Source code for neuroconv.datainterfaces.ecephys.maxwell.maxonedatainterface

import os
from pathlib import Path
from platform import system
from typing import Optional

from pydantic import DirectoryPath, FilePath

from ..baserecordingextractorinterface import BaseRecordingExtractorInterface


[docs]class MaxOneRecordingInterface(BaseRecordingExtractorInterface): # pragma: no cover """ Primary data interface class for converting MaxOne data. Using the :py:class:`~spikeinterface.extractors.MaxwellRecordingExtractor`. """ display_name = "MaxOne Recording" associated_suffixes = (".raw", ".h5") info = "Interface for MaxOne recording data." ExtractorName = "MaxwellRecordingExtractor"
[docs] @staticmethod def auto_install_maxwell_hdf5_compression_plugin( hdf5_plugin_path: Optional[DirectoryPath] = None, download_plugin: bool = True ) -> None: """ If you do not yet have the Maxwell compression plugin installed, this function will automatically install it. Parameters ---------- hdf5_plugin_path : string or Path, optional Path to your systems HDF5 plugin library. Uses the home directory by default. download_plugin : boolean, default: True Whether to force download of the decompression plugin. It's a very lightweight install but does require an internet connection. This is left as True for seamless passive usage and should not impact performance. """ from neo.rawio.maxwellrawio import auto_install_maxwell_hdf5_compression_plugin auto_install_maxwell_hdf5_compression_plugin(hdf5_plugin_path=hdf5_plugin_path, force_download=download_plugin)
def __init__( self, file_path: FilePath, hdf5_plugin_path: Optional[DirectoryPath] = None, download_plugin: bool = True, verbose: bool = False, es_key: str = "ElectricalSeries", ) -> None: """ Load and prepare data for MaxOne. Parameters ---------- file_path : string or Path Path to the .raw.h5 file. hdf5_plugin_path : string or Path, optional Path to your systems HDF5 plugin library. Uses the home directory by default. download_plugin : boolean, default: True Whether to force download of the decompression plugin. It's a very lightweight install but does require an internet connection. This is left as True for seamless passive usage and should not impact performance. verbose : boolean, default: True Allows verbosity. es_key : str, default: "ElectricalSeries" The key of this ElectricalSeries in the metadata dictionary. """ if system() != "Linux": raise NotImplementedError( "The MaxOneRecordingInterface has not yet been implemented for systems other than Linux." ) hdf5_plugin_path = os.environ.get( "HDF5_PLUGIN_PATH", hdf5_plugin_path or Path.home() / "hdf5_plugin_path_maxwell", ) os.environ["HDF5_PLUGIN_PATH"] = str(hdf5_plugin_path) if download_plugin: self.auto_install_maxwell_hdf5_compression_plugin(hdf5_plugin_path=hdf5_plugin_path) super().__init__(file_path=file_path, verbose=verbose, es_key=es_key)
[docs] def get_metadata(self) -> dict: metadata = super().get_metadata() maxwell_version = self.recording_extractor.neo_reader.raw_annotations["blocks"][0]["maxwell_version"] metadata["Ecephys"]["Device"][0].update(description=f"Recorded using Maxwell version '{maxwell_version}'.") return metadata