Clicking a category or group tag filtered the homepage table in place while rewriting the address bar to the category URL, so one URL rendered two different pages and readers never reached category pages with their intros and guides; category pages also showed a redundant self-referential filter bar.
Co-Authored-By: Claude <noreply@anthropic.com>
Covers Pydantic for API input and config, Pandera for dataframes, and jsonschema for JSON Schema validation.
Co-Authored-By: Claude <noreply@anthropic.com>
Covers pytest as the default, a how-to-choose item per README subcategory, and a guide on Hypothesis, Playwright, tox/Nox, mocks, and coverage.
Co-Authored-By: Claude <noreply@anthropic.com>
Covers librosa for audio analysis, MoviePy for scripted video editing, VidGear for real-time video, Mutagen vs tinytag for tag read/write and licensing, and beets as a CLI tag/organization tool.
Co-Authored-By: Claude <noreply@anthropic.com>
Covers OpenCV as the default, Ultralytics YOLO for detection/segmentation/pose models and its AGPL-3.0/Enterprise License terms, Kornia for GPU-batch vision ops, FiftyOne for dataset curation, and pytesseract vs EasyOCR for OCR.
Co-Authored-By: Claude <noreply@anthropic.com>
Ruff for linting and formatting, a type checker alongside it, and pre-commit to run them, with per-tool guidance sourced from each project's own docs.
Co-Authored-By: Claude <noreply@anthropic.com>
Explains when to pick Wagtail (developer-defined page types) versus
django CMS (editors composing pages live), based on each project's
own documentation.
Co-Authored-By: Claude <noreply@anthropic.com>
Category pages sorted rows by downloads, which buried editorial leads (the Django ORM showed as row 7 of 7, tkinter as 14 of 14) against CONTRIBUTING.md's "position is the marker"; the full intro in the hero pushed the table 2.5 screens down on a phone; and every page repeated the same "Search every project in one place" H2.
Co-Authored-By: Claude <noreply@anthropic.com>
Every entry description row carried aria-hidden="true", hiding descriptions that sighted readers can see from screen readers.
Co-Authored-By: Claude <noreply@anthropic.com>
Category page titles read "ORM Python Libraries", but people search "python orm", so the title word order didn't match the query.
Co-Authored-By: Claude <noreply@anthropic.com>
7 of its 13 items carried links while the other intros' lists carry none, and the entry table already links every project.
Co-Authored-By: Claude <noreply@anthropic.com>
The list had 22 items, which in the upcoming layout sits above the entry table and would push it 4-5 phone screens down; it now has one item per README subcategory, in README order, reusing the intro's own wording.
Co-Authored-By: Claude <noreply@anthropic.com>
The old intro told every FastAPI app to use SQLModel, which SQLModel's own docs don't claim beyond simple cases, and leaned on API names from SQLAlchemy's 2.0 rename (Mapped, mapped_column).
Co-Authored-By: Claude <noreply@anthropic.com>
The old intro's lead used the same "For a Python X library, use …" template as other category pages, and it carried version-bound usage tips (Qt Widgets vs Qt Quick, pyside6-uic, per-toolkit threading helpers) instead of each project's recommended setup.
Co-Authored-By: Claude <noreply@anthropic.com>
The AI and Agents category page had no intro, so readers got a 35-project list with no guidance on which library to pick for building an agent, serving a model, or fine-tuning.
Co-Authored-By: Claude <noreply@anthropic.com>
Category pages carried no text of their own beyond the README one-line description, and most meta descriptions fell back to a generic "Explore N curated Python projects" line, which correlated with weak search rankings for category queries. This adds optional per-category intro markdown files rendered under the H1, with the first paragraph used as the meta description and links opening in a new tab, starting with the ORM category.
Co-Authored-By: Claude <noreply@anthropic.com>
Renamed or dissolved category slugs (e.g. /categories/web-servers/rpc/, /categories/code-analysis/code-linters/) returned 404, Search Console lists 7 of them, and each audit re-home was dropping the old URL's ranking.
Co-Authored-By: Claude <noreply@anthropic.com>
The uv-audit bug had no automated guard: pypi_name_overrides.json is
a manual registry, so a wrong-package mapping is only caught if
someone already suspects it.
Two broader checks were measured against the real list and rejected.
Checking that PyPI metadata links back to the entry's GitHub repo
would not have caught uv-audit, since that package declares no
home_page or project_urls, landing it in a 26-entry bucket of
packages that simply don't declare a repo (numba, selenium, pyglet,
etc.), plus 10 benign cases of orgs moving or splitting bindings.
Flagging display-name/repo-name mismatches yields 46 hits, all
legitimate python-X-repo-to-X-package pairs, with uv-build sitting
among them despite being a real Astral package with the identical
shape to uv-audit.
What discriminates is the bundled marker itself: a "(part of X)"
entry ships inside something else and has no package of its own, so
the sweep must never query it. This test walks the real README and
requires a null override for every bundled entry whose normalized
name is PyPI-shaped. Verified it fails with exactly the uv-audit
message when that override is removed, and passes with it restored,
across the three current bundled entries with no false positives. It
runs offline, fitting the existing network-less CI.
Co-Authored-By: Claude <noreply@anthropic.com>
Renaming the entry from "uv audit" to "uv-audit" made the name PyPI-shaped: normalize() leaves spaces alone, so "uv audit" failed PYPI_NAME_RE and collect_names skipped it, but "uv-audit" passes, so the next sweep would have queried PyPI for it.
A uv-audit package does exist on PyPI, but it is version 0.1.9 by Alekse Marusich of rocshers, an unrelated third-party tool whose summary ("uv Tool for checking dependencies for vulnerabilities") is close enough to be mistaken for Astral's built-in uv audit subcommand. Without the override the entry would have shown that stranger's download count and lost its Bundled badge.
The sweep now writes uv-audit as NOT_FOUND, which load_downloads skips, so the badge is unaffected.
Co-Authored-By: Claude <noreply@anthropic.com>
azure-sdk-for-python and google-cloud-python were rendering "Not on
PyPI", which is misleading. Both do ship on PyPI, just as many
per-service packages (azure-identity, azure-storage-blob,
google-cloud-storage, etc.) rather than under the repo name.
pypi_name_overrides.json already recorded that distinction in its
reason field; those two entries now carry an optional "badge" value
that build.py reads into the PyPI Downloads column. The other sixteen
no-count entries (cpython, renpy, agent skill repos, etc.) keep
"Not on PyPI" since that remains accurate for them.
Co-Authored-By: Claude <noreply@anthropic.com>
django.db.models, geodjango, httpx.URL and uv audit were rendering
"Not on PyPI" alongside eighteen genuinely standalone projects that
simply are not packaged on PyPI, conflating two different reasons for
a missing download count.
These four entries now carry a "(part of X)" description prefix in
README.md, mirroring the existing "(Python standard library)"
convention. build.py reads that prefix into a bundled flag that both
templates render as a "Bundled" badge.
The prefix approach was chosen over a separate data file so README.md
stays the single source of content truth, and over inferring from the
entry name because geodjango is neither dotted nor spaced and would
have been missed. Redundant tail wording was trimmed from the
httpx.URL, geodjango and uv audit descriptions now that the prefix
names the parent.
The new entry format is documented in CONTRIBUTING.md and the
vocabulary in CONTEXT.md.
Co-Authored-By: Claude <noreply@anthropic.com>
Standard-library entries rendered "Not on PyPI" in the PyPI Downloads
column, which read like missing data rather than a deliberate category
— the build already forces downloads to None for them so they never
pick up a same-named PyPI backport. They now render a "Stdlib" badge
instead, while genuine non-PyPI entries keep "Not on PyPI".
The filter tag is renamed to match, so the source-type value, the row
filter tag, and the synthetic category heading all read Stdlib now.
The literal "Built-in" strings scattered through build.py are routed
through the existing BUILTIN_FILTER constant so the label lives in one
place. The category page slug stays "built-in" so the public URL
/categories/built-in/ does not break.
Co-Authored-By: Claude <noreply@anthropic.com>
Kicker now mirrors the hero kicker ("The definitive list that
answers..."), replacing the old "field guide" line changed in
8b14b4f. Subtitle now matches the current tagline ("An opinionated
guide to the best Python frameworks, libraries, and tools.") on one
line. PNG regenerated from the SVG.
Co-Authored-By: Claude <noreply@anthropic.com>
Entries now end with "(PyPI downloads/month: N, GitHub stars: M)" where
known, replacing the stars-only note, since download counts are the
list's stated primary evidence signal. annotate_entries_with_stars is
renamed to annotate_entries_with_stats and looks downloads up by the
first link's display name, skipping category-index bullets (which link
into the site itself) and Built-in entries (which would otherwise hit
same-named PyPI backports like logging or asyncio).
The intro now mirrors the README subtitle verbatim with the
project/category totals on their own line below, and the "opinionated
catalog" wording is gone since the shortlist ADR is literally titled
"shortlist, not a catalog".
Co-Authored-By: Claude <noreply@anthropic.com>
Replaces the em dash in the PyPI Downloads column with a source-badge
pill labeled "Not on PyPI", reusing the existing badge style used by
the stars column for visual consistency. Sorting is unaffected since
non-numeric cells already parse as missing.
Co-Authored-By: Claude <noreply@anthropic.com>
Downloads is now the default sort, so it sits directly after the
project name in both the index and category table templates. The
source-type badge stays in the stars cell.
Co-Authored-By: Claude <noreply@anthropic.com>
Entries with a download count now sort first (descending), with
stars, then Built-in, then name as fallback tiers for entries that
lack a count. main.js mirrors this in its default activeSort, clean
URL check, and third-click reset target. Sorting by stars remains one
header click away.
Co-Authored-By: Claude <noreply@anthropic.com>
Every entry is now {"package": str|null, "reason": str|null} instead of
a bare string/null. Reasons are required for null packages, explaining
why the name must never be queried (squatted name, stdlib module,
monorepo umbrella, GitHub-only project, and so on). Reasons are
optional for remaps and kept only on the six non-obvious ones: pytorch
(squatter), jinja (jinja is Jinja1), strawberry (unrelated bookmarking
service), django-rules (abandoned fork), django-rest-framework (dead
alias), and devpi (deprecated metapackage); plain publishes-as-X
remaps get a null reason.
load_overrides() in the clickpy fetcher now extracts the package field
from each entry; resolve() and the pepy/bigquery cross-check scripts
are unchanged since they consume load_overrides()'s output.
Co-Authored-By: Claude <noreply@anthropic.com>
Every queried name now resolves 447/447. Adds 23 explicit null
overrides so squatters can never silently attach a PyPI number to
these names later: stdlib-named entries (concurrent-futures, difflib,
mimetypes, sqlite3, tkinter, tomllib, zoneinfo), interpreters
(micropython, pypy), monorepo umbrellas (azure-sdk-for-python,
google-cloud-python), self-hosted or distro-installed projects (odoo,
cloud-init, warehouse), GitHub-only projects (thealgorithms,
geodjango, django-db-models, django-ai-plugins, graphify,
sentry-skills, social-engineer-toolkit, trailofbits-skills), and
httpx-url (a class within httpx, not a package).
Caveat: graphify and django-ai-plugins are young projects that may
legitimately publish to PyPI later — flip their null to a remap
during a future audit if they do.
Co-Authored-By: Claude <noreply@anthropic.com>
autobahn-python publishes as autobahn (7.1M/mo), pangu-py as pangu, and
strawberry-django as strawberry-graphql-django (1.5M/mo). httpx.URL is
left unmapped deliberately since it's a class within the httpx package,
not a package of its own.
Co-Authored-By: Claude <noreply@anthropic.com>
Sourced from website/data/pypi_downloads.tsv the same way
github_stars.json feeds the stars column. The new sortable column
sits between GitHub Stars and Last Commit on the homepage and
category pages, formatted with thousands separators like stars, with
an em dash when no PyPI data exists. Rows are matched by normalized
README display name; Built-in entries never show counts since
same-named PyPI packages are stdlib backports (e.g. the asyncio
package).
Below 960px the column hides and the count moves into the expand
row, mirroring the existing Last Commit treatment. main.js gains the
downloads sort branch and URL param.
The deploy workflow fetches the TSV via the new
make fetch_pypi_downloads target with a daily actions/cache
fallback, mirroring the stars fetch, but non-fatal: the column
degrades to dashes when the fetch fails, unlike stars which the
build requires.
Co-Authored-By: Claude <noreply@anthropic.com>
A pypi.org identity sweep of all 438 cached rows (project_urls/home_page
vs entry GitHub URL) found download counts were looked up by README
display name, so entries whose name differs from the canonical package
silently measured squatters or dead predecessors: pytorch measured a
squatter (169,737/mo vs torch's 94M), jinja measured Jinja1 (3,168 vs
jinja2's 736M), django-rest-framework a dead alias package (real:
djangorestframework), django-rules an abandoned fork (real: rules),
strawberry an unrelated bookmarking service (real: strawberry-graphql),
devpi a deprecated metapackage (mapped to devpi-server).
New curated website/data/pypi_name_overrides.json maps normalized
README name to the real package, or null for projects not
pip-installable whose name is squatted or a relic (cpython, pyenv,
renpy, python-patterns, winpython); also maps mem0 to mem0ai, fasthtml
to python-fasthtml, and playwright-python to playwright.
All three fetch scripts resolve names through it; the clickpy TSV
cache gains a package column recording what each row actually
measured. .gitignore switches website/data/ to website/data/* with a
negation so the curated overrides file is tracked while caches stay
ignored.
Co-Authored-By: Claude <noreply@anthropic.com>
Capture parser quirks worth knowing before editing README.md:
everything above is ignored, new subcategories need no
parser change, a standalone all-bold paragraph becomes a Thematic
Group marker, prose after leaks into llms.txt, and the
build's "Total entries" figure counts sub-items rather than just
entries.
Co-Authored-By: Claude <noreply@anthropic.com>
Note that data/github_stars.json is gitignored and fetched by CI at
deploy time, so local runs are preview-only and should never be
committed; entries removed from README.md just leave harmless orphan
keys behind.
Co-Authored-By: Claude <noreply@anthropic.com>
The pypi downloads sweep looks up counts by README display name; when
the display name differs from the canonical package, the row silently
measures an unrelated squatter or a dead predecessor. Document the
failure mode in both the fetcher's docstring and the audit skill so
famous entries with off-looking counts get identity-verified before
being cited.
Co-Authored-By: Claude <noreply@anthropic.com>
Adds a header row (name, downloads, fetched_at) to data/pypi_downloads.tsv
and stamps every row with the sweep date, so audits can tell evidence age
and skip re-fetching when the cache is less than 7 days old. The sweep
itself costs ~1s, so freshness is checked by the reader (audit-the-list
skill) instead of skip logic in the fetch script. SKILL.md documents the
7-day freshness rule.
Co-Authored-By: Claude <noreply@anthropic.com>
Authored by the parallel fetch-scripts session (its intended 0cffb5f
never landed; the content rode into an audit commit by accident and is
extracted here): fetch_pypi_downloads_via_clickpy.py becomes the
flagless full-README sweep and sole writer of data/pypi_downloads.tsv;
new fetch_pypi_downloads_via_bigquery.py is a print-only cross-check
taking explicit names behind a 400 GB billing cap; audit-the-list
SKILL.md step 2 updated to match.
Co-Authored-By: Claude <noreply@anthropic.com>
fetch_pypi_downloads.py becomes fetch_pypi_downloads_via_clickpy.py and
fetch_pepy_downloads.py becomes fetch_pypi_downloads_via_pepy.py, ahead
of splitting the BigQuery path into its own file. Updates the usage
strings, the cross-file import, and the audit-the-list skill's
references to match. Pure rename, no behavior change.
Co-Authored-By: Claude <noreply@anthropic.com>
Vinta ChenandClaude