Regular expression to capture 3 strings in quotation marks separated by space

ok, I guess the code says it all.

//I have this string.
var str = '"Watch out" "for the" "rock!"'
//this is one of the开发者_Go百科 many patterns that I tried
res=str.match(/"(.*)" "(.*)" "(.*)"/g)

I want an array like: res=["Watch out","for the","rock!"] How can I do that?

Like this:

var str = '"Watch out" "for the" "rock!"'
res=str.match(/"(.*?)" "(.*?)" "(.*?)"/)
res.shift()

You need to remove the g flag because with that flag you will get a list of matches and not the list of groups for a single match. Also, you should specify your groups as non-greedy (quantifier *?), otherwise they might match too much. Finally, remove the first entry from the array using res.shift() - the first entry is the entire matched string and you want only the groups.

/"[^"]*"\s?/g should be a better regex for splitting up any amount of values separated by quotations and spaces.

res = str.match(/^"([^"]*)" "([^"]*)" "([^"]*)"$/).slice(1)

About regular expression I can suggest you the following web site: RegEx LIB. It contains many regular expressions (the link I sent you is directly for String category).
Moreover you can test your regular expressions (also using Client Side engine - Javascript) at REG Tester.

I am sure there you can find your wished regular expression and you can also refer to it everytime you will need a regulare expression.

I use it every time I need RegEX in my projects.

Expanding on Ktash's answer, but using a capture group (the brackets) to get only the text within the quotation marks:

/"([^"]*)"\s?/g;

The only problem is that match() doesn't return capture groups with a global modifier. So have to use in split() or RegExp.exec() which gets more messy:

Option 1 - Using RegExp.exec()

str = '"Watch out" "for the" "rock!"'
re = /"([^"]*)"\s?/g;
result = [];
match = re.exec(str);
while (match != null) {
     result.push(match[1]);
     match = re.exec(str);
}

Each call to re.exec(str) returns the captured group until none are remaining when it will return null. If you called it again, it would start again, i.e:

> str = '"Watch out" "for the" "rock!"'
  ""Watch out" "for the" "rock!""
> re = /"([^"]*)"\s?/g;
  /"([^"]*)"\s?/g;
> re.exec(str)
  [""Watch out" ", "Watch out"]
> re.exec(str)
  [""for the" ", "for the"]
> re.exec(str)
  [""rock!"", "rock!"]
> re.exec(str)
  null
> re.exec(str)
  [""Watch out" ", "Watch out"]

As per all the other answers here, the results returned are an array containing the whole match and then the capture group, hence why match[1] is used.

Option 2 - Using split()

str = '"Watch out" "for the" "rock!"'
re = /"([^"]*)"\s?/; //The global identifier isn't needed
result = str.split(re)
result = result.filter(function(e){if (e!==""){return true}}) //Remove blanks from array

.split() accepts a regex, however the results of str.split(re) are:

["", "Watch out", "", "for the", "", "rock!", ""]

so it is necessary to use the filter() function to remove the blank entries. I think the split() approach is more elegant, but I think Internet Explorer (at least up to 8) lacks the filter() function.