Fix/parse tree bug

blabla/sentence_processor/sentence_processing_engine.py

@@ -2,6 +2,7 @@
 from blabla.sentence_processor.yngve_tree import YngveNode
 from blabla.sentence_processor.pos_tag_counting_engine import PosTagCounter
 from blabla.utils.exceptions import *
+from blabla.utils.utils import get_nltk_tree
 import blabla.utils.settings as settings
 from blabla.utils.global_params import *
 import os
@@ -140,9 +141,8 @@ def const_parse_tree(self):
 				parseTree (CoreNLP): The CoreNLP constituency parse tree object
 		"""
 		document = self.client.annotate(self._sent)
-		document = json.loads(document.text)
-		pt = document["sentences"][0]["parse"]
-		parseTree = Tree.fromstring(pt)
+		parseTree = document.sentence[0].parseTree
+		parseTree = get_nltk_tree(parseTree)
 		return parseTree
 
 	def num_words(self):

blabla/utils/utils.py

@@ -0,0 +1,18 @@
+"""A file containing helper functions"""
+
+from nltk.tree import Tree
+from typing import Any
+
+
+def get_nltk_tree(parse_tree: Any) -> Tree:
+    """Given a parse_tree obtained from CoreNLP, this function recursively generates an NLTK
+    Tree based on the parse_tree information.
+    Args:
+        parse_tree (CoreNLP_pb2.ParseTree): The CoreNLP formatted parse tree
+    Returns:
+        (Tree): The parse tree converted into NLTK format
+    """
+    return Tree(
+        parse_tree.value,
+        [get_nltk_tree(child) for child in parse_tree.child]
+    ) if parse_tree.child else parse_tree.value

setup.py

@@ -30,7 +30,7 @@
     ],
     download_url="https://github.com/novoic/blabla/archive/v0.2.2.tar.gz",
     install_requires=[
-        "stanza==1.0.0",
+        "stanza>=1.0.0,<2.0.0",
         "flask==1.1.2",
         "jsonpickle==1.4",
         "anytree==2.8.0",