CODEWITHSUNDEEP
pythonxmljsonserializationtkinter
Basj
Basj

Reputation: 46343

Save objects into JSON or XML file in Python

I create objects (using tkinter widgets) that I wish to save in a JSON or XML file, so that I can recover them after startup.

from Tkinter import *

class Texte:
    def __init__(self, ax, ay, txt):
        self.entry = Entry(root,bd=0,font=("Purisa",int(15)))
        self.entry.insert(0, txt)
        self.x = ax
        self.y = ay 
        self.entry.place(x=self.x,y=self.y)

root = Tk()

a = Texte(10, 20, 'blah')
b = Texte(20, 70, 'blah2')

# here the user will modify the entries' x, y, txt, etc.

L = [a,b]

# here save the list L (containing the Texte objects) into a JSON file or XML so that I can recover them after restart 

root.mainloop()

How can save and restore these objects with JSON or XML ?

(I'm a bit lost with http://docs.python.org/2/library/json.html right now.)

Upvotes: 4

Views: 9882

Answers (3)

Andreas Klebinger
Andreas Klebinger

Reputation: 407

If you never modify these objects outside of your Application why the requirement for JSON or XML? If you don't change your data outside of your application you can use the pickle Module to serialize and deserialize the object into either binary data or a ASCII string and save these.

For details see: http://docs.python.org/2/library/pickle.html

There is also a 3rd party library which enables storage of classes as JSON. http://jsonpickle.github.io/

I haven't used it yet myself though so not sure how human readable the output is, but if all you want to do is store it as file and reload it after a application restart I don't see any advantage of JSON/XML over using Pickle.

Edit: As others pointed out you can use cPickle over pickle for better performance. The Algorithm is the same so you use them the exact same way.

Upvotes: 3

aIKid
aIKid

Reputation: 28322

It's mentioned in the docs, use json.dump.

Example of use:

import json

data = {'a':1, 'b':2}
with open('my_json.txt', 'w') as fp:
    json.dump(data, fp)

In your case, you can't convert the object itself to a json format. Save the information only:

data = {'a':(10, 20, 'blah'), 'b':(20, 70, 'blah2')
with open('my_json.txt', 'w') as fp:
     json.dump(data, fp)

And when you load it back:

with open('my_json.txt') as fp:
    data = json.loads(fp)
    a = Texte(*data['a'])
    b = Texte(*data['b'])

Upvotes: 6

Jon
Jon

Reputation: 12954

Please see update: json is simplejson.

json.dump(for file export) and json.dumps(for string export) can be very useful if you have a simple object. However, if you need to save a more complex data structure like a dictionary filled with dictionaries {'a':{...}, 'b':2} json from the standard library struggles.

For such cases a tool like simplejson can be useful. http://simplejson.readthedocs.org/en/latest/

>>> import simplejson as json
>>> json.dumps(['foo', {'bar': ('baz', None, 1.0, 2)}])
'["foo", {"bar": ["baz", null, 1.0, 2]}]'
>>> print(json.dumps("\"foo\bar"))
"\"foo\bar"
>>> print(json.dumps(u'\u1234'))
"\u1234"

As you need to save a complex list L to json - I would go with simplejson. Something like:

import simplejson as json
with open('Texte.json', 'w') as texte_file:
    json.dump(L, texte_file)`

Update: simplejson is json from the standard library (added in 2.6 to the standard library). For more details see: What are the differences between json and simplejson Python modules?.

Upvotes: 1

Related Questions