miscs

annotating_by_label.py

@@ -12,8 +12,7 @@ import annotating_with_checks
 
 from utils import natural_key
 
-# Roughly 10 A4 pages at 100 DPI
-MAX_HEIGHT_PX = 20000 # Can be increased by 10%.
+MAX_HEIGHT_PX = 25000 # Can be increased by 10%.
 
 def render_item(item):
     student_id, label, content = item
@@ -134,15 +133,32 @@ def main():
         shutil.rmtree(bgnot_dir)
     os.makedirs(bgnot_dir, exist_ok=True)
 
+    used_prefixes = set()
+
+    previous_prefix = None
     for line in lines:
         labels = [l.strip() for l in line.split(',') if l.strip()]
         safe_labels = [l.replace(":", "").strip() for l in line.split(',') if l.strip()]
         if not labels:
             continue
 
-        prefix = os.path.commonprefix(safe_labels).strip()
-        if not prefix:
-            prefix = "Group"
+        base_prefix = os.path.commonprefix(safe_labels).strip()
+        if not base_prefix:
+            base_prefix = "Group"
+
+        unique_prefix = base_prefix
+        if unique_prefix[-1] == "i":
+            unique_prefix = unique_prefix[:-1]
+        counter = 2
+        while unique_prefix in used_prefixes:
+            unique_prefix = f"{base_prefix}-{counter}"
+            counter += 1
+        if counter == 2 and previous_prefix and previous_prefix in unique_prefix:
+            unique_prefix = f"{previous_prefix}-{counter}"
+        elif counter == 2:
+            previous_prefx = unique_prefix
+
+        used_prefixes.add(unique_prefix)
 
         items_to_render = []
         for sid, lbls in results.items():
@@ -201,7 +217,7 @@ def main():
             batches = batches2
 
         for i, batch in enumerate(batches, 1):
-            save_batch(batch, prefix, i, root_dir, args.overwrite)
+            save_batch(batch, unique_prefix, i, root_dir, args.overwrite)
 
 if __name__ == "__main__":
     main()