docs: correct the install instructions and repoint place resolution - #234
Conversation
The README told readers to run `pip install -U "bblocks>=3"`, which fails because 3.0.0 is not published yet, and pointed them at bblocks-places, whose final release now directs users to resolvekit. Co-authored-by: Claude <noreply@anthropic.com>
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #234 +/- ##
=======================================
Coverage 97.99% 97.99%
=======================================
Files 31 31
Lines 2986 2986
=======================================
Hits 2926 2926
Misses 60 60 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: ae6a6b240d
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".
| The DSA importer (`get_dsa`) parses PDF tables and needs the `pdf` extra: | ||
|
|
||
| ```bash | ||
| pip install "bblocks[pdf]" |
There was a problem hiding this comment.
Add the
pdf extra before documenting it
In this commit's tree, pyproject.toml declares only the all and data-importers extras, while camelot-py remains a core dependency. Consequently, this command cannot select the promised PDF dependency set and the documentation incorrectly tells users that get_dsa requires an extra that the released package metadata does not provide. Add the extra in the same change or retain the existing installation guidance until it exists.
Useful? React with 👍 / 👎.
The README told readers to run
pip install -U "bblocks>=3". That command fails today, because 3.0.0 is unpublished and PyPI serves 2.2.0. The migration section around it described 3.x in the present tense as though it had shipped, so the section now describes 2.x as the installable release and 3.0.0 as pending. It has to move back to the present tense when 3.0.0 is tagged.The README also pointed readers at
bblocks-placesfor place resolution. Its final release, 0.0.6, directs users toresolvekit, so both mentions point there instead.bblockshas no dependency onresolvekitand still usescountry_converterinternally, so the guidance is written for consumers doing their own resolution. The Basic Usage snippet calledfrom bblocks import places, an import this package has never had, and now callsrk.bulk. Its install line asks forresolvekit[pandas], sincebulk()needs the pandas integration.The
pdfextra thatget_dsa()requires is documented in the README, the getting-started page and the DSA importer page. That extra arrives in #233, so merge this after it..codegraph/joins the other local tooling entries in.gitignore.Two documentation problems are left alone.
mkdocs build --strictfails on an orphanedimporters/dsa.mdthat the nav does not list, and on a dead comtrade.un.org link inimporters/baci.md. Both predate these changes, and no docs build runs in CI to catch them.Testing:
mkdocs build --stricton Python 3.12, which reports the two pre-existing failures above and no new warnings.Co-authored-by: Claude noreply@anthropic.com