Update ppt_chunker.py

ppt_chunker.py

@@ -1,3 +1,4 @@
+import re
 import uuid
 import pandas as pd
 from io import StringIO
@@ -185,7 +186,12 @@ def build_prompt_conv():
         }
     ]
 
-#def multiline_string_to_json(multiline_string: str):
+def find_next_word_after_spaces(input_string):
+    match = re.search(r'\s{2,}(\S+)', input_string)
+    
+    if match:
+        return match.group(1)
+    return None
         
 
 def ppt_chunker(file_like, llm):
@@ -208,12 +214,12 @@ def ppt_chunker(file_like, llm):
         if elem.category == 'Table':
             if current_chunk == '':
                 lines = elem.text.split('\n')
-                result = [line.split() for line in lines]
+                result = [find_next_word_after_spaces(line) for line in lines]
                 print(f'TAB : {pd.DataFrame(result)}')
                 current_chunk = elem.text
             else:
                 lines = elem.text.split('\n')
-                result = [line.split() for line in lines]
+                result = [find_next_word_after_spaces(line) for line in lines]
                 print(f'TAB : {pd.DataFrame(result)}')
                 current_chunk += '\n' + elem.text
             continue