Update ppt_chunker.py

ppt_chunker.py

@@ -173,14 +173,19 @@ def ppt_chunker(file_like, llm):
 
     ids = []
     chunks = []
-    current_chunk = []
+    current_chunk = ''
 
     for elem in elements:
         if elem.category == 'PageBreak':
             ids.append(int(uuid.UUID(elem.id)))
             chunks.append(current_chunk)
-            current_chunk = []
-        current_chunk.append(clean(elem.text, extra_whitespace=True, dashes=True, bullets=True, lowercase=True, trailing_punctuation=True))
+            current_chunk = ''
+            continue
+
+        if current_chunk == '':
+            current_chunk = clean(elem.text, extra_whitespace=True, dashes=True, bullets=True, lowercase=True, trailing_punctuation=True))
+        else:
+            current_chunk += '\n' + clean(elem.text, extra_whitespace=True, dashes=True, bullets=True, lowercase=True, trailing_punctuation=True))
 
     for chunk in chunks:
         print(f' TEXT : {chunk}')