正则表达式解析对象文字

Question

Im trying to parse out a JavaScript object literal from a script block on a page. Here's the example I have of the data:

//End Update for CASE00370003 2011/08/22
stores[0] = {
'fullName' : 'Bobs Commons',
'street1' : '23 Chestnut Commons Dr'
};
//Some more comments
stores[1] = {
'fullName' : 'Gove Wood',
'street1' : '65 Lake Rd'
};

So far I've come up with:

/^(stores\[)(\d){1,2}(])(.|\n)*};$/m

However, the string ends with "};" will grab the last occurrence of the }; on stores[1], so each won't be broken individually. Thanks!

Answer 1

Seems that what you need is 'non-greedy' matches ( *? instead of * ) - it captures the shortest match rather than the longest. So the regex will look like:

/^(stores\[)(\d){1,2}(])(.|\n)*?};$/m

After suggesting that I still have to note that you in general should not do that with regular expressions at all :). Your approach will break if the object string properties contain '}' (eg, 'fullName' : 'Bobs Comm}ons').

Answer 2

Additionally to the @amakhrov answer above:

you may perform consecutive searches of closing curly brace unless JSON.parse(literal_string); executes without exceptions

UPD: missed the important fact, that for that it needs to be a valid JSON

Answer 3

PREPARE TO HAVE YOUR MIND BLOWN.

JSON (the thing several answers have mentioned) --->

J ava- S cript O bject N otation.

Cool, right? Since JS' declaration syntax is so useful, it became a neutral data format. That's the format your web request is returning, and JavaScript even has special libraries for parsing it. I think you may have just been unclear on that part.