def __init__(self, labels=None):
# TODO Call the superclass with *args and **kwargs, then update everything to using ResourceBase
labels = labels or FilterSet()
# JSON-encoded list of label names, or space-delimited string of label names.
if isinstance(labels, str):
try:
labels = FilterSet(
Label(label) for label in json.loads(labels))
except json.decoder.JSONDecodeError:
labels = FilterSet(
Label(label) for label in labels.strip().split())
elif isinstance(labels, LabelSet):
labels = FilterSet(labels)
# Labels can be some other iterable than a list, but each label must be a Label or string.
elif hasattr(labels, "__iter__"):
labels = FilterSet(
label if isinstance(label, Label) else Label(label)
for label in labels)
else:
raise InvalidLabelException(
"Labels must be expressed as a whitespace-delimited string or an iterable of strings or Label instances."
)
super().__init__(labels)
def test_one(self):
"""Test that the `one` method returns one result without mutating its `FilterContainer` instance."""
filterables = FilterSet({FilterableThing(a=3, b=2), FilterableThing(a=3, b=99), FilterableThing(a=77)})
self.assertEqual(filterables.one(a__equals=77), FilterableThing(a=77))
# Test that `one` doesn't mutate the `FilterContainer` instance (i.e. `one` can be called again on the same
# instance and give the same result).
self.assertEqual(filterables.one(a__equals=77), FilterableThing(a=77))
def test_filter_with_filterables_of_differing_attributes_fails_if_setting_enabled(self):
"""Test filtering with filterables of differing attributes raises an error if any filterables lack the
filtered-for attribute when `ignore_items_without_attribute` is False.
"""
filter_set = FilterSet({FilterableThing(a=3), FilterableThing(b=90), FilterableThing(a=77)})
for kwarg in {"a__gt": 2}, {"b__equals": 90}, {"b__equals": 0}:
with self.assertRaises(AttributeError):
filter_set.filter(**kwarg, ignore_items_without_attribute=False)
def test_filter_with_filterables_of_differing_attributes(self):
"""Test filtering with filterables of differing attributes ignores the filterables lacking the filtered-for
attribute.
"""
filterables = {FilterableThing(a=3), FilterableThing(b=90), FilterableThing(a=77)}
filter_set = FilterSet(filterables)
self.assertEqual(filter_set.filter(a__gt=2), {FilterableThing(a=3), FilterableThing(a=77)})
self.assertEqual(filter_set.filter(b__equals=90), {FilterableThing(b=90)})
self.assertEqual(filter_set.filter(b__equals=0), set())
def __init__(
self,
path=None,
files=None,
recursive=False,
hypothetical=False,
id=None,
name=None,
tags=None,
labels=None,
):
super().__init__(name=name, id=id, tags=tags, labels=labels)
self.path = path or os.getcwd()
self.files = FilterSet()
self._recursive = recursive
self._hypothetical = hypothetical
self._cloud_metadata = {}
if files:
if not any((isinstance(files, list), isinstance(
files, set), isinstance(files, tuple))):
raise InvalidInputException(
"The `files` parameter of a `Dataset` must be an iterable of `Datafile` instances, strings, or "
f"dictionaries. Received {files!r} instead.")
self.files = self._instantiate_datafiles(files)
return
if storage.path.is_cloud_path(self.path):
self._instantiate_from_cloud(path=self.path)
else:
self._instantiate_from_local_directory(path=self.path)
if self._hypothetical:
logger.debug("Ignored stored metadata for %r.", self)
return
if self.metadata(include_sdk_version=False) != {
"name": name or self.name,
"id": id or self.id,
"tags": TagDict(tags),
"labels": LabelSet(labels),
}:
logger.warning(
"Overriding metadata given at instantiation with stored metadata for %r - set `hypothetical` to `True` "
"at instantiation to avoid this.",
self,
)
def test_ordering_and_checking_constant_increment_raises_error_if_increment_not_constant(self):
"""Test ordering by nested attributes while checking for a constant increment results in an error if the
increment is not the one given.
"""
things = [FilterableThing(a=1), FilterableThing(a=0), FilterableThing(a=5)]
with self.assertRaises(exceptions.BrokenSequenceException):
FilterSet(things).order_by("a", check_constant_increment=1)
def test_ordering_and_checking_start_value_raises_error_when_start_value_different(self):
"""Test ordering by nested attributes while checking the start value results in an error if the start value is
not the one given.
"""
things = [FilterableThing(a=1), FilterableThing(a=0)]
with self.assertRaises(exceptions.BrokenSequenceException):
FilterSet(things).order_by("a", check_start_value=3)
def test_order_by_list_attribute(self):
"""Test that ordering by list attributes orders by the size of the list."""
cats = [
FilterableThing(previous_names=["Scatta", "Catta"]),
FilterableThing(previous_names=["Kitty"]),
FilterableThing(previous_names=[]),
]
sorted_filter_set = FilterSet(cats).order_by("previous_names")
self.assertEqual(sorted_filter_set, FilterList(reversed(cats)))
def test_order_by_nested_attributes(self):
"""Test ordering by nested attributes."""
cats = [
FilterableThing(name=FilterableThing(first="Miaow", last="Miaow")),
FilterableThing(name=FilterableThing(first="Kitty", last="Miaow")),
]
sorted_filter_set = FilterSet(cats).order_by("name__first")
self.assertEqual(sorted_filter_set, FilterList([cats[1], cats[0]]))
def _instantiate_datafiles(self, files):
"""Instantiate and add the given files to a `FilterSet`.
:param iter(str|dict|octue.resources.datafile.Datafile) files:
:return octue.resources.filter_containers.FilterSet:
"""
with concurrent.futures.ThreadPoolExecutor() as executor:
return FilterSet(
executor.map(self._instantiate_datafile, copy.deepcopy(files)))
def test_filter_name_filters_include_extension(self):
"""Ensures that filters applied to the name will catch terms in the extension"""
files = [
Datafile(path="path-within-dataset/a_test_file.csv"),
Datafile(path="path-within-dataset/a_test_file.txt"),
]
self.assertEqual(
Dataset(files=files).files.filter(
name__icontains="txt").pop().path,
FilterSet({files[1]}).pop().local_path)
def _set_metadata(self, metadata):
"""Set the dataset's metadata.
:param dict metadata:
:return None:
"""
if "files" in metadata:
# There's no need to provide the hypothetical parameter here as this method is only used for
# non-hypothetical datasets.
self.files = FilterSet(
Datafile(path=path) for path in metadata["files"])
for attribute in self._METADATA_ATTRIBUTES:
if attribute not in metadata:
continue
if attribute == "id":
self._set_id(metadata["id"])
continue
setattr(self, attribute, metadata[attribute])
def _instantiate_from_local_directory(self, path):
"""Instantiate the dataset from a local directory.
:param str path: the path to a local directory
:return None:
"""
self.files = FilterSet()
for level, (directory_path, _, filenames) in enumerate(os.walk(path)):
for filename in filenames:
if filename == METADATA_FILENAME:
continue
if not self._recursive and level > 0:
break
self.files.add(
Datafile(path=os.path.join(directory_path, filename),
hypothetical=self._hypothetical))
if not self._hypothetical:
self._use_local_metadata()
def _instantiate_from_cloud(self, path):
"""Instantiate the dataset from a cloud directory.
:param str path: the cloud path to a directory in cloud storage
:return None:
"""
if not self._hypothetical:
self._use_cloud_metadata()
if not self.files:
bucket_name = storage.path.split_bucket_name_from_cloud_path(
path)[0]
self.files = FilterSet(
Datafile(path=storage.path.generate_gs_path(
bucket_name, blob.name),
hypothetical=self._hypothetical)
for blob in GoogleCloudStorageClient().scandir(
path,
recursive=self._recursive,
filter=(lambda blob:
(not blob.name.endswith(METADATA_FILENAME) and
SIGNED_METADATA_DIRECTORY not in blob.name)),
))
def test_filtering_with_multiple_filters(self):
"""Test that multiple filters can be specified in FilterSet.filter at once."""
filterables = {FilterableThing(a=3, b=2), FilterableThing(a=3, b=99), FilterableThing(a=77)}
self.assertEqual(FilterSet(filterables).filter(a__equals=3, b__gt=80), {FilterableThing(a=3, b=99)})
def test_instantiation_from_filter_set_of_labels(self):
"""Test that a LabelSet can be instantiated from a FilterSet of labels."""
label_set = LabelSet(
labels=FilterSet({Label(
"a"), Label("b-c"), Label("d-e-f")}))
self.assertEqual(label_set, self.LABEL_SET)
def test_order_by_in_reverse(self):
"""Test ordering in reverse works correctly."""
cats = [FilterableThing(age=5), FilterableThing(age=3), FilterableThing(age=4)]
sorted_filter_set = FilterSet(cats).order_by("age", reverse=True)
self.assertEqual(sorted_filter_set, FilterList([cats[0], cats[2], cats[1]]))
class Dataset(Labelable, Taggable, Serialisable, Identifiable, Hashable,
Metadata, CloudPathable):
"""A representation of a dataset. The default usage is to provide the path to a local or cloud directory and create
the dataset from the files it contains. Alternatively, the `files` parameter can be provided and only those files
are included. Either way, the `path` parameter should be explicitly set to something meaningful.
:param str|None path: the path to the dataset (defaults to the current working directory if none is given)
:param iter(str|dict|octue.resources.datafile.Datafile)|None files: the files belonging to the dataset
:param bool recursive: if `True`, include in the dataset all files in the subdirectories recursively contained within the dataset directory
:param bool hypothetical: if `True`, ignore any metadata stored for this dataset locally or in the cloud and use whatever is given at instantiation
:param str|None id: an optional UUID to assign to the dataset (defaults to a random UUID if none is given)
:param str|None name: an optional name to give to the dataset (defaults to the dataset directory name)
:param dict|octue.resources.tag.TagDict|None tags: key-value pairs with string keys conforming to the Octue tag format (see `TagDict`)
:param iter(str)|octue.resources.label.LabelSet|None labels: space-separated string of labels relevant to the dataset
:return None:
"""
_ATTRIBUTES_TO_HASH = ("files", )
_METADATA_ATTRIBUTES = ("id", "name", "tags", "labels")
# Paths to files are added to the serialisation in `Dataset.to_primitive`.
_SERIALISE_FIELDS = (*_METADATA_ATTRIBUTES, "path")
def __init__(
self,
path=None,
files=None,
recursive=False,
hypothetical=False,
id=None,
name=None,
tags=None,
labels=None,
):
super().__init__(name=name, id=id, tags=tags, labels=labels)
self.path = path or os.getcwd()
self.files = FilterSet()
self._recursive = recursive
self._hypothetical = hypothetical
self._cloud_metadata = {}
if files:
if not any((isinstance(files, list), isinstance(
files, set), isinstance(files, tuple))):
raise InvalidInputException(
"The `files` parameter of a `Dataset` must be an iterable of `Datafile` instances, strings, or "
f"dictionaries. Received {files!r} instead.")
self.files = self._instantiate_datafiles(files)
return
if storage.path.is_cloud_path(self.path):
self._instantiate_from_cloud(path=self.path)
else:
self._instantiate_from_local_directory(path=self.path)
if self._hypothetical:
logger.debug("Ignored stored metadata for %r.", self)
return
if self.metadata(include_sdk_version=False) != {
"name": name or self.name,
"id": id or self.id,
"tags": TagDict(tags),
"labels": LabelSet(labels),
}:
logger.warning(
"Overriding metadata given at instantiation with stored metadata for %r - set `hypothetical` to `True` "
"at instantiation to avoid this.",
self,
)
@property
def name(self):
"""Get the name of the dataset
:return str:
"""
if self._name:
return self._name
if self.exists_in_cloud:
return storage.path.split(self.path)[-1].split("?")[0]
return os.path.split(os.path.abspath(os.path.split(self.path)[-1]))[-1]
@name.setter
def name(self, name):
"""Set the name of the dataset.
:param str name:
:return None:
"""
self._name = name
@property
def exists_locally(self):
"""Return `True` if the dataset exists locally.
:return bool:
"""
return not self.exists_in_cloud
@property
def all_files_are_in_cloud(self):
"""Do all the files of the dataset exist in the cloud?
:return bool:
"""
return all(file.exists_in_cloud for file in self.files)
@property
def _metadata_path(self):
"""Get the path to the dataset's metadata file.
:return str:
"""
if self.exists_in_cloud:
if storage.path.is_url(self.path):
return self.path
return storage.path.join(self.path, METADATA_FILENAME)
return os.path.join(self.path, METADATA_FILENAME)
def __iter__(self):
yield from self.files
def __len__(self):
return len(self.files)
def __contains__(self, item):
return item in self.files
def __enter__(self):
"""Enter the dataset metadata updating context.
:return Dataset:
"""
return self
def __exit__(self, *args):
"""Update the cloud or local metadata for the dataset.
:return None:
"""
self.update_metadata()
def upload(self, cloud_path=None, update_cloud_metadata=True):
"""Upload a dataset to the given cloud path.
:param str|None cloud_path: cloud path to store dataset at (e.g. `gs://bucket_name/path/to/dataset`)
:param bool update_cloud_metadata: if `True`, update the metadata of the dataset in the cloud at upload time
:return str: cloud path for dataset
"""
cloud_path = self._get_cloud_location(cloud_path)
files_and_paths = []
for datafile in self.files:
if self.exists_in_cloud:
path_type = "cloud_path"
else:
path_type = "local_path"
datafile_path_relative_to_dataset = self._datafile_path_relative_to_self(
datafile, path_type=path_type)
files_and_paths.append((
datafile,
storage.path.join(
cloud_path,
*datafile_path_relative_to_dataset.split(os.path.sep)),
))
def upload_datafile(iterable_element):
"""Upload a datafile to the given cloud path.
:param tuple(octue.resources.datafile.Datafile, str) iterable_element:
:return str:
"""
datafile = iterable_element[0]
cloud_path = iterable_element[1]
datafile.upload(cloud_path=cloud_path)
return datafile.cloud_path
# Use multiple threads to significantly speed up file uploads by reducing latency.
with concurrent.futures.ThreadPoolExecutor() as executor:
for path in executor.map(upload_datafile, files_and_paths):
logger.debug("Uploaded datafile to %r.", path)
self.path = cloud_path
if update_cloud_metadata:
# If the dataset's metadata has been changed locally, update it in the cloud.
local_metadata = self.metadata(use_octue_namespace=False)
if self._cloud_metadata != local_metadata:
self.update_cloud_metadata()
return cloud_path
def update_metadata(self):
"""If the dataset is cloud-based, update its cloud metadata; otherwise, update its local metadata.
:return None:
"""
if self.exists_in_cloud:
self.update_cloud_metadata()
return
self.update_local_metadata()
def update_cloud_metadata(self):
"""Create or update the cloud metadata file for the dataset.
:return None:
"""
GoogleCloudStorageClient().upload_from_string(
string=json.dumps(
{"dataset": self.to_primitive(include_files=False)},
cls=OctueJSONEncoder),
cloud_path=self._metadata_path,
)
def update_local_metadata(self):
"""Create or update the local octue metadata file with the dataset's metadata.
:return None:
"""
existing_metadata_records = load_local_metadata_file(
self._metadata_path)
existing_metadata_records["dataset"] = self.to_primitive(
include_files=False)
os.makedirs(self.path, exist_ok=True)
overwrite_local_metadata_file(data=existing_metadata_records,
path=self._metadata_path)
def generate_signed_url(self, expiration=datetime.timedelta(days=7)):
"""Generate a signed URL for the dataset. This is done by uploading a uniquely named metadata file containing
signed URLs to the datasets' files and returning a signed URL to that metadata file.
:param datetime.datetime|datetime.timedelta expiration: the amount of time or date after which the URL should expire
:return str: the signed URL for the dataset
"""
if not self.exists_in_cloud:
raise CloudLocationNotSpecified(
f"{self!r} must exist in the cloud for a signed URL to be generated for it."
)
signed_metadata = self.to_primitive()
signed_metadata["files"] = [
datafile.generate_signed_url(expiration=expiration)
for datafile in self.files
]
path_to_signed_metadata_file = storage.path.join(
self.path, SIGNED_METADATA_DIRECTORY, coolname.generate_slug())
storage_client = GoogleCloudStorageClient()
storage_client.upload_from_string(
string=json.dumps(signed_metadata, cls=OctueJSONEncoder),
cloud_path=path_to_signed_metadata_file,
)
return storage_client.generate_signed_url(
cloud_path=path_to_signed_metadata_file, expiration=expiration)
def add(self, datafile, path_in_dataset=None):
"""Add a datafile to the dataset. If the datafile's location is outside the dataset, it is copied to the dataset
root or to the `path_in_dataset` if provided.
:param octue.resources.datafile.Datafile datafile: the datafile to add to the dataset
:param str|None path_in_dataset: if provided, set the datafile's local path to this path within the dataset
:raise octue.exceptions.InvalidInputException: if the datafile is not a `Datafile` instance
:return None:
"""
if not isinstance(datafile, Datafile):
raise InvalidInputException(
f"{datafile!r} must be of type `Datafile` to add it to the dataset."
)
if self.exists_in_cloud:
new_cloud_path = storage.path.join(
self.path, path_in_dataset or datafile.name)
# Add a cloud datafile to a cloud dataset.
if datafile.exists_in_cloud:
if datafile.cloud_path != new_cloud_path and not datafile.cloud_path.startswith(
self.path):
datafile.upload(new_cloud_path)
self.files.add(datafile)
return
# Add a local datafile to a cloud dataset.
datafile.upload(new_cloud_path)
self.files.add(datafile)
return
new_local_path = os.path.join(self.path, path_in_dataset
or datafile.name)
# Add a cloud datafile to a local dataset.
if datafile.exists_in_cloud:
datafile.download(local_path=new_local_path)
self.files.add(datafile)
return
# Add a local datafile to a local dataset.
if datafile.local_path != new_local_path and not datafile.local_path.startswith(
self.path):
datafile.local_path = new_local_path
self.files.add(datafile)
def get_file_by_label(self, label):
"""Get a single datafile from a dataset by filtering for files with the provided label.
:param str label: the label to filter for
:raise octue.exceptions.UnexpectedNumberOfResultsException: if zero or more than one results satisfy the filters
:return octue.resources.datafile.DataFile:
"""
return self.files.one(labels__contains=label)
def download(self, local_directory=None):
"""Download all files in the dataset into the given local directory. If no path to a local directory is given,
the files will be downloaded to temporary locations.
:param str|None local_directory:
:return None:
"""
if not self.exists_in_cloud:
raise CloudLocationNotSpecified(
f"You can only download files from a cloud dataset. This dataset's path is {self.path!r}."
)
local_directory = local_directory or tempfile.TemporaryDirectory().name
files_and_paths = []
for file in self.files:
if not file.exists_in_cloud:
continue
path_relative_to_dataset = self._datafile_path_relative_to_self(
file, path_type="cloud_path")
local_path = os.path.abspath(
os.path.join(local_directory,
*path_relative_to_dataset.split("/")))
files_and_paths.append((file, local_path))
def download_datafile(iterable_element):
"""Download a datafile to the given path.
:param tuple(octue.resources.datafile.Datafile, str) iterable_element:
:return str:
"""
datafile = iterable_element[0]
local_path = iterable_element[1]
return datafile.download(local_path=local_path)
# Use multiple threads to significantly speed up files downloads by reducing latency.
with concurrent.futures.ThreadPoolExecutor() as executor:
for path in executor.map(download_datafile, files_and_paths):
logger.debug("Downloaded datafile to %r.", path)
logger.info("Downloaded %r to %r.", self, local_directory)
def to_primitive(self, include_files=True):
"""Convert the dataset to a dictionary of primitives, converting its files into their paths for a lightweight
serialisation.
:param bool include_files: if `True`, include the `files` parameter in the dictionary
:return dict:
"""
serialised_dataset = super().to_primitive()
if self.exists_in_cloud:
path_type = "cloud_path"
else:
path_type = "local_path"
if include_files:
serialised_dataset["files"] = sorted(
getattr(datafile, path_type) for datafile in self.files)
return serialised_dataset
def _instantiate_from_cloud(self, path):
"""Instantiate the dataset from a cloud directory.
:param str path: the cloud path to a directory in cloud storage
:return None:
"""
if not self._hypothetical:
self._use_cloud_metadata()
if not self.files:
bucket_name = storage.path.split_bucket_name_from_cloud_path(
path)[0]
self.files = FilterSet(
Datafile(path=storage.path.generate_gs_path(
bucket_name, blob.name),
hypothetical=self._hypothetical)
for blob in GoogleCloudStorageClient().scandir(
path,
recursive=self._recursive,
filter=(lambda blob:
(not blob.name.endswith(METADATA_FILENAME) and
SIGNED_METADATA_DIRECTORY not in blob.name)),
))
def _instantiate_from_local_directory(self, path):
"""Instantiate the dataset from a local directory.
:param str path: the path to a local directory
:return None:
"""
self.files = FilterSet()
for level, (directory_path, _, filenames) in enumerate(os.walk(path)):
for filename in filenames:
if filename == METADATA_FILENAME:
continue
if not self._recursive and level > 0:
break
self.files.add(
Datafile(path=os.path.join(directory_path, filename),
hypothetical=self._hypothetical))
if not self._hypothetical:
self._use_local_metadata()
def _instantiate_datafiles(self, files):
"""Instantiate and add the given files to a `FilterSet`.
:param iter(str|dict|octue.resources.datafile.Datafile) files:
:return octue.resources.filter_containers.FilterSet:
"""
with concurrent.futures.ThreadPoolExecutor() as executor:
return FilterSet(
executor.map(self._instantiate_datafile, copy.deepcopy(files)))
def _instantiate_datafile(self, file):
"""Instantiate a datafile from multiple input formats.
:param str|dict|octue.resources.datafile.Datafile file:
:return octue.resources.datafile.Datafile:
"""
if isinstance(file, Datafile):
return file
if isinstance(file, str):
return Datafile(path=file, hypothetical=self._hypothetical)
return Datafile.deserialise(file)
def _get_cloud_metadata(self):
"""Get the cloud metadata for the given dataset if a dataset metadata file has previously been uploaded.
:return None:
"""
if storage.path.is_url(self.path):
try:
self._cloud_metadata = requests.get(self.path).json()
except requests.exceptions.ConnectionError:
pass
return
storage_client = GoogleCloudStorageClient()
if not storage_client.exists(cloud_path=self._metadata_path):
return
self._cloud_metadata = json.loads(
storage_client.download_as_string(
cloud_path=self._metadata_path)).get("dataset", {})
def _use_cloud_metadata(self):
"""Update the dataset instance's metadata from the metadata file located in its cloud directory. If no metadata
is stored for the dataset, do nothing.
:return None:
"""
self._get_cloud_metadata()
if not self._cloud_metadata:
return
self._set_metadata(self._cloud_metadata)
def _use_local_metadata(self):
"""Update the dataset instance's metadata from the local metadata records file. If no metadata is stored for the
dataset, do nothing.
:return None:
"""
local_metadata = load_local_metadata_file(self._metadata_path)
dataset_metadata = local_metadata.get("dataset", {})
if not dataset_metadata:
return
self._set_metadata(dataset_metadata)
def _set_metadata(self, metadata):
"""Set the dataset's metadata.
:param dict metadata:
:return None:
"""
if "files" in metadata:
# There's no need to provide the hypothetical parameter here as this method is only used for
# non-hypothetical datasets.
self.files = FilterSet(
Datafile(path=path) for path in metadata["files"])
for attribute in self._METADATA_ATTRIBUTES:
if attribute not in metadata:
continue
if attribute == "id":
self._set_id(metadata["id"])
continue
setattr(self, attribute, metadata[attribute])
def _datafile_path_relative_to_self(self, datafile, path_type):
"""Get the path of the given datafile relative to the dataset.
:param octue.resources.datafile.Datafile datafile: the datafile to get the relative path for
:param str path_type: the datafile path type to use to calculate the relative path - one of "cloud_path" or "local_path"
:return str: the relative path
"""
if storage.path.is_url(self.path):
dataset_path = self.path.split(SIGNED_METADATA_DIRECTORY)[0].strip(
"/")
else:
dataset_path = self.path
datafile_path = getattr(datafile, path_type)
if storage.path.is_url(datafile_path):
datafile_path = datafile_path.split("?")[0]
return storage.path.relpath(datafile_path, dataset_path)
def test_order_by_with_int_attribute(self):
"""Test ordering a FilterSet by an integer attribute returns an appropriately ordered FilterList."""
cats = [FilterableThing(age=5), FilterableThing(age=4), FilterableThing(age=3)]
sorted_filter_set = FilterSet(cats).order_by("age")
self.assertEqual(sorted_filter_set, FilterList(reversed(cats)))
def test_ordering_and_checking_start_value(self):
"""Test ordering by nested attributes and checking the start value."""
things = [FilterableThing(a=1), FilterableThing(a=0)]
sorted_filter_set = FilterSet(things).order_by("a", check_start_value=0)
self.assertEqual(sorted_filter_set, FilterList([things[1], things[0]]))
def test_ordering_by_a_non_existent_attribute(self):
"""Ensure an error is raised if ordering is attempted by a non-existent attribute."""
filter_set = FilterSet([FilterableThing(age=5), FilterableThing(age=4), FilterableThing(age=3)])
with self.assertRaises(exceptions.InvalidInputException):
filter_set.order_by("dog-likeness")
def test_order_by_with_string_attribute(self):
"""Test ordering a FilterSet by a string attribute returns an appropriately ordered FilterList."""
cats = [FilterableThing(name="Zorg"), FilterableThing(name="James"), FilterableThing(name="Princess Carolyn")]
sorted_filter_set = FilterSet(cats).order_by("name")
self.assertEqual(sorted_filter_set, FilterList([cats[1], cats[2], cats[0]]))
def test_one_with_no_filter(self):
"""Test that the `one` method can be used with no filters on a FilterSet with only one item to return that item."""
filterable_thing = FilterableThing(a=3, b=2)
filterables = FilterSet({filterable_thing})
self.assertEqual(filterables.one(), filterable_thing)
def test_ordering_and_checking_constant_increment(self):
"""Test ordering by nested attributes while checking for a constant increment."""
things = [FilterableThing(a=1), FilterableThing(a=0), FilterableThing(a=2)]
sorted_filter_set = FilterSet(things).order_by("a", check_constant_increment=1)
self.assertEqual(sorted_filter_set, FilterList([things[1], things[0], things[2]]))
def test_one_fails_if_more_than_one_result(self):
"""Test that the `one` method raises an error if there is more than one result."""
filterables = FilterSet({FilterableThing(a=3, b=2), FilterableThing(a=3, b=99), FilterableThing(a=77)})
with self.assertRaises(exceptions.UnexpectedNumberOfResultsException):
filterables.one(a__equals=3)
def test_instantiation_from_filter_set_of_strings(self):
"""Test that a LabelSet can be instantiated from a FilterSet of strings."""
label_set = LabelSet(labels=FilterSet({"a", "b-c", "d-e-f"}))
self.assertEqual(label_set, self.LABEL_SET)
def test_error_raised_if_any_items_are_not_filterable(self):
"""Test that an error is raised if any items in the FilterSet are not of type Filterable."""
filter_set = FilterSet([1, 2, 3])
with self.assertRaises(TypeError):
filter_set.filter(a__equals=2)
