cvpl_tools/tools/fs.py

View source at fs.py.

APIs

cvpl_tools.tools.fs.cdir_init(cache_url: str | RDirFileSystem) → CDirQueryResult

Returned CDirQueryResult is the same as calling cdir_query()

Function behavior: 1. If directory does not exist, make a new empty directory 2. If directory exists but timestamp is not found (not committed), then remove it and make a new directory 3. If commit timestamp is found, do nothing

Parameters:

cache_url – Pointing to the directory location

cvpl_tools.tools.fs.cdir_commit(cache_url: str | RDirFileSystem)

Create a commit timestamp

cvpl_tools.tools.fs.cdir_query(cache_url: str | RDirFileSystem) → CDirQueryResult

Query the commit status of a directory

1. If the url points to a directory that does not exists or does not have ‘.cache_url_timestamp’ file created under, then the returned query object will have query.commit = None 2. Otherwise the directory is a committed directory, and the returned query object will contain the time stamp of commit in its commit attribute query.commit

Parameters:

cache_url – The location to be queried

Returns:

A CDirQueryResult containing the commit status of the given cache url will be returned

async cvpl_tools.tools.fs.save(file: str | RDirFileSystem, im, storage_options: dict = None)

Save an image object into given path

Supported im object types: - np.ndarray - dask.Array - cvpl_tools.im.ndblock.NDBlock

Parameters:

  • file – The full/relative path to the directory to be saved to

  • im – Object to be saved

  • storage_options – Specifies options in saving method and saved file format -preferred_chunksize (tuple[int, …]): chunk sizes to save as; will rechunk if different from current size; only applies to dask arrays. -multiscale (int): The number of downsample layers for save ome-zarr; only applies if the image is a dask image -compressor: The compressor to use to compress array or chunks

cvpl_tools.tools.fs.load(file: str | RDirFileSystem, storage_options: dict = None)

Load an image from the given directory.

The image is one saved by cvpl_tools.tools.fs.save()

Parameters:

  • file – Full path to the directory to be read from

  • storage_options – Specifies options in saving method and saved file format - compressor (numcodecs.abc.Codec, optional): Compressor used to compress the chunks

Returns:

Recreated image; this method attempts to keep meta and content of the loaded image stays the same as when they are saved

cvpl_tools.tools.fs.display(file: str | RDirFileSystem, context_args: dict)

Display an image in the viewer; supports numpy or dask ome zarr image

The image is one saved by cvpl_tools.tools.fs.save()

Parameters:

  • file – Full path to the directory to be read from

  • context_args – contains viewer and arguments passed to the viewer’s add image functions - viewer (napari.Viewer, optional): Only display if a viewer is provided - is_label (bool, optional): defaults to False; if True, use viewer’s add_labels() instead of add_image() to display the array - layer_args (dict, optional)