        ----------
+        path: PathLike | str
+            Path to the audio file.
+
+        Returns
+        -------
+        tuple[int,int,int]:
+            Sample rate, number of frames and channels of the MSEED file.
+
+        """
+        _require_obspy()
+        import obspy  # type: ignore[import-not-found]
+
+        metadata = obspy.read(pathname_or_url=path, headonly=True)
+        sample_rate = {trace.meta.sampling_rate for trace in metadata.traces}
+        if len(sample_rate) != 1:
+            msg = "Inconsistent sampling rates in MSEED file."
+            raise ValueError(msg)
+
+        frames = sum(trace.meta.npts for trace in metadata.traces)
+        return (
+            int(sample_rate.pop()),
+            frames,
+            1,
+        )
+
+    def read(
+        self,
+        path: PathLike | str,
+        start: int = 0,
+        stop: int | None = None,
+    ) -> np.ndarray:
+        """Read the content of a MSEED file.
+
+        Parameters
+        ----------
+        path: PathLike | str
+            Path to the audio file.
+        start: int
+            First frame to read.
+        stop: int
+            Frame after the last frame to read.
+
+        Returns
+        -------
+        np.ndarray:
+            A ``(channel * frames)`` array containing the MSEED data.
+
+        """
+        _require_obspy()
+        import obspy  # type: ignore[import-not-found]  # noqa: PLC0415
+
+        file_content = obspy.read(path)
+
+        data = np.concatenate([trace.data for trace in file_content])
+        return data[start:stop]

osekit/audio_backend/soundfile_backend.py

@@ -0,0 +1,84 @@
+from os import PathLike
+
+import numpy as np
+import soundfile as sf
+
+
+class SoundFileBackend:
+    """Backend for reading conventional audio files (WAV, FLAC, MP3...)."""
+
+    def __init__(self) -> None:
+        """Instantiate a SoundFileBackend."""
+        self._file: sf.SoundFile | None = None
+
+    def close(self) -> None:
+        """Close the currently opened file."""
+        if self._file is None:
+            return
+        self._file.close()
+        self._file = None
+
+    def info(self, path: PathLike | str) -> tuple[int, int, int]:
+        """Return the sample rate, number of frames and channels of the audio file.
+
+        Parameters
+        ----------
+        path: PathLike | str
+            Path to the audio file.
+
+        Returns
+        -------
+        tuple[int,int,int]:
+            Sample rate, number of frames and channels of the audio file.
+
+        """
+        self._switch(path)
+        return (
+            self._file.samplerate,
+            self._file.frames,
+            self._file.channels,
+        )
+
+    def read(
+        self,
+        path: PathLike | str,
+        start: int = 0,
+        stop: int | None = None,
+    ) -> np.ndarray:
+        """Read the content of an audio file.
+
+        If the audio file is not the current opened file,
+        the current opened file is switched.
+
+        Parameters
+        ----------
+        path: PathLike | str
+            Path to the audio file.
+        start: int
+            First frame to read.
+        stop: int
+            Frame after the last frame to read.
+
+        Returns
+        -------
+        np.ndarray:
+            A ``(channel * frames)`` array containing the audio data.
+
+        """
+        self._switch(path)
+        self._file.seek(start)
+        return self._file.read(stop - start)
+
+    def _close(self) -> None:
+        if self._file is None:
+            return
+        self._file.close()
+        self._file = None
+
+    def _open(self, path: PathLike | str) -> None:
+        self._file = sf.SoundFile(path, "r")
+
+    def _switch(self, path: PathLike | str) -> None:
+        if self._file is None or self._file.name != str(path):
+            self._close()
+            self._open(path)

osekit/config.py

@@ -0,0 +1,28 @@
+import logging
+import stat
+
+from osekit.logging_context import LoggingContext
+
+TIMESTAMP_FORMAT_AUDIO_FILE = "%Y-%m-%dT%H:%M:%S.%f%z"
+TIMESTAMP_FORMAT_EXPORTED_FILES_UNLOCALIZED = "%Y_%m_%d_%H_%M_%S_%f"
+TIMESTAMP_FORMAT_EXPORTED_FILES_LOCALIZED = "%Y_%m_%d_%H_%M_%S_%f%z"
+TIMESTAMP_FORMATS_EXPORTED_FILES = [
+    TIMESTAMP_FORMAT_EXPORTED_FILES_LOCALIZED,
+    TIMESTAMP_FORMAT_EXPORTED_FILES_UNLOCALIZED,
+]
+
+FPDEFAULT = 0o664  # Default file permissions
+DPDEFAULT = stat.S_ISGID | 0o775  # Default directory permissions
+
+global_logging_context = LoggingContext()
+print_logger = logging.getLogger("printer")
+
+resample_quality_settings = {
+    "downsample": "QQ",
+    "upsample": "MQ",
+}
+
+multiprocessing = {
+    "is_active": False,
+    "nb_processes": None,
+}

osekit/core_api/__init__.py

@@ -0,0 +1,3 @@
+from osekit.audio_backend.audio_file_manager import AudioFileManager
+
+audio_file_manager = AudioFileManager()