Add standard pre-commit config (#40)

* Add standard pre-commit config

* Add automated pre-commit fixes

* Ignore line too long

.pre-commit-config.yaml

@@ -1,6 +1,22 @@
 repos:
--   repo: https://github.com/python/black
-    rev: 22.12.0
-    hooks:
-    - id: black
-      pass_filenames: true
+    - repo: https://github.com/pre-commit/pre-commit-hooks
+      rev: v4.4.0
+      hooks:
+          - id: check-docstring-first
+          - id: check-executables-have-shebangs
+          - id: check-merge-conflict
+          - id: end-of-file-fixer
+          - id: mixed-line-ending
+            args: [--fix=lf]
+          - id: requirements-txt-fixer
+          - id: trailing-whitespace
+    - repo: https://github.com/charliermarsh/ruff-pre-commit
+      rev: v0.0.240
+      hooks:
+        - id: ruff
+          args: [ --config=pyproject.toml ]
+    - repo: https://github.com/psf/black
+      rev: 23.1.0
+      hooks:
+          - id: black
+            args: [--config=pyproject.toml]

README.md

@@ -8,24 +8,24 @@ Utilities and scripts for the generation of cleaned-up data for the `bg-atlasapi
 ### To contribute
 1) Fork this repo
 
-2) Clone your repo 
+2) Clone your repo
 ```bash
 git clone https://github.com/USERNAME/bg-atlasgen
 ```
 
 3) Install an editable version
 ```bash
 cd bg-atlasgen
-pip install -e . 
+pip install -e .
 ```
-4) Create a script to package your atlas, and place into 
+4) Create a script to package your atlas, and place into
 `bg_atlasgen/atlas_scripts`. Please see other scripts for examples.
 
-Your script should contain everything required to run. The raw data should be 
+Your script should contain everything required to run. The raw data should be
 hosted on a publicly accessible repository so that anyone can run the script
  to recreate the atlas.
 
-If you need to add any dependencies, please add them as an extra in the 
+If you need to add any dependencies, please add them as an extra in the
 setup.py file, e.g.:
 
 ```python

bg_atlasgen/atlas_scripts/admba_3d_dev_mouse.py

@@ -2,26 +2,23 @@
 
 import dataclasses
 import json
+import multiprocessing as mp
 import time
 import zipfile
-
 from os import listdir, path
+from pathlib import Path
 from typing import Tuple
 
-import pandas as pd
 import numpy as np
-import multiprocessing as mp
-
-from rich.progress import track
-from pathlib import Path
-
+import pandas as pd
 from bg_atlasapi import utils
-from bg_atlasgen.mesh_utils import create_region_mesh, Region
-from bg_atlasgen.wrapup import wrapup_atlas_from_data
 from bg_atlasapi.structure_tree_util import get_structures_tree
-
+from rich.progress import track
 from skimage import io
 
+from bg_atlasgen.mesh_utils import Region, create_region_mesh
+from bg_atlasgen.wrapup import wrapup_atlas_from_data
+
 PARALLEL = True
 
 
@@ -116,7 +113,6 @@ def create_meshes(download_dir_path, structures, annotated_volume, root_id):
     smooth = False  # smooth meshes after creation
     start = time.time()
     if PARALLEL:
-
         pool = mp.Pool(mp.cpu_count() - 2)
 
         try:

bg_atlasgen/atlas_scripts/allen_cord.py

@@ -1,31 +1,29 @@
 __version__ = "1"
 
 import json
+import multiprocessing as mp
 import time
-import tifffile
 import zipfile
+from pathlib import Path
+from random import choices
 
-
-import pandas as pd
 import numpy as np
-import multiprocessing as mp
-from random import choices
-from loguru import logger
-from rich.progress import track
-from pathlib import Path
+import pandas as pd
+import tifffile
 
 # import sys
-
 # sys.path.append("./")
-
 from bg_atlasapi import utils
+from bg_atlasapi.structure_tree_util import get_structures_tree
+from loguru import logger
+from rich.progress import track
+
 from bg_atlasgen.mesh_utils import (
-    create_region_mesh,
     Region,
+    create_region_mesh,
     inspect_meshes_folder,
 )
 from bg_atlasgen.wrapup import wrapup_atlas_from_data
-from bg_atlasapi.structure_tree_util import get_structures_tree
 
 PARALLEL = True
 TEST = False
@@ -123,7 +121,7 @@ def create_meshes(download_dir_path, structures, annotated_volume, root_id):
     nodes = list(tree.nodes.values())
     if TEST:
         logger.info(
-            f"Creating atlas in test mode: selecting 10 random regions for mesh creation"
+            "Creating atlas in test mode: selecting 10 random regions for mesh creation"
         )
         nodes = choices(nodes, k=10)
 
@@ -289,7 +287,6 @@ def create_atlas(working_dir):
 
 
 if __name__ == "__main__":
-
     # Generated atlas path:
     bg_root_dir = Path.home() / "brainglobe_workingdir" / "allen_cord_smooth"
     bg_root_dir.mkdir(exist_ok=True, parents=True)

bg_atlasgen/atlas_scripts/allen_mouse.py

@@ -1,15 +1,15 @@
 __version__ = "2"
 
+from pathlib import Path
+
 from allensdk.api.queries.ontologies_api import OntologiesApi
 from allensdk.api.queries.reference_space_api import ReferenceSpaceApi
 from allensdk.core.reference_space_cache import ReferenceSpaceCache
-
+from bg_atlasapi import descriptors
 from requests import exceptions
-from pathlib import Path
 from tqdm import tqdm
 
 from bg_atlasgen.wrapup import wrapup_atlas_from_data
-from bg_atlasapi import descriptors
 
 
 def create_atlas(working_dir, resolution):
@@ -83,7 +83,7 @@ def create_atlas(working_dir, resolution):
         ]
 
     # Wrap up, compress, and remove file:0
-    print(f"Finalising atlas")
+    print("Finalising atlas")
     output_filename = wrapup_atlas_from_data(
         atlas_name=ATLAS_NAME,
         atlas_minor_version=__version__,

bg_atlasgen/atlas_scripts/azba_zfish.py

@@ -10,30 +10,27 @@
 __version__ = "1"
 
 import csv
-import time
-import tifffile
+import multiprocessing as mp
 import tarfile
-from random import choices
+import time
+from pathlib import Path
 
 import numpy as np
-import multiprocessing as mp
-
+import tifffile
+from bg_atlasapi import utils
+from bg_atlasapi.structure_tree_util import get_structures_tree
 from rich.progress import track
-from pathlib import Path
+
 from bg_atlasgen.mesh_utils import (
     Region,
     create_region_mesh,
-    inspect_meshes_folder,
 )
-from bg_atlasapi.structure_tree_util import get_structures_tree
 from bg_atlasgen.wrapup import wrapup_atlas_from_data
-from bg_atlasapi import utils
 
 PARALLEL = False  # Disable for debugging mesh creation
 
 
 def create_atlas(working_dir, resolution):
-
     # metadata
     ATLAS_NAME = "azba_zfish"
     SPECIES = "Danio rerio"
@@ -135,7 +132,6 @@ def create_atlas(working_dir, resolution):
     smooth = True
 
     if PARALLEL:
-
         print("Multiprocessing mesh creation...")
         pool = mp.Pool(int(mp.cpu_count() / 2))
 
@@ -159,7 +155,6 @@ def create_atlas(working_dir, resolution):
             pass
 
     else:
-
         print("Multiprocessing disabled")
         # nodes = list(tree.nodes.values())
         # nodes = choices(nodes, k=10)

bg_atlasgen/atlas_scripts/example_mouse.py

@@ -1,18 +1,17 @@
 __version__ = "2"
 
+from pathlib import Path
+
 from allensdk.api.queries.ontologies_api import OntologiesApi
 from allensdk.api.queries.reference_space_api import ReferenceSpaceApi
 from allensdk.core.reference_space_cache import ReferenceSpaceCache
-
 from requests import exceptions
-from pathlib import Path
 from tqdm import tqdm
 
 from bg_atlasgen.wrapup import wrapup_atlas_from_data
 
 
 def create_atlas(working_dir, resolution):
-
     # Specify information about the atlas:
     RES_UM = resolution  # 100
     ATLAS_NAME = "example_mouse"

bg_atlasgen/atlas_scripts/humanatlas.py

@@ -1,25 +1,25 @@
 import json
-from rich.progress import track
-import pandas as pd
-import numpy as np
-import time
 import multiprocessing as mp
+import time
 from pathlib import Path
+
+import numpy as np
+import pandas as pd
 import treelib
-from brainio import brainio
 import urllib3
 from allensdk.core.structure_tree import StructureTree
+from bg_atlasapi.structure_tree_util import get_structures_tree
+from brainio import brainio
+from rich.progress import track
 
 # import sys
-
 # sys.path.append("./")
 from bg_atlasgen.mesh_utils import (
-    create_region_mesh,
     Region,
+    create_region_mesh,
     inspect_meshes_folder,
 )
 from bg_atlasgen.wrapup import wrapup_atlas_from_data
-from bg_atlasapi.structure_tree_util import get_structures_tree
 
 
 def prune_tree(tree):

bg_atlasgen/atlas_scripts/kim_developmental_ccf_mouse.py

@@ -1,24 +1,21 @@
 __version__ = "1"
 
 import json
+import multiprocessing as mp
 import time
-import tarfile
+import zipfile
+from pathlib import Path
 
-import pandas as pd
+import imio
 import numpy as np
-import multiprocessing as mp
-
+import pandas as pd
+from bg_atlasapi import utils
+from bg_atlasapi.structure_tree_util import get_structures_tree
 from rich.progress import track
-from pathlib import Path
 from scipy.ndimage import zoom
 
-from bg_atlasapi import utils
-from bg_atlasgen.mesh_utils import create_region_mesh, Region
+from bg_atlasgen.mesh_utils import Region, create_region_mesh
 from bg_atlasgen.wrapup import wrapup_atlas_from_data
-from bg_atlasapi.structure_tree_util import get_structures_tree
-import imio
-import zipfile
-import os
 
 PARALLEL = True  # disable parallel mesh extraction for easier debugging
 
@@ -47,7 +44,6 @@ def get_structure_id_path_from_id(id, id_dict, root_id):
         return structure_id_path
 
     while True:
-
         parent = int(id_dict[id])
         structure_id_path.insert(0, parent)
 
@@ -141,7 +137,6 @@ def create_atlas(
 
     structures = []
     for row in range(df.shape[0]):
-
         entry = {
             "acronym": df["Acronym"][row],
             "id": int(df["ID"][row]),  # from np.int for JSON serialization
@@ -184,15 +179,13 @@ def create_atlas(
         node.data = Region(is_label)
 
     if mesh_creation == "generate":
-
         closing_n_iters = 2
         decimate_fraction = 0.04
         smooth = False  # smooth meshes after creation
 
         start = time.time()
 
         if PARALLEL:
-
             pool = mp.Pool(mp.cpu_count() - 2)
 
             try: