fix: Adaptation for the Course Authoring problems structure

xmodule/capa/capa_problem.py

@@ -548,7 +548,14 @@ def find_correct_answer_text(self, answer_id):
             return answer_id[0]
         if xml_element.tag == 'optioninput':
             return xml_element.xpath('@correct')[0]
-        return ', '.join(xml_element.xpath('*[@correct="true"]/text()'))
+
+        answers_list = xml_element.xpath('*[@correct="true"]//text()')
+        cleaned_answers = [answer for answer in answers_list if re.sub(r'[^a-zA-Z0-9\s]', '', answer).strip()]
+
+        if not cleaned_answers:
+            cleaned_answers = xml_element.getparent().xpath('@answer')
+
+        return ', '.join(cleaned_answers)
 
     def find_question_label(self, answer_id):
         """
@@ -625,7 +632,9 @@ def generate_default_question_label():
             if questiontext_elem is not None and questiontext_elem.tag in LABEL_ELEMS:
                 question_text = questiontext_elem.text
             else:
-                question_text = generate_default_question_label()
+                # Course authoring store a question inside <p> tag within the problem
+                problem_title = xml_elem.xpath('p') or xml_elem.getparent().xpath('p')
+                question_text = problem_title[0].text if len(problem_title) == 1 else generate_default_question_label()
 
         return question_text
 

xmodule/capa/inputtypes.py

@@ -542,7 +542,9 @@ def extract_choices(element, i18n, text_only=False):
                 if not text_only:
                     text = stringify_children(choice)
                 else:
-                    text = choice.text
+                    text_string = choice.xpath('string()')
+                    cleaned_text = re.sub(r'[^a-zA-Z0-9\s]', '', text_string).strip()
+                    text = cleaned_text or text_string
                 choices.append((choice.get("name"), text))
             else:
                 if choice.tag != 'compoundhint':