Turn values from string to integers in JSON file python

Question

I'm trying to change values in a JSON file from strings to integers, my issue is that the keys are row numbers so I can't call by key name (as they will change consistently). The values that need changing are within the "sharesTraded" object. Below is my JSON file:

{
    "lastDate": {
        "30": "04/04/2022",
        "31": "04/04/2022",
        "40": "02/03/2022",
        "45": "02/01/2022"
    },
    "transactionType": {
        "30": "Automatic Sell",
        "31": "Automatic Sell",
        "40": "Automatic Sell",
        "45": "Sell"
    },
    "sharesTraded": {
        "30": "29,198",
        "31": "105,901",
        "40": "25,000",
        "45": "1,986"
    }
}

And here is my current code:

import json

data = json.load(open("AAPL22_trades.json"))

dataa = data['sharesTraded']
dataa1 = dataa.values()
data1 = [s.replace(',', '') for s in dataa1]
data1 = [int(i) for i in data1] 


open("AAPL22_trades.json", "w").write(
    json.dumps(data1, indent=4))

However, I need the integer values to replace the string values. Instead, my code just replaces the entire JSON with the integers. I imagine there is something extra at the end that I'm missing because the strings have been changed to integers but applying it to the JSON is missing.

Why do you write out data1 back to the file? Surely you meant to modify just the nested dict within data, and then just write data back to the file (through json.dump()) — quamrana
– quamrana, Commented Dec 4, 2022 at 17:08

tdelaney · Accepted Answer · 2022-12-04 17:32:55Z

2

You have a couple of problems. First, since you only convert the values to a list, you loose the information about which key is associated with the values. Second, you write that list back to the file, loosing all of the other data too.

You could create a new dictionary with the modified values and assign that back to the original.

import json

data = json.load(open("AAPL22_trades.json"))
data['sharesTraded'] = {k:int(v.replace(',', '')) 
    for k,v in data['sharesTraded'].items()}
json.dump(data, open("AAPL22_trades.json", "w"), indent=4)

edited Dec 4, 2022 at 17:32

answered Dec 4, 2022 at 17:23

tdelaney

77.9k6 gold badges91 silver badges129 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

quamrana Over a year ago

Did you mean: json.dump(data, open("AAPL22_trades.json", "w"), indent=4)?

tdelaney Over a year ago

@quamrana - oh, right. The actual data. Fixed.

quamrana Over a year ago

btw do you regard having open() as a parameter, just a hack or a valid alternative to with open(...?

tdelaney Over a year ago

@quamrana - yes, I'm old-school that way. File objects close the underlying file when they are deleted. Here, I'm trusting that json.dump doesn't keep a reference to the file, so it closes when the function exits. Implementations like the now-defunct jython do background garbage collection so the file object isn't really deleted when the function exits. And in a more complicated case, you may have circular references that get the python garbage collector involved, with the same problematic result. But when doing things where you don't have those risks, I'm comfortable skipping with.

tdelaney Over a year ago

@quamrana - Most python objects are deleted when refcount goes to zero. Garbage collection happens when refcount goes wrong. And to me, that is a strong suggestion that you've got a bug that should be addressed. There are legitimate cases where python can't call an object's __del__, which is good argument for using with. But if __del__ isn't called, that's crud building up in your program which would be a problem for anything long-running.

Collectives™ on Stack Overflow

Turn values from string to integers in JSON file python

1 Answer 1

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related