Update ppt_chunker.py

ppt_chunker.py

@@ -14,7 +14,7 @@ def process_chunk(chunk, nlp):
         current = chunk[1][i]
         current_text = current[1]
         
-        if (type(current) is tuple) and current_text.isupper() and (current[0] == 'Title'):
+        if (type(current) is list) and current_text.isupper() and (current[0] == 'Title'):
             tokens = nlp(current_text)
             
             try:
@@ -22,7 +22,7 @@ def process_chunk(chunk, nlp):
             except IndexError:
                 continue
                 
-            if (type(next) is tuple) and next[1].isupper() and (next[0] == 'Title'):
+            if (type(next) is list) and next[1].isupper() and (next[0] == 'Title'):
                 if tokens[-1].pos_ in ["ADP", 'PART', 'PRON', 'DET', "AUX", 'SCONJ', 'CONJ', "CCONJ"]:
                     chunk[1][i+1][1] = current_text + ' ' + next[1]
                     marked.append(i)
@@ -53,7 +53,7 @@ def ppt_chunk(file_like, model):
             if elem.category == "ListItem":
                 list_items.append(elem.text)
             else:
-                current_chunk.append((elem.category, elem.text))
+                current_chunk.append([elem.category, elem.text])
 
     sr = time.time()
     for chunk in chunks: