jupyter-cache

v0.6.1 2023-04-22

A patch release to fix compatibility with sqlalchemy <1.4.

• FIX: compatibility with SQLAlchemy < 1.4.0 #105 @DimitriPapadopoulos

v0.6.0 2023-04-21

This is a minor release to improve our packaging infrastructure and to support several new versions of dependencies.

Breaking changes

• ‼️ BREAKING: Drop Python 3.7, add Python 3.11, unpin myst-nb in docs #96 (@choldgraf)

Updated versions

• Update nbclient requirement from <0.6,>=0.2 to >=0.2,<0.8 #103 (@choldgraf)
• UPDATE: SQLAlchemy 2.0 #93 (@jzluo, @choldgraf)
• 🔧 MAINTAIN: setuptools -> flit #82 (@chrisjsewell)

Contributors to this release

The following people contributed discussions, new ideas, code and documentation contributions, and review.
See our definition of contributors.

(GitHub contributors page for this release)

@AakashGfude (activity) | @choldgraf (activity) | @chrisjsewell (activity) | @jstac (activity) | @jzluo (activity) | @kloczek (activity) | @pre-commit-ci (activity)

What's Changed

♻️ REFACTOR: package API/CLI/documentation (#74)

This release includes major re-writes to key parts of the package,
to improve the user interface, and add additional functionality for reading and executing notebooks.

Key changes:

1. stage/staging is now rephrased to notebook, plus the addition of project, i.e. you add notebooks to a project, then execute them.
2. notebook read_data is specified per notebook in the project, allowing for multiple types of file to be read/executed via the CLI (e.g. text-based notebooks via https://jupytext.readthedocs.io).
   Before, the read functions were passed directly to the API methods.
3. The executor can be specified with jbcache execute --executor, and a parallel notebook executor has been added.
4. Improved execution status indicator in jbcache project list and other CLI improvements.
5. Re-write of documentation, including better front page, with quick start guide and better logo.

Dependencies have also been restructured, such that the CLI dependencies (click, tabulate) are now required,
whereas nbdime is now optional (since it is only used for optional notebook diffing).

‼️ Breaking:

The name of the SQL table nbstage has been changed to nbproject, and read_data/exec_data fields have been added to the nbproject.
This means that reading will fail for caches creted using older versions of the package.
However, the version of jupyter-cache, used to create the cache, is now recorded, allowing for the possibility of future automated migrations.

Full Changelog: https://github.com/executablebooks/jupyter-cache/compare/v0.4.3...v0.5.0

See CHANGELOG.md

See CHANGELOG.md

See CHANGELOG.md

Improved

• Moved execution functionality from nbconvert to nbclient
• Fixed UTF8 encoding (for Windows OS), thanks to @phaustin

Fixed

• Moved testing from Travis CI to GitHub Actions (and added tests for Windows OS)

This allows for text-based notebooks to be staged, then converted on the fly before execution (e.g. using jupytext)