"string indices must be integers" error when loading in JSON

Question

I want to access items from a new dictionary called conversations by implementing a for loop.

{" conversations": [
{"tag": "greeting",
"user": ["Hi", " What's your name?", " How are you?", "Hello", "Good day"],
"response": ["Hello, my name is Rosie. Nice to see you", "Good to see you again", " How can I help you?"],
"context_set": ""
},
{"tag": "Good-bye",
"user": ["Bye", "See you", "Goodbye"],
"response": ["Thanks for visiting our company", "Have a nice day", "Good-bye."]
},
{"tag": "thanks",
"user": ["Thanks", "Thank you", "That's helpful", "Appreciated your service" ],
"response": ["Glad to help!", "My pleasure", "You’re welcome."]
}
]
}

The code I use to load the dictionary in a notebook is

    file_name = 'dialogue.txt'
    with open(file_name, encoding='utf8') as f:
        for line in f:
        print(line.strip())
        dialogue_text = f.read()

This line of code does not return any results when trying to access the dictionary.

  for k in dialogue_text:
  print(k)

My intention is to write this code by implementing tokenization and stemming, but it returned an error

words = []
labels = []
docs_x = []
docs_y = []

for conversation in dialogue_text["conversations"]:
    for user in dialogue_text["user"]:
        words = nltk.word_tokenize(user)
        words.extend(words)
        docs_x.append(words)
        docs_y.append(intent["tag"])if intent["tag"] not in labels:
        labels.append(intent["tag"])words = [stemmer.stemWord(w.lower()) for w in words if w 
     != "?"]
     words = sorted(list(set(words)))labels = sorted(labels)

Error Message:

     TypeError                                 Traceback (most recent call last)
     <ipython-input-12-d42234f8e809> in <module>()
     10 docs_y = []
     11 
     ---> 12 for conversation in dialogue_text["conversations"]:
     13     for user in dialogue_text["user"]:
     14         words = nltk.word_tokenize(user)

     TypeError: string indices must be integers

What code should I write to resolve this issue?

Answer 1

Try this out:

import json

dialogue_text = json.load(open("dialogue.txt", encoding='utf8'))

for conversation in dialogue_text[" conversations"]:
    for user in conversation['user']:
        print(user)

Output:

Hi
What's your name?
 How are you?
Hello
Good day
Bye
See you
Goodbye
Thanks
Thank you
That's helpful
Appreciated your service