Merge pull request #139 from Imageomics/copilot/fix-138

Fix UTF-8 encoding issues in file operations to prevent Windows crashes and add Windows CI testing

Author: johnbradley

.github/workflows/run-tests.yml

@@ -4,10 +4,11 @@ on: [push]
 
 jobs:
   build:
-    runs-on: ubuntu-latest
+    runs-on: ${{ matrix.os }}
 
     strategy:
       matrix:
+        os: [ubuntu-latest, windows-latest]
         python-version: ['3.10', '3.11', '3.12', '3.13']
 
     steps:

src/bioclip/__main__.py

@@ -202,7 +202,7 @@ def parse_args(input_args=None):
 
 def create_classes_str(cls_file_path):
     """Reads a file with one class per line and returns a comma separated string of classes"""
-    with open(cls_file_path, 'r') as cls_file:
+    with open(cls_file_path, 'r', encoding="utf-8") as cls_file:
         cls_str = [item.strip() for item in cls_file.readlines()]
     return ",".join(cls_str)
 

src/bioclip/predict.py

@@ -348,7 +348,7 @@ def get_txt_names(self) -> List[List[str]]:
             List[List[str]]: A list of lists, where each inner list contains names corresponding to the text embeddings.
         """
         txt_names_json = self.get_cached_datafile("embeddings/txt_emb_species.json")
-        with open(txt_names_json) as fd:
+        with open(txt_names_json, encoding="utf-8") as fd:
             txt_names = json.load(fd)
         return txt_names