hamid / umlaut-fixer

This script addresses(find and replace) a common but often overlooked issue in text processing: the distinction between combining umlauts and precomposed umlauts

umlaut-fixer

This script addresses(find and replace) a common but often overlooked issue in text processing: the distinction between combining umlauts and precomposed umlauts. I've also covered fully this matter in this article.

Overview

Combining umlauts consist of a base letter followed by a diacritic mark, representing characters like "ü". In contrast, precomposed umlauts, such as "ü", combine both elements into a single character. This subtle difference can lead to inconsistencies in text formatting and searching, particularly in environments that do not uniformly support combining characters.

Potential Problems

Inconsistent Formatting: Some systems may apply text styles (like bold or italic) only to the base character, leaving the diacritic mark unchanged. Search and Match Issues: Searching for a specific character may yield incomplete results if the search query does not account for both umlaut types. Encoding and Compatibility: Diverse encoding standards may treat these characters differently, leading to potential data loss or corruption when transferring text between systems.

1) ü = 1 character 

2) ü = 2 character -> u + ¨

How to fix that?

1. You can easily find the "ü" and replace it with "ü" and also for the other umlaut-based character.
2. Use this script to run it manually or run it automatically in your pipeline.

Features

• Dual Mode Operation: Offers both 'find' and 'replace' modes to identify occurrences of combining umlauts and convert them into their precomposed counterparts.
• Recursive Path Search: Scans files recursively within the specified directory, ensuring comprehensive coverage.
• Exclusion of Specific Directories: Automatically ignores directories such as node_modules, preventing unnecessary processing of irrelevant files.
• Cross-Platform Compatibility: Designed to run on Linux, macOS, and Windows (via WSL), catering to a wide range of environments and use cases.

Installation and Running

Linux/macOS

Download the script file, umlaut_fixer.sh Make the script executable:

chmod +x umlaut_fixer.sh

Run the script interactively:

./umlaut_fixer.sh

Windows

For Windows users, running the script requires the Windows Subsystem for Linux (WSL). Install WSL, navigate to the directory containing the script, and follow the Linux/macOS instructions above.

Non-Interactive Mode (CI/CD Mode)

The script supports a non-interactive mode for seamless integration into CI/CD pipelines or for automation purposes. To use this feature, you can pass command-line arguments to the script:

To perform a replacement operation without interactive prompts, use:

./umlaut_fixer.sh -r /path/to/directory

This mode is particularly useful for automated text processing tasks, where manual intervention is impractical or impossible.