Merge pull request #41 from quantifyearth/mwd-misc

Small misc fixes

Author: mdales

CHANGES.md

@@ -1,3 +1,14 @@
+## v1.1.2 (12/01/2026)
+
+### Changed
+
+* Changed highest supported GDAL from 3.11.x to 3.12.x.
+* Clean up species info data when exporting it.
+
+### Fixed
+
+* Instruct mypy to ignore GDAL for typing.
+
 ## v1.1.1 (02/12/2025)
 
 ### Changed

aoh/aohcalc.py

@@ -61,7 +61,7 @@ def aohcalc(
 
     # We drop the geometry as that's a lot of data, more than the raster often
     species_info = filtered_species_info.drop('geometry', axis=1)
-    manifest = {k: v[0] for (k, v) in species_info.items()}
+    manifest = {k: v[0].item() if hasattr(v[0], 'item') else v[0] for (k, v) in species_info.items()}
 
     species_id = filtered_species_info.id_no.values[0]
     try:

aoh/cleaning.py

@@ -15,40 +15,40 @@ def tidy_data(
     """Tidy up the data as per Daniele Busana et al"""
 
     # Lower elevation higher than upper elevation
-    if not pd.isna(row.elevation_lower) and not pd.isna(row.elevation_upper):
-        if row.elevation_lower > row.elevation_upper:
-            row.elevation_lower = elevation_min
-            row.elevation_upper = elevation_max
+    if not pd.isna(row['elevation_lower']) and not pd.isna(row['elevation_upper']):
+        if row['elevation_lower'] > row['elevation_upper']:
+            row['elevation_lower'] = elevation_min
+            row['elevation_upper'] = elevation_max
 
     # Missing lower and/or upper elevation
-    if pd.isna(row.elevation_lower):
-        row.elevation_lower = elevation_min
-        if not pd.isna(row.elevation_upper) and row.elevation_upper < elevation_min:
-            row.elevation_upper = elevation_max
-    if pd.isna(row.elevation_upper):
-        row.elevation_upper = elevation_max
-        if row.elevation_lower > elevation_max:
-            row.elevation_lower = elevation_min
+    if pd.isna(row['elevation_lower']):
+        row['elevation_lower'] = elevation_min
+        if not pd.isna(row['elevation_upper']) and row['elevation_upper'] < elevation_min:
+            row['elevation_upper'] = elevation_max
+    if pd.isna(row['elevation_upper']):
+        row['elevation_upper'] = elevation_max
+        if row['elevation_lower'] > elevation_max:
+            row['elevation_lower'] = elevation_min
 
     # Lower elevation < -500 and/or upper elevation > 9000
-    row.elevation_lower = max(elevation_min, row.elevation_lower)
-    row.elevation_upper = min(elevation_max, row.elevation_upper)
+    row['elevation_lower'] = max(elevation_min, row['elevation_lower'])
+    row['elevation_upper'] = min(elevation_max, row['elevation_upper'])
 
     # Small difference (<50m) between lower and upper elevation
-    elevation_diff = row.elevation_upper - row.elevation_lower
+    elevation_diff = row['elevation_upper'] - row['elevation_lower']
     if elevation_diff < elevation_seperation:
         spare = elevation_seperation - elevation_diff
         adjust = math.ceil(spare / 2)
-        row.elevation_lower -= adjust
-        row.elevation_upper += adjust
-
-        if row.elevation_lower < elevation_min:
-            adjust = elevation_min - row.elevation_lower
-            row.elevation_lower += adjust
-            row.elevation_upper += adjust
-        elif row.elevation_upper > elevation_max:
-            adjust = row.elevation_upper - elevation_max
-            row.elevation_lower -= adjust
-            row.elevation_upper -= adjust
+        row['elevation_lower'] -= adjust
+        row['elevation_upper'] += adjust
+
+        if row['elevation_lower'] < elevation_min:
+            adjust = elevation_min - row['elevation_lower']
+            row['elevation_lower'] += adjust
+            row['elevation_upper'] += adjust
+        elif row['elevation_upper'] > elevation_max:
+            adjust = row['elevation_upper'] - elevation_max
+            row['elevation_lower'] -= adjust
+            row['elevation_upper'] -= adjust
 
     return row

aoh/validation/validate_map_prevalence.py

@@ -3,6 +3,7 @@
 
 import argparse
 from pathlib import Path
+from typing import cast
 
 import pandas as pd
 
@@ -22,7 +23,7 @@ def generate_results_summary(aoh_df: pd.DataFrame, outliers: pd.DataFrame) -> st
     )
 
     # Count species by class
-    class_counts = aoh_df.groupby('class_name').size().to_dict()
+    class_counts = cast(dict[str,int], aoh_df.groupby('class_name').size().to_dict())
     outlier_counts = outliers.groupby('class_name').size().to_dict()
 
     for class_name in sorted(class_counts.keys()):

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "aoh"
-version = "1.1.1"
+version = "1.1.2"
 description = "A library for calculating Area of Habitat for species distribution mapping"
 authors = [
     {name = "Michael Dales", email = "mwd24@cam.ac.uk"}
@@ -32,7 +32,7 @@ dependencies = [
     "yirgacheffe>=1.11.1,<2.0",
     "zenodo_search",
     "pandas>=2.0,<3.0",
-    "gdal[numpy]>=3.8,<3.12",
+    "gdal[numpy]>=3.8,<3.13",
     "tomli",
 ]
 
@@ -83,3 +83,7 @@ packages = ["aoh", "aoh.summaries", "aoh.validation"]
 
 [tool.setuptools.package-data]
 aoh = ["py.typed"]
+
+[[tool.mypy.overrides]]
+module = "osgeo.*"
+ignore_missing_imports = true