Bug fixes for geoh5py and dataclasses

* fix: update matplotlib cmap for deprecation

* fix: geoh5py block model was flipped

* fix: changing dataclasses to use default factory to prevent shared objects

* ci: changing lint rules

* style: style fixes by ruff and autoformatting by black

---------

Co-authored-by: lachlangrose <7371904+lachlangrose@users.noreply.github.com>

.github/workflows/linter.yml

@@ -2,8 +2,6 @@ name: "✅ Linter"
 
 on:
   push:
-    branches:
-      - master
     paths:
       - '**.py'
 
@@ -47,6 +45,6 @@ jobs:
           ruff check ${{env.PROJECT_FOLDER}} --fix
       - uses: stefanzweifel/git-auto-commit-action@v5
         with:
-          push_options: --force
+          push_options: 
           commit_message: "style: style fixes by ruff and autoformatting by black"
 

LoopStructural/datatypes/_point.py

@@ -1,4 +1,4 @@
-from dataclasses import dataclass
+from dataclasses import dataclass, field
 import numpy as np
 
 from typing import Optional, Union
@@ -10,9 +10,9 @@
 
 @dataclass
 class ValuePoints:
-    locations: np.ndarray
-    values: np.ndarray
-    name: str
+    locations: np.ndarray = field(default_factory=lambda: np.array([[0, 0, 0]]))
+    values: np.ndarray = field(default_factory=lambda: np.array([0]))
+    name: str = "unnamed"
     properties: Optional[dict] = None
 
     def to_dict(self):
@@ -108,9 +108,9 @@ def from_dict(cls, d, flatten=False):
 
 @dataclass
 class VectorPoints:
-    locations: np.ndarray
-    vectors: np.ndarray
-    name: str
+    locations: np.ndarray = field(default_factory=lambda: np.array([[0, 0, 0]]))
+    vectors: np.ndarray = field(default_factory=lambda: np.array([[0, 0, 0]]))
+    name: str = "unnamed"
     properties: Optional[dict] = None
 
     def to_dict(self):

LoopStructural/datatypes/_structured_grid.py

@@ -1,19 +1,19 @@
 from typing import Dict
 import numpy as np
-from dataclasses import dataclass
+from dataclasses import dataclass, field
 from LoopStructural.utils import getLogger
 
 logger = getLogger(__name__)
 
 
 @dataclass
 class StructuredGrid:
-    origin: np.ndarray
-    step_vector: np.ndarray
-    nsteps: np.ndarray
-    cell_properties: Dict[str, np.ndarray]
-    properties: Dict[str, np.ndarray]
-    name: str
+    origin: np.ndarray = field(default_factory=lambda: np.array([0, 0, 0]))
+    step_vector: np.ndarray = field(default_factory=lambda: np.array([1, 1, 1]))
+    nsteps: np.ndarray = field(default_factory=lambda: np.array([10, 10, 10]))
+    cell_properties: Dict[str, np.ndarray] = field(default_factory=dict)
+    properties: Dict[str, np.ndarray] = field(default_factory=dict)
+    name: str = "default_grid"
 
     def to_dict(self):
         return {

LoopStructural/datatypes/_surface.py

@@ -1,4 +1,4 @@
-from dataclasses import dataclass
+from dataclasses import dataclass, field
 from typing import Optional
 import numpy as np
 import io
@@ -9,8 +9,8 @@
 
 @dataclass
 class Surface:
-    vertices: np.ndarray
-    triangles: np.ndarray
+    vertices: np.ndarray = field(default_factory=lambda: np.array([[0, 0, 0]]))
+    triangles: np.ndarray = field(default_factory=lambda: np.array([[0, 0, 0]]))
     normals: Optional[np.ndarray] = None
     name: str = 'surface'
     values: Optional[np.ndarray] = None

LoopStructural/export/geoh5.py

@@ -78,8 +78,10 @@ def add_structured_grid_to_geoh5(filename, structured_grid, overwrite=True, grou
         if structured_grid.cell_properties is not None:
             for k, v in structured_grid.cell_properties.items():
                 data[k] = {
-                    'association': "CELL",
-                    "values": np.rot90(v.reshape(structured_grid.nsteps - 1, order="F")).flatten(),
+                    "association": "CELL",
+                    "values": np.flipud(
+                        np.rot90(v.reshape(structured_grid.nsteps - 1, order="F"), 1)
+                    ).flatten(),
                 }
         block = geoh5py.objects.BlockModel.create(
             workspace,

LoopStructural/interpolators/__init__.py

@@ -115,4 +115,5 @@ class InterpolatorType(IntEnum):
 
 from ._interpolator_factory import InterpolatorFactory
 from ._interpolator_builder import InterpolatorBuilder
+
 # from ._api import LoopInterpolator

LoopStructural/modelling/features/_cross_product_geological_feature.py

@@ -1,5 +1,4 @@
-"""
-"""
+""" """
 
 import numpy as np
 from typing import Optional

LoopStructural/modelling/features/_projected_vector_feature.py

@@ -1,5 +1,4 @@
-"""
-"""
+""" """
 
 import numpy as np
 from typing import Optional

LoopStructural/modelling/features/fold/__init__.py

@@ -1,6 +1,4 @@
-"""
-
-"""
+""" """
 
 from ._fold import FoldEvent
 from ._svariogram import SVariogram

LoopStructural/modelling/input/process_data.py

@@ -78,7 +78,9 @@ def __init__(
         self.contacts = contacts
         self._contact_orientations = None
         self.contact_orientations = contact_orientations
-        self._fault_orientations = pd.DataFrame(columns=["X", "Y", "Z", "gx", "gy", "gz", "coord", "feature_name"])
+        self._fault_orientations = pd.DataFrame(
+            columns=["X", "Y", "Z", "gx", "gy", "gz", "coord", "feature_name"]
+        )
         self.fault_orientations = fault_orientations
         self._fault_locations = None
         self.fault_locations = fault_locations
@@ -313,7 +315,7 @@ def fault_properties(self, fault_properties):
                 pts = self.fault_orientations.loc[
                     self.fault_orientations["feature_name"] == fname, ["gx", "gy", "gz"]
                 ]
-                if len(pts)>0:
+                if len(pts) > 0:
                     fault_properties.loc[
                         fname,
                         ["avgNormalEasting", "avgNormalNorthing", "avgNormalAltitude"],

LoopStructural/utils/_surface.py

@@ -63,7 +63,7 @@ def fit(
         self,
         values: Optional[Union[list, int, float]],
         name: Optional[Union[List[str], str]] = None,
-        local=False
+        local=False,
     ) -> surface_list:
         """Extract isosurfaces from the interpolator
 

LoopStructural/utils/colours.py

@@ -18,12 +18,14 @@ def random_colour(n: int = 1, cmap='tab20'):
         List of colours in the form of (r,g,b,a) tuples
     """
     from matplotlib import colormaps as cm
+
     colours = []
     for _i in range(n):
         colours.append(cm.get_cmap(cmap)(rng.random()))
 
     return colours
 
+
 def random_hex_colour(n: int = 1, cmap='tab20'):
     """
     Generate a list of random colours
@@ -46,4 +48,4 @@ def random_hex_colour(n: int = 1, cmap='tab20'):
     for _i in range(n):
         colours.append(cm.get_cmap(cmap)(rng.random()))
 
-    return [f'#{int(c[0]*255):02x}{int(c[1]*255):02x}{int(c[2]*255):02x}' for c in colours]
+    return [f'#{int(c[0]*255):02x}{int(c[1]*255):02x}{int(c[2]*255):02x}' for c in colours]

examples/1_basic/plot_3_model_visualisation.py

@@ -2,12 +2,12 @@
 
 1c. Visualising models
 ===============================
-The following tutorial will demonstrate how to use the Loop structural visualisation module. 
+The following tutorial will demonstrate how to use the Loop structural visualisation module.
 This module provides a wrapper for the lavavu model that is written by
-Owen Kaluza. 
+Owen Kaluza.
 
 Lavavu allows for interactive visualisation of 3D models within a jupyter
-notebook environment. 
+notebook environment.
 
 """
 

examples/3_fault/fault_network.py

@@ -1,7 +1,7 @@
 """
 3b. Modelling a fault network in LoopStructural
 ===============================================
-Uses GeologicalModel, ProcessInputData and Loop3DView from LoopStructural library. 
+Uses GeologicalModel, ProcessInputData and Loop3DView from LoopStructural library.
 Also using geopandas to read a shapefile, pandas, matplotlib and numpy."""
 
 import LoopStructural

examples/4_advanced/_5_using_logging.py

@@ -1,7 +1,7 @@
 """
 1e. Using logging
 ===============================
-LoopStructural has a number of levels of logging incorporated in the code to allow 
+LoopStructural has a number of levels of logging incorporated in the code to allow
 for recording and debugging the models.
 The python logging module allows for 5 levels of messages to be returned to the user:
 1. Debug messages
@@ -11,14 +11,14 @@
 5. Critical messages
 
 LoopStructural uses all of these logging levels to report the various aspects of the model
-building process. 
-Generally, the user only needs to be aware of the warning and error messages. 
+building process.
+Generally, the user only needs to be aware of the warning and error messages.
 
 By default the warning, error and critical messages are returned to the console and will appear to
-the user. 
+the user.
 All messages except for debug are recorded to a file :code:`default-loop-structural-logfile.log`.
 
-Lets have a look at the logging from the Claudius model. 
+Lets have a look at the logging from the Claudius model.
 """
 
 from LoopStructural import GeologicalModel

examples/4_advanced/_7_local_weights.py

@@ -1,11 +1,11 @@
 """
 ============================
-1f. Local data weighting 
+1f. Local data weighting
 ============================
 LoopStructural primarily uses discrete interpolation methods (e.g. finite differences on a regular grid,
 or linear/quadratic on tetrahedral meshes). The interpolation is determined by combining a regularisation
-term and the data weights. The default behaviour is for every data point to be weighted equally, however 
-it is also possible to vary these weights per datapoint. 
+term and the data weights. The default behaviour is for every data point to be weighted equally, however
+it is also possible to vary these weights per datapoint.
 
 """
 

examples/4_advanced/_model_from_geological_map.py

@@ -1,17 +1,17 @@
 """
 4.a Building a model using the ProcessInputData
 ===============================================
-There is a disconnect between the input data required by 3D modelling software and a geological map. 
-In LoopStructural the geological model is a collection of implicit functions that can be mapped to 
-the distribution of stratigraphic units and the location of fault surfaces. Each implicit function 
-is approximated from the observations of the stratigraphy, this requires grouping conformable geological 
-units together as a singla implicit function, mapping the different stratigraphic horizons to a value of 
-the implicit function and determining the relationship with geological structures such as faults. 
-In this tutorial the **ProcessInputData** class will be used to convert geologically meaningful datasets to input for LoopStructural. 
+There is a disconnect between the input data required by 3D modelling software and a geological map.
+In LoopStructural the geological model is a collection of implicit functions that can be mapped to
+the distribution of stratigraphic units and the location of fault surfaces. Each implicit function
+is approximated from the observations of the stratigraphy, this requires grouping conformable geological
+units together as a singla implicit function, mapping the different stratigraphic horizons to a value of
+the implicit function and determining the relationship with geological structures such as faults.
+In this tutorial the **ProcessInputData** class will be used to convert geologically meaningful datasets to input for LoopStructural.
 The **ProcessInputData** class uses:
-* stratigraphic contacts* stratigraphic orientations* stratigraphic thickness* stratigraphic order 
-To build a model of stratigraphic horizons and:* fault locations* fault orientations * fault properties* fault edges 
-To use incorporate faults into the geological model. """
+* stratigraphic contacts* stratigraphic orientations* stratigraphic thickness* stratigraphic order
+To build a model of stratigraphic horizons and:* fault locations* fault orientations * fault properties* fault edges
+To use incorporate faults into the geological model."""
 
 ##############################
 # Imports