def structural(p):
print(len(p))
dataclasses.fields(p)
dataclasses.asdict(p)
dataclasses.astuple(p)
dataclasses.replace(p)
def insert(self, product: Product, *, overwrite=False):
table = tables[product.__class__]
c = self.connection.cursor()
data = astuple(product)
if len(data) > 1:
part_2 = ", ?" * (len(data) - 1)
part_2 = f"(?{part_2})"
else:
part_2 = "?"
if not overwrite:
part_1 = f"INSERT INTO {table} VALUES "
order = part_1 + part_2
c.execute(order, data)
else:
part_1 = f"INSERT OR REPLACE INTO {table} VALUES "
order = part_1 + part_2
c.execute(order, data)
self.connection.commit()
def copy_files(exp_params: ExperimentParameters) -> Tuple[Path, Path]:
_, measure, strategy, weight, learning_based, auc = astuple(exp_params)
filepath = exp_params.filepath
path: Path = filepath.parent
file, filename = filepath.name, filepath.stem
if learning_based:
new_folder_path = path / f"{filename}_{measure}_{strategy}_{weight}_{auc}"
else:
new_folder_path = path / f"{filename}_{measure}"
new_folder_path.mkdir()
new_filepath = new_folder_path / file
copyfile(str(filepath), new_filepath)
old_clustered_filepath = path / f"{filename}_{measure}_{strategy}_{weight}_{auc}_clustered.arff"
new_clustered_filepath = new_folder_path / f"{filename}.clus"
copyfile(str(old_clustered_filepath), new_clustered_filepath)
old_clustered_filepath.unlink()
return new_filepath, new_clustered_filepath
def dataclassToJson(objectClass: dataclass,
objects: List[dataclass],
filePrefix: str,
sort=True) -> None:
try:
sort and objects.sort()
except TypeError: # '<' not supported between instances of 'objectClass' and 'objectClass'
print(f'Warning: {objectClass.__name__} is not sorted.')
json.dump([asdict(o) for o in objects],
open(f'{filePrefix}.json', 'w'),
indent=4)
json.dump(
{
'keys': [f.name for f in fields(objectClass)],
'values': [astuple(o) for o in objects]
},
open(f'{filePrefix}.min.json', 'w'),
separators=(',', ':'))
def __handle_single_requirement(
self, package: PackageInfo,
requirement: Requirement) -> Tuple[bool, str]:
package_name, latest_version, _, current_version, _ = astuple(package)
if requirement.ignore:
message = Messages.IGNORED.format(package=requirement.package)
return False, message
if not all([latest_version, current_version]):
message = Messages.CANNOT_FETCH.format(package=package_name,
version=requirement.version)
return False, message
if latest_version > current_version:
return self._is_rotten(package)
message = Messages.NOT_ROTTEN.format(package=requirement.package,
version=str(current_version))
return False, message
def as_path(self) -> SVGPath:
*shape_fields, x, y, w, h, rx, ry = dataclasses.astuple(self)
path = SVGPath()
path.M(x + rx, y)
path.H(x + w - rx)
if rx > 0:
path.A(rx, ry, x + w, y + ry)
path.V(y + h - ry)
if rx > 0:
path.A(rx, ry, x + w - rx, y + h)
path.H(x + rx)
if rx > 0:
path.A(rx, ry, x, y + h - ry)
path.V(y + ry)
if rx > 0:
path.A(rx, ry, x + rx, y)
path.end()
path._copy_common_fields(*shape_fields)
return path
def CompileSpec(*args, **kwargs):
"""
CompileSpec specifies the model information.
Example:
cs = CompileSpec(
inputs=(
TensorSpec(
shape=[1, 3, 224, 224],
),
),
outputs=(
TensorSpec(
shape=[1, 1000],
),
),
backend=CoreMLComputeUnit.CPU,
allow_low_precision=True,
),
"""
return astuple(_CompileSpec(*args, **kwargs))
def simulate(initial_state, config, intervention=None, seed=None):
"""Simulate a run of the Zika simulator model.
The simulation starts at initial_state at time 0, and evolves the state
using dynamics whose parameters are specified in config.
Parameters
----------
initial_state: `whynot.simulators.zika.State`
Initial State object, which is used as x_{t_0} for the simulator.
config: `whynot.simulators.zika.Config`
Config object that encapsulates the parameters that define the dynamics.
intervention: `whynot.simulators.zika.Intervention`
Intervention object that specifies what, if any, intervention to perform.
seed: int
Seed to set internal randomness. The simulator is deterministic, so
the seed parameter is ignored.
Returns
-------
run: `whynot.dynamics.Run`
Rollout of the model.
"""
# Simulator is deterministic, so seed is ignored
# pylint: disable-msg=unused-argument
t_eval = np.arange(
config.start_time, config.end_time + config.delta_t, config.delta_t
)
solution = odeint(
dynamics,
y0=dataclasses.astuple(initial_state),
t=t_eval,
args=(config, intervention),
rtol=config.rtol,
atol=config.atol,
)
states = [initial_state] + [State(*state) for state in solution[1:]]
return wn.dynamics.Run(states=states, times=t_eval)
def add_to_list(
current_list: HmrcTransactionLog,
date_index: datetime.date,
symbol: str,
quantity: Decimal,
amount: Decimal,
fees: Decimal,
) -> None:
"""Add entry to given transaction log."""
if date_index not in current_list:
current_list[date_index] = {}
if symbol not in current_list[date_index]:
current_list[date_index][symbol] = HmrcTransactionData(
quantity=Decimal(0), amount=Decimal(0), fees=Decimal(0))
current_quantity, current_amount, current_fees = astuple(
current_list[date_index][symbol])
current_list[date_index][symbol] = HmrcTransactionData(
quantity=current_quantity + quantity,
amount=current_amount + amount,
fees=current_fees + fees,
)
def get_line_segment(self, image):
""" Returns the line segment that intersects the image. """
# TODO: Fix for when line more closely aligns with y-axis
m, n = image.shape[:2]
a, b, c = astuple(self)
if abs(a) < 0.01:
print("Returning line segment parallel to y-axis.")
x0, y0 = (0, int(-c / b))
x1, y1 = (n, int(-c / b))
return ((x0, y0), (x1, y1))
if abs(b) < 0.01:
print("Returning line segment parallel to x-axis.")
x0, y0 = (int(-c / a), 0)
x1, y1 = (int(-c / a), m)
return ((x0, y0), (x1, y1))
x0, y0 = (0, int(-c / b))
x1, y1 = (n, int(-(a * n + c) / b))
return ((x0, y0), (x1, y1))
def insert_tile(tile: Tile) -> int:
sql = f"""INSERT INTO tiles({','.join([*asdict(tile)])})
VALUES ({','.join(['%s' for _ in range(len(astuple(tile)))])})
RETURNING id"""
conn = None
tileId: int = -1
try:
conn = connect_db()
cur = conn.cursor()
cur.execute(sql, astuple(tile))
tileId = cur.fetchone()[0]
conn.commit()
cur.close()
return tileId
except (Exception, psycopg2.DatabaseError) as error:
print("insert tile: ", error)
finally:
if conn is not None:
conn.close()
def step(self, op, num):
x, y, wpdx, wpdy = dataclasses.astuple(self)
if op in DIRS:
dx, dy = DIRS[op]
wpdx += dx * num
wpdy += dy * num
elif op == "F":
x += wpdx * num
y += wpdy * num
else:
assert op in "LR"
assert num in [90, 180, 270]
if num == 180:
wpdx *= -1
wpdy *= -1
elif (op, num) in [("R", 90), ("L", 270)]:
wpdx, wpdy = wpdy, -wpdx
elif (op, num) in [("L", 90), ("R", 270)]:
wpdx, wpdy = -wpdy, wpdx
return self.__class__(x, y, wpdx, wpdy)
def write_results(self, prices: List):
logging.info("Writing %s lines", len(prices))
conn = None
try:
# read database configuration
# connect to the PostgreSQL database
conn = psycopg2.connect(host="postgres", database="fokko", user="******", password="******")
# create a new cursor
cur = conn.cursor()
for price in prices:
# execute the INSERT statement
cur.execute(self.sql, astuple(price))
# commit the changes to the database
conn.commit()
# close communication with the database
cur.close()
except (Exception, psycopg2.DatabaseError) as error:
print(error)
finally:
if conn is not None:
conn.close()
def step(self, op, num):
x, y, dirx, diry = dataclasses.astuple(self)
if op in DIRS:
dx, dy = DIRS[op]
x += dx * num
y += dy * num
elif op == "F":
x += dirx * num
y += diry * num
else:
assert op in "LR"
assert num in [90, 180, 270]
if num == 180:
dirx *= -1
diry *= -1
elif (op, num) in [("R", 90), ("L", 270)]:
dirx, diry = RTURNS[dirx, diry]
elif (op, num) in [("L", 90), ("R", 270)]:
dirx, diry = LTURNS[dirx, diry]
return self.__class__(x, y, dirx, diry)
def merge_data(parsed: List[CountyResult],
gdf: gpd.GeoDataFrame) -> gpd.GeoDataFrame:
"""
Merge together the parsed data from `parsed_data` with the GeoDataFrame
from the Census. Note that this standardizes for the column headers the
javascript app expects.
"""
df = pd.DataFrame(
[dataclasses.astuple(row) for row in parsed],
columns=[field.name for field in dataclasses.fields(CountyResult)],
)
gdf = gdf.merge(df.drop(columns=["state"]), left_on="id", right_on="fips")
gdf = gdf.rename(
columns={
"clinton_vote": "dem",
"trump_vote": "gop",
"stein_vote": "grn",
"johnson_vote": "lib",
"mcmullin_vote": "una",
})
return gdf
def encode_fsobject_memo(memo: FilesystemObjectMemo) -> bytes:
# Return a representation of *memo* as marshal-encoded tuple.
if not isinstance(memo, FilesystemObjectMemo):
raise TypeError
if memo.stat is None: # filesystem object did not exist
return marshal.dumps(()) # != b''
t = dataclasses.astuple(memo.stat)
if not all(isinstance(f, int) for f in t):
raise TypeError
if not stat.S_ISLNK(memo.stat.mode) and memo.symlink_target is not None:
raise ValueError
if stat.S_ISLNK(memo.stat.mode) and not isinstance(memo.symlink_target, str):
raise TypeError
return marshal.dumps((
memo.stat.mode, memo.stat.size, memo.stat.mtime_ns, memo.stat.uid, memo.stat.gid,
memo.symlink_target))
def normalize_point_cluster(self,
point_cluster: PointCluster) -> PointCluster:
"""
normalize point-cluster such that,
- jimage of cluster[0] is (0, 0, 0)
- points are in ascending order
"""
# sort sites by (site_index, jimage)
tuple_points = sorted(astuple(p) for p in point_cluster.points)
points = [
DerivativeSite(site_index, jimage)
for site_index, jimage in tuple_points
]
offset = np.array(points[0].jimage)
shifed_new_points = [
DerivativeSite(
p.site_index,
tuple((np.array(p.jimage, dtype=int) - offset).tolist()))
for p in points
]
return PointCluster(shifed_new_points)
def eval_features_once(self, pool, layers, tile_size=512):
"""Computes the set of feature maps for an image."""
img_size = np.array(self.img.shape[-2:])
ntiles = (img_size - 1) // tile_size + 1
tile_size = img_size // ntiles
if np.prod(ntiles) > 1:
print('Using %dx%d tiles of size %dx%d.' %
(ntiles[1], ntiles[0], tile_size[1], tile_size[0]))
features = {}
for layer in layers:
scale, channels = self.layer_info(layer)
shape = (channels, ) + tuple(np.int32(np.ceil(img_size / scale)))
features[layer] = np.zeros(shape, dtype=np.float32)
for y in range(ntiles[0]):
for x in range(ntiles[1]):
xy = np.array([y, x])
start = xy * tile_size
end = start + tile_size
if y == ntiles[0] - 1:
end[0] = img_size[0]
if x == ntiles[1] - 1:
end[1] = img_size[1]
tile = self.img[:, start[0]:end[0], start[1]:end[1]]
pool.ensure_healthy()
pool.request(
FeatureMapRequest(start, SharedNDArray.copy(tile), layers))
pool.reset_next_worker()
for _ in range(np.prod(ntiles)):
start, feats_tile = astuple(pool.resp_q.get())
for layer, feat in feats_tile.items():
scale, _ = self.layer_info(layer)
start_f = start // scale
end_f = start_f + np.array(feat.array.shape[-2:])
features[layer][:, start_f[0]:end_f[0],
start_f[1]:end_f[1]] = feat.array
feat.unlink()
return features
def _make_fundamental(obj, repeatable):
if repeatable:
if isinstance(obj, str):
# https://medium.com/@bdov_/https-medium-com-bdov-python-objects-part-iii-string-interning-625d3c7319de
# CPython implementation of marshal.dumps(): https://github.com/python/cpython/blob/master/Python/marshal.c
return b's' + obj.encode()
elif isinstance(obj, bytes):
return b'b' + obj
if is_immutable_fundamental(obj):
return obj
r = repeatable
if isinstance(obj, collections.abc.Mapping
): # note: loses order of collections.OrderedDict
if r:
return tuple(
sorted((_make_fundamental(k, r), _make_fundamental(v, r))
for k, v in obj.items()))
return {
_make_fundamental(k, r): _make_fundamental(v, r)
for k, v in obj.items()
}
if isinstance(obj, (set, frozenset)):
obj = frozenset(_make_fundamental(k, r) for k in obj)
if not r:
return obj
return tuple(sorted(obj))
if isinstance(obj, collections.abc.Iterable):
return tuple(_make_fundamental(k, r) for k in obj)
if dataclasses.is_dataclass(obj) and not isinstance(obj, type):
return dataclasses.astuple(obj)
raise TypeError
def write_records_as_csv(records: list, outfile: str) -> None:
"""
Write the records as a csv file
Args:
records: a list of dataclass objects or namedtuples
(anything with a '_fields' attribute)
outfile: the path to save the csv file
"""
r0 = records[0]
if dataclasses.is_dataclass(r0):
column_names = [field.name for field in dataclasses.fields(r0)]
records = [dataclasses.astuple(rec) for rec in records]
elif hasattr(r0, "_fields"):
column_names = r0._fields
else:
raise TypeError("records should be a namedtuple or a dataclass")
f = open(outfile, 'w', newline='', encoding='utf-8')
w = _csv.writer(f)
w.writerow(column_names)
for record in records:
w.writerow(record)
f.close()
def run_worker(args):
(
environment_name,
species_name,
batch_num,
worker_num,
num_workers,
) = args
# Go through every replay and sum up stats
env_class = get_env_module(environment_name)
replay_directory = find_batch_directory(environment_name, species_name, batch_num)
ws = WorkerStats()
for agent_replay in iter_replay_data(
replay_directory,
env_class.State,
worker_num,
num_workers,
):
ws.total_mcts_considerations += agent_replay.total_mcts_considerations()
ws.num_games += 1
ws.num_positions += len(agent_replay.positions)
return astuple(ws)
def fetchall(conn: Any,
table: str,
where: Optional[Dataclass],
columns: Union[str, List[str]] = "*") -> Optional[Generator]:
"""Run a SELECT on a table. Returns all the results as a tuple!"""
if not any(astuple(where)):
return None
where_ = {
TABLEMAP[table].get(k): v
for k, v in asdict(where).items()
if v is not None and k in TABLEMAP[table]
}
sql = ""
if columns != "*":
sql = f"SELECT {', '.join(columns)} FROM {table} "
else:
sql = f"SELECT * FROM {table} "
if where:
sql += "WHERE "
sql += " AND ".join(f"{k} = " "%s" for k in where_.keys())
with conn.cursor() as cursor:
cursor.execute(sql, tuple(where_.values()))
return (DATACLASSMAP[table](*record) for record in cursor.fetchall())
def test_TH1Axis_determination(logging_mixin, create_hist_axis_range,
axis_type, axis, hist_to_test, test_root_hists):
""" Test TH1 axis determination in the HistAxisRange object. """
import ROOT
axis_map = {
"x_axis": ROOT.TH1.GetXaxis,
"y_axis": ROOT.TH1.GetYaxis,
"z_axis": ROOT.TH1.GetZaxis,
}
axis = axis_map[axis]
# Get the HistAxisRange object
obj, object_args = create_hist_axis_range
# Insert the proepr axis type
obj.axis_type = axis_type
# Determine the test hist
hist = dataclasses.astuple(test_root_hists)[hist_to_test]
# Check that the axis retrieved by the specified function is the same
# as that retrieved by the HistAxisRange object.
# NOTE: GetZaxis() (for example) is still valid for a TH1. It is a minimal axis
# object with 1 bin. So it is fine to check for equivalnce for axes that
# don't really make sense in terms of a hist's dimensions.
assert axis(hist) == obj.axis(hist)
def __getitem__(self: "CacheDict[KT, VT]", key: KT, /) -> VT:
log.debug("get [%r] key: [%r]", ReprWrapper(self), key)
if not isinstance(key, self.mapping.KeyType):
raise CacheDictKeyTypeException(
{
"key": key,
"key_type": type(key),
"KT": self._get_key_type_mapping(),
},
)
select_stmt = self.mapping.select_statement()
cursor = self._execute(select_stmt, dataclasses.astuple(key), op="select")
res: typing.Optional[CacheDictRow[KT, VT]] = cursor.fetchone()
if not res:
raise CacheDictNoSuchKeyException(
{"key": key, "table": self.mapping.table_ident},
)
elif res.value is None:
raise CacheDictNoneReturnedException(
{"key": key, "table": self.mapping.table_ident},
)
else:
return res.value
def __call__(self, sample: Sample):
image, landmarks = dataclasses.astuple(sample)
h, w = image.shape[:2]
if isinstance(self.output_size, int):
if h > w:
new_h, new_w = self.output_size * (h / w), self.output_size
else:
new_h, new_w = self.output_size, self.output_size * (w / h)
else:
new_h, new_w = self.output_size
new_h, new_w = int(new_h), int(new_w)
new_img = transform.resize(image, (new_h, new_w))
# h and w are swapped for landmarks because for images,
# x and y axis are axis 1 and axis 0 respectively
new_landmarks = landmarks * [new_w / w, new_h / h]
return Sample(
image=new_img,
landmarks=new_landmarks
)
def simulate(initial_state, config, intervention=None, seed=None):
"""Simulate a run of the Chen et al. model opioid epidemic model.
Parameters
----------
initial_state: whynot.simulator.opioid.State
Initial state of the dynamics
config: whynot.simulator.opioid.Config
Config object to determine simulation dynamics.
intervention: whynot.simulator.opioid.Intervention
(Optional) Intervention object to determine what, if any,
intervention to perform during the rollout of the dynamics.
seed: int
The simulator is deterministic, so the seed parameter is ignored.
Returns
-------
run: whynot.dynamics.Run
Run object produced by running simulate for the opioid simulator
"""
# pylint: disable-msg=unused-argument
# pylint:disable-msg=no-member
t_eval = np.arange(config.start_time, config.end_time + config.delta_t,
config.delta_t)
solution = odeint(
dynamics,
y0=dataclasses.astuple(initial_state),
t=t_eval,
args=(config, intervention),
rtol=1e-4,
atol=1e-4,
)
states = [initial_state] + [State(*state) for state in solution[1:]]
return wn.dynamics.Run(states=states, times=t_eval)
def _async_process_discovered_usb_device(self, device: USBDevice) -> None:
"""Process a USB discovery."""
_LOGGER.debug("Discovered USB Device: %s", device)
device_tuple = dataclasses.astuple(device)
if device_tuple in self.seen:
return
self.seen.add(device_tuple)
matched = [matcher for matcher in self.usb if _is_matching(device, matcher)]
if not matched:
return
service_info = UsbServiceInfo(
device=device.device,
vid=device.vid,
pid=device.pid,
serial_number=device.serial_number,
manufacturer=device.manufacturer,
description=device.description,
)
sorted_by_most_targeted = sorted(matched, key=lambda item: -len(item))
most_matched_fields = len(sorted_by_most_targeted[0])
for matcher in sorted_by_most_targeted:
# If there is a less targeted match, we only
# want the most targeted match
if len(matcher) < most_matched_fields:
break
discovery_flow.async_create_flow(
self.hass,
matcher["domain"],
{"source": config_entries.SOURCE_USB},
service_info,
)
async def test_unordered_UTXOs():
api = API()
api.initialize()
mempool = MemPool(coin, api)
event = Event()
async with OldTaskGroup() as group:
await group.spawn(mempool.keep_synchronized, event)
await event.wait()
await group.cancel_remaining()
# Check the default dict is handled properly
prior_len = len(mempool.hashXs)
assert await mempool.unordered_UTXOs(os.urandom(HASHX_LEN)) == []
assert prior_len == len(mempool.hashXs)
# Test all hashXs
utxos = api.UTXOs()
for hashX in api.hashXs:
mempool_result = await mempool.unordered_UTXOs(hashX)
our_result = utxos.get(hashX, [])
assert set(our_result) == {
dataclasses.astuple(mr)
for mr in mempool_result
}
def voxelize_batch(b: batched_example.BatchedExample,
voxelization_config: configuration.VoxelizationConfig):
"""Voxelizes a batched example with the given settings."""
with t.no_grad():
voxel_content_fn = {
configuration.TaskType.SEMANTIC:
batched_example.VoxelContentSemanticLabel(b.mesh_labels),
configuration.TaskType.FG_BG:
batched_example.voxel_content_1
}[voxelization_config.task_type]
resolution = dataclasses.astuple(voxelization_config.resolution)
res_mul = voxelization_config.voxelization_image_resolution_multiplier
res_cons = voxelization_config.conservative_rasterization
depth_mul = voxelization_config.voxelization_projection_depth_multiplier
b = batched_example.voxelize(
b,
resolution=resolution,
voxel_content_fn=voxel_content_fn,
sub_grid_sampling=voxelization_config.sub_grid_sampling,
image_resolution_multiplier=res_mul,
conservative_rasterization=res_cons,
projection_depth_multiplier=depth_mul)
return b
def add_to_map(self, url: str, action: Method, fut: asyncio.Task,
**kwargs) -> None:
if fut.exception():
_logger.warning(
f"Can't {action.value} on {url}: {fut.exception()}")
self.hook_add_to_map_error(url=url,
action=action,
fut=fut,
**kwargs)
return
sub_name = kwargs.pop("sub_name")
if action == Method.ADD_SUB:
vs = VirtualSubscription(**kwargs)
self.real_map[url][sub_name] = vs
self.name_to_subscription[url][sub_name] = fut.result()
if action == Method.ADD_MI:
nodes = kwargs["nodes"]
vs = self.real_map[url][sub_name]
vs.subscribe_data_change(nodes, *astuple(kwargs["node_attr"]))
for node, handle in zip(nodes, fut.result()):
if isinstance(handle, ua.StatusCode):
# a StatusCode is returned, the request has failed.
vs.unsubscribe([node])
_logger.info(f"Node {node} subscription failed: {handle}")
# The node is invalid, remove it from both maps
if handle.name == "BadNodeIdUnknown":
_logger.warning(
f"WARNING: Abandoning {node} because it returned {handle} from {url}"
)
real_vs = self.ha_client.ideal_map[url][sub_name]
real_vs.unsubscribe([node])
continue
self.node_to_handle[url][node] = handle
self.hook_add_to_map(fut=fut, url=url, action=action, **kwargs)
def save(self, cursor: _cursor):
data = astuple(self)
names = ",".join(['"%s"' % f.name for f in fields(self)])
template = ",".join(["%s"] * len(data))
values = cursor.mogrify(template, data).decode()
sql = f"""
insert into {self._table} ({names}) values ({values}) on conflict do nothing
"""
if hasattr(self, 'id'):
sql += " returning *"
cursor.execute(sql)
# If the entry is found, correct the auto-generated object.id with the value from database.
if hasattr(self, 'id'):
result = cursor.fetchone()
# nothing was inserted
if not result:
sql = f"select id from {self._table} where {self._uniq_field}=%s"
cursor.execute(sql, (getattr(self, self._uniq_field), ))
result2 = cursor.fetchone()
self.id = result2[0]
import dataclasses
from typing import Type, Union
class A:
pass
dataclasses.fields(<warning descr="'dataclasses.fields' method should be called on dataclass instances or types">A</warning>)
dataclasses.fields(<warning descr="'dataclasses.fields' method should be called on dataclass instances or types">A()</warning>)
dataclasses.asdict(<warning descr="'dataclasses.asdict' method should be called on dataclass instances">A()</warning>)
dataclasses.astuple(<warning descr="'dataclasses.astuple' method should be called on dataclass instances">A()</warning>)
dataclasses.replace(<warning descr="'dataclasses.replace' method should be called on dataclass instances">A()</warning>)
@dataclasses.dataclass
class B:
pass
dataclasses.fields(B)
dataclasses.fields(B())
dataclasses.asdict(B())
dataclasses.astuple(B())
dataclasses.replace(B())
dataclasses.asdict(<warning descr="'dataclasses.asdict' method should be called on dataclass instances">B</warning>)
dataclasses.astuple(<warning descr="'dataclasses.astuple' method should be called on dataclass instances">B</warning>)
dataclasses.replace(<warning descr="'dataclasses.replace' method should be called on dataclass instances">B</warning>)
def unknown(p):
dataclasses.fields(p)
dataclasses.asdict(p)
dataclasses.astuple(p)
def union1(p: Union[A, B]):
dataclasses.fields(p)
dataclasses.asdict(p)
dataclasses.astuple(p)
dataclasses.replace(p)
def union2(p: Union[Type[A], Type[B]]):
dataclasses.fields(p)
dataclasses.asdict(<warning descr="'dataclasses.asdict' method should be called on dataclass instances">p</warning>)
dataclasses.astuple(<warning descr="'dataclasses.astuple' method should be called on dataclass instances">p</warning>)
dataclasses.replace(<warning descr="'dataclasses.replace' method should be called on dataclass instances">p</warning>)
def __str__(self):
"""Format items separated by a separator."""
items = (str(x) for x in dataclasses.astuple(self) if x is not None)
return self._sep.join(items)
def __iter__(self):
return iter(dataclasses.astuple(self))
