v1.4.1

zaidmukaddam/scirav1.4.1Feb 22, 2026by amirivojdan

AI Summary

This release focuses on bug fixes and refactoring to improve the robustness of Persian text processing within the Shekar component.

Key Highlights

Fixed Sentence Tokenizer to correctly handle Persian quotations and multi-sentence structures.
Improved Spellchecker accuracy and stability.
Refactored Spacing Normalizer into four distinct modules for better maintainability and clarity.

New Features

Refactored Spacing Normalizer into four modular components: OrthographicSpacingNormalizer, PunctuationSpacingNormalizer, WordSpacingNormalizer, and VerbalSpacingNormalizer.

Full Release Notes

# What’s New in v1.4.1

This release focuses on bug fixes and refactoring for more robust Persian text processing in Shekar.

## Highlights

- **Fixed Sentence Tokenizer Bug** – handles Persian quotations and multi-sentence structures correctly.  
- **Fixed Spellchecker Bug** – improved accuracy and stability.  
- **Refactored Spacing Normalizer** – split into four well-defined modules:  
  - `OrthographicSpacingNormalizer`  
  - `PunctuationSpacingNormalizer`  
  - `WordSpacingNormalizer`  
  - `VerbalSpacingNormalizer`  
  Enhances clarity, maintainability, and modularity of preprocessing pipelines.