Enhance datetime formatting for Git: handle very old dates by returning a minimum date of 1980-01-01

Author: marcarl

exporters/git/batch_processor.py

@@ -15,7 +15,7 @@
 from exporters.git.git_utils import checkout_branch, push_to_target_repository
 
 
-def process_files_with_git_batch(json_files, output_dir, verbose, predocs, batch_size=10):
+def process_files_with_git_batch(json_files, output_dir, verbose, predocs, batch_size=100):
     """Process files with git batch workflow, using same branch but pushing after each batch."""
     # Clone target repository once for all batches
     repo_dir, original_cwd = clone_target_repository_to_temp(verbose=verbose)
@@ -66,7 +66,8 @@ def process_files_with_git_batch(json_files, output_dir, verbose, predocs, batch
                 print(f"Misslyckades med att pusha till target repository")
 
     except Exception as e:
-        print(f"Oväntat fel vid git batch processing: {e}")
+        print(f"Fel vid git batch processing: {e}")
+        raise  # Re-raise the exception so temporal processing errors are visible
     finally:
         # Always change back to original directory
         os.chdir(original_cwd)

temporal/apply_temporal.py

@@ -6,7 +6,7 @@
 relativt till ett angivet target datum.
 
 Regler:
-1. Sektioner med selex:status="upphavd" tas bort helt
+1. Sektioner med selex:status="upphavd" eller "gammal upphord" tas bort helt
 2. Sektioner med selex:status="ikraft" och selex:ikraft_datum > target_date tas bort helt
 3. Sektioner med selex:status="ikraft" och selex:ikraft_datum <= target_date får sina temporal attribut borttagna
 4. Sektioner med selex:upphor_datum som är <= target_date tas bort helt
@@ -75,7 +75,7 @@ def apply_temporal(markdown_text: str, target_date: str, verbose: bool = False)
             # Kontrollera status-attribut
             if status_match:
                 status_value = status_match.group(1)
-                if "upphavd" in status_value:
+                if "upphavd" in status_value or "upphord" in status_value:
                     should_remove = True
                     remove_reason = f"status '{status_value}'"
                 elif "ikraft" in status_value and ikraft_match:

util/datetime_utils.py

@@ -33,9 +33,17 @@ def format_datetime_for_git(dt_str: Optional[str]) -> Optional[str]:
         else:
             # Just date, add midnight time
             dt = datetime.fromisoformat(dt_str + 'T00:00:00')
+        
+        # Git/GitHub has problems with very old dates, use 1980-01-01 as minimum
+        if dt.year < 1980:
+            return "1980-01-01T00:00:00"
+        
         return dt.strftime('%Y-%m-%dT%H:%M:%S')
     except (ValueError, AttributeError):
         # Fallback: try to add time to basic date format
         if dt_str and len(dt_str) == 10:  # YYYY-MM-DD format
+            year = int(dt_str[:4])
+            if year < 1980:
+                return "1980-01-01T00:00:00"
             return dt_str + 'T00:00:00'
         return dt_str