Add Match-the-Columns Question Type

backend/Generator/match_columns.py

@@ -0,0 +1,76 @@
+import spacy
+import random
+from nltk.tokenize import sent_tokenize
+
+nlp = spacy.load("en_core_web_sm")
+
+def extract_pairs_from_text(text, max_pairs=6):
+    doc = nlp(text)
+    pairs = []
+    seen_terms = set()
+    seen_definitions = set()
+
+    for sent in doc.sents:
+        sent_doc = nlp(sent.text)
+
+        for ent in sent_doc.ents:
+            term = ent.text.strip()
+            definition = sent.text.strip()
+            if (
+                term and
+                len(term.split()) <= 4 and
+                term.lower() not in seen_terms and
+                definition not in seen_definitions and
+                len(definition) > len(term) + 10
+            ):
+                pairs.append({"term": term, "definition": definition})
+                seen_terms.add(term.lower())
+                seen_definitions.add(definition)
+                break
+
+        if len(pairs) >= max_pairs:
+            break
+
+    if len(pairs) < 3:
+        for sent in doc.sents:
+            sent_doc = nlp(sent.text)
+            for chunk in sent_doc.noun_chunks:
+                term = chunk.text.strip()
+                definition = sent.text.strip()
+                if (
+                    term and
+                    len(term.split()) <= 4 and
+                    term.lower() not in seen_terms and
+                    definition not in seen_definitions and
+                    len(definition) > len(term) + 10
+                ):
+                    pairs.append({"term": term, "definition": definition})
+                    seen_terms.add(term.lower())
+                    seen_definitions.add(definition)
+                    break
+            if len(pairs) >= max_pairs:
+                break
+
+    return pairs[:max_pairs]
+
+
+def generate_match_columns(payload):
+    text = payload.get("input_text", "")
+    max_pairs = payload.get("max_questions", 6)
+
+    pairs = extract_pairs_from_text(text, max_pairs)
+
+    if not pairs:
+        return {"pairs": [], "left_column": [], "right_column": []}
+
+    left_column = [p["term"] for p in pairs]
+    right_column = [p["definition"] for p in pairs]
+
+    shuffled_right = right_column.copy()
+    random.shuffle(shuffled_right)
+
+    return {
+        "pairs": pairs,
+        "left_column": left_column,
+        "right_column": shuffled_right
+    }

backend/server.py

@@ -25,6 +25,7 @@
 from httplib2 import Http
 from oauth2client import client, file, tools
 from mediawikiapi import MediaWikiAPI
+from Generator.match_columns import generate_match_columns
 
 app = Flask(__name__)
 CORS(app)
@@ -490,6 +491,23 @@ def get_transcript():
 
     return jsonify({"transcript": transcript_text})
 
+
+@app.route("/get_match_columns", methods=["POST"])
+def get_match_columns():
+    data = request.get_json()
+    input_text = data.get("input_text", "")
+    use_mediawiki = data.get("use_mediawiki", 0)
+    max_questions = data.get("max_questions", 6)
+    input_text = process_input_text(input_text, use_mediawiki)
+
+    result = generate_match_columns({
+        "input_text": input_text,
+        "max_questions": max_questions
+    })
+
+    return jsonify(result)
+
+
 if __name__ == "__main__":
     os.makedirs("subtitles", exist_ok=True)
     app.run()