Replace string in list with python

Question

trying to replace all elements named 'number' to 'numbr' in the data list but doesn't get it working.

Edit: So each key number should be renamed to numbr. Values stay as they are.

What am I doing wrong?

Thank you for your help!

data = [{'address': {
 'city': 'city A',
 'company_name': 'company A'},
 'amount': 998,
 'items': [{'description': 'desc A1','number': 'number A1'}],
 'number': 'number of A',
 'service_date': {
 'type': 'DEFAULT',
 'date': '2015-11-18'},
 'vat_option': 123},
 {'address': {
 'city': 'city B',
 'company_name': 'company B'},
 'amount': 222,
 'items': [{'description': 'desc B1','number': 'number B1'},
 {'description': 'desc B2','number': 'number B2'}],
 'number': 'number of B',
 'service_date': {
 'type': 'DEFAULT',
 'date': '2015-11-18'},
 'vat_option': 456}
 ]
def replace(l, X, Y):
 for i,v in enumerate(l):
 if v == X:
 l.pop(i)
 l.insert(i, Y)
replace(data, 'number', 'numbr')
print data

• RomanPerekhrest: it should rename only keys? Post the expected result
• thowi: This is the result - as you can see, no change: [{'vat_option': 123, 'items': [{'description': 'desc A1', 'number': 'number A1'}], 'number': 'number of A', 'amount': 998, 'address': {'city': 'city A', 'company_name': 'company A'}, 'service_date': {'date': '2015年11月18日', 'type': 'DEFAULT'}}, {'vat_option': 456, 'items': [{'description': 'desc B1', 'number': 'number B1'}, {'description': 'desc B2', 'number': 'number B2'}], 'number': 'number of B', 'amount': 222, 'address': {'city': 'city B', 'company_name': 'company B'}, 'service_date': {'date': '2015年11月18日', 'type': 'DEFAULT'}}]
• RomanPerekhrest: I've asked for the expected result, not the actual
• thowi: Sorry. Expected result is: [{'vat_option': 123, 'items': [{'description': 'desc A1', 'numbr': 'number A1'}], 'numbr': 'number of A', 'amount': 998, 'address': {'city': 'city A', 'company_name': 'company A'}, 'service_date': {'date': '2015年11月18日', 'type': 'DEFAULT'}}, {'vat_option': 456, 'items': [{'description': 'desc B1', 'numbr': 'number B1'}, {'description': 'desc B2', 'numbr': 'number B2'}], 'numbr': 'number of B', 'amount': 222, 'address': {'city': 'city B', 'company_name': 'company B'}, 'service_date': {'date': '2015年11月18日', 'type': 'DEFAULT'}}]
• thowi: So each key number -> numbr. Values stay as they are.

Answer 1

The following is a recursive replace implementation that replaces p1 by p2 in any string it encounters in the s object, recursing through lists, sets, tuples, dicts (both keys and values):

def nested_replace(s, p1, p2):
 if isinstance(s, basestring): # Python2
 # if isinstance(s, (str, bytes)): # Python3
 return s.replace(p1, p2)
 if isinstance(s, (list, tuple, set)):
 return type(s)(nested_replace(x, p1, p2) for x in s)
 if isinstance(s, dict):
 return {nested_replace(k, p1, p2): nested_replace(v, p1, p2) for k, v in s.items()}
 return s
>>> from pprint import pprint
>>> pprint(nested_replace(data, 'number', 'numbr'))
[{'address': {'city': 'city A', 'company_name': 'company A'},
 'amount': 998,
 'items': [{'description': 'desc A1', 'numbr': 'numbr A1'}],
 'numbr': 'numbr of A',
 'service_date': {'date': '2015-11-18', 'type': 'DEFAULT'},
 'vat_option': 123},
 {'address': {'city': 'city B', 'company_name': 'company B'},
 'amount': 222,
 'items': [{'description': 'desc B1', 'numbr': 'numbr B1'},
 {'description': 'desc B2', 'numbr': 'numbr B2'}],
 'numbr': 'numbr of B',
 'service_date': {'date': '2015-11-18', 'type': 'DEFAULT'},
 'vat_option': 456}]

• thowi: Thank you!!! That's a good start because it works fine! Awesome :) Is there a chance to only replace the keys? Current result is: 'numbr': 'numbr of B', it would be better to have it like: 'numbr': 'number of B',
• user2390182: Sure, remove the block dealing with lists and stuff. And in the dict block, replace nested_replace(v, p1, p2) by a simple v. You will lose the keys of nested dicts though...
• thowi: Hey, sorry to bother you again - can you please take a look here? I think I got you wrong, because replacing is not working :( def nested_replace(s, p1, p2): if isinstance(s, basestring): # Python2 return s.replace(p1, p2) # if isinstance(s, (list, tuple, set)): # return type(s)(nested_replace(x, p1, p2) for x in s) if isinstance(s, dict): return {nested_replace(k, p1, p2): v for k, v in s.items()} return s

Answer 2

eval function is anti pattern, but I think eval is best solution here

data1 = eval(repr(data).replace('number', 'numbr'))

• thowi: Hey, thank you! Thats a smart, small solution :D But I'd indeed like to replace only the key, value can stay as is.

Answer 3

If you are trying to replace both keys and values this will work.

from json import dumps, loads
 data = [{'address': {
 'city': 'city A',
 'company_name': 'company A'},
 'amount': 998,
 'items': [{'description': 'desc A1','number': 'number A1'}],
 'number': 'number of A',
 'service_date': {
 'type': 'DEFAULT',
 'date': '2015-11-18'},
 'vat_option': 123},
 {'address': {
 'city': 'city B',
 'company_name': 'company B'},
 'amount': 222,
 'items': [{'description': 'desc B1','number': 'number B1'},
 {'description': 'desc B2','number': 'number B2'}],
 'number': 'number of B',
 'service_date': {
 'type': 'DEFAULT',
 'date': '2015-11-18'},
 'vat_option': 456}
 ]
data_string = dumps(data)
data = loads(data_string.replace('number', 'numbr')

• thowi: Hey, thank you! Thats a smart, small solution :D But I'd indeed like to replace only the key, value can stay as is.