| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259 |
- [project]
- name = "archivebox"
- version = "0.9.0rc1"
- requires-python = ">=3.13"
- description = "Self-hosted internet archiving solution."
- authors = [{name = "Nick Sweeting", email = "[email protected]"}]
- license = {text = "MIT"}
- readme = "README.md"
- keywords = ["internet archiving", "web archiving", "digipres", "warc", "preservation", "backups", "archiving", "web", "bookmarks", "puppeteer", "browser", "download"]
- classifiers = [
- "Development Status :: 4 - Beta",
- "Environment :: Console",
- "Environment :: Web Environment",
- "Framework :: Django",
- "Intended Audience :: Developers",
- "Intended Audience :: Education",
- "Intended Audience :: End Users/Desktop",
- "Intended Audience :: Information Technology",
- "Intended Audience :: Legal Industry",
- "Intended Audience :: System Administrators",
- "License :: OSI Approved :: MIT License",
- "Natural Language :: English",
- "Operating System :: OS Independent",
- "Programming Language :: Python :: 3",
- "Programming Language :: Python :: 3.13",
- "Programming Language :: Python :: 3.14",
- "Topic :: Internet :: WWW/HTTP",
- "Topic :: Internet :: WWW/HTTP :: Indexing/Search",
- "Topic :: Internet :: WWW/HTTP :: WSGI :: Application",
- "Topic :: Sociology :: History",
- "Topic :: Software Development :: Libraries :: Python Modules",
- "Topic :: System :: Archiving",
- "Topic :: System :: Archiving :: Backup",
- "Topic :: System :: Recovery Tools",
- "Topic :: Utilities",
- "Typing :: Typed",
- ]
- dependencies = [
- ### Django libraries
- "setuptools>=74.1.0", # for: django 5 on python >=3.12, distutils is no longer in stdlib but django 5.1 expects distutils (TODO: check if this can be removed eventually)
- "django>=6.0",
- "daphne>=4.2.0", # ASGI server for Django (no channels needed - websockets not used)
- "django-ninja>=1.5.1",
- "django-extensions>=3.2.3",
- "django-signal-webhooks>=0.3.0",
- "django-admin-data-views>=0.4.1",
- "django-object-actions>=4.3.0",
- "django-taggit==6.1.0", # TODO: remove this in favor of KVTags only
- ### State Management
- "python-statemachine>=2.3.6",
- ### CLI / Logging
- "click>=8.1.7", # for: nicer CLI command + argument definintions
- "rich>=13.8.0", # for: pretty CLI output
- "rich-click>=1.8.4", # for: pretty CLI command help text & output
- "ipython>=8.27.0", # for: archivebox shell (TODO: replace with bpython?)
- ### Host OS / System
- "supervisor>=4.2.5", # for: archivebox server starting daphne and workers
- "psutil>=6.0.0", # for: monitoring orchestractor,actors,workers,etc. and machine.models.Process
- "platformdirs>=4.3.6", # for: finding a xdg-config dir to store tmp/lib files in
- "py-machineid>=0.6.0", # for: machine/detect.py calculating unique machine guid
- "atomicwrites==1.4.1", # for: config file writes, index.json file writes, etc. (TODO: remove this deprecated lib in favor of archivebox.filestore.util/os.rename/os.replace)
- "python-crontab>=3.2.0", # for: archivebox schedule (TODO: remove this in favor of our own custom archivebox scheduler)
- "croniter>=3.0.3", # for: archivebox schedule (TODO: remove this in favor of our own custom archivebox scheduler)
- ### Base Types
- "pydantic>=2.8.0", # for: archivebox.api (django-ninja), archivebox.config (pydantic-settings), and archivebox.index.schema (pydantic)
- "pydantic-settings>=2.5.2", # for: archivebox.config
- "python-benedict[io,parse]>=0.33.2", # for: dict replacement all over the codebase to allow .attr-style access
- "base32-crockford>=0.3.0", # for: encoding UUIDs in base32
- ### Static Typing
- "mypy-extensions>=1.0.0", # for: django-stubs type hints (TODO: remove in favor of pylance/pyright?)
- "django-stubs>=5.0.4", # for: vscode type hints on models and common django APIs
- ### API clients
- "requests>=2.32.3", # for: fetching title, static files, headers (TODO: replace with httpx?)
- "sonic-client>=1.0.0",
- "pocket>=0.3.6", # for: importing URLs from Pocket API
- ### Parsers
- "feedparser>=6.0.11", # for: parsing pocket/pinboard/etc. RSS/bookmarks imports
- "dateparser>=1.2.0", # for: parsing pocket/pinboard/etc. RSS/bookmark import dates
- "tzdata>=2024.2", # needed for dateparser {TZ: UTC} on some systems: https://github.com/ArchiveBox/ArchiveBox/issues/1553
- "w3lib>=2.2.1", # used for parsing content-type encoding from http response headers & html tags
- ### Extractor dependencies (optional binaries detected at runtime via shutil.which)
- "yt-dlp>=2024.1.0", # for: media extractor
- ### Binary/Package Management
- "abx-pkg>=0.1.0", # for: detecting, versioning, and installing binaries via apt/brew/pip/npm
- "gallery-dl>=1.31.1",
- ### UUID7 backport for Python <3.14
- "uuid7>=0.1.0; python_version < '3.14'", # for: uuid7 support on Python 3.13 (provides uuid_extensions module)
- "pytest-django>=4.11.1",
- ]
- [project.optional-dependencies]
- sonic = [
- # sonic client lib now included by default, sonic group is now a no-op:
- # "sonic-client>=1.0.0",
- # to use sonic make sure you have a sonic server running in docker (archivebox/sonic) or locally:
- # echo "deb [signed-by=/usr/share/keyrings/valeriansaliou_sonic.gpg] https://packagecloud.io/valeriansaliou/sonic/debian/ bookworm main" > /etc/apt/sources.list.d/valeriansaliou_sonic.list
- # curl -fsSL https://packagecloud.io/valeriansaliou/sonic/gpgkey | gpg --dearmor -o /usr/share/keyrings/valeriansaliou_sonic.gpg
- # apt install sonic
- ]
- ldap = [
- # python-ldap depends on the openldap bindings which provide no prebuilt wheels becuase they link against tons of other system packages
- # apt install build-essential python3-dev python3-ldap libsasl2-dev libldap2-dev libssl-dev
- "python-ldap>=3.4.3",
- "django-auth-ldap>=4.1.0",
- ]
- debug = [
- # packages needed for running with DEBUG=True
- "django-debug-toolbar>=4.4.6",
- "djdt_flamegraph>=0.2.13",
- "ipdb>=0.13.13",
- "requests-tracker>=0.3.3",
- "django-autotyping>=0.5.1",
- ]
- all = [
- "archivebox[sonic,ldap,debug]"
- ]
- [dependency-groups]
- dev = [
- ### BUILD
- "uv>=0.4.26",
- "pip>=24.2",
- "setuptools>=75.1.0",
- "wheel>=0.44.0",
- "bumpver>=2023.1129",
- #"homebrew-pypi-poet>=0.10.0", # for: generating archivebox.rb brewfile list of python packages
- ### DOCS
- "recommonmark>=0.7.1",
- "sphinx>=8.1.3",
- "sphinx-rtd-theme>=2.0.0",
- "myst-parser>=4.0.0",
- "sphinx-autodoc2>=0.5.0",
- "linkify-it-py>=2.0.3",
- ### DEBUGGING
- "django-debug-toolbar>=4.4.6",
- "requests-tracker>=0.3.3",
- "djdt_flamegraph>=0.2.13",
- "ipdb>=0.13.13",
- "logfire[django]>=0.51.0",
- "opentelemetry-instrumentation-django>=0.47b0",
- "opentelemetry-instrumentation-sqlite3>=0.47b0",
- "viztracer>=0.17.0", # usage: viztracer ../.venv/bin/archivebox manage check
- # "snakeviz", # usage: python -m cProfile -o flamegraph.prof ../.venv/bin/archivebox manage check
- ### TESTING
- "pytest>=8.3.3",
- "bottle>=0.13.1",
- ### LINTING
- "ruff>=0.6.6",
- "flake8>=7.1.1",
- "mypy>=1.11.2",
- ]
- [tool.uv]
- environments = ["sys_platform == 'darwin'", "sys_platform == 'linux'"]
- package = true
- # compile-bytecode = true
- [tool.uv.pip]
- python-version = "3.13"
- # compile-bytecode = true
- [build-system]
- requires = ["pdm-backend"]
- build-backend = "pdm.backend"
- # https://github.com/astral-sh/uv/issues/3957
- [tool.setuptools]
- packages = ["archivebox"]
- package-dir = {"archivebox" = "archivebox"}
- [tool.ruff]
- line-length = 140
- target-version = "py313"
- src = ["archivebox"]
- exclude = ["*.pyi", "typings/", "migrations/"]
- # https://docs.astral.sh/ruff/rules/
- [tool.ruff.lint]
- ignore = ["E731", "E303", "E266", "E241", "E222"]
- [tool.pytest.ini_options]
- testpaths = [ "tests" ]
- DJANGO_SETTINGS_MODULE = "archivebox.core.settings"
- [tool.mypy]
- mypy_path = "archivebox,archivebox/typings"
- namespace_packages = true
- explicit_package_bases = true
- # follow_imports = "silent"
- # ignore_missing_imports = true
- # disallow_incomplete_defs = true
- # disallow_untyped_defs = true
- # disallow_untyped_decorators = true
- # exclude = "tests/.*"
- plugins = ["mypy_django_plugin.main"]
- [tool.django-stubs]
- django_settings_module = "core.settings"
- [tool.pyright]
- include = [
- "archivebox",
- ]
- exclude = [
- ".venv",
- "**/*.pyi",
- "**/__init__.pyi",
- "**/node_modules",
- "**/__pycache__",
- "**/migrations",
- ]
- stubPath = "./archivebox/typings"
- venvPath = "."
- venv = ".venv"
- # ignore = ["src/oldstuff"]
- # defineConstant = { DEBUG = true }
- reportMissingImports = true
- reportMissingTypeStubs = false
- pythonVersion = "3.13"
- pythonPlatform = "Linux"
- [project.scripts]
- archivebox = "archivebox.cli:main"
- [project.urls]
- Homepage = "https://github.com/ArchiveBox/ArchiveBox"
- Source = "https://github.com/ArchiveBox/ArchiveBox"
- Documentation = "https://github.com/ArchiveBox/ArchiveBox/wiki"
- "Bug Tracker" = "https://github.com/ArchiveBox/ArchiveBox/issues"
- Changelog = "https://github.com/ArchiveBox/ArchiveBox/releases"
- Roadmap = "https://github.com/ArchiveBox/ArchiveBox/wiki/Roadmap"
- Community = "https://github.com/ArchiveBox/ArchiveBox/wiki/Web-Archiving-Community"
- Demo = "https://demo.archivebox.io"
- Donate = "https://github.com/ArchiveBox/ArchiveBox/wiki/Donations"
- [tool.bumpver]
- current_version = "v0.8.5rc53"
- version_pattern = "vMAJOR.MINOR.PATCH[PYTAGNUM]"
- commit_message = "bump version {old_version} -> {new_version}"
- tag_message = "{new_version}"
- tag_scope = "default"
- pre_commit_hook = ""
- post_commit_hook = ""
- commit = true
- tag = true
- push = true
- [tool.bumpver.file_patterns]
- "pyproject.toml" = [
- 'current_version = "{version}"',
- 'version = "{pep440_version}"',
- ]
|