ENH: add the lite_data dataset (JupyterLite split 1/5)

doc/api/datasets.rst

@@ -30,6 +30,7 @@ Datasets
    hf_sef.data_path
    kiloword.data_path
    limo.load_data
+   lite_data.data_path
    misc.data_path
    mtrf.data_path
    multimodal.data_path

doc/changes/dev/14128.newfeature.rst

@@ -0,0 +1 @@
+Add the :func:`mne.datasets.lite_data.data_path` fetcher for the curated data subset used by the browser documentation, by `Natneal B`_.

doc/documentation/datasets.rst

@@ -541,6 +541,29 @@ the people in the scene were unrecognizable.
 
     * :ref:`tut-eyetrack-heatmap`
 
+.. _lite-data:
+
+JupyterLite data
+================
+:func:`mne.datasets.lite_data.data_path`
+
+A small curated archive holding the data files needed to run the tutorials and
+examples in the browser, taken from the ``sample``, ``kiloword``, ``erp_core``,
+``mtrf`` and ``eegbci`` datasets. The files are unchanged and keep the same
+checksums as the full datasets. It extracts to ``MNE-lite-data/``, keeping each
+file under its original dataset folder (``MNE-sample-data/``,
+``MNE-kiloword-data/``, ...).
+
+Those datasets ship as separate multi-GB archives, so without this the
+documentation build would download several gigabytes to serve a handful of
+files.
+
+The ``somato`` dataset is not included, so the somatosensory tutorials and
+examples are not available in the browser.
+
+This exists for the documentation build; for analysis, use the individual
+dataset fetchers above.
+
 References
 ==========
 

mne/datasets/__init__.pyi

@@ -19,6 +19,7 @@ __all__ = [
     "hf_sef",
     "kiloword",
     "limo",
+    "lite_data",
     "misc",
     "mtrf",
     "multimodal",
@@ -48,6 +49,7 @@ from . import (
     hf_sef,
     kiloword,
     limo,
+    lite_data,
     misc,
     mtrf,
     multimodal,

mne/datasets/config.py

@@ -209,6 +209,16 @@
     config_key="MNE_DATASETS_SAMPLE_PATH",
 )
 
+# Curated subset of sample (plus a few files from kiloword/erp_core/mtrf/eegbci)
+# used by the JupyterLite browser docs; see mne/datasets/lite_data/.
+MNE_DATASETS["lite_data"] = dict(
+    archive_name="MNE-lite-data.tar.gz",
+    hash="md5:5f9c4fffed32e79bc2bc2061bf22ce99",
+    url="https://osf.io/download/a8qbx",
+    folder_name="MNE-lite-data",
+    config_key="MNE_DATASETS_LITE_DATA_PATH",
+)
+
 MNE_DATASETS["somato"] = dict(
     archive_name="MNE-somato-data.tar.gz",
     hash="md5:9a191907b326b9402341ee7a0d1240d8",

mne/datasets/lite_data/__init__.py

@@ -0,0 +1,7 @@
+# Authors: The MNE-Python contributors.
+# License: BSD-3-Clause
+# Copyright the MNE-Python contributors.
+
+"""Curated data subset for the JupyterLite browser documentation."""
+
+from .lite_data import data_path, get_version

mne/datasets/lite_data/lite_data.py

@@ -0,0 +1,49 @@
+# Authors: The MNE-Python contributors.
+# License: BSD-3-Clause
+# Copyright the MNE-Python contributors.
+
+"""Curated data subset used by the JupyterLite browser documentation.
+
+``lite_data`` holds the data files needed to run the tutorials and examples in
+the browser, taken from ``sample``, ``kiloword``, ``erp_core``, ``mtrf`` and
+``eegbci``. The files are unchanged and keep the same checksums as the full
+datasets. It extracts to ``MNE-lite-data/`` with each file under its original
+dataset folder (``MNE-sample-data/``, ``MNE-kiloword-data/``, ...), so paths
+match.
+
+Those datasets ship as separate multi-GB archives, so without this the
+documentation build would download several gigabytes to serve a handful of
+files.
+
+The ``somato`` dataset is not included, so the somatosensory tutorials and
+examples do not run in the browser.
+"""
+
+from ...utils import verbose
+from ..utils import _data_path_doc, _download_mne_dataset, _get_version, _version_doc
+
+
+@verbose
+def data_path(
+    path=None, force_update=False, update_path=True, download=True, *, verbose=None
+):  # noqa: D103
+    return _download_mne_dataset(
+        name="lite_data",
+        processor="untar",
+        path=path,
+        force_update=force_update,
+        update_path=update_path,
+        download=download,
+    )
+
+
+data_path.__doc__ = _data_path_doc.format(
+    name="lite_data", conf="MNE_DATASETS_LITE_DATA_PATH"
+)
+
+
+def get_version():  # noqa: D103
+    return _get_version("lite_data")
+
+
+get_version.__doc__ = _version_doc.format(name="lite_data")

mne/datasets/tests/test_datasets.py

@@ -322,3 +322,18 @@ def test_fetch_uncompressed_file(tmp_path):
     )
     fetch_dataset(dataset_dict, path=None, force_update=True)
     assert (tmp_path / "foo" / "LICENSE.foo").is_file()
+
+
+def test_lite_data():
+    """Test the lite_data curated dataset is registered correctly."""
+    from mne.datasets import lite_data
+    from mne.datasets.config import MNE_DATASETS
+
+    assert "lite_data" in MNE_DATASETS
+    cfg = MNE_DATASETS["lite_data"]
+    assert cfg["archive_name"] == "MNE-lite-data.tar.gz"
+    assert cfg["hash"].startswith("md5:")
+    assert cfg["url"].startswith("https://osf.io/")
+    assert cfg["config_key"] == "MNE_DATASETS_LITE_DATA_PATH"
+    assert callable(lite_data.data_path)
+    assert callable(lite_data.get_version)

mne/utils/config.py

@@ -170,6 +170,7 @@ def set_memmap_min_size(memmap_min_size):
     "MNE_DATASETS_TESTING_PATH": "str, path for testing data",
     "MNE_DATASETS_VISUAL_92_CATEGORIES_PATH": "str, path for visual_92_categories data",
     "MNE_DATASETS_KILOWORD_PATH": "str, path for kiloword data",
+    "MNE_DATASETS_LITE_DATA_PATH": "str, path for lite_data data",
     "MNE_DATASETS_FIELDTRIP_CMC_PATH": "str, path for fieldtrip_cmc data",
     "MNE_DATASETS_PHANTOM_KIT_PATH": "str, path for phantom_kit data",
     "MNE_DATASETS_PHANTOM_4DBTI_PATH": "str, path for phantom_4dbti data",