Content plan by paddymul · Pull Request #638 · buckaroo-data/buckaroo

paddymul · 2026-03-20T20:37:34Z

No description provided.

github-actions · 2026-03-20T20:39:25Z

📦 TestPyPI package published

pip install --index-strategy unsafe-best-match --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ buckaroo==0.13.2.dev23491757659

or with uv:

uv pip install --index-strategy unsafe-best-match --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ buckaroo==0.13.2.dev23491757659

MCP server for Claude Code

claude mcp add buckaroo-table -- uvx --from "buckaroo[mcp]==0.13.2.dev23491757659" --index-strategy unsafe-best-match --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ buckaroo-table

📖 Docs preview

🎨 Storybook preview

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 38a51e33e6

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-03-20T20:42:18Z

docs/source/articles/embedding-guide.rst

+The HTML file references ``static-embed.js`` and ``static-embed.css``.
+These are included in the buckaroo package under ``buckaroo/static/`` —
+copy them alongside your HTML or serve them from a web server.


Stop telling users to copy assets that the wheel doesn't ship

Anyone following this section on a clean install will generate HTML that never renders: to_html() hardcodes static-embed.js/css, but the package build still only ships widget.js, compiled.css, standalone.js, and standalone.css (see pyproject.toml's tool.hatch.build.artifacts). The same missing files also break the new DDD pages on ReadTheDocs, because scripts/generate_ddd_static_html.py copies from buckaroo/static/ and only warns when static-embed.* is absent.

Useful? React with 👍 / 👎.

chatgpt-codex-connector · 2026-03-20T20:42:18Z

scripts/generate_ddd_static_html.py

+    ('weird-types-polars', 'Weird Types (Polars → Pandas)',
+     pl_df_with_weird_types_as_pandas(),


Generate the DDD Polars page from a real Polars DataFrame

This entry bypasses the Polars static-embed path by converting the sample to pandas before calling to_html(). buckaroo.artifact.prepare_buckaroo_artifact() only switches to PolarsBuckarooWidget for an actual pl.DataFrame, so the published weird-types-polars.html page won't exercise the Polars serializer/analysis at all and can miss regressions like Duration/Decimal formatting while the docs still claim Polars embedding works.

Useful? React with 👍 / 👎.

github-actions · 2026-03-20T21:10:06Z

📖 Docs preview

https://buckaroo-data.readthedocs.io/en/638/

Key pages on this branch:

This file belongs on the content-plan branch (PR #638), not here. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* docs: Dastardly DataFrame Dataset post with static embeds - DDD article with inline iframe embeds for 10 tricky DataFrames - Generator script to produce static embed HTML at RTD build time - RTD config: build JS bundle, generate DDD pages, copy to output - Fix: coerce Period/Interval/Timedelta to strings in parquet b64 (hyparquet can't decode pandas Arrow extension types) - Tests for weird-type parquet roundtrip - Ship static-embed.js/css in wheel (pyproject.toml artifacts) - Docs preview link in TestPyPI PR comment - BuckarooCompare article - Content plan outline Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: add voice/personality to DDD post, add type coverage research Incorporate colorful wording from buckaroo_writing notes into the DDD blog post intro ("wonderfully variant splendor", "weirdest DataFrames in the wild", "hard fought experience"). Add Buckaroo's displayability philosophy. New research doc tests every pandas (classic, extension, arrow-backed) and polars dtype through both JSON and parquet serialization paths, documenting which are covered by the DDD and test suite. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: add full dtype coverage table to DDD post Cross-engine table showing Yes/not tested/fail for every dtype across pandas-classic, pandas-arrow, and polars. Footnote noting that serialization complexity warrants its own follow-up post. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: add parquet, JS type, and display columns to dtype table Expand the dtype coverage table with three new columns showing the full pipeline: what Parquet physical type each dtype maps to, what JS type hyparquet produces after decode, and how buckaroo's display formatter renders it. Footnotes explain BigInt handling, coercion for types without native Parquet equivalents, and the two serialization failures. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: credit Cecil Curry / beartype for DDD inspiration The naming and early shape of the DDD was influenced by an exchange with leycec on beartype#529. Added a shout-out in the intro. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: update copyright years to include 2026 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: link first DDD mention to source on GitHub main Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: show actual rendered values in dtype table Buckaroo display column Replace generic type names with example values buckaroo actually displays: 1,234 for integers, 1d 2h 3m 4s for durations, 68656c6c6f for binary, etc. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: fix dtype table — actual display values, footnotes, no blank cells - Replace generic type names with actual buckaroo display values - Add footnotes for BigInt (stringified, no commas) and Period (time span) - Fill all blank cells with — for dtypes that don't exist in that engine - Clarify Period label as "Period (time span)" - Fix Nullable int/float/bool row with proper values Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * WIP * docs: remove content-plan.md from ddd-post branch This file belongs on the content-plan branch (PR #638), not here. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: address inline review comments on DDD article - Shorten Cecil Curry praise (remove "makes open source worth doing") - Add LLM-generated DataFrames and inherited pipelines to "why this matters" - Fix "three values" → "three non-numeric values" for infinity/NaN section - Remove duplicate None mention from MultiIndex rows section - Add note about planned MultiIndex-both-axes improvements - Update footnote 1: "putting together this table exposed areas that need work" - Fix comma→period typo in footnote 1 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

* docs: Dastardly DataFrame Dataset post with static embeds - DDD article with inline iframe embeds for 10 tricky DataFrames - Generator script to produce static embed HTML at RTD build time - RTD config: build JS bundle, generate DDD pages, copy to output - Fix: coerce Period/Interval/Timedelta to strings in parquet b64 (hyparquet can't decode pandas Arrow extension types) - Tests for weird-type parquet roundtrip - Ship static-embed.js/css in wheel (pyproject.toml artifacts) - Docs preview link in TestPyPI PR comment - BuckarooCompare article - Content plan outline Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: add voice/personality to DDD post, add type coverage research Incorporate colorful wording from buckaroo_writing notes into the DDD blog post intro ("wonderfully variant splendor", "weirdest DataFrames in the wild", "hard fought experience"). Add Buckaroo's displayability philosophy. New research doc tests every pandas (classic, extension, arrow-backed) and polars dtype through both JSON and parquet serialization paths, documenting which are covered by the DDD and test suite. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: add full dtype coverage table to DDD post Cross-engine table showing Yes/not tested/fail for every dtype across pandas-classic, pandas-arrow, and polars. Footnote noting that serialization complexity warrants its own follow-up post. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: add parquet, JS type, and display columns to dtype table Expand the dtype coverage table with three new columns showing the full pipeline: what Parquet physical type each dtype maps to, what JS type hyparquet produces after decode, and how buckaroo's display formatter renders it. Footnotes explain BigInt handling, coercion for types without native Parquet equivalents, and the two serialization failures. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: credit Cecil Curry / beartype for DDD inspiration The naming and early shape of the DDD was influenced by an exchange with leycec on beartype#529. Added a shout-out in the intro. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: update copyright years to include 2026 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: link first DDD mention to source on GitHub main Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: show actual rendered values in dtype table Buckaroo display column Replace generic type names with example values buckaroo actually displays: 1,234 for integers, 1d 2h 3m 4s for durations, 68656c6c6f for binary, etc. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: fix dtype table — actual display values, footnotes, no blank cells - Replace generic type names with actual buckaroo display values - Add footnotes for BigInt (stringified, no commas) and Period (time span) - Fill all blank cells with — for dtypes that don't exist in that engine - Clarify Period label as "Period (time span)" - Fix Nullable int/float/bool row with proper values Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * WIP * docs: remove content-plan.md from ddd-post branch This file belongs on the content-plan branch (PR #638), not here. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * docs: address inline review comments on DDD article - Shorten Cecil Curry praise (remove "makes open source worth doing") - Add LLM-generated DataFrames and inherited pipelines to "why this matters" - Fix "three values" → "three non-numeric values" for infinity/NaN section - Remove duplicate None mention from MultiIndex rows section - Add note about planned MultiIndex-both-axes improvements - Update footnote 1: "putting together this table exposed areas that need work" - Fix comma→period typo in footnote 1 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * build: auto-create static stubs in hatch hook, remove manual touch hacks The hatch build hook (initialize2 → initialize) now auto-creates empty stub files in buckaroo/static/ when the real JS artifacts haven't been built. This eliminates the need for manual mkdir/touch blocks in CI and ReadTheDocs. - scripts/hatch_build.py: rename initialize2 → initialize (was dead code), create stubs for editable installs, build JS for wheel builds - pyproject.toml: use glob for artifacts (buckaroo/static/*.js|*.css) instead of incomplete file list - checks.yml: remove 4 separate mkdir/touch stub blocks - .readthedocs.yaml: remove pnpm install for js-core (stubs handle it) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * build: stop tracking standalone.js/css in git These are build artifacts produced by full_build.sh. They're already in .gitignore but were force-tracked. The glob artifact pattern in pyproject.toml ensures they're still included in the wheel. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: build all static assets in hatch hook pnpm fallback The pnpm fallback only built buckaroo-widget (widget.js/css) but not compiled.css, standalone.js/css, or static-embed.js/css. A bare `uv build` would produce an incomplete wheel. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

- Post 1: Dastardly DataFrame Dataset with inline iframe embeds - Post 3: Static Embedding & the Incredible Shrinking Widget - Post 5: Buckaroo Embedding Guide - Post 8: BuckarooCompare — Diff Your DataFrames - Script to generate DDD static HTML pages at docs build time - RTD config runs generate_ddd_static_html.py before copying extra-html - Fleshed out content-plan.md with 9-post publishing sequence Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

- Touch empty static files (compiled.css, widget.js, etc.) before running generate_ddd_static_html.py so anywidget import succeeds without a full JS build - Widen RST table columns to fit "Buckaroo static embed" row Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Adds a step to the CheckDocs job that comments on PRs with the ReadTheDocs preview URL and links to key article pages. Uses the same create-or-update pattern as the TestPyPI comment. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Converts so-you-want-to-write-a-dataframe-viewer to RST with a proper list-table for the comparison of open source DataFrame viewers. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Updates based on research into each project: - Buckaroo: Static Export → Yes - Perspective: Both server/browser, Arrow serialization, Jupyter compatible - DTale: JSON confirmed, No static export, uses react-virtualized - Marimo: JSON confirmed, not Jupyter compatible, not anywidget - ipydatagrid: No static export (confirmed broken), Lumino DataGrid - Mito: Endo (custom) table viewer - iTables: anywidget optional - Add quak (manzt): DuckDB-backed, Arrow, anywidget - Add hyperlinks to all project GitHub repos Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

…mparison as published Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

DDD article, pyproject.toml, and readthedocs.yaml were modified by earlier commits on this branch but the canonical versions were merged to main via PRs 641 and 642. Reset to main to avoid regressions. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

- Remove Python code block for column renaming (not needed) - Drop "smaller payloads" reason — not true for Parquet (column names stored once in metadata, not per-row) - Clarify caching is buckaroo's LRU in resolveDFData.ts, not hyparquet - Add TypeScript type flow diagram (string → ArrayBuffer → rows → cells) - Add "why it ended up this way" section: evolution from default AG-Grid + pandas JSON (slow) to Parquet - Explain BigInt flow: fast INT64 on Python side, hyparquet decodes as BigInt, buckaroo stringifies only when > MAX_SAFE_INTEGER - Explain duration flow: whole column coerced to string in Python, parsed back to human-readable on JS side Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

- Drop nonexistent "Hytable" from intro - Add Panel Tabulator and Streamlit st.dataframe to comparison table - Add hyperlinks throughout prose: AG-Grid, Perspective, glide-data-grid, tanstack-table, react-data-grid, anywidget, JS Jabber podcast - Update glide-data-grid description (actively maintained, not abandoned) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

…ting - Add before/after comparison: 4 jobs in 6 min → 23 jobs in 7 min - Add section on dual dependency testing strategy (min pinned + max versions) - Explain why pandas/pyarrow/polars compatibility testing needs fast CI - Link to Depot open source sponsorship program Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Dec 24 (pre-Depot): 3 jobs. Now: 23 jobs (20 added since sponsorship). Itemized breakdown of what was added: 6 Playwright suites, 8 Python matrix jobs, MCP integration, smoke tests, screenshots, docs, TestPyPI. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

…speed Same-pipeline benchmarks show Depot isn't measurably faster than GitHub runners (I/O-bound workload). The real value is consistent provisioning (no Monday afternoon queue delays) and no minute quotas, which gave confidence to invest in CI optimization and grow from 3 to 23 jobs. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

- "How I made Buckaroo fast" — do less, not optimize more - "Testing Buckaroo" — unit, integration, Playwright, screenshots, smoke, MCP, dual deps, DDD as test suite - Mark Depot article as draft with pending CTO input Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Article now includes a 4-scenario comparison table with real data: - GitHub Sunday night (3m15s), Monday sequential (5m19s), Monday parallel (5m58s), Depot Monday parallel (4m18s) - Key finding: per-job Depot is slightly slower, but consistent provisioning (19s stagger vs 114s) wins on critical path - GitHub ranges 3m–6m+, Depot is 4m–4m30s regardless New scripts: - ci_critical_path.sh: critical path for a single run - ci_list_runs.sh: list runs for a PR or branch - ci_all_timings.sh: JSON output of all timing data per run - ci_timing_table.py: formatted comparison table from JSON Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Complete rewrite with controlled benchmark results: - 21 runs across 6 scenarios (cold/warm, parallel/sequential, Sunday/Monday) - GitHub Monday: 7m46s ±143s. Depot Monday: 4m03s ±20s. - Key finding: per-job Depot is slightly slower, but consistent provisioning (all jobs start in 20s vs 1-7 min stagger) wins - Includes reproduction steps with scripts - Restructured: benchmark → results → analysis → before/after → advice Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

- Fix "GitHub Actions is" grammar - Explain per-job slowness: Depot has per-runner provisioning overhead, but provisions all in parallel; GitHub provisions sequentially from pool - Verify Monday cache write speed (GH 0.8s vs Depot 2.1s, confirmed) - Fix "Monday morning" typo - Remove "no minute quotas" point — reads as "Depot is great if free" - Renumber value prop list (now 2 items: consistency + confidence) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

The scariest part was transferring paddymul/buckaroo to the buckaroo-data org (required for Depot's open source program). Feared losing GitHub stars — turns out GitHub's transfer preserves everything (stars, issues, PRs, forks, redirects). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

paddymul temporarily deployed to testpypi March 20, 2026 20:38 — with GitHub Actions Inactive

chatgpt-codex-connector bot reviewed Mar 20, 2026

View reviewed changes

paddymul temporarily deployed to testpypi March 20, 2026 20:43 — with GitHub Actions Inactive

paddymul force-pushed the content-plan branch from 9f2aa02 to 0ef6141 Compare March 20, 2026 20:56

paddymul temporarily deployed to testpypi March 20, 2026 20:57 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 20, 2026 21:02 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 20, 2026 21:10 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 20, 2026 21:13 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 20, 2026 21:21 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 20, 2026 21:28 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 20, 2026 21:32 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 20, 2026 21:39 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 20, 2026 21:42 — with GitHub Actions Inactive

paddymul force-pushed the content-plan branch from 8965d35 to 6c85569 Compare March 20, 2026 21:49

paddymul temporarily deployed to testpypi March 20, 2026 21:50 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 20, 2026 21:55 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 21, 2026 03:22 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 21, 2026 03:36 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 21, 2026 16:10 — with GitHub Actions Inactive

paddymul added a commit that referenced this pull request Mar 21, 2026

docs: remove content-plan.md from ddd-post branch

3dcf1f3

This file belongs on the content-plan branch (PR #638), not here. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

paddymul temporarily deployed to testpypi March 21, 2026 16:28 — with GitHub Actions Inactive

paddymul and others added 5 commits March 21, 2026 13:00

docs: add initial content plan draft

3a12c86

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

fix: remove unused imports in generate_ddd_static_html.py

068713d

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

paddymul and others added 4 commits March 21, 2026 13:00

docs: convert dataframe viewer article from markdown to RST

350cd0e

Converts so-you-want-to-write-a-dataframe-viewer to RST with a proper list-table for the comparison of open source DataFrame viewers. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

docs: update content plan — mark DDD, types-to-display, and viewer co…

ad927ca

…mparison as published Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

paddymul force-pushed the content-plan branch from b9f59b8 to 17b2637 Compare March 21, 2026 17:01

paddymul temporarily deployed to testpypi March 21, 2026 17:02 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 21, 2026 17:20 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 21, 2026 21:50 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 22, 2026 21:22 — with GitHub Actions Inactive

docs: remove Windows job from Depot article

0688c58

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

paddymul temporarily deployed to testpypi March 22, 2026 21:46 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 22, 2026 22:17 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 22, 2026 22:28 — with GitHub Actions Inactive

docs: fix timing — 3.5 min critical path, not 7 min (excludes Windows)

eb2642a

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

paddymul temporarily deployed to testpypi March 22, 2026 22:39 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 23, 2026 12:05 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 23, 2026 15:25 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 23, 2026 17:59 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 24, 2026 13:13 — with GitHub Actions Inactive

paddymul temporarily deployed to testpypi March 24, 2026 13:25 — with GitHub Actions Inactive

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Content plan#638

Content plan#638
paddymul wants to merge 29 commits intomainfrom
content-plan

paddymul commented Mar 20, 2026

Uh oh!

github-actions bot commented Mar 20, 2026 •

edited

Loading

Uh oh!

chatgpt-codex-connector bot left a comment

Uh oh!

chatgpt-codex-connector bot Mar 20, 2026

Uh oh!

chatgpt-codex-connector bot Mar 20, 2026

Uh oh!

github-actions bot commented Mar 20, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

		('weird-types-polars', 'Weird Types (Polars → Pandas)',
		pl_df_with_weird_types_as_pandas(),

Conversation

paddymul commented Mar 20, 2026

Uh oh!

github-actions bot commented Mar 20, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

📦 TestPyPI package published

MCP server for Claude Code

📖 Docs preview

🎨 Storybook preview

Uh oh!

chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector bot Mar 20, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector bot Mar 20, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions bot commented Mar 20, 2026

📖 Docs preview

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

github-actions bot commented Mar 20, 2026 •

edited

Loading