Obtaining a JSON string from python list that contains sub-lists of objects?

Question

I have a class called Data like this:

class Data:

    def __init__(self, ticker, comments, submissions):
        self.ticker = ticker
        self.comments = comments
        self.submissions = submissions

Where ticker is a string, comments is a list of objects of type Comment and submissions is a list of objectf of type Submission. Comment and Submission have their own fields.

Now I have a list of objects of type Data

I want to itrate through the list and get a JSON String containing all the elements and print it to a file.

My code:

    json_string = json.dumps([ob.__dict__ for ob in data_list])
    f = open("data.json", "w")
    f.write(json_string)
    f.close()

This throws an error of type:

TypeError: Object of type Comment is not JSON serializable

I can't figure out what I'm doing wrong here, does anyone know?

Edit:

Comment class:

class Comment:

def __init__(self, author_name, body, ups):
    self.author_name = author_name
    self.body = body
    self.ups = ups

All fields are string/int

how does the class Comment look like?

yedpodtrzitko
– yedpodtrzitko

2020-08-23 15:23:32 +00:00
Commented Aug 23, 2020 at 15:23 — yedpodtrzitko
– yedpodtrzitko, Commented Aug 23, 2020 at 15:23
I added to the question

Isvoran Andrei
– Isvoran Andrei

2020-08-23 15:27:39 +00:00
Commented Aug 23, 2020 at 15:27 — Isvoran Andrei
– Isvoran Andrei, Commented Aug 23, 2020 at 15:27

Gray Hat · Accepted Answer · 2020-08-23 15:47:02Z

Using default=lambda x: x.__dict__ should help you. It will convert any non serializable objects and you don't have to modify a lot of previous code

import json
# rest of your code
with open("file.json", "w+", encoding="utf-8") as file:
    json.dump(datalist, file, default=lambda x: x.__dict__) #datalist is a list in my case

EDIT :

This was the complete code in my case when testing it:

import json
class Data:
    def __init__(self, ticker="string", comment=[], submissions=[]):
        self.ticker = ticker
        self.comments = comments
        self.submissions = submissions


class Comment:
    def __init__(self, author_name="", body="", ups=1):
        self.author_name = author_name
        self.body = body
        self.ups = ups

class Submission:
    def __init__(self, author_name="", body="", ups=1):
        self.author_name = author_name
        self.body = body
        self.ups = ups

comments = [Comment(ups=i) for i in range(10)]
submissions = [Submission(ups=2*i) for i in range(10)]

datalist = [Data(comment=comments, submissions=submissions) for i in range(5)]

with open("file.json", "w+", encoding="utf-8") as file:
    json.dump(datalist, file, default=lambda x: x.__dict__)

yedpodtrzitko · Accepted Answer · 2020-08-23 15:38:09Z

Classes can't be serialized by default. So either you have to serialize it manually, the same way a you do with Data class, or use a custom json encoder.

manually:

class Data:
   ...
   def to_json(self):
       res = self.__dict__
       res['comments'] = self.comments.__dict__
       return res

however this solution is not really flexible, so rather use a custom JSON encoder, which will automatically handle all objects it encounters during serialization:

# from top of my head something like this:
from json import JSONEncoder
class MyEncoder(JSONEncoder):
    def default(self, o):
        # handle instance of `Data` during json.dump 
        if isinstance(o, Data):
             return o.__dict__
        # handle instance of `Comment` during json.dump
        if isinstance(o, Comment):
             return o.__dict__

        return super().default(o)


json.dumps(data_list, cls=MyEncoder) # custom encoder should handle it

Collectives™ on Stack Overflow

Obtaining a JSON string from python list that contains sub-lists of objects?

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related