charfinder 1.1.9

A CLI and library for normalized Unicode character search with fuzzy matching.

🔎 charfinder

charfinder is a modern terminal and Python-based tool for searching and exploring Unicode characters by name — supporting both exact and advanced fuzzy matching — with Unicode normalization, efficient caching, structured logging.

Designed for both technical and non-technical users, CharFinder enables reliable Unicode search in terminals, scripts, automation workflows, and applications. It offers transparency and precise control over matching behavior, making it suitable for developer tooling, data pipelines, chatbots, and messaging interfaces.

---

📚 Table of Contents

1. 🎥 Demo Video

2. ✨ Features

3. 📦 Project Structure

   • 3.1 📂 Structure
   • 3.2 🧱 Architecture

4. 🌐 Unicode and Normalization

5. 🎯 Matching Engine (Exact & Fuzzy)

6. 🚀 Usage

   • 6.1 Installation
   • 6.2 💻 CLI Usage
     • Demo
   • 6.3 🐍 Python Library Usage

7. 🧱 Internals and Architecture

   • 7.1 Architecture Overview

   • 7.2 Key Components

     • Caching
     • Environment Management
     • Logging

8. 🧪 Testing

   • Running Tests
   • Code Quality Enforcement
   • Coverage Policy
   • Test Layers

9. 👨‍💼 Developer Guide

   • 🔨 Cloning & Installation
   • 🔧 Makefile Commands
   • 🗒️ Onboarding Tips

10. ⚡ Performance

11. 🚧 Limitations and Known Issues

12. 📖 Documentation

13. 🙏 Acknowledgments

14. 🧾 License

---

🎥 1. Demo Video

https://github.com/user-attachments/assets/e19b0bbd-d99b-401b-aa29-0092627f376b

To see another demo of CLI usage, see subsection Demo

---

✨ 2. Features

CharFinder is a feature-rich Unicode character search tool, designed for both CLI and Python library usage. It combines exact and fuzzy matching with fast caching, robust environment management, and beautiful CLI output.

🔍 Unicode Character Search

• Search Unicode characters by name:

  • Exact match (substring or word_subset)
  • Fuzzy match with configurable thresholds and algorithms

• Supported fuzzy algorithms:

  • simple_ratio — SequenceMatcher-based (from difflib)
  • normalized_ratio — Normalized variant of simple_ratio
  • levenshtein_ratio — Based on python-Levenshtein
  • token_sort_ratio — Word-order invariant (from rapidfuzz)
  • hybrid_score — Aggregates multiple algorithms

• Hybrid fuzzy matching:

  • Combines multiple algorithms using an aggregation function: mean, median, max, or min

📉 Unicode Normalization

• All matching is performed after Unicode normalization.
• Matching is case-insensitive, accent-insensitive, and format-insensitive
• Input and character names are normalized using configurable Unicode profiles (--normalization-profile)
• Alternate names (from UnicodeData.txt) are supported

🔄 Caching

• Unicode name cache:

  • Built on first run
  • Stored as a local JSON file for fast reuse

• LRU cache:

  • Internal normalization results are LRU-cached for performance

📊 Logging

• Rotating file logs under logs/{ENV}/
• Console logging:
  • INFO level by default
  • DEBUG level with --debug
• Each log record includes the current environment (DEV, UAT, PROD)
• Logging is modular and test-friendly

🔧 Environment-aware Behavior

• .env support with layered resolution logic
• Environment-specific behavior:
  • Log directory changes by environment
  • Test mode activates .env.test

📚 See config_environment.md

💻 CLI Features

• Rich CLI with argcomplete tab completion

• Color output:

  • Modes: auto, always, never
  • Colors used for result rows, headers, and logs

• Advanced CLI options:

  • Matching behavior:

    • --fuzzy — Enable fuzzy matching
    • --threshold — Set similarity threshold (0.0–1.0)
    • --fuzzy-algo — Select fuzzy algorithm (e.g., token_sort_ratio)
    • --fuzzy-match-mode — Choose fuzzy match mode: first, all, or hybrid
    • --hybrid-agg-fn — Set aggregation function: mean, median, min, or max
    • --exact-match-mode — Specify exact match logic: word_subset or substring

  • Output control:

    • --color — Control color output: auto, always, or never
    • --verbose — Display formatted results in the console
    • --debug — Enable full diagnostics: dotenv resolution, config state, match algorithms and scores

• Detailed CLI help with examples

📚 See cli_architecture.md and for examples see the subsection demo

🐍 Python Library Usage

• Import and use the core API:

  • find_chars() — Yields formatted result rows
  • find_chars_raw() — Returns structured data (for scripting / JSON)

• Fully type-annotated

• CLI dependencies are not required for library usage

📚 See core_logic.md

🧪 Testability & Quality

• Code quality enforcement:

  • ruff (lint & format), mypy (type-check)

• High test coverage

• CLI tested via subprocess integration tests

• Modular conftest.py with reusable fixtures

• Clean pytest + coverage + pre-commit workflow

📚 See unit_test_design.md

📑 Modern Packaging & Tooling

• pyproject.toml-based (PEP 621)
• GitHub Actions CI pipeline:
  • Python 3.10 to 3.13
  • Lint (Ruff), type-check (MyPy), test, coverage
• Easy publishing to PyPI

---

3. 📦 Project Structure

CharFinder follows a clean, layered architecture to ensure separation of concerns, maintainability, and testability.

The project is structured for ease of contribution and for flexible usage as both:

• A CLI tool (charfinder command).
• An importable Python library.

3.1 📂 Structure

The project is organized as follows:

charfinder/
├── .github/workflows/               # GitHub Actions CI pipeline
├── .pre-commit-config.yaml          # Pre-commit hooks
├── publish/                         # Sample config for PyPI/TestPyPI
├── .env.sample                      # Sample environment variables
├── LICENSE.txt
├── Makefile                         # Automation tasks
├── MANIFEST.in                      # Files to include in sdist
├── pyproject.toml                   # PEP 621 build + dependencies
├── README.md                        # Project documentation (this file)
├── docs/                            # Detailed documentation (.md files)
├── data/                            # Downloaded UnicodeData and cache
│   ├── UnicodeData.txt              # Standard Unicode name definitions
│   └── cache/                       # Local character name cache
├── src/charfinder/                  # Main package code
│   ├── __init__.py                  # Package version marker
│   ├── __main__.py                  # Enables `python -m charfinder`
│   ├── fuzzymatchlib.py             # Fuzzy matching algorithm registry
│   ├── validators.py                # Input validation logic
│   │
│   ├── cli/                         # CLI logic (modularized)
│   │   ├── __init__.py
│   │   ├── args.py                  # CLI argument definitions
│   │   ├── cli_main.py              # CLI main entry point
│   │   ├── diagnostics.py           # Diagnostics and debugging info
│   │   ├── diagnostics_match.py     # Match strategy explanation
│   │   ├── handlers.py              # CLI command handlers
│   │   ├── parser.py                # CLI parser and argument preprocessing
│   │   └── utils_runner.py          # CLI runner and echo utilities
│   │
│   ├── config/                      # Configuration and constants
│   │   ├── __init__.py
│   │   ├── aliases.py               # Alias mappings for fuzzy algorithms
│   │   ├── constants.py             # Default values and valid options
│   │   ├── settings.py              # Environment/config management
│   │   └── types.py                 # Shared type definitions
│   │
│   ├── core/                        # Core Unicode search logic
│   │   ├── __init__.py
│   │   ├── core_main.py             # Public API entry point for core logic
│   │   ├── finders.py               # Output routing and formatting
│   │   ├── handlers.py              # Search coordination and config builder
│   │   ├── matching.py              # Exact and fuzzy matching logic
│   │   ├── name_cache.py            # Unicode name cache builder
│   │   └── unicode_data_loader.py   # UnicodeData.txt loader and parser
│   │
│   ├── utils/                       # Shared utilities
│   │   ├── __init__.py
│   │   ├── formatter.py             # Terminal and log formatting
│   │   ├── logger_helpers.py        # Custom logging helpers
│   │   ├── logger_setup.py          # Logger setup/teardown
│   │   ├── logger_styles.py         # Logging color/style definitions
│   │   └── normalizer.py            # Unicode normalization logic
│
└── tests/                           # Unit, integration, and manual tests
    ├── cli/                         # CLI interface and argument handling tests
    ├── config/                      # Tests for constants, settings, types, aliases
    ├── core/                        # Core Unicode search, cache, and matching logic
    ├── utils/                       # Terminal formatting, normalization, and logger utilities
    ├── helpers/                     # Internal testing utilities (not test files)
    ├── manual/                      # Manual testing and usage examples
    │   └── demo.ipynb               # Interactive demo notebook
    ├── test_fuzzymatchlib.py        # Tests for fuzzy algorithm registry and scoring
    ├── test_validators.py           # Input validation and config resolution logic
    └── conftest.py                  # Shared test fixtures and environment isolation

3.2 🧱 Architecture

CharFinder implements a layered architecture with clear boundaries:

📚 See section Internals and Architecture, and following documentatoins:

• docs/cli_architecture.md
• docs/core_logic.md
• docs/environment_config.md
• docs/logging_system.md
• docs/caching.md

---

4. 🌐 Unicode & Normalization

Unicode is the global standard for encoding text, defining unique code points for every letter, symbol, emoji, and script. It enables CharFinder to search across more than 140,000 characters—covering everything from Latin letters to CJK ideograms and emojis.

Why It Matters for CharFinder

• ✅ Multilingual coverage: Supports scripts from all major languages and symbol sets.
• ✅ Emoji and symbol support: All emoji and symbols are part of Unicode and fully searchable.
• ✅ Alternate name discovery: CharFinder indexes official names and alternate names (from field 10 of UnicodeData.txt) to support queries like "underscore", "slash", or "period".

---

🔄 Normalization

Characters that look the same can be encoded in different ways. For example:

• é (U+00E9) vs. é (e + U+0301) are visually identical but distinct Unicode sequences.

To ensure consistent matching, CharFinder applies Unicode normalization, case folding, whitespace cleanup, and optional accent/diacritic stripping depending on the selected profile.

You can customize this behavior using the --normalization-profile CLI argument:

Profile	Unicode Form	Strip Accents	Collapse Whitespace	Remove Zero-Width	Transformation Summary
raw	—	❌	❌	❌	No changes
light	NFC	❌	✅	❌	Trim + collapse spaces + .upper()
medium	NFC, NFKD	❌	✅	❌	light + Unicode normalization
aggressive	NFC, NFKD	✅	✅	✅	medium + remove diacritics + zero-width characters

The default profile is aggressive, which offers the most robust matching by removing visual and encoding differences.

---

🔍 Normalization in Action

Input	Codepoints	Normalized	Matches?
café	U+0063 U+0061 U+0066 U+00E9	CAFE	✅
café	U+0063 U+0061 U+0066 U+0065 U+0301	CAFE	✅
CAFÉ	U+0043 U+0041 U+0046 U+00C9	CAFE	✅
CAFÉ	U+0043 U+0041 U+0046 U+0045 U+0301	CAFE	✅
𝒸𝒶𝓻é (italic math)	U+1D4B8 U+1D4B6 U+1D4FB U+00E9	CARE	✅ (fallback)
ｃａｆｅ́ (fullwidth)	U+FF43 U+FF41 U+FF46 U+FF45 U+0301	CAFE	✅ (folded)

Even though the second input uses a decomposed form (e + combining acute), CharFinder normalizes and folds it to ensure a stable match.

---

🧪 Terminal Example with Emoji

CharFinder correctly matches Unicode emoji and symbols. For example:

Note: Composite emoji like 👩‍💻 (woman technologist) are grapheme clusters, not individual Unicode code points, and are not listed in UnicodeData.txt. CharFinder focuses on official single-codepoint characters.

📚 See unicode_normalization.md

---

5. 🎯 Matching Engine (Exact + Fuzzy)

CharFinder uses a layered and configurable matching strategy to identify Unicode characters by name. It starts with exact matching for speed and precision, then optionally falls back to fuzzy matching if no exact hits are found or if --prefer-fuzzy is enabled.

🔹 Exact Matching

• Fast string comparisons using two match modes substring or word-subset.
• Controlled via --exact-match-mode (default: word-subset).
• Ideal for full or partial queries that directly appear in character names.

🔸 Fuzzy Matching

Fuzzy matching recovers from typos, partial input, or scrambled tokens. It supports following match modes:

• Single-algorithm mode (--fuzzy-match-mode=single): uses the algorithm specified by --fuzzy-algo (e.g., token_subset_ratio, token_sort_ratio, levenshtein_ratio, etc.)
• Hybrid mode (--fuzzy-match-mode=hybrid): combines multiple algorithms using weighted scores and an aggregation function (mean [default], median, max, min)
• Controlled via --fuzzy-match-mode (default: hybrid).

Fuzzy control options:

• --fuzzy, --prefer-fuzzy — enable fallback or hybrid behavior
• --fuzzy-algo — select algorithm for single mode
• --fuzzy-match-mode {single, hybrid} — control fuzzy strategy
• --threshold — set minimum similarity score

Matching behavior can also be influenced by environment variables. See sample.env

⚙️ Normalization

Matching is applied after Unicode normalization, which includes case folding, accent removal, and Unicode normalization. This is configurable via --normalization-profile.

📚 See matching.md for full logic, algorithm details, and internal representation.

---

6. 🚀 Usage

The following usage guide shows how to install, run, and integrate CharFinder both via its command-line interface (CLI) and as a Python library. Whether you are an end user, developer, or automator, CharFinder is designed to fit seamlessly into your workflow.

6.1 Installation

👤 For Users

PyPI (Recommended)

pip install charfinder

GitHub (Development Version)

pip install git+https://github.com/berserkhmdvhb/charfinder.git

👨‍💼 For Developers

Clone and Install in Editable Mode

git clone https://github.com/berserkhmdvhb/charfinder.git
cd charfinder
make develop

Alternatively:

python -m venv .venv
source .venv/bin/activate  # or .venv\Scripts\activate on Windows
pip install -e .[dev]

---

6.2 💻 CLI Usage

CharFinder provides a CLI for exploring Unicode characters.

Basic Example

charfinder heart

Example output:

U+2764      ❤     HEAVY BLACK HEART  (\u2764)

Full Help

charfinder --help

CLI Options

Option	Description
-q, --query	Provide search query as an option (alternative to positional query)
--fuzzy	Enable fuzzy search if no exact matches are found
--prefer-fuzzy	Include fuzzy results even if exact matches are found (hybrid mode)
--threshold	Set fuzzy match threshold (0.0 to 1.0); applies to all algorithms
--fuzzy-algo	Select fuzzy algorithm: token_sort_ratio (default), simple_ratio, normalized_ratio, levenshtein
--fuzzy-match-mode	Fuzzy match mode: single, hybrid (default)
--hybrid-agg-fn	Aggregation function for hybrid mode: mean (default), median, max, min
--exact-match-mode	Exact match strategy: word-subset (default), substring
--normalization-profile	Normalization level: aggressive (default), medium, light, raw
--format	Output format: text (default) or json
--color	Color output mode: auto (default), always, never
--show-score	Display match scores alongside results (enabled by default for JSON output)
-v, --verbose	Enable terminal output (stdout/stderr); defaults to enabled in CLI, disabled in tests
--debug	Show detailed diagnostics, including config, strategy, and environment
--version	Show installed version of CharFinder

Advanced CLI Tips

• Use --fuzzy and --threshold for typo tolerance.
• Use --format json for scripting and automation.
• Enable diagnostics with --debug or by setting CHARFINDER_DEBUG_ENV_LOAD=1.

Demo

Basic Example

Usage of --verbose or -v flag

Usage of --debug for diagnostics

Fuzzy Match Example

Usage --format to export JSON Output

📚 See cli_architecture.md.

---

6.3 🐍 Python Library Usage

CharFinder can also be used as a pure Python library:

Example: Basic Search

from charfinder.core.core_main import find_chars

for line in find_chars("snowman"):
    print(line)

Example: Fuzzy Search with Options

from charfinder.core.core_main import find_chars

for line in find_chars(
    "snwmn",
    fuzzy=True,
    threshold=0.6,
    fuzzy_algo="rapidfuzz",
    fuzzy_match_mode="single",
    exact_match_mode="word-subset",
    agg_fn="mean",
):
    print(line)

Example: Raw Results (for Scripting)

from charfinder.core.core_main import find_chars_raw

results = find_chars_raw("grinning", fuzzy=True, threshold=0.7)

for item in results:
    print(item)

📚 See core_logic.md.

---

7. 🧱 Internals and Architecture

CharFinder is built with a layered, modular architecture designed for clarity, testability, and extensibility. It supports robust CLI interaction and Python API usage.

7.1 Architecture Overview

The system is structured into clearly defined layers:

1. Core Logic Layer (core/)

• Implements the core Unicode search engine: exact/fuzzy matching, scoring, and normalization.

• Fully decoupled from CLI and formatting logic.

• Key modules:

  • finders.py — main search orchestrator
  • matching.py — scoring logic for fuzzy and exact matches, uses matching library fuzzymatchlib.py
  • name_cache.py — Unicode name caching, loading, and saving
  • unicode_data_loader.py — parses and validates UnicodeData.txt and alternate names

📚 See core_logic.md, matching.md

2. Finder API Layer (core/core_main.py)

• Exposes public APIs: find_chars(), find_chars_with_info(), etc.
• Orchestrates validation, normalization, and config setup
• Consumed by CLI and external Python usage

3. CLI Layer (cli/)

• Argument parsing (args.py, parser.py)
• Execution and output routing (cli_main.py, handlers.py)
• Output formatting (formatter.py, utils_runner.py)
• Fully testable and modular CLI engine

📚 See cli_architecture.md

4. Diagnostics Layer (cli/diagnostics.py, cli/diagnostics_match.py)

• Provides structured debug output for:

  • Matching decisions, fallback logic, algorithm insights

• Activated via --debug or CHARFINDER_DEBUG_ENV_LOAD=1

📚 See debug_diagnostics.md

5. Utilities Layer (utils/)

• Shared helpers:

  • normalizer.py — normalization, folding, and caching
  • logger_helpers.py, logger_setup.py — terminal and file-based logging utilities
  • formatter.py, logger_styles.py — console output styling

6. Configuration Layer (config/)

• Centralized configuration:

  • settings.py — dotenv loading, environment mode detection, paths, log config
  • constants.py — global constant values (defaults, exit codes, env var names)
  • types.py — shared types and protocols for core and CLI usage
  • aliases.py — fuzzy algorithm aliases and canonical name resolution

  📚 See config_constants.md, config_environment.md, config_types_protocols.md

7. Validation Layer (validators.py)

• Core + CLI shared validation

• Ensures consistent input handling:

  • Fuzzy algorithm names, match modes, thresholds, color modes
  • CLI/environment/default priority resolution

📚 See validators.md

---

7.2 Key Components

🔁 Caching

CharFinder uses layered caching:

• In-Memory:

  • cached_normalize() — memoizes normalization results for performance

• Persistent:

  • unicode_name_cache.json stores normalized character name mappings
  • Auto-rebuilt from UnicodeData.txt + alternates if missing or outdated

📚 See caching.md

---

⚙️ Environment Management

Supports predictable, override-friendly config loading:

• Runtime modes: DEV, UAT, PROD, TEST

• Load order:

  1. DOTENV_PATH if explicitly set
  2. .env from project root
  3. Fallback to system environment

→ Enable CHARFINDER_DEBUG_ENV_LOAD=1 for detailed trace

📚 See config_environment.md

---

📋 Logging

Flexible logging system supports development, testing, and production:

• Rotating file logs per environment: logs/{ENV}/charfinder.log
• Console output respects --verbose and --debug
• Color detection adjusts automatically for terminals and scripts
• Logging setup via setup_logging() in logger_setup.py

📚 See logging_system.md

---

🧪 8. Testing

CharFinder has a comprehensive test suite covering core logic, CLI integration, caching, environment handling, and logging.

Testing Layer (tests/)

• Unit tests (core, CLI, utils)
• Integration tests (via CLI subprocess)
• Logging behavior tests
• All tests isolated and environment-aware
• High test coverage using pytest
• Test isolation enforced via Pytest fixtures and .env cleanup

Running Tests

Run the full test suite:

make test

Run only failed or last tests:

make test-fast

Run tests with coverage:

make coverage

Generate HTML coverage report:

make coverage-html

Code Quality Enforcement

make lint-all

Applies Ruff formatting, Ruff checking, and MyPy static type checks. This runs all of the following commands:

Linting and Formatting

make lint-ruff

which is equivalent to

ruff check src/ tests/

make fmt

which is equivalent to

ruff format src/ tests/

Static Type Checks

make type-check

which is equivalent to

mypy src/ tests/

Coverage Policy

• Target: 100% coverage on all Python files under src/
• CLI integration tests cover all major CLI scenarios via subprocess.run
• Logging behaviors, .env loading, and edge cases are all tested

Test Layers

• Unit tests: test core logic in isolation (core, caching, normalization, settings, utils)
• CLI integration tests: test full CLI entrypoint via subprocess
• Logging tests: test rotating logging, suppression, environment filtering
• Settings tests: test different .env and environment variable scenarios

📚 See unit_test_design.md

---

👨‍💻 9. Developer Guide

🔨 Cloning & Installation

For Users:

git clone https://github.com/berserkhmdvhb/charfinder.git
cd charfinder
python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
make install

For Developers (Contributors):

git clone https://github.com/berserkhmdvhb/charfinder.git
cd charfinder
python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
make develop

🔧 Makefile Commands

Command	Description
make install	Install the package in editable mode
make develop	Install with all dev dependencies
make fmt	Auto-format code using Ruff
make fmt-check	Check code formatting (dry run)
make lint-ruff	Run Ruff linter
make type-check	Run MyPy static type checker
make lint-all	Run formatter, linter, and type checker
make lint-all-check	Dry run: check formatting, lint, and types
make test	Run all tests using Pytest
make test-file FILE=...	Run a single test file or keyword
make test-file-function FILE=... FUNC=...	Run a specific test function
make test-fast	Run only last failed tests
make test-coverage	Run tests and show terminal coverage summary
make test-coverage-xml	Run tests and generate XML coverage report
make test-cov-html	Run tests with HTML coverage report and open it
make test-coverage-rep	Show full line-by-line coverage report
make test-coverage-file FILE=...	Show coverage for a specific file
make check-all	Run format-check, lint, and full test suite
make test-watch	Auto-rerun tests on file changes
make precommit	Install pre-commit hook
make precommit-check	Dry run all pre-commit hooks
make precommit-run	Run all pre-commit hooks
make env-check	Show Python and environment info
make env-debug	Show debug-related env info
make env-clear	Unset CHARFINDER_* and DOTENV_PATH environment variables
make env-show	Show currently set CHARFINDER_* and DOTENV_PATH variables
make env-example	Show example env variable usage
make dotenv-debug	Show debug info from dotenv loader
make safety	Check dependencies for vulnerabilities
make check-updates	List outdated pip packages
make check-toml	Check pyproject.toml for syntax validity
make clean-logs	Remove DEV log files
make clean-cache	Remove cache files
make clean-coverage	Remove coverage data
make clean-build	Remove build artifacts
make clean-pyc	Remove .pyc and pycache files
make clean-all	Remove all build, test, cache, and log artifacts
make build	Build package for distribution
make publish-test	Upload to TestPyPI
make publish	Upload to PyPI
make upload-coverage	Upload coverage report to Coveralls

📝 Onboarding Tips

• Always use make develop to install full dev dependencies.
• Run make check-all before pushing changes, or equivalently, run make lint-all-check and make test-coverage.
• Validate .env loading with make dotenv-debug.

---

⚡ 10. Performance

📚 See performance.md

---

🚧 11. Limitations and Known Issues

📚 See limitations_issues.md

---

📖 12. Documentation

This project includes detailed internal documentation to help both developers and advanced users understand its design, architecture, and internals.

The following documents are located in the docs/ directory:

Document	Description
caching.md	Explanation of cache layers: Unicode name cache, cached_normalize(), performance considerations.
cli_architecture.md	Overview of CLI modules, their flow, entry points, and command routing logic.
config_constants.md	Centralized constants used across the project: default values, valid input sets, exit codes, environment variable names, normalization profiles, hybrid scoring weights, and logging defaults.
config_environment.md	Detailed explanation of environment variable handling and .env resolution priorities and scenarios
config_types_protocols.md	Project-wide types, Protocol interfaces, and their role in extensibility and static typing.
core_logic.md	Core logic and library API (find_chars, find_chars_raw): processing rules, transformations, architecture.
debug_diagnostics.md	Debug and diagnostic output systems: --debug, CHARFINDER_DEBUG_ENV_LOAD, dotenv introspection.
logging_system.md	Logging architecture: setup, structured logging, rotating files, and environment-based folders.
matching.md	Detailed explanation of exact and fuzzy matching algorithms and options. Includes mode combinations and flowcharts.
unicode_normalization.md	Unicode normalization explained: what is used (NFC), why, and implications for search.
packaging.md	Packaging and publishing: pyproject.toml, build tools, versioning strategy, and PyPI release process.
unit_test_design.md	Testing layers: unit tests, CLI integration tests, coverage strategy.
validators.md	Centralized validation logic shared across CLI and core. Type safety, fallbacks, source-aware behavior.

These documents serve both as developer onboarding materials and technical audit references.

---

🙏 13. Acknowledgments

Special thanks to Luciano Ramalho @ramalho, author of Fluent Python.

The original charfinder function in his book (Chapter 4: Unicode Text Versus Bytes) directly inspired the creation of this project — both in concept and in name.

Luciano also provided critical early feedback through GitHub issues, which shaped major improvements, and the overall evolution of release v1.1.6. His insights on alternate Unicode names, query flexibility, and CLI UX were invaluable.

---

🧾 14. License

MIT License © 2025 berserkhmdvhb