Merge branch 'init/search-lindera'

.github/workflows/main.yml

@@ -72,6 +72,7 @@ jobs:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
       - name: 'Configure dependencies'
         run: |
+          echo '3.11' > .python-version
           uv sync --frozen --all-extras
       - name: 'Run tests'
         run: |

docs/components/index.rst

@@ -0,0 +1,12 @@
+==========
+Components
+==========
+
+Index of toybox components that are used in ``conf.py`` directly.
+Subpages have overview, usage and **demo**.
+
+.. toctree::
+   :maxdepth: 1
+   :glob:
+
+   */index

docs/components/lindera_search/index.rst

@@ -0,0 +1,45 @@
+==============================
+atsphinx.toybox.lindera_search
+==============================
+
+.. note:: Added by v2024.12.10
+
+Overview
+========
+
+This is to use Lindera search components of Sphinx.
+
+Usage
+=====
+
+Install requires
+----------------
+
+This extension requires optional installation.
+Please add 'lindera-search' option when you install it.
+
+.. code-block:: console
+
+   pip install 'atsphinx-toybox[lindera-search]'
+
+Set configuration
+-----------------
+
+Add this into your ``conf.py`` of Sphinx.
+
+.. code-block:: python
+   :caption: conf.py
+
+   html_search_options = {
+       "type": "atsphinx.toybox.lindera_search.LinderaSplitter",
+   }
+
+Demo
+====
+
+.. todo: Write it.
+
+Refs
+====
+
+* `lindera-py <https://pypi.org/project/lindera-py/>`_

docs/index.rst

@@ -12,4 +12,5 @@ This is attakei's alpha-level Sphinx-extensions.
 
    guide
    extensions/index
+   components/index
    changes

pyproject.toml

@@ -42,6 +42,9 @@ Documentation = "https://atsphinx.github.io/toybox"
 sass = [
     "httpx>=0.28.0",
 ]
+lindera-search = [
+    "lindera-py>=0.38.4",
+]
 
 [build-system]
 requires = ["hatchling"]
@@ -77,5 +80,5 @@ only-includes = ["src/atsphinx"]
 exclude = 'conf\.py$'
 
 [[tool.mypy.overrides]]
-module = ['docutils', 'docutils.*']
+module = ['docutils', 'docutils.*', 'lindera_py']
 ignore_missing_imports = true

src/atsphinx/toybox/lindera_search/__init__.py

@@ -0,0 +1,16 @@
+"""Tokenize override by Lindera."""
+
+from lindera_py import Segmenter, Tokenizer, load_dictionary
+from sphinx.search.ja import BaseSplitter
+
+
+class LinderaSplitter(BaseSplitter):
+    """Simple splitter class using Lindera as tokeniser."""
+
+    def __init__(self, options: dict[str, str]) -> None:  # noqa: D107
+        self.dictionary = load_dictionary("ipadic")
+        self.segmenter = Segmenter("normal", self.dictionary)
+        self.tokenizer = Tokenizer(self.segmenter)
+
+    def split(self, input: str) -> list[str]:  # noqa: D102
+        return [token.text for token in self.tokenizer(input)]