def repr_fn_with_debug(val): # Wrap repr_fn to add debug info
try:
return repr_fn(val)
except Exception as e: # pylint: disable=broad-except
err_msg = (f'HTML formatting of column {name} failed:\n'
f' * feature: {feature}\n'
f' * input: {val!r}\n')
py_utils.reraise(e, prefix=err_msg)
def _try_import(module_name):
"""Try importing a module, with an informative error message on failure."""
try:
mod = importlib.import_module(module_name)
return mod
except ImportError:
err_msg = ("Tried importing %s but failed. See setup.py extras_require. "
"The dataset you are trying to use may have additional "
"dependencies.")
utils.reraise(err_msg)
def _try_import(module_name):
"""Try importing a module, with an informative error message on failure."""
try:
mod = importlib.import_module(module_name)
return mod
except ImportError as e:
err_msg = ("Failed importing {name}. This likely means that the dataset "
"requires additional dependencies that have to be "
"manually installed (usually with `pip install {name}`). See "
"setup.py extras_require.").format(name=module_name)
utils.reraise(e, suffix=err_msg)
def _serialize_example(
example: Any, features: features_lib.FeaturesDict,
serializer: example_serializer.ExampleSerializer) -> str:
try:
encoded_example = features.encode_example(example)
except Exception as e: # pylint: disable=broad-except
py_utils.reraise(e,
prefix='Failed to encode example:\n',
suffix=f'{example}\n')
return serializer.serialize_example(encoded_example)
def _reraise_with_list_builders(
e: Exception,
ns_name: Optional[str],
builder_name: str,
) -> NoReturn:
"""Add the list of available builders to the DatasetNotFoundError."""
# Should optimize to only filter through given namespace
all_datasets = list_builders()
all_datasets_str = '\n\t- '.join([''] + all_datasets)
error_string = f'Available datasets:{all_datasets_str}\n'
error_string += textwrap.dedent("""
Check that:
- if dataset was added recently, it may only be available
in `tfds-nightly`
- the dataset name is spelled correctly
- dataset class defines all base class abstract methods
- the module defining the dataset class is imported
""")
# Add close matches
name = f'{ns_name}:{builder_name}' if ns_name else builder_name
close_matches = difflib.get_close_matches(name, all_datasets, n=1)
if close_matches:
error_string += f'\nDid you meant: {close_matches[0]}'
raise py_utils.reraise(e, suffix=error_string)
def _dataset_name_and_kwargs_from_name_str(
name_str: str,
) -> Tuple[str, Dict[str, Value]]:
"""Extract kwargs from name str."""
err_msg = textwrap.dedent(
f"""\
Parsing builder name string {name_str} failed.
The builder name string must be of the following format:
dataset_name[/config_name][:version][/kwargs]
Where:
* dataset_name and config_name are string following python variable naming.
* version is of the form x.y.z where {{x,y,z}} can be any digit or *.
* kwargs is a comma list separated of arguments and values to pass to
builder.
Examples:
my_dataset
my_dataset:1.2.*
my_dataset/config1
my_dataset/config1:1.*.*
my_dataset/config1/arg1=val1,arg2=val2
my_dataset/config1:1.2.3/right=True,foo=bar,rate=1.2
"""
)
res = _NAME_REG.match(name_str)
if not res:
raise ValueError(err_msg)
name = res.group('dataset_name')
# Normalize the name to accept CamelCase
name = camelcase_to_snakecase(name)
kwargs = _kwargs_str_to_kwargs(res.group('kwargs'))
try:
for attr in ['config', 'version']:
val = res.group(attr)
if val is None:
continue
if attr in kwargs:
raise ValueError('Dataset %s: cannot pass %s twice.' % (name, attr))
kwargs[attr] = val
return name, kwargs
except Exception as e: # pylint: disable=broad-except
py_utils.reraise(e, prefix=err_msg) # pytype: disable=bad-return-type
