From 686fc7e540c4a8994375a73e29b2938d62a1ebc4 Mon Sep 17 00:00:00 2001 From: kdeldycke <159718+kdeldycke@users.noreply.github.com> Date: Tue, 11 Aug 2026 11:40:21 +0000 Subject: [PATCH] Format `pyproject.toml` --- pyproject.toml | 558 ++++++++++++++++++++++++------------------------- 1 file changed, 276 insertions(+), 282 deletions(-) diff --git a/pyproject.toml b/pyproject.toml index 110ff0e3..1b14a27c 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -2,6 +2,34 @@ # Docs: https://packaging.python.org/en/latest/guides/writing-pyproject-toml/ name = "blog" version = "1.0.0" +# Kept byte-identical to the GitHub repository description: `lint-repo` compares the +# two with a plain `!=`, with no normalization, so a stray emoji shortcode or a +# trailing full stop on one side is a failed check. Change both or neither. +description = "My personal blog, and the yak-shaving that keeps two decades of URLs alive." +readme = "readme.md" +keywords = [ + "atom-feed", + "blog-posts", + "blog-source", + "creative-commons", + "dev-blog", + "markdown-blog", + "myst", + "myst-markdown", + "myst-parser", + "pelican", + "pelican-blog", + "pelican-site", + "personal-blog", + "personal-site", + "python-blog", + "static-blog", + "static-website", + "technical-blog", + "wrangler", + "yak-shaving", +] +license = { text = "CC-BY-NC-SA-4.0" } # A floor rather than a compatibility claim, and the only Python control in the repo. # Nothing here needs 3.14 and the corpus built fine on 3.12, but this blog has one # contributor and one pipeline, so the range never describes anyone else's @@ -22,87 +50,215 @@ version = "1.0.0" # The cost is one upstream pin per bump. 3.14 needed pelican-search 1.1.1, whose # predecessor capped rtoml below the release that first shipped cp314 wheels. Expect # to go hunting for the equivalent when 3.15 lands. -requires-python = ">= 3.14" -# Kept byte-identical to the GitHub repository description: `lint-repo` compares the -# two with a plain `!=`, with no normalization, so a stray emoji shortcode or a -# trailing full stop on one side is a failed check. Change both or neither. -description = "My personal blog, and the yak-shaving that keeps two decades of URLs alive." -license = { text = 'CC-BY-NC-SA-4.0' } -readme = "readme.md" -keywords = [ - 'atom-feed', - 'blog-posts', - 'blog-source', - 'creative-commons', - 'dev-blog', - 'markdown-blog', - 'myst', - 'myst-markdown', - 'myst-parser', - 'pelican', - 'pelican-blog', - 'pelican-site', - 'personal-blog', - 'personal-site', - 'python-blog', - 'static-blog', - 'static-website', - 'technical-blog', - 'wrangler', - 'yak-shaving', -] +requires-python = ">=3.14" classifiers = [ - # See: https://pypi.org/pypi?%3Aaction=list_classifiers - 'Development Status :: 5 - Production/Stable', - 'Environment :: Web Environment', - 'Framework :: Pelican', - 'License :: Other/Proprietary License', - 'Natural Language :: English', - 'Natural Language :: French', - 'Operating System :: OS Independent', - "Programming Language :: Python :: 3", - "Programming Language :: Python :: 3.12", - 'Programming Language :: Python :: Implementation :: CPython', - 'Programming Language :: Other', - 'Topic :: Internet :: WWW/HTTP :: Dynamic Content :: News/Diary', - 'Topic :: Text Processing :: Markup :: HTML', - 'Topic :: Text Processing :: Markup :: Markdown', - 'Typing :: Typed', + # See: https://pypi.org/pypi?%3Aaction=list_classifiers + "Development Status :: 5 - Production/Stable", + "Environment :: Web Environment", + "Framework :: Pelican", + "License :: Other/Proprietary License", + "Natural Language :: English", + "Natural Language :: French", + "Operating System :: OS Independent", + "Programming Language :: Other", + "Programming Language :: Python :: 3 :: Only", + "Programming Language :: Python :: 3.10", + "Programming Language :: Python :: 3.11", + "Programming Language :: Python :: 3.12", + "Programming Language :: Python :: 3.13", + "Programming Language :: Python :: 3.14", + "Programming Language :: Python :: Implementation :: CPython", + "Topic :: Internet :: WWW/HTTP :: Dynamic Content :: News/Diary", + "Topic :: Text Processing :: Markup :: HTML", + "Topic :: Text Processing :: Markup :: Markdown", + "Typing :: Typed", ] dependencies = [ - "pelican[Markdown]>=4.10.0", - "typogrify>=2.0.7", - "plumage>=5.0.0", - "pelican-seo>=1.2.2", - "pelican-search>=1.1.0", - # "pelican-similar-posts ~= 1.0.0", # XXX Broken! - "pelican-sitemap>=1.1.0", - "pelican-neighbors>=1.2.0", - # "gensim ~= 4.3.0", # XXX Broken! + "pelican[Markdown]>=4.10", + "pelican-neighbors>=1.2", + "pelican-search>=1.1", + "pelican-seo>=1.2.2", + # "pelican-similar-posts ~= 1.0.0", # XXX Broken! + "pelican-sitemap>=1.1", + "plumage>=5", + "typogrify>=2.0.7", + # "gensim ~= 4.3.0", # XXX Broken! +] +optional-dependencies.test = [ + "pytest>=8.3.2", + # More pytest plugins at: https://docs.pytest.org/en/latest/reference/plugin_list.html + "pytest-github-actions-annotate-failures>=0.2", + "pytest-randomly>=3.15", + # Most of this suite waits on the network rather than on a CPU: every rule in + # `_redirects` and every rule in `_headers` is one live request to production. That + # is the workload xdist is best at, and it takes the local run from about four + # minutes to under three. + # + # The psutil extra is what makes `--numprocesses=logical` available, should the + # split ever want tuning beyond physical cores. + "pytest-xdist[psutil]>=3.7", + "requests>=2.32.3", ] +optional-dependencies.typing = [ "types-requests>=2.32.0.20240622" ] +urls.Funding = "https://github.com/sponsors/kdeldycke" +urls.Homepage = "https://kevin.deldycke.com" +urls.Repository = "https://github.com/kdeldycke/blog" + +[tool.uv] +# Both pins below are repomatic's, written by `repomatic init uv` and left alone +# afterwards. Floor uv rather than cap it: astral-sh/setup-uv reads `required-version` +# directly and installs the newest release satisfying it, so a ceiling would break every +# machine the day the next minor lands. +required-version = ">=0.11.15" +# Cooldown before a newly published package can be resolved, held equal to +# `[tool.repomatic] minimum-release-age` so the window here never excludes a version the +# `uvx` installs in CI would pick. Until this landed the cooldown came from +# ~/.config/uv/uv.toml alone, which is one machine's config and does not travel: a fresh +# clone resolved with no window at all. +exclude-newer = "1 week" +# The theme is my own, published by me: the cooldown above guards against a compromised +# upstream, which does not apply here, and it otherwise holds each release back for a +# week. A zero duration rather than a fixed date, so the next release is picked up without +# editing this file. Not a repomatic-owned key: the sync leaves it untouched. +exclude-newer-package = { plumage = "0 days" } +package = false -[project.optional-dependencies] -test = [ - "pytest>=8.3.2", - # More pytest plugins at: https://docs.pytest.org/en/latest/reference/plugin_list.html - "pytest-github-actions-annotate-failures>=0.2.0", - "pytest-randomly>=3.15.0", - # Most of this suite waits on the network rather than on a CPU: every rule in - # `_redirects` and every rule in `_headers` is one live request to production. That - # is the workload xdist is best at, and it takes the local run from about four - # minutes to under three. - # - # The psutil extra is what makes `--numprocesses=logical` available, should the - # split ever want tuning beyond physical cores. - "pytest-xdist[psutil]>=3.7", - "requests>=2.32.3", +[tool.typos] +default.extend-ignore-re = [ + # error: `ba` should be `by`, `be` + # --> ./content/2011/postgresql-commands.md:47:95 + # | + # 47 | $ psql --tuples-only --no-align -d database_id -c "SELECT id FROM res_users;" | sed ':a;N;$!ba;s/\n/ /g' + # | ^^ + # | + # Issue reported at: https://github.com/crate-ci/typos/issues/642#issuecomment-2058343059 + "!ba;", + # error: `classe` should be `class`, `classes` + # --> ./content/2010/convert-lotus-notes-nsf-files-mbox-nlconverter.md:120:77 + # | + # 120 | pywintypes.com_error: (-2147221231, 'ClassFactory ne peut pas fournir la classe demand\xe9e', None, None) + # | ^^^^^^ + # | + " ne peut pas fournir la classe ", + # error: `hda` should be `had` + # --> content/2005/qemu-commands.md:19:29 + # | + # 19 | $ qemu -cdrom /dev/cdrom -hda /home/kevin/qemu-disk-image.qcow -boot d + # | ^^^ + # | + # Issue reported at: https://github.com/crate-ci/typos/issues/998 + " -hda ", + # error: `espace` should be `escape` + # --> ./content/2005/easy-mirroring-without-raid-the-poor-mans-disk-array.md:10:116 + # | + # 10 | ... in USB enclosures]({filename}/2005/creer-un-espace-de-stockage-fiable-avec-raid-5-et-lvm-sous-linux.md) ... + # | ^^^^^^ + # | + "creer-un-espace-de-stockage-fiable-avec-raid-5-et-lvm-sous-linux.md", + # error: `cylindres` should be `cylinders` + # --> ./content/2005/how-to-mount-a-file-system-binary-image.md:28:31 + # | + # 28 | 9 têtes, 56 secteurs/piste, 0 cylindres, total 0 sectors + # | ^^^^^^^^^ + # | + " cylindres, ", + # error: `THM` should be `THEM` + # --> ./content/2011/nichrome-preview-behind-the-scene-video.md:57:3 + # | + # 57 | `.THM` preview files generated by the camera for each `.MOV` clip. + # | ^^^ + # | + ".THM", + # error: `ND` should be `AND` + # --> ./content/2010/making-info-tgv-android-app-video.md:28:414 + # | + # 28 | ... so I attached a variable ND-filter to the lens ... + # | ^^ + # | + " ND-filter", + # error: `Shure` should be `Sure` + # --> ./content/2010/making-info-tgv-android-app-video.md:68:154 + # | + # 68 | ... me his Shure Beta57A microphone ... + # | ^^^^^ + # | + "Shure ", + # error: `shure` should be `sure` + # --> ./content/2010/making-info-tgv-android-app-video.md:70:13 + # | + # 70 | ![]({attach}shure-beta-57a-microphone-on-gorillapod.jpg) + # | ^^^^^ + # | + "shure-", + # error: `hda` should be `had` + # --> ./content/2006/hardware-commands.md:25:28 + # | + # 25 | - Low-level format of the `hda` device: + # | ^^^ + # | + "`hda`", + # error: `dbe` should be `be` + # --> ./content/2008/hp-w2207h-external-display-on-intel-945.md:37:9 + # | + # 37 | Load "dbe" # Double-Buffering Extension + # | ^^^ + # | + ' "dbe" ', + # error: `abitrate` should be `arbitrate` + # --> ./content/2006/video-commands.md:297:140 + # | + # 297 | ... vcodec=mpeg4:vbitrate=400:v4mv:mbd=2:trell:autoaspect:dia=2:acodec=mp3:abitrate=32:vpass=1 -vf hqdn3d -o output.avi + # | ^^^^^^^^ + # | + ":abitrate=", +] # Sections of text to ignore. +# Exclude French articles. +files.extend-exclude = [ + "content/2004/mise-en-place-paserelle-adsl-mandrake-10.md", + "content/2005/creer-un-espace-de-stockage-fiable-avec-raid-5-et-lvm-sous-linux.md", + "content/2005/script-de-reconnection-automatique-ppp.md", + "content/2006/ajouter-une-entree-jack-universelle-sur-un-telephone-sagem-my700xi.md", + "content/2009/cool-cavemen-creative-commons-moi.md", + "content/2010/comment-supprimer-des-comptes-mails-secondaires-free-fr.md", + "content/2012/ovh-configuration-modem-thomson-tg585-ipoe.md", ] -typing = ["types-requests>=2.32.0.20240622"] -[project.urls] -"Homepage" = 'https://kevin.deldycke.com' -"Repository" = 'https://github.com/kdeldycke/blog' -"Funding" = "https://github.com/sponsors/kdeldycke" +[tool.mypy] +warn_redundant_casts = true +warn_unused_ignores = true +warn_return_any = true +warn_unreachable = true +pretty = true +warn_unused_configs = true +overrides = [ + # Pelican ships no py.typed marker, so importing it is an error under the settings above. + # Silenced here rather than with an inline ignore on each import, because every local + # workaround in pelican_patches.py has to reach into the engine to hook onto it. Drop + # this once Pelican starts advertising its own types. + { module = [ "pelican.*" ], ignore_missing_imports = true }, +] + +[tool.pytest] +# The tests import pelicanconf and pelican_patches, which sit at the repository root. +# This project is not a package (see [tool.uv] below), so there is nothing installed for +# them to be imported from: pytest only puts the test file's own folder on sys.path, and +# from tests/ that no longer covers the root. Needed for the `pytest` console script CI +# runs, which unlike `python -m pytest` does not prepend the working directory. +ini_options.pythonpath = [ "." ] +# https://docs.pytest.org/en/latest/customize.html#pyproject-toml +# `--numprocesses=auto` splits the run across cores. It pays off because the suite is +# mostly live HTTP against production, so workers spend their time waiting rather than +# competing, and it costs a fraction of a second on the offline files. +# +# It only works because every test ID is deterministic. xdist compares what each worker +# collected and aborts the whole run when they disagree, which is what the generated +# redirect cases used to cause: they drew their placeholder values from the global +# random state at collection time, so every worker invented different URLs. `token()` +# in `test_redirects.py` seeds off the rule instead. Keep it that way, or this flag +# turns every run into "Different tests were collected between gw0 and gw1". +ini_options.addopts = [ "--durations=10", "--numprocesses=auto" ] +# Make sure tests that are expected to fail do not resurect and start working all of a sudden. +ini_options.xfail_strict = true [tool.lychee] # Since v0.24 lychee reads this table straight out of pyproject.toml, so repomatic runs no @@ -111,34 +267,34 @@ typing = ["types-requests>=2.32.0.20240622"] # entries are therefore restated below rather than inherited. Everything from `archive.ph` # down is a verbatim copy; only the first entry is specific to this repo. exclude = [ - # Pelican's internal link markers are resolved at build time, so they only ever exist - # in the Markdown source. lychee reads that source and rewrites `]({attach}foo.jpg)` - # into `file://.../%7Battach%7Dfoo.jpg`, a path nothing on disk answers to. Left alone - # that is ~690 phantom failures across 164 posts, enough to bury a real one. Matching - # the encoded braces rather than the literal `{attach}` is deliberate: the check runs - # after lychee has already percent-encoded them. - "%7B(attach|author|category|filename|index|static|tag)%7D", - # GitHub issue comment fragments are dynamically loaded and can't be verified. - "^https://github.com/.+/issues/[0-9]+#issuecomment-.*$", - # Release binary download URLs redirect to the GitHub CDN (objects.githubusercontent.com). - "^https://github.com/.+/releases/.+/download/.*$", - # Allow redirects of generic DOI URLs to their specific Zenodo release page. - "^https://doi.org/.+/zenodo.*$", - # Skip archive.ph and Sci-Hub because they restrict access to crawlers. - # Skip HN because of rate-limiting. - # See: https://github.com/lycheeverse/lychee/issues/989#issuecomment-1587208730 - # https://github.com/lycheeverse/lychee/pull/1147 - "archive\\.ph", - "sci-hub\\.st", - "ycombinator\\.com", - "x\\.com", - # Same story as x.com above, and this corpus predates the rename: 29 links still point - # at the old domain, which answers a crawler with a login wall. Not in the bundled - # config, which covers repositories whose prose is younger than the rename. - "twitter\\.com", - # VirusTotal analysis pages are a JS app that rate-limits bots, and the - # binaries page links one per released binary. - "^https://www\\.virustotal\\.com/gui/.*$", + # Pelican's internal link markers are resolved at build time, so they only ever exist + # in the Markdown source. lychee reads that source and rewrites `]({attach}foo.jpg)` + # into `file://.../%7Battach%7Dfoo.jpg`, a path nothing on disk answers to. Left alone + # that is ~690 phantom failures across 164 posts, enough to bury a real one. Matching + # the encoded braces rather than the literal `{attach}` is deliberate: the check runs + # after lychee has already percent-encoded them. + "%7B(attach|author|category|filename|index|static|tag)%7D", + # GitHub issue comment fragments are dynamically loaded and can't be verified. + "^https://github.com/.+/issues/[0-9]+#issuecomment-.*$", + # Release binary download URLs redirect to the GitHub CDN (objects.githubusercontent.com). + "^https://github.com/.+/releases/.+/download/.*$", + # Allow redirects of generic DOI URLs to their specific Zenodo release page. + "^https://doi.org/.+/zenodo.*$", + # Skip archive.ph and Sci-Hub because they restrict access to crawlers. + # Skip HN because of rate-limiting. + # See: https://github.com/lycheeverse/lychee/issues/989#issuecomment-1587208730 + # https://github.com/lycheeverse/lychee/pull/1147 + "archive\\.ph", + "sci-hub\\.st", + "ycombinator\\.com", + "x\\.com", + # Same story as x.com above, and this corpus predates the rename: 29 links still point + # at the old domain, which answers a crawler with a login wall. Not in the bundled + # config, which covers repositories whose prose is younger than the rename. + "twitter\\.com", + # VirusTotal analysis pages are a JS app that rate-limits bots, and the + # binaries page links one per released binary. + "^https://www\\.virustotal\\.com/gui/.*$", ] [tool.repomatic] @@ -149,7 +305,6 @@ exclude = [ # PACKAGE_ONLY and was never migrated to the narrower `is_python_package` metadata # key (fixed upstream, pending release). Drop this once that ships. bumpversion.sync = false - # The release lane used to be listed here too: changelog, changelog.yaml, release.yaml # and publish-pypi-action. RepoScope.PACKAGE_ONLY now keeps all four away from a # `[tool.uv] package = false` project on its own, so naming them only restated what @@ -157,9 +312,8 @@ bumpversion.sync = false # debug.yaml is the one that stays: it is PYTHON_ONLY, and this repo really is a Python # project. It just has nothing to debug, being a Pelican toolchain rather than a package. exclude = [ - "workflows/debug.yaml", + "workflows/debug.yaml", ] - # `sync-gitignore` rebuilds .gitignore from gitignore.io and appends this key, in that # order and nothing else: the file on disk is never read. So the Pelican rules below only # survive by being declared here, and until they were, they sat in the generated file @@ -170,8 +324,7 @@ exclude = [ # tree this repo does not have. `.cc-writes/` is repomatic's own addition rather than a # bundled default, and is worth carrying: Claude Code drops that staging directory into # whichever tree it is launched from. -[tool.repomatic.gitignore] -extra-content = """ +gitignore.extra-content = """ # Claude Code local files. .claude/scheduled_tasks.lock .claude/settings.local.json @@ -191,22 +344,20 @@ seo_report.html # only works because extra-content is appended after the generated rules. !scripts/ """ - # The default filter is written for a repo whose prose lives in docs/ and readme.md. Here # the prose is the product: every post is under content/, so without it below a new article # ships without ever having its links checked. An override rather than extra_paths because # extra_paths only appends, and two of the defaults need removing, not supplementing: # changelog.md and citation.cff name files this repo does not have. An override replaces # the default outright, so what follows is the entire filter. -[tool.repomatic.workflow.paths] -"docs.yaml" = [ - ".github/workflows/docs.yaml", - "content/**", - "docs/**", - "pyproject.toml", - "readme.md", - "uv.lock", - "**/*.py", +workflow.paths."docs.yaml" = [ + ".github/workflows/docs.yaml", + "content/**", + "docs/**", + "pyproject.toml", + "readme.md", + "uv.lock", + "**/*.py", ] # Same reasoning, and the same reason the default misses: source_paths derives from # [project] name, so the stock filter watches blog/**, a directory this repo does not @@ -220,169 +371,12 @@ seo_report.html # pelican_patches.py joined the list when the local Pelican workarounds moved out of # pelicanconf.py: it runs on every build, so a change to it has to re-run the tests. # The suite itself is globbed, since pytest discovers it the same way. -"tests.yaml" = [ - ".github/workflows/tests.yaml", - "content/**", - "pelican_patches.py", - "pelicanconf.py", - "pyproject.toml", - "tests/**", - "uv.lock", -] - -[tool.uv] -package = false -# Both pins below are repomatic's, written by `repomatic init uv` and left alone -# afterwards. Floor uv rather than cap it: astral-sh/setup-uv reads `required-version` -# directly and installs the newest release satisfying it, so a ceiling would break every -# machine the day the next minor lands. -required-version = ">=0.11.15" -# Cooldown before a newly published package can be resolved, held equal to -# `[tool.repomatic] minimum-release-age` so the window here never excludes a version the -# `uvx` installs in CI would pick. Until this landed the cooldown came from -# ~/.config/uv/uv.toml alone, which is one machine's config and does not travel: a fresh -# clone resolved with no window at all. -exclude-newer = "1 week" -# The theme is my own, published by me: the cooldown above guards against a compromised -# upstream, which does not apply here, and it otherwise holds each release back for a -# week. A zero duration rather than a fixed date, so the next release is picked up without -# editing this file. Not a repomatic-owned key: the sync leaves it untouched. -exclude-newer-package = { plumage = "0 days" } - -[tool.mypy] -warn_unused_configs = true -warn_redundant_casts = true -warn_unused_ignores = true -warn_return_any = true -warn_unreachable = true -pretty = true - -# Pelican ships no py.typed marker, so importing it is an error under the settings above. -# Silenced here rather than with an inline ignore on each import, because every local -# workaround in pelican_patches.py has to reach into the engine to hook onto it. Drop -# this once Pelican starts advertising its own types. -[[tool.mypy.overrides]] -module = ["pelican.*"] -ignore_missing_imports = true - -[tool.pytest.ini_options] -# https://docs.pytest.org/en/latest/customize.html#pyproject-toml -# `--numprocesses=auto` splits the run across cores. It pays off because the suite is -# mostly live HTTP against production, so workers spend their time waiting rather than -# competing, and it costs a fraction of a second on the offline files. -# -# It only works because every test ID is deterministic. xdist compares what each worker -# collected and aborts the whole run when they disagree, which is what the generated -# redirect cases used to cause: they drew their placeholder values from the global -# random state at collection time, so every worker invented different URLs. `token()` -# in `test_redirects.py` seeds off the rule instead. Keep it that way, or this flag -# turns every run into "Different tests were collected between gw0 and gw1". -addopts = ["--durations=10", "--numprocesses=auto"] -# The tests import pelicanconf and pelican_patches, which sit at the repository root. -# This project is not a package (see [tool.uv] below), so there is nothing installed for -# them to be imported from: pytest only puts the test file's own folder on sys.path, and -# from tests/ that no longer covers the root. Needed for the `pytest` console script CI -# runs, which unlike `python -m pytest` does not prepend the working directory. -pythonpath = ["."] -# Make sure tests that are expected to fail do not resurect and start working all of a sudden. -xfail_strict = true - -[tool.typos] -default.extend-ignore-re = [ - # error: `ba` should be `by`, `be` - # --> ./content/2011/postgresql-commands.md:47:95 - # | - # 47 | $ psql --tuples-only --no-align -d database_id -c "SELECT id FROM res_users;" | sed ':a;N;$!ba;s/\n/ /g' - # | ^^ - # | - # Issue reported at: https://github.com/crate-ci/typos/issues/642#issuecomment-2058343059 - "!ba;", - # error: `classe` should be `class`, `classes` - # --> ./content/2010/convert-lotus-notes-nsf-files-mbox-nlconverter.md:120:77 - # | - # 120 | pywintypes.com_error: (-2147221231, 'ClassFactory ne peut pas fournir la classe demand\xe9e', None, None) - # | ^^^^^^ - # | - " ne peut pas fournir la classe ", - # error: `hda` should be `had` - # --> content/2005/qemu-commands.md:19:29 - # | - # 19 | $ qemu -cdrom /dev/cdrom -hda /home/kevin/qemu-disk-image.qcow -boot d - # | ^^^ - # | - # Issue reported at: https://github.com/crate-ci/typos/issues/998 - " -hda ", - # error: `espace` should be `escape` - # --> ./content/2005/easy-mirroring-without-raid-the-poor-mans-disk-array.md:10:116 - # | - # 10 | ... in USB enclosures]({filename}/2005/creer-un-espace-de-stockage-fiable-avec-raid-5-et-lvm-sous-linux.md) ... - # | ^^^^^^ - # | - "creer-un-espace-de-stockage-fiable-avec-raid-5-et-lvm-sous-linux.md", - # error: `cylindres` should be `cylinders` - # --> ./content/2005/how-to-mount-a-file-system-binary-image.md:28:31 - # | - # 28 | 9 têtes, 56 secteurs/piste, 0 cylindres, total 0 sectors - # | ^^^^^^^^^ - # | - " cylindres, ", - # error: `THM` should be `THEM` - # --> ./content/2011/nichrome-preview-behind-the-scene-video.md:57:3 - # | - # 57 | `.THM` preview files generated by the camera for each `.MOV` clip. - # | ^^^ - # | - ".THM", - # error: `ND` should be `AND` - # --> ./content/2010/making-info-tgv-android-app-video.md:28:414 - # | - # 28 | ... so I attached a variable ND-filter to the lens ... - # | ^^ - # | - " ND-filter", - # error: `Shure` should be `Sure` - # --> ./content/2010/making-info-tgv-android-app-video.md:68:154 - # | - # 68 | ... me his Shure Beta57A microphone ... - # | ^^^^^ - # | - "Shure ", - # error: `shure` should be `sure` - # --> ./content/2010/making-info-tgv-android-app-video.md:70:13 - # | - # 70 | ![]({attach}shure-beta-57a-microphone-on-gorillapod.jpg) - # | ^^^^^ - # | - "shure-", - # error: `hda` should be `had` - # --> ./content/2006/hardware-commands.md:25:28 - # | - # 25 | - Low-level format of the `hda` device: - # | ^^^ - # | - "`hda`", - # error: `dbe` should be `be` - # --> ./content/2008/hp-w2207h-external-display-on-intel-945.md:37:9 - # | - # 37 | Load "dbe" # Double-Buffering Extension - # | ^^^ - # | - ' "dbe" ', - # error: `abitrate` should be `arbitrate` - # --> ./content/2006/video-commands.md:297:140 - # | - # 297 | ... vcodec=mpeg4:vbitrate=400:v4mv:mbd=2:trell:autoaspect:dia=2:acodec=mp3:abitrate=32:vpass=1 -vf hqdn3d -o output.avi - # | ^^^^^^^^ - # | - ":abitrate=", -] # Sections of text to ignore. -# Exclude French articles. -files.extend-exclude = [ - "content/2004/mise-en-place-paserelle-adsl-mandrake-10.md", - "content/2005/creer-un-espace-de-stockage-fiable-avec-raid-5-et-lvm-sous-linux.md", - "content/2005/script-de-reconnection-automatique-ppp.md", - "content/2006/ajouter-une-entree-jack-universelle-sur-un-telephone-sagem-my700xi.md", - "content/2009/cool-cavemen-creative-commons-moi.md", - "content/2010/comment-supprimer-des-comptes-mails-secondaires-free-fr.md", - "content/2012/ovh-configuration-modem-thomson-tg585-ipoe.md", +workflow.paths."tests.yaml" = [ + ".github/workflows/tests.yaml", + "content/**", + "pelican_patches.py", + "pelicanconf.py", + "pyproject.toml", + "tests/**", + "uv.lock", ]