Registry / serialization / number-parser

number-parser

library0.3.2pypypi✓ verified 20d ago

number-parser is a Python library designed to convert numbers written in natural language into their equivalent numeric forms. It supports cardinal numbers in English, Hindi, Spanish, Ukrainian, and Russian, and also handles ordinal numbers and simple fractions in English. The current version is 0.3.2, and it maintains an active, though somewhat irregular, release cadence with several updates in early 2023.

serialization data

pip install number-parser

github ↗package ↗

INSTALL

IMPORT

SIG · NUMBER-PARSER

number-parser

serializationpythonv0.3.2

Install

1.9s avg

Import

16ms

Disk

17MB

Pass rate

10/ 10

Env Coverage10 / 10

glibc

3.9–3.13

musl

3.9–3.13

Install & Compatibility

Where this runs

tested against v0.3.2 · pip install

no network on importno background threads

Install × environment matrix

Each cell = how many times install + import succeeded across repeated harness runs. Partial = flaky.
glibc = Debian/Ubuntu slim · musl = Alpine Linux

musl

py 3.10–3.95 runs

installs and imports cleanly · install 0.0s · import 0.018s · 19.1MB

glibc

py 3.10–3.95 runs

installs and imports cleanly · install 1.9s · import 0.014s · 20MB

17MB installed

● package 17MB

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

parse

✓ from number_parser import parse

✗ from number_parser.parser import parse

An import bug was fixed in v0.3.2; direct import from `number_parser` is the correct and stable path.

parse_number

✓ from number_parser import parse_number

parse_ordinal

✓ from number_parser import parse_ordinal

Supports English only.

parse_fraction

✓ from number_parser import parse_fraction

Supports English only.

Demonstrates how to use `parse` for in-place number conversion within a string and `parse_number` for converting a standalone number written in words.

from number_parser import parse # Convert numbers in a sentence text = "I have two hats and thirty seven coats." parsed_text = parse(text) print(f"Original: {text}\nParsed: {parsed_text}") # Convert a single number in words number_in_words = "two thousand and twenty" parsed_number = parse_number(number_in_words) print(f"Number '{number_in_words}' parsed to: {parsed_number}")

Debug

Known issues

breakingPrior to version 0.3.2, users might have encountered an import bug (#91), which could lead to incorrect or failed imports of the parsing functions. This was fixed in version 0.3.2.

fix

Upgrade to number-parser >= 0.3.2. Ensure imports are directly from `number_parser` (e.g., `from number_parser import parse`) and not from internal modules like `number_parser.parser`.

affects: <0.3.2

gotchaInconsistent whitespace handling around sentence separators following numbers was an issue in versions prior to 0.3.1, potentially affecting parsing accuracy in specific contexts.

fix

Upgrade to number-parser >= 0.3.1 to benefit from the fix for inconsistent whitespace handling.

affects: <0.3.1

gotchaThe library's language support is granular: cardinal numbers are supported in multiple languages, but ordinal numbers and simple fractions are currently supported only for English. Attempting to parse non-English ordinals or fractions might yield unexpected results or errors.

fix

Consult the documentation for specific language support for each parsing function (e.g., `parse_ordinal` and `parse_fraction` are English-only). Ensure your input language matches the supported features.

affects: All versions

gotchaParsing natural language numbers can be inherently ambiguous, especially across different locales and contexts (e.g., 'one hundred thousand' vs. 'one, hundred thousand'). While the library aims for robust parsing, complex or unusual phrasing might not always be interpreted as expected.

fix

Test the library with a representative sample of your natural language number inputs. Be mindful of potential ambiguities and, if necessary, preprocess inputs or provide clearer phrasing.

affects: All versions

Errors

Common errors & fixes

ModuleNotFoundError: No module named 'number_parser'

The `number-parser` library is not installed in the current Python environment or there's a typo in the import statement.

fix

Ensure the library is installed using pip: `pip install number-parser`. Then, import correctly as `from number_parser import parse` (note the underscore in `number_parser`).

parse_number('not_a_number')

The `parse_number` function is designed to convert a single number written in words (e.g., 'two thousand') to its integer form. Providing a string that does not represent a valid number in this format will result in `None` being returned, which users might perceive as an error if they expect an exception or a default numeric value.

fix

Check the return value of `parse_number` for `None` to handle unparseable input gracefully. For parsing numbers embedded in text, use the `parse` function instead. Example: `result = parse_number('not_a_number')` then `if result is None: print('Input could not be parsed.')`

AttributeError: module 'number_parser' has no attribute 'parse_string'

Users might guess function names based on common parsing patterns or misremember the API. The `number-parser` library uses `parse`, `parse_number`, `parse_ordinal`, and `parse_fraction` for its core functionalities, not `parse_string`.

fix

Refer to the official documentation or examples for the correct function names. For general string parsing, use `number_parser.parse()`. For single number parsing, use `number_parser.parse_number()`.

number-parser 'language not supported'

While `number-parser` supports several languages (English, Hindi, Spanish, Ukrainian, and Russian), attempting to use an unsupported language code with functions that accept a `language` parameter (e.g., `parse_number('one', language='french')`) will lead to incorrect parsing or unexpected behavior as the library will default to English or fail to recognize numbers in the unsupported language.

fix

Ensure that the `language` parameter is set to one of the currently supported locales ('en', 'hi', 'es', 'uk', 'ru'). If a language is not supported, the library may not be suitable for that specific use case.

Upgrade

Version history

0.3.2latest on PyPI · released Mar 28, 2023

Audit

Dependencies

No dependency data recorded yet.

Agent activity

5 hits · last 30 days

node

Resources

githubhttps://github.com/scrapinghub/number-parser ↗

packagehttps://pypi.org/project/number-parser/ ↗