Normality

3.1.0 · active · verified Thu Apr 16

Normality is a Python micro-package (current version 3.1.0) that provides a small set of text normalization functions for easier re-use. It accepts unicode or utf-8 encoded text and removes various classes of characters, such as diacritics and punctuation. This library is actively maintained and is useful as a preparation step for further text analysis.

Common errors

```
ImportError: No module named 'icu'
```
cause The `pyicu` dependency, which is mandatory since `normality` version 3.0, is not installed in the environment.

fix Install the `pyicu` package: `pip install pyicu`.
```
AttributeError: module 'normality' has no attribute 'normaltest'
```
cause `normality` is a text processing library, not a statistical library. Users are likely trying to call `scipy.stats.normaltest` mistakenly on the `normality` module.

fix If you intend to perform statistical normality tests, import from `scipy.stats`: `from scipy.stats import normaltest`. If you want text normalization, use functions like `normality.normalize()`.

Warnings

breaking As of version 3.0, `normality` requires `pyicu` as a mandatory dependency. If `pyicu` cannot be installed, users should revert to `normality < 3.0.0` to avoid `ImportError` or runtime issues.
Fix: Ensure `pyicu` is installed in your environment (`pip install pyicu`) or downgrade to `normality==2.*`.
gotcha The `normality` library for text normalization is often confused with statistical 'normality tests' (e.g., Shapiro-Wilk, D'Agostino-Pearson, Anderson-Darling) found in libraries like SciPy (`scipy.stats.normaltest`). These are entirely different concepts; `normality` focuses on text string manipulation, not statistical analysis of data distributions.
Fix: Verify the intended purpose: `normality` (pudo/normality) for text, `scipy.stats.normaltest` (or similar) for statistical distribution testing.

Install

pip install normality Install latest version

Imports

normalize
```
from normality import normalize
```
slugify
```
from normality import slugify
```
collapse_spaces
```
from normality import collapse_spaces
```
ascii_text
```
from normality import ascii_text
```
latinize_text
```
from normality import latinize_text
```

Quickstart

This quickstart demonstrates the core normalization functions of the `normality` library, including `normalize` for general text cleaning, `slugify` for creating URL-friendly slugs, and `collapse_spaces` for standardizing whitespace.

from normality import normalize, slugify, collapse_spaces

text = normalize('Nie wieder "Grüne Süppchen" kochen!')
print(f"Normalized: {text}")
# Expected: nie wieder grune suppchen kochen

slug = slugify('My first blog post!')
print(f"Slugified: {slug}")
# Expected: my-first-blog-post

spaced_text = 'this \n\n\r\nhas\tlots of \nodd spacing.'
cleaned_text = collapse_spaces(spaced_text)
print(f"Collapsed spaces: {cleaned_text}")
# Expected: this has lots of odd spacing.

view raw JSON →