Exceptions and logging

ests.exceptions

Exceptions

All library exceptions inherit the base class EstsError and one of the built-in Python classes, so they can be caught both by their esTS name and by the familiar built-in type - existing code with except ValueError keeps working.

Exception	Built-in class	When raised
EstsError	Exception	Base class, never raised itself
SourceTypeError	TypeError	The data source is neither a string nor a Doc, a string or a Doc is passed where a list of words is expected, the tokenizer is not callable or returns a non-iterable object
SourceError	ValueError	The source has no words, no sentences or no texts, lacks the annotation a statistic needs (the parts of speech, the lemmas, the dependency parse), is a string longer than the max_length of the pipeline, or no unit is left after culling
ParameterError	ValueError	A threshold, window, segment size, number of items, logarithm base or confidence level is out of range; an unknown preset, scale, metric name, measure or variant
UnknownStatError	ParameterError, KeyError	An unknown statistic is requested by name, as in DiversityStats.windowed
DatasetNotFoundError	OSError	The model of spaCy is not installed or a dataset is not downloaded; the message shows the command that brings what is missing
DataFileError	ValueError	A dataset file is corrupted, has an unexpected format or cannot be decoded
DownloadError	RuntimeError	The file could not be downloaded or failed the checksum verification

The classes are available from ests and from ests.exceptions.

Note

DatasetNotFoundError is what a statistic of Universal Dependencies raises when the model es_core_news_sm is not installed, which is the first thing a new reader meets: MorphStats("El gato duerme") without the model says how to download it. DataFileError and DownloadError are reserved for the dataset loaders of the coming releases; nothing raises them yet.

Example

from ests import EstsError, ParameterError, WordsExtractor

try:
    WordsExtractor(ngram_range=(2, 1))
except ParameterError as e:
    print(e)
# The lower N-gram bound is greater than the upper

try:
    WordsExtractor(tokenizer=42).extract("El gato duerme.")
except EstsError as e:
    print(type(e).__name__)
# SourceTypeError

Logging

The library prints nothing on its own: its messages go to the ests logger, which has a NullHandler by default, so they are silent. To see them, configure logging in your application:

Example

import logging

logging.basicConfig(level=logging.INFO, format="%(name)s: %(message)s")

Methods that print by design, such as the print_stats() methods of the statistics classes, write to standard output; logging does not affect them.