pyproject.toml

[project]
name = "chunklet-py"
version = "2.3.2"
description = "High-fidelity context-aware chunking and interactive visualization for RAG. Advanced segmentation for code and documents, because your LLM is only as smart as the fragments you feed it."
authors = [
  { name = "speedyk_005", email = "speedy40115719@gmail.com" }
]

license = "MIT"
readme = "README.md"
requires-python = ">=3.11"

classifiers = [
  "Development Status :: 5 - Production/Stable",
  "Intended Audience :: Developers",
  "Intended Audience :: Science/Research",
  "Intended Audience :: Information Technology",
  "Operating System :: OS Independent",
  "Programming Language :: Python :: 3.11",
  "Programming Language :: Python :: 3.12",
  "Programming Language :: Python :: 3.13",
  "Programming Language :: Python :: 3.14",
  "Topic :: Text Processing",
  "Topic :: Scientific/Engineering :: Artificial Intelligence",
  "Topic :: Software Development :: Libraries :: Python Modules"
]

keywords = [
  "nlp", "chunking", "text-splitting", "sentence-splitting", "llm", "rag", "ai", "multilingual",
  "text processing", "natural language processing", "data processing",
  "information retrieval", "semantic search", "document processing", "document-chunker",
  "code chunking", "code-chunker", "source code analysis", "programming languages",
  "code structure", "chunk-visualization",
]

dependencies = [
  "loguru>=0.7.3,<1.0",
  "pysbd>=0.3.4,<1.0",
  "sentsplit>=1.0.8,<2.0",

  # Rust binding (1.0+) for faster performance on supported platforms, legacy for Android temporarily.
  "sentencex<=0.6.1; sys_platform == 'linux' and (platform_machine == 'aarch64' or platform_machine == 'armv7l')",
  "sentencex>=1.0.0,<2.0; sys_platform != 'linux' or (platform_machine != 'aarch64' and platform_machine != 'armv7l')",

  "indic-nlp-library>=0.92,<1.0",
  "py3langid>=0.3.0,<1.0",
  "mpire>=2.10.2,<3.0",
  "more-itertools>=10.6.0,<12.0",
  "regex>=2025.7.29",
  "pydantic>=2.11.0,<3.0",
  "typer>=0.19.0,<1.0",
  "charset-normalizer>=3.4.2,<4.0",

  # sentsplit depends on deprecated pkg_resources removed in setuptools 81.0.0
  "setuptools<81",
]

[project.optional-dependencies]
structured-document = [
  "pdfminer.six>=20250324",
  "python-docx>=1.2.0,<2.0",
  "mammoth>=1.9.1,<2.0",
  "ebooklib>=0.19,<1.0",
  "docutils>=0.21.2,<1.0",
  "markdownify>=1.1.0,<2.0",
  "striprtf>=0.0.29,<1.0",

  # Lastest version of 2.x.x series
  "pylatexenc==2.10",
  
  "odfpy>=1.4.1,<2.0",
  "openpyxl>=3.1.5,<4.0",
  "tabulate2>=1.10.2,<2.0",
]

# Kept for compatibility
document = ["chunklet-py[structured-document]"]

code = [
  "defusedxml>=0.7.1,<1.0",
  "littletree>=0.8.7,<1.0",
]

visualization = [
  "uvicorn>=0.35.0,<1.0",
  "fastapi>=0.116.0,<1.0",
  "python-multipart>=0.0.20,<1.0",
  "msgpack>=1.1.0,<2.0",

  # Major changes happens often
  "aiofiles==25.1.0",
]

viz = ["chunklet-py[visualization]"]

all = ["chunklet-py[structured-document,code,visualization]"]

dev = [
  "pytest>=8.3.5",
  "pytest-cov>=6.2.1",
  "pytest-mock>=3.14.1",
  "pytest-timeout>=2.3.1",
  "ruff>=0.14.14",
]

docs = [
  "mike>=2.1.2",
  "mkdocs-material>=9.6.20",
  "mkdocs-api-autonav>=0.4.0",
  "mkdocstrings-python>=1.17.0",
  "pymdown-extensions>=10.16.1",
]

dev-all = ["chunklet-py[dev,docs,structured-document,code,visualization]"]

[project.urls]
Homepage = "https://github.com/speedyk-005/chunklet-py"
Repository = "https://github.com/speedyk-005/chunklet-py"
Documentation = "https://speedyk-005.github.io/chunklet-py/"
Issues = "https://github.com/speedyk-005/chunklet-py/issues"
Changelog = "https://github.com/speedyk-005/chunklet-py/blob/main/CHANGELOG.md"

[project.scripts]
chunklet = "chunklet.cli:app"

[build-system]
requires = [
    "setuptools>=64",
    "wheel", # Essential for building dependencies that use legacy setup.py (like odfpy)
]
build-backend = "setuptools.build_meta"

[tool.setuptools]
package-dir = { "" = "src" }
packages = [
    "chunklet",
    "chunklet.common",
    "chunklet.document_chunker",
    "chunklet.document_chunker.converters",
    "chunklet.document_chunker.processors",
    "chunklet.code_chunker",
    "chunklet.sentence_splitter",
    "chunklet.visualizer",
]
include-package-data = true

[tool.setuptools.package-data]
chunklet = ["visualizer/static/**/*"]

[tool.pytest.ini_options]
minversion = "7.4"
filterwarnings = [
    "ignore::DeprecationWarning:websockets.*",
    "ignore::DeprecationWarning:pkg_resources.*",
]

# disable coverage report to prevent racing with multiprocessing
# addopts = "--cov=chunklet --cov-report=term-missing:skip-covered --durations=10"
addopts = "--doctest-modules --timeout=60"
testpaths = ["tests", "src/chunklet/common/dotdict.py"]
python_files = "test_*.py"

[tool.coverage.run]
omit = [
    "src/chunklet/cli.py",
    "src/chunklet/common/logging_utils.py",
    "src/chunklet/base_chunker.py",
    "src/chunklet/**/__init__.py",
]


[tool.ruff]
target-version = "py310"
line-length = 88

[tool.ruff.lint]
# E, W (Flake8), F (Pyflakes), I (Isort), B (Bugbear)
select = ["E", "F", "I", "W", "B"]
ignore = ["E501", "E203"]

[tool.ruff.lint.per-file-ignores]

# Ignore unused imports in all __init__.py files
"src/chunklet/**/__init__.py" = ["F401", "F403"]

# Disable all linting for the heavy regex file
"src/chunklet/code_chunker/patterns.py" = ["ALL"]

# Ignore warnings about typer.Option(...)
"src/chunklet/cli.py" = ["B008"]

# Ignore warnings about File(...)
"src/chunklet/visualizer/visualizer.py" = ["B008"]

# Ignore E402 - warnings.filterwarnings() must be called before imports to suppress pkg_resources
"src/chunklet/sentence_splitter/sentence_splitter.py" = ["E402"]

[tool.ruff.format]
# Replaces Black exclude logic
exclude = [
    ".git", ".venv", "build", "dist",
    "src/chunklet/code_chunker/patterns.py"
]