utils

plantdb.commons.utils Link

This module contains utility functions.

fsdb_file_from_local_file Link

fsdb_file_from_local_file(path)

Creates a temporary fsdb.File object from a local file.

Parameters:

Name	Type	Description	Default
`path`	`Path or str`	The file path to use to create the temporary local database.	required

Returns:

Type	Description
`File`	The temporary `fsdb.File`.

Source code in plantdb/commons/utils.py

def fsdb_file_from_local_file(path):
    """Creates a temporary ``fsdb.File`` object from a local file.

    Parameters
    ----------
    path : pathlib.Path or str
        The file path to use to create the temporary local database.

    Returns
    -------
    plantdb.commons.fsdb.File
        The temporary ``fsdb.File``.
    """
    from plantdb.commons.fsdb import FSDB
    from plantdb.commons.fsdb import Scan
    from plantdb.commons.fsdb import Fileset
    from plantdb.commons.fsdb import File
    from plantdb.commons.fsdb.core import MARKER_FILE_NAME
    path = Path(path)
    dirname, fname = path.parent, path.name
    id = Path(fname).stem
    with tempfile.TemporaryDirectory() as tmpdir:
        # Initialise a temporary `FSDB`:
        Path(f"{tmpdir}/{MARKER_FILE_NAME}").touch()  # add the db marker file
        db = FSDB(tmpdir)
        # Initialize a `Scan` instance:
        scan = Scan(db, "tmp")
        # Initialize a `Fileset` instance:
        fileset = Fileset(scan, dirname)
        # Initialize a `File` instance & return it:
        f = File(db=db, fileset=fileset, f_id=id)
        f.filename = fname
        f.metadata = None
    return f

is_radians Link

is_radians(angles)

Guess if the sequence of angles is in radians.

Parameters:

Name	Type	Description	Default
`angles`	`list of float`	Sequence of angle values.	required

Returns:

Type	Description
`bool`	`True` if the sequence is in radians, else `False.

Notes

This assumes that the angles can not be greater than 360 degrees or its equivalent in radians.

Source code in plantdb/commons/utils.py

def is_radians(angles):
    """Guess if the sequence of angles is in radians.

    Parameters
    ----------
    angles : list of float
        Sequence of angle values.

    Returns
    -------
    bool
        `True` if the sequence is in radians, else `False.

    Notes
    -----
    This assumes that the angles can not be greater than 360 degrees or its equivalent in radians.
    """
    from math import radians
    if all([angle < radians(360) for angle in angles]):
        return True
    else:
        return False

locate_task_filesets Link

locate_task_filesets(scan, tasks)

Map the task names to task filesets.

Parameters:

Name	Type	Description	Default
`scan`	`Scan`	A `Scan` instance from a local plant database (FSDB).	required
`tasks`	`list of str`	A list of task names to look up in the scan's list of filesets.	required

Returns:

Type	Description
`dict`	A task indexed dictionary of fileset ids, value may be "None" if no matching fileset was found.

Notes

If more than one fileset id matches a task name, only the first one (found) will be returned!

Examples:

>>> from plantdb.commons.utils import locate_task_filesets
>>> from plantdb.commons.fsdb import FSDB
>>> from plantdb.commons.test_database import test_database
>>> db = test_database('real_plant_analyzed')
>>> db.connect()
>>> scan = db.get_scan('real_plant_analyzed')
>>> tasks_fs = locate_task_filesets(scan, ['Masks', 'PointCloud', 'UnknownTask'])
>>> print(tasks_fs)
{'Masks': 'Masks_1__0__1__0____channel____rgb_5619aa428d', 'PointCloud': 'PointCloud_1_0_1_0_10_0_7ee836e5a9', 'UnknownTask': 'None'}
>>> db.disconnect()

Source code in plantdb/commons/utils.py

def locate_task_filesets(scan, tasks):
    """Map the task names to task filesets.

    Parameters
    ----------
    scan : plantdb.commons.fsdb.Scan
        A ``Scan`` instance from a local plant database (FSDB).
    tasks : list of str
        A list of task names to look up in the scan's list of filesets.

    Returns
    -------
    dict
        A task indexed dictionary of fileset ids, value may be "None" if no matching fileset was found.

    Notes
    -----
    If more than one fileset id matches a task name, only the first one (found) will be returned!

    Examples
    --------
    >>> from plantdb.commons.utils import locate_task_filesets
    >>> from plantdb.commons.fsdb import FSDB
    >>> from plantdb.commons.test_database import test_database
    >>> db = test_database('real_plant_analyzed')
    >>> db.connect()
    >>> scan = db.get_scan('real_plant_analyzed')
    >>> tasks_fs = locate_task_filesets(scan, ['Masks', 'PointCloud', 'UnknownTask'])
    >>> print(tasks_fs)
    {'Masks': 'Masks_1__0__1__0____channel____rgb_5619aa428d', 'PointCloud': 'PointCloud_1_0_1_0_10_0_7ee836e5a9', 'UnknownTask': 'None'}
    >>> db.disconnect()
    """
    # List all filesets in the scan dataset:
    fs_list = scan.list_filesets()
    # Find the fileset corresponding to the given list of tasks, if any:
    fileset_names = {}
    for task in tasks:
        try:
            # TODO: could be improved by using the saved config ('pipeline.toml') and recreate the name hash from luigi...
            fileset_names[task] = [fs for fs in fs_list if fs.startswith(task)][0]
        except IndexError:
            fileset_names[task] = "None"
    return fileset_names

partial_match Link

partial_match(reference, target, fuzzy=False)

Partial matching of a reference dictionary against a target, potentially using regexp.

Parameters:

Name	Type	Description	Default
`reference`	`dict`	The reference dictionary with partial information to test against the target.	required
`target`	`dict`	The target dictionary.	required
`fuzzy`	`bool`	Whether to use fuzzy matching or not, that is the use of regular expressions.	`False`

Returns:

Type	Description
`bool`	Whether the partial match is found or not.

Examples:

>>> from plantdb.commons.utils import partial_match
>>> ref = {"object": {"environment":"virtual"}}
>>> target = {'object': {'environment': 'virtual', 'plant_id': 'arabidopsis000', 'species': 'Arabidopsis Thaliana'}, 'scanner': {'workspace': {'x': [-200, 200], 'y': [-200, 200], 'z': [10, 1000]}}}
>>> partial_match(ref, target)
True
>>> ref = {"object": {"species":"Arabidopsis.*"}}
>>> partial_match(ref, target, fuzzy=True)
True
>>> ref = {"species":"Arabidopsis.*"}
>>> partial_match(ref, target, fuzzy=True)
False

Source code in plantdb/commons/utils.py

def partial_match(reference, target, fuzzy=False):
    """Partial matching of a reference dictionary against a target, potentially using regexp.

    Parameters
    ----------
    reference : dict
        The reference dictionary with partial information to test against the target.
    target : dict
        The target dictionary.
    fuzzy : bool
        Whether to use fuzzy matching or not, that is the use of regular expressions.

    Returns
    -------
    bool
        Whether the partial match is found or not.

    Examples
    --------
    >>> from plantdb.commons.utils import partial_match
    >>> ref = {"object": {"environment":"virtual"}}
    >>> target = {'object': {'environment': 'virtual', 'plant_id': 'arabidopsis000', 'species': 'Arabidopsis Thaliana'}, 'scanner': {'workspace': {'x': [-200, 200], 'y': [-200, 200], 'z': [10, 1000]}}}
    >>> partial_match(ref, target)
    True
    >>> ref = {"object": {"species":"Arabidopsis.*"}}
    >>> partial_match(ref, target, fuzzy=True)
    True
    >>> ref = {"species":"Arabidopsis.*"}
    >>> partial_match(ref, target, fuzzy=True)
    False

    """
    from re import match
    if isinstance(reference, dict) and isinstance(target, dict):
        return all(
            key in target and partial_match(value, target[key], fuzzy)
            for key, value in reference.items()
        )
    elif isinstance(reference, list) and isinstance(target, list):
        return len(reference) <= len(target) and all(
            any(partial_match(ref_item, target_item, fuzzy) for target_item in target)
            for ref_item in reference
        )
    elif fuzzy and isinstance(reference, str) and isinstance(target, str):
        return bool(match(reference, target))
    else:
        return reference == target

read_image_from_file Link

read_image_from_file(filename)

Read an image from a file and return it.

Parameters:

Name	Type	Description	Default
`filename`	`str or Path`	The path to the image to read.	required

Returns:

Name	Type	Description
`image`	`Image`	The loaded image.

Examples:

>>> from plantdb.commons.utils import read_image_from_file
>>> from plantdb.commons.test_database import test_database
>>> db = test_database('real_plant')
>>> db.connect()
>>> scan = db.get_scan('real_plant')
>>> fileset = scan.get_fileset('images')
>>> img_path = fileset.get_file('00000_rgb').path()
>>> image = read_image_from_file(img_path)
>>> print(image.size)
(1440, 1080)
>>> db.disconnect()

Source code in plantdb/commons/utils.py

def read_image_from_file(filename):
    """Read an image from a file and return it.

    Parameters
    ----------
    filename : str or pathlib.Path
        The path to the image to read.

    Returns
    -------
    image : PIL.Image.Image
        The loaded image.

    Examples
    --------
    >>> from plantdb.commons.utils import read_image_from_file
    >>> from plantdb.commons.test_database import test_database
    >>> db = test_database('real_plant')
    >>> db.connect()
    >>> scan = db.get_scan('real_plant')
    >>> fileset = scan.get_fileset('images')
    >>> img_path = fileset.get_file('00000_rgb').path()
    >>> image = read_image_from_file(img_path)
    >>> print(image.size)
    (1440, 1080)
    >>> db.disconnect()
    """
    image = Image.open(filename)
    image.load()
    return image

tmpdir_from_fileset Link

tmpdir_from_fileset(fileset)

Creates a temporary directory to host the Fileset object and write files.

Parameters:

Name	Type	Description	Default
`fileset`	`Fileset`	The fileset to use to create the temporary local database.	required

Returns:

Type	Description
`TemporaryDirectory`	The temporary directory hosting the fileset and file(s), if any.

Source code in plantdb/commons/utils.py

def tmpdir_from_fileset(fileset):
    """Creates a temporary directory to host the ``Fileset`` object and write files.

    Parameters
    ----------
    fileset : plantdb.commons.fsdb.Fileset
        The fileset to use to create the temporary local database.

    Returns
    -------
    tempfile.TemporaryDirectory
        The temporary directory hosting the fileset and file(s), if any.
    """
    tmpdir = tempfile.TemporaryDirectory()
    for f in fileset.get_files():
        filepath = Path(tmpdir.name) / f.filename
        to_file(f, filepath)
    return tmpdir

to_file Link

to_file(dbfile, path)

Write a dbfile to a file in the filesystem.

Parameters:

Name	Type	Description	Default
`dbfile`	`File`	The `File` instance to save under given `path`.	required
`path`	`Path or str`	The file path to use to save the `dbfile`.	required

Source code in plantdb/commons/utils.py

def to_file(dbfile, path):
    """Write a `dbfile` to a file in the filesystem.

    Parameters
    ----------
    dbfile : plantdb.commons.fsdb.File
        The ``File`` instance to save under given `path`.
    path : pathlib.Path or str
        The file path to use to save the `dbfile`.
    """
    b = dbfile.read_raw()
    path = Path(path)
    with path.open(mode="wb") as fh:
        fh.write(b)
    return