Lektor Index Pages Plugin

This Lektor plugin can be used to generate “index pages” for a blog or similar collection of pages. These index pages list the blog posts segregated by some key, with each index page containing only those posts which match that key.

Examples of what this can be used for include:

• Category Indexes: A set of index pages, one for each category, which lists all the posts in that category. (Multi-valued index keys are also supported, so that each post can appear on more than a single index page: e.g. keyword indexes.)

• Date Indexes: A set of index pages, one for each year (say), which list all the posts in that year. (Sub-indexes are supported subindexes — e.g., each year index may have as children a sequence of month indexes.)

Behind the scenes, judicious caching of indexing results, and careful control of Lektor’s dependency tracking prevent all this from slowing the build process down too excruciatingly much.

Project Links

• Documentation
• Github
• PyPI

Author

Jeff Dairiki dairiki@dairiki.org

Changelog

Release 1.1.0 (2023-06-16)

• Drop support for Lektor 3.2.x
• Test under python 3.11
• Add type annotations
• Convert packaging from setuptools to PDM
• Disuse pkg_resources in favor of importlib.metadata
• Add pre-commit configuration
• Run black and reorder-python-imports on code

Release 1.0 (2022-01-28)

• Drop support for python 2.7 and 3.6.
• Fix deprecation warning from jinja2. Jinja2 version 3 is now required.

Documentation

• Documentation clarifications, updates and fixes. (PR #2 — Thank you Bart Van Loon!)

• Add missing requirement recommonmark to docs/requirements.txt.

• Add docs enviroment to tox.ini to test that docs will build cleanly.

Testing

• Test under python 3.10 and lektor<3.3

Release 0.1 (2021-02-05)

No code changes.

Update development status classifier to "stable".

Release 0.1a3 (2020-05-08)

API changes

• Added a key field on the index virtual source object. It is an alias to _id, but is syntactically more self-explanatory.

• The keys configuration key has been renamed to key.

• When the key expression is being evaluted, the record whose key(s) is(are) to be be computed is now available in the jinja context as item rather than this.

Documentation

• Documentation moved from README to Sphinx docs at RTFD.io

Release 0.1a2 (2020-05-06)

API changes

• The record argument has been dropped from the (jinja) global index_pages function. (Since indexes can not have multiple parents, it is not necessary.)

• The parent configuration key has been renamed to parent_path.

• The slug configuration key has been renamed to slug_format.

• The attributes config section has been renamed to fields.

  Though they are not quite like regular Lektor Record fields, they are more field-like than attribute-like. (I.e. access is via getitem rather than getattr.)

Release 0.1a1 (2020-05-05)

Initial release.