Add codespell support (config, workflow to detect/not fix) and make it fix few typos

.codespellrc

@@ -0,0 +1,8 @@
+[codespell]
+# Ref: https://github.com/codespell-project/codespell#using-a-config-file
+# TODO: go through gene_names.json manually since might have some true positives
+skip = .git*,*.css,.codespellrc,gene_names.json,requirements*.txt,region_aliases.json
+check-hidden = true
+# embedded images and otherwise too long lines
+ignore-regex = ^\s*"image/\S+": ".*|.{300,}|\bFOM\b
+# ignore-words-list = 

.github/workflows/codespell.yml

@@ -0,0 +1,25 @@
+# Codespell configuration is within .codespellrc
+---
+name: Codespell
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+permissions:
+  contents: read
+
+jobs:
+  codespell:
+    name: Check for spelling errors
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Annotate locations with typos
+        uses: codespell-project/codespell-problem-matcher@v1
+      - name: Codespell
+        uses: codespell-project/actions-codespell@v2

docs/conf.py

@@ -122,7 +122,7 @@ def is_allen_api_microarray_service_available():
             shell=False,
         )
 
-# napolean settings
+# napoleon settings
 napoleon_google_docstring = False
 napoleon_use_param = True
 napoleon_use_ivar = True
@@ -213,7 +213,7 @@ def is_allen_api_microarray_service_available():
 html_theme = "sphinx_book_theme"
 html_show_sourcelink = False
 html_show_sphinx = True
-html_logo = "_static/siibra-python.jpeg"  # overriden by logo in html_theme_options
+html_logo = "_static/siibra-python.jpeg"  # overridden by logo in html_theme_options
 html_favicon = "_static/siibra_favicon.ico"
 html_permalinks = False
 

docs/create_preconfiguration.rst

@@ -14,7 +14,7 @@ How to structure jsons?
 =======================
 In ``config_schema`` folder within
 `siibra-python <https://github.com/FZJ-INM1-BDA/siibra-python>`_, the existing
-schemes are provied as json files. These are meant to be a guide, particularly
+schemes are provided as json files. These are meant to be a guide, particularly
 for the minimum requirements.
 
 Step-by-step
@@ -37,7 +37,7 @@ I. Determine the type
     * Volume of interest
     * Parcellation-based activity timeseries (tabular)
 
-**Not listed:** If the data is not fitting to any of the catagories and would
+**Not listed:** If the data is not fitting to any of the categories and would
 like to see your data in siibra, then a new schema is needed as otherwise CI
 will fail. It is expected that each new schema should be as general as possible.
 For example, instead of a schema for regional BOLD data, a schema for activity
@@ -82,7 +82,7 @@ IV. Use/share your preconfiguration
 -----------------------------------
 
 You can now use this json to load your desired data as a siibra object and take
-adventage its functionalities. You can also share this json with your colleagues
+advantage its functionalities. You can also share this json with your colleagues
 to collabrate.
 
 At this stage, if you'd like to see your data in siibra-toolbox as default,

docs/developer.rst

@@ -219,7 +219,7 @@ Adding data to siibra-toolsuite
 
 Anatomical Anchor
 -----------------
-Each feautre instance requires an anatomical anchor. This could be a parcellation
+Each feature instance requires an anatomical anchor. This could be a parcellation
 (as in ``RegionalConnecticity``), a region, a region and a location, or a location.
 Using the anatomical anchor siibra can determine the semantic and spatial
 relationship between different ``AtlasConcepts``.

docs/download_maps.ipynb

@@ -6,7 +6,7 @@
    "source": [
     "# Downloading parcellation and probability maps\n",
     "\n",
-    "`siibra` can facilitate structured, automatized and reprodicible download of parcellation maps. In this example, we demonstrate how to download all maximum probability and probabilistic maps of *Julich-Brain* - more then 300 structures available in different spaces. \n",
+    "`siibra` can facilitate structured, automated and reproducible download of parcellation maps. In this example, we demonstrate how to download all maximum probability and probabilistic maps of *Julich-Brain* - more then 300 structures available in different spaces. \n",
     "\n",
     "As often, we start by loading the human atlas."
    ]

e2e/core/test_parcellation.py

@@ -65,4 +65,4 @@ def test_region_id_uniqueness(parcellation: Parcellation):
             duplicates.add(region.id)
         ids.add(region)
 
-    assert len(duplicates) == 0, f"Folowing regions a duplicate IDs in {parcellation.name}:\n{duplicates}"
+    assert len(duplicates) == 0, f"Following regions a duplicate IDs in {parcellation.name}:\n{duplicates}"

e2e/core/test_region.py

@@ -121,5 +121,5 @@ def test_boundingbox(space):
 
     v_l = hoc1_l.get_regional_mask(space)
     v_r = hoc1_r.get_regional_mask(space)
-    assert bbox_l == v_l.get_boundingbox(clip=True, background=0.0), "Boundingbox of regional mask should be the same as bouding mask of the regions"
-    assert bbox_r == v_r.get_boundingbox(clip=True, background=0.0), "Boundingbox of regional mask should be the same as bouding mask of the regions"
+    assert bbox_l == v_l.get_boundingbox(clip=True, background=0.0), "Boundingbox of regional mask should be the same as bounding mask of the regions"
+    assert bbox_r == v_r.get_boundingbox(clip=True, background=0.0), "Boundingbox of regional mask should be the same as bounding mask of the regions"

e2e/features/image/test_image.py

@@ -18,7 +18,7 @@
 def test_feature_has_datasets(feature: Image):
     if feature.name in PRERELEASE_FEATURES_W_NO_DATASET:
         if len(feature.datasets) > 0:
-            pytest.fail(f"Feature '{feature}' was listed as prerelase previosly but now have dataset information. Please update `PRERELEASE_FEATURES_W_NO_DATASET`")
+            pytest.fail(f"Feature '{feature}' was listed as prerelease previously but now have dataset information. Please update `PRERELEASE_FEATURES_W_NO_DATASET`")
         pytest.skip(f"Feature '{feature}' has no datasets yet as it is a prerelease data.")
     assert len(feature.datasets) > 0, f"{feature} has no datasets"
 

e2e/features/test_get.py

@@ -65,10 +65,10 @@ def test_subclass_count(fid):
 
 def test_querying_with_volume():
     # Use features with location anchors only. Because hybrid ones will also
-    # employ sementic links between regions, potentially changing the result.
+    # employ semantic links between regions, potentially changing the result.
     region = siibra.get_region("julich 2.9", "ca1")
     volume = region.get_regional_mask('mni152')
     profiles_region = siibra.features.get(region, "BigBrainIntensityProfile")[0]
     profiles_volume = siibra.features.get(volume, "BigBrainIntensityProfile")[0]
-    # the ids will be diffent but the content has to be the same. Even the order.
+    # the ids will be different but the content has to be the same. Even the order.
     assert [p.location for p in profiles_region] == [p.location for p in profiles_volume]

examples/02_maps_and_templates/002_accessing_templates.py

@@ -78,7 +78,7 @@
 # or to fetch from a concrete resource format.
 # Meshes are provided as a dictionary with an Nx3 array of N vertices,
 # and an Mx3 array of M triangles defined from the vertices.
-# we can pre-check wether a volume provides image or mesh data
+# we can pre-check whether a volume provides image or mesh data
 # explicitly using `provides_mesh` and `provides_image`
 assert icbm_tpl.provides_mesh
 icbm_mesh = icbm_tpl.fetch(format='mesh')

examples/02_maps_and_templates/007_adding_custom_parcellation.py

@@ -49,7 +49,7 @@
 
 # The text file with label mappings has a custom format. We provide a tsv
 # decoder to extract the list of region/label pairs since the txt file is tab
-# seperated.
+# separated.
 volume_info = conn.get("AICHA/AICHA_vol1.txt", decode_func=siibra.retrieval.requests.DECODERS['.tsv'])
 volume_info
 
@@ -84,8 +84,8 @@
 
 
 # %%
-# On the other hand, some features are only anchored to a sementic region
-# object and the link to the custome region is not directly known to siibra.
+# On the other hand, some features are only anchored to a semantic region
+# object and the link to the custom region is not directly known to siibra.
 # However, siibra circumvents this by comparing volumes of these regions to
 # assign a link between them.
 volume = region.get_regional_mask('mni152')

examples/03_data_features/003_cell_distributions.py

@@ -83,7 +83,7 @@
 view.add_markers([tuple(location)])
 
 # %%
-# Now let us look into BigBrain intenstiy profiles for V1 left and dispaly the
+# Now let us look into BigBrain intenstiy profiles for V1 left and display the
 # gray matter mesh coordinates on the region mask.
 v1left = siibra.get_region("julich 2.9", "v1 left")
 cf = siibra.features.get(v1left, "BigBrainIntensityProfile")[0]

examples/05_anatomical_assignment/002_activation_maps.py

@@ -66,7 +66,7 @@
 # assign cytoarchitectonic regions to this functional map. Since we are here
 # usually interested in correlations of the modes, we filter the result by
 # significant (positive) correlations.
-# To assigne an image, we first need to createa Volume which has to have a
+# To assign an image, we first need to create Volume which has to have a
 # space defined.
 volume = siibra.volumes.volume.from_nifti(img, difumo_maps.space, "fusiform posterior")
 with siibra.QUIET:  # suppress progress output

siibra/__init__.py

@@ -108,7 +108,7 @@ def set_feasible_download_size(maxsize_gbyte):
 
 def set_cache_size(maxsize_gbyte: int):
     """
-    siibra runs maintainance on its local cache to keep it under a predetermined
+    siibra runs maintenance on its local cache to keep it under a predetermined
     size of 2 gigabytes. This method changes the cache size.
 
     Parameters
@@ -129,9 +129,9 @@ def warm_cache(level=WarmupLevel.INSTANCE):
     Preload preconfigured siibra concepts.
 
     Siibra relies on preconfigurations that simplify integrating various
-    concepts such as parcellations, refernce spaces, and multimodal data
+    concepts such as parcellations, reference spaces, and multimodal data
     features. By preloading the instances, siibra commits all preconfigurations
-    to the memory at once instead of commiting them when required.
+    to the memory at once instead of committing them when required.
     """
     Warmup.warmup(level)
 

siibra/commons.py

@@ -139,7 +139,7 @@ def __iter__(self) -> Iterator[T]:
         return (w for w in self._elements.values())
 
     def __contains__(self, key: Union[str, T]) -> bool:
-        """Test wether the given key or element is defined by the registry."""
+        """Test whether the given key or element is defined by the registry."""
         if isinstance(key, str):
             return key in self._elements
         return key in [item for _, item in self._elements.values()]
@@ -554,7 +554,7 @@ def connected_components(
 
     Note
     ----
-    `Uses skimage.measure.label()` to determine foreground compenents.
+    `Uses skimage.measure.label()` to determine foreground components.
 
     Parameters
     ----------
@@ -568,7 +568,7 @@ def connected_components(
     Yields
     ------
     Generator[Tuple[int, np.ndarray], None, None]
-        tuple of integer label of the component and component as an nd.array in
+        tuple of integer label of the component and component as an ndarray in
         the shape of the original image.
     """
     from skimage import measure
@@ -644,11 +644,11 @@ def MI(arr1, arr2, nbins=100, normalized=True):
     assert (all(arr.size > 0) for arr in [arr1, arr2])
 
     # compute the normalized joint 2D histogram as an
-    # empirical measure of the joint probabily of arr1 and arr2
+    # empirical measure of the joint probability of arr1 and arr2
     pxy, _, _ = np.histogram2d(arr1.ravel(), arr2.ravel(), bins=nbins)
     pxy /= pxy.sum()
 
-    # extract the empirical propabilities of intensities
+    # extract the empirical probabilities of intensities
     # from the joint histogram
     px = np.sum(pxy, axis=1)  # marginal for x over y
     py = np.sum(pxy, axis=0)  # marginal for y over x

siibra/core/concept.py

@@ -145,9 +145,9 @@ def urls(self) -> List[str]:
     @property
     def authors(self):
         return [
-            contributer['name']
+            contributor['name']
             for ds in self.datasets
-            for contributer in ds.contributors
+            for contributor in ds.contributors
         ]
 
     @property

siibra/core/parcellation.py

@@ -221,7 +221,7 @@ def _split_group_spec(self, spec: str):
         """
         spec = re.sub(r'Group: *', '', spec)
         for substr in re.findall(r'\(.*?\)', spec):
-            # temporarilty replace commas inside brackets with a placeholder
+            # temporarily replace commas inside brackets with a placeholder
             # because these are not region spec delimiters
             spec = spec.replace(substr, re.sub(r', *', '##', substr))
         # process the comma separated substrings
@@ -354,7 +354,7 @@ def find_regions(
 ):
     """
     Find regions matching the given region specification across all parcellation
-    instances in the registery.
+    instances in the registry.
 
     Parameters
     ----------

siibra/core/region.py

@@ -133,8 +133,8 @@
         Example
         -------
         >>> region = siibra.get_region("monkey", "PG")
-        >>> for assesment in region.get_related_regions():
-        >>>    print(assesment)
+        >>> for assessment in region.get_related_regions():
+        >>>    print(assessment)
         'PG' is homologous to 'Area PGa (IPL)'
         'PG' is homologous to 'Area PGa (IPL) left'
         'PG' is homologous to 'Area PGa (IPL) right'
@@ -481,7 +481,7 @@
                     volume.merge(descendant_volumes),
                     label=1
                 )
-                name += f"'{result.space}' (built by merging the mask {threshold_info} of its decendants)"
+                name += f"'{result.space}' (built by merging the mask {threshold_info} of its descendants)"
             else:
                 raise e
         result._name = name
@@ -497,7 +497,7 @@
         maptype: MapType = MapType.LABELLED,
     ) -> Union[volume.FilteredVolume, volume.Volume, volume.Subvolume]:
         """
-        Get a volume reprsenting this region in the given space and MapType.
+        Get a volume representing this region in the given space and MapType.
 
         Note
         ----
@@ -540,7 +540,7 @@
 
     def mapped_in_space(self, space, recurse: bool = False) -> bool:
         """
-        Verifies wether this region is defined by an explicit map in the given space.
+        Verifies whether this region is defined by an explicit map in the given space.
 
         Parameters
         ----------
@@ -931,8 +931,8 @@
     Example
     -------
     >>> region = siibra.get_region("monkey", "PG")
-    >>> for assesment in siibra.core.region.get_related_regions(region):
-    >>>    print(assesment)
+    >>> for assessment in siibra.core.region.get_related_regions(region):
+    >>>    print(assessment)
     'PG' is homologous to 'Area PGa (IPL)'
     'PG' is homologous to 'Area PGa (IPL) left'
     'PG' is homologous to 'Area PGa (IPL) right'

siibra/core/structure.py

@@ -76,7 +76,7 @@ def assign(self, other: "BrainStructure") -> assignment.AnatomicalAssignment:
         # Two cases:
         # 1) self is location, other is location -> look at spatial intersection/relationship, do it here
         # 2) self is location, other is region -> get region map, then call again. do it here
-        # If self is region -> Region overwrite this method, adressed there
+        # If self is region -> Region overwrite this method, addressed there
 
         assert not isinstance(self, _region.Region)  # method is overwritten by Region!
         if (self, other) in self._ASSIGNMENT_CACHE:

siibra/experimental/contour.py

@@ -46,7 +46,7 @@ def crop(self, voi: boundingbox.BoundingBox):
 
         if cropped is not None and not isinstance(cropped, point.Point):
             assert isinstance(cropped, pointcloud.PointCloud)
-            # Identifiy contour splits are by discontinuouities ("jumps")
+            # Identify contour splits are by discontinuouities ("jumps")
             # of their labels, which denote positions in the original contour
             jumps = np.diff([self.labels.index(lb) for lb in cropped.labels])
             splits = [0] + list(np.where(jumps > 1)[0] + 1) + [len(cropped)]

siibra/experimental/plane3d.py

@@ -33,7 +33,7 @@ def __init__(self, point1: point.Point, point2: point.Point, point3: point.Point
         The plane's reference space is defined by the first point.
         """
         self.space = point1.space
-        # normal is the cross product of two arbitray in-plane vectors
+        # normal is the cross product of two arbitrary in-plane vectors
         n = np.cross(
             (point2.warp(self.space) - point1).coordinate,
             (point3.warp(self.space) - point1).coordinate,
@@ -107,7 +107,7 @@ def intersect_mesh(self, mesh: dict):
         )[0]
         faces = mesh["faces"][face_indices]
 
-        # for each of N selected faces, indicate wether we cross the plane
+        # for each of N selected faces, indicate whether we cross the plane
         # as we go from vertex 2->0, 0->1, 1->2, respectively.
         # This gives us an Nx3 array, where forward crossings are identified by 1,
         # and backward crossings by -1.
@@ -160,7 +160,7 @@ def intersect_mesh(self, mesh: dict):
         face_id = 0  # index of the mesh face to consider
         while len(face_indices) > 0:
 
-            # continue the contour with the next foward edge intersection
+            # continue the contour with the next forward edge intersection
             p = fwd_intersections[face_id]
             points.append(p)
             # Remember the ids of the face and start-/end vertices for the point

siibra/explorer/url.py

@@ -211,7 +211,7 @@ def get_hash(full_string: str):
         for char in full_string:
             # overflowing is expected and in fact the whole reason why convert number to int32
 
-            # in windows, int32((0 - min_int32) << 5), rather than overflow to wraper around, raises OverflowError
+            # in windows, int32((0 - min_int32) << 5), rather than overflow to wrapper around, raises OverflowError
             shifted_5 = int32(
                 (return_val - min_int32) if return_val > max_int32 else return_val << 5
             )

siibra/features/anchor.py

@@ -88,7 +88,7 @@ def parcellations(self) -> List[Parcellation]:
 
     @property
     def space(self) -> Space:
-        # may be overriden by derived classes, e.g. in features.VolumeOfInterest
+        # may be overridden by derived classes, e.g. in features.VolumeOfInterest
         return None if self.location is None else self.location.space
 
     @property

siibra/features/connectivity/regional_connectivity.py

@@ -275,7 +275,7 @@
 
         regions = [r for r in matrix.index if matches(r, region)]
         if len(regions) == 0:
-            raise ValueError(f"Invalid region specificiation: {region}")
+            raise ValueError(f"Invalid region specification: {region}")
         elif len(regions) > 1:
             raise ValueError(f"Region specification {region} matched more than one profile: {regions}")
         else: