验证整个字符串并获得所有匹配项

Question

I'd like to validate my string input if it it contains digits separated by spaces. I tried using the regex

^(\d)(?:\s(\d))*$

and was able to check the validity via the test(String) method, but when I use str.split(regex) it does not give me the correct matches.

For example, I have the following input:

2 3 4 5

When I try to use the split method, I get

["", "2", "5", ""]

I was expecting the result to be

["2", "3", "4", "5"]

Any suggestion is greatly appreciated, thanks!

Answer 1

Your ^(\\d)(?:\\s(\\d))*$ regex splits the string correctly as ["", "2", "5", ""] because:

• The first (\\d) captures a digit at the beginning of the string, and inserts it into the resulting array. Since the digit is found at the beginning, the empty string before the string start and the first digit appears as the first array element, and 2 is the second
• The (?:\\s(\\d))* construct is a * -quantified non-capturing group that matches 0+ sequenes of a whitespace followed with a digit, and the digit is captured . That is, each time a space+digit is matched, the Group 2 slot is re-written with a new value, and the last one is kept that way, which is 5 . Since the right-hand split border happened at the end of the string, the empty string is added to the resulting array.

If your string is already validated, you need no /\\d+/g , you may just split with a space (note that capturing groups are redundant in your regex then):

 var s = "2 3 4 5"; if (/^\\d(?:\\s\\d)*$/.test(s)) { console.log(s.split(" ")); } // var s = "2 3 4 a"; won't show any output 

Answer 2

Use the regular expression \\d+ with the global option g (which tells .match to return all matches instead of just the first one):

"2 3 4 5".match(/\\d+/g)

Result: ["2", "3", "4", "5"]

---

Edit : To test the validity of the input string (requirement: digits separated by spaces), use:

(/^(\\d )+\\d?$/).test("1 2 3 4 a")

The (\\d )+ matches any number of repetitions of a digit followed by a space, and \\d? matches a possible trailing digit without a following whitespace character.