from mongo.helpers import get_mongo_collection mongoUri = "mongodb://speechRecoUser:speech!reco@localhost/archSpeechReco" dbName = "archSpeechReco" colName = "moviesMeta" col = get_mongo_collection(colName, dbName, mongoUri) pipeline = [{'$match': {'gcTextReco.words': {'$exists': True}}}, {'$project': {'gcTextReco.words': 1}}] col.aggregate(pipeline) words = col.aggregate(pipeline) words_dict = dict() for w in words: ww = w['gcTextReco']['words'] words_dict[w['_id']] = " ".join([e['word'] for e in ww]) for key, value in words_dict.items(): try: col.update_one({"_id": key}, {"$set": {"gcTextReco.transcript_fix": value}}) except Exception as e: print(e) else: print(f"mongo update OK {key}")