Skip to content

Changelog

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[Unreleased]

[1.0.3] - 2026-08-20

Fixed

  • __version__ now reads from package metadata so it stays in sync with pyproject.toml.
  • Release workflow now correctly extracts the CHANGELOG section for GitHub releases.

[1.0.2] - 2026-08-20

Fixed

  • Apply ruff format to slugany/_steps.py so the CI format check passes.

[1.0.1] - 2026-08-20

Fixed

  • CLI stdio UTF-8 reconfiguration now uses getattr to satisfy mypy and contextlib.suppress to keep ruff happy.
  • Wrapped the long test list in test_mixed_case_idempotency_all_styles to respect the 100-character line length.

[1.0.0] - 2026-08-20

Added

  • Emoji text mode: replaces emojis with textual descriptions via _EMOJI_TEXT mapping table (100+ emojis)
  • Emoji keep mode: preserves emojis in output with allow_unicode=True validation
  • Greek-to-Latin confusable mappings in _CONFUSABLES (α→a, β→b, γ→g, etc.)
  • deconfuse() public function for standalone Unicode homoglyph replacement
  • deconfuse exported from slugany package (__all__)
  • Slugifier class: stateless slugifier bound to a fixed configuration with Slugifier.style() factory
  • Slugifier.__call__ with TypeError on non-string input
  • Slugifier.__repr__ for debugging
  • Slugifier.config read-only property exposing the frozen SlugConfig
  • CSS-safe slug prefixing for camel, pascal, and train styles
  • _apply_css_safe pipeline step (before _apply_case_style)
  • py.typed marker (PEP 561) for type checker discovery
  • "Typing :: Typed" classifier in pyproject.toml
  • CI coverage enforcement at 100% (--cov-fail-under=100)
  • Trusted publishing via pypa/gh-action-pypi-publish@release/v1 with attestations
  • Mkdocstrings API documentation with Google-style docstrings (Args, Returns, Raises, Examples)
  • Comprehensive guide documentation: migration, performance, contracts
  • README with badges (PyPI, CI, coverage, Python versions, License), 3-way comparison table (python-slugify vs unicode-slugify vs slugany)
  • FastAPI/Pydantic integration examples in README
  • Slugifier builder pattern examples in README
  • CLI examples in README
  • 478 tests with 100% coverage

Changed

  • _handle_emoji now supports three modes: strip (remove), text (replace with description), keep (preserve with allow_unicode=True)
  • _EMOJI_RE regex expanded with additional ranges (U+2300-23FF, U+2B00-2BFF, U+1F000-1F0FF, etc.)
  • Pipeline expanded to 18 steps (added _apply_css_safe and second _truncate/_trim_separators pass)
  • _CONFUSABLES expanded with Greek uppercase and lowercase mappings
  • All public function docstrings completed with Google-style Args, Returns, Raises, Examples sections
  • README test count badge updated to 478
  • Performance documentation updated with pipeline diagram, short-circuit optimization, and Slugifier reuse section
  • Migration documentation expanded with full parameter mapping table and intentional differences section
  • Contracts documentation expanded with case-insensitive stopwords, CSS-safe output, and multi-character separator contracts

Fixed

  • _handle_emoji multi-character emoji matching: iterates over each character in match group for _EMOJI_TEXT lookup
  • emoji_mode='keep' now raises ValueError when allow_unicode=False instead of silently preserving emojis
  • Greek letters αβγ now deconfused to abg before NFKD fallback (previously produced empty string)
  • Emoji text mode correctly replaces emojis with descriptions before transliteration step

[0.2.0] - 2026-08-20

Added

  • Language auto-detection: lang='auto' detects dominant language from Unicode code-point ranges (es, pt, de, fr, it) and selects the appropriate transliteration table
  • Optional Pydantic Slug type: Annotated[str, BeforeValidator] that auto-slugifies string values in Pydantic models; falls back to str when pydantic is not installed
  • _LANG_DETECT_RANGES table in _tables.py for language detection

Fixed

  • Pipeline order: _trim_separators now runs before _truncate to avoid trailing separators after truncation
  • word_boundary truncation at exact separator position: returns full truncated text instead of losing a word when the boundary falls exactly at a separator
  • SlugConfig.from_kwargs now rejects bool for max_length (Python bool is a subclass of int)
  • is_slug with empty separator: avoids catastrophic backtracking (ReDoS) by using a simple character-class match
  • CLI now catches ValueError and TypeError from slugify and prints to stderr with exit code 1 instead of an unhandled traceback

[0.1.0] - 2026-08-19

Added

  • slugify() function with keyword-only API and TypeError on non-string input
  • slugify_batch() for batch processing of multiple texts
  • is_slug() validator with allow_unicode support
  • SlugConfig frozen dataclass with from_kwargs() factory and style presets
  • Transliteration tables for es, pt, de, fr, it
  • Smart punctuation normalization (curly quotes, em-dash, NBSP, zero-width, bullets)
  • HTML entity decoding (& → &)
  • Emoji stripping via regex
  • Confusable character detection (Cyrillic → Latin)
  • Case styles: kebab, snake, camel, pascal, dot, train, filename, url
  • Stopwords removal
  • Pre/post replacements (dict or iterable of tuples)
  • max_length with optional word_boundary truncation
  • allow_unicode mode to preserve non-ASCII characters
  • fallback parameter for empty slug results
  • css_safe mode to prefix digit-leading slugs
  • emoji_mode (strip, text, keep)
  • html_entities and smart_punctuation toggles
  • lru_cache built-in (maxsize=512) with cache_info() and cache_clear()
  • Pipeline architecture with 16 pure-function steps
  • CLI with argparse, auto-stdin detection, and batch mode
  • py.typed marker (PEP 561)
  • CI/CD: lint (ruff), typecheck (mypy), test (3.11/3.12/3.13), release, docs
  • MkDocs Material documentation with mkdocstrings API reference
  • Idempotency, determinism, ASCII output, and no-double-separator contracts
  • SlugConfig API reference page in documentation
  • CONTRIBUTING.md with development setup, PR process, and code style guidelines
  • SECURITY.md with vulnerability reporting policy
  • CODE_OF_CONDUCT.md based on Contributor Covenant 2.1
  • GitHub issue templates (bug report, feature request)
  • GitHub pull request template with checklist
  • Dependabot configuration for pip and GitHub Actions
  • Documentation URL in pyproject.toml project URLs
  • __version__ attribute exported from package

Changed

  • slugify() and slugify_batch() parameters are now properly typed (no more Any sentinel pattern) — IDE autocompletion and type checking now work correctly for all 14 keyword arguments
  • slugify_batch() now accepts explicit keyword arguments instead of **kwargs: Any
  • SlugConfig.from_kwargs() no longer uses type: ignore comments — validation constants extracted to module-level frozensets
  • _STYLE_PRESETS and validation constants moved before SlugConfig class definition
  • is_slug() now raises TypeError for non-string separator input
  • CLI _build_kwargs return type changed from dict[str, object] to dict[str, Any] to eliminate type: ignore on slugify calls
  • pyproject.toml description translated to English; license changed to SPDX expression with license-files
  • README.md and docs/index.md translated from Spanish to English
  • mkdocs.yml site description translated to English
  • Development status classifier updated from Alpha to Beta
  • Removed redundant re.UNICODE flag from _UNICODE_ALNUM_RE (default in Python 3)
  • _CONFIG_DEFAULTS now computed from dataclass fields instead of hardcoded values (DRY)
  • Fixed pipeline step count in documentation (16, not 17)
  • Fixed benchmark numbers in performance documentation
  • Fixed incorrect emoji_mode examples in README
  • README expanded with advanced examples, documentation links, and contributing section
  • docs/index.md rewritten as a proper landing page with feature list and navigation
  • CONTRIBUTING.md updated with full dev setup (dev+docs) and Code of Conduct reference
  • CI/CD workflows now use pip dependency caching

Fixed

  • Non-ASCII stopwords not transliterated before matching
  • css_safe + max_length idempotency (pipeline reorder)
  • train + max_length idempotency (double truncate)
  • Unicode case boundary splitting for camel/pascal/train styles
  • Case boundary splitting: digit-to-lowercase should not split
  • Camel/pascal/train idempotency with consecutive uppercase letters
  • Missing Cyrillic homoglyphs in _CONFUSABLES (в, к, м, н, т)
  • Missing emoji ranges in _EMOJI_RE (U+2300-23FF, U+2B00-2BFF, etc.)
  • Replacements double-application (pre/post split by ASCII/non-ASCII keys)
  • CSS safe idempotency with camel/pascal/train styles
  • Type validation for bool/int/str fields in from_kwargs
  • Stopwords matching: non-alphanumeric chars stripped consistently
  • Deconfuse skipped when allow_unicode=True
  • Emoji mode text/keep preserving emojis through transliteration
  • HTML entities, smart punctuation, transliteration interactions

Removed

  • _verify.py — redundant sanity-check script duplicated by the test suite
  • docs/changelog.md — duplicate of root CHANGELOG.md; mkdocs now references the original