Most reusable way to publish a list as JSON

Question

I have a list that I want to publish as JSON, so that others can easily reuse it:

product1: price1
product2: price2
product3: price3
product4: price4

What is the most reusable way to publish this data as JSON?
productX and priceX are strings.

---

I thought I would use the same format as this JSON list of country names/codes, but it does not even validate on JSONLint:

Parse error on line 2:
[    {        name: 'Afghanistan',
--------------^
Expecting 'STRING', '}'

For some reason everyone seems happy to use though, so I might be misunderstanding something...
Sample:

[ 
  {name: 'Afghanistan', code: 'AF'}, 
  {name: 'Albania', code: 'AL'}
]

Answer 1

Your sample data as well as the provided list of country codes are valid JavaScript, but they are not JSON. To make them valid JSON, put all strings in double quotes:

[ 
  {"name": "Afghanistan", "code": "AF"}, 
  {"name": "Albania", "code": "AL"}
]

This answer on StackOverflow gives a nice overview of the differences between JavaScript and JSON.

Regarding your initial question, your sample data would look like this:

[
    {"product1": "price1"},
    {"product2": "price2"},
    {"product3": "price3"},
    {"product4": "price4"}
]

In most cases¹ it's probably better to make your prices actual numbers instead of strings that just look like numbers:

[
    {"product1": 123.45},
    {"product2": 234.56},
    {"product3": 345.67},
    {"product4": 456.78}
]

If you think about adding more properties to your products in the future, you should probably use this instead:

[
    {
        "name": "product1",
        "price": 123.45
    },
    {
        "name": "product2",
        "price": 234.56
    },
    {
        "name": "product3",
        "price": 345.67
    },
    {
        "name": "product4",
        "price": 456.78
    }
]

---

¹ As Walter Tross pointed out in the comments, there are edge cases when rounding errors can occur. For example, if you have a product that costs 456789.99, it might happen that the computer program that reads your JSON will actually calculate the price as 456790.00. There is currently no simple solution for this problem. If you think this might be relevant for you, you might want to read up on some related discussions and the general problem of the precision of floating point numbers.