We Keep Coding

How to rewrite this RegExp to not use deprecated API?

I have the following RegExp myRegexp, that matches numbers in a string:
var myRegexp = new RegExp('[0-9]+');
Then I have the following code that extracts numbers from a string and returns an array:
var string = '123:456';
var nums = new Array();
while(myRegexp.test(string)) {
nums.length++;
nums[nums.length - 1] = RegExp.lastMatch;
string = RegExp.rightContext;
}
Should return an array of two elements: "123", and "456".
However, RegExp.lastMatch and RegExp.rightContext are deprecated/non-standard API, and not portable. How can I rewrite this logic using portable JS API?
Thanks,

To match all numbers in a string, you'd simply use string.match(/\d/g); to match all single digits in a separate array entry, or string.match(/\d+/g); to match as numbers. There's no need for any of the things you've tried to use…
let string = "2kdkane2kdkie83kdkdk303ldld";
let match = string.match(/\d+/g);
let match1 = string.match(/\d/g);
console.log('numbers:', match);
console.log('single digits:', match1);

Use the g flag to perform a global match which will find all matches without having to repeatedly test the string.
let s = '123:456'
const regexp = new RegExp(/\d+/g);
let nums = s.match(regexp);
console.log(nums);

match regular expression - JavaScript

So I have the following url:
var oURL = "https://graph.facebook.com/#{username}/posts?access_token=#{token}";
I want to take username and token out of it;
I tried:
var match = (/#\{(.*?)\}/g.exec(oURL));
console.log(match);
but it is giving me:
["#{username}", "username", index: 27, input: "https://graph.facebook.com/#{username}/posts?access_token=#{token}"
Why isn't catching token?
Thanks

The problem is that exec only returns the first match from the given index whenever called.
Returns
If the match succeeds, the exec() method returns an array and updates
properties of the regular expression object. The returned array has
the matched text as the first item, and then one item for each
capturing parenthesis that matched containing the text that was
captured.
If the match fails, the exec() method returns null.
You would need to loop, continuously matching again to find all the matches.
var matches = [],
match,
regex = /#\{(.*?)\}/g,
oURL = "https://graph.facebook.com/#{username}/posts?access_token=#{token}";
while (match = regex.exec(oURL)) {
matches.push(match)
}
console.log(matches)
However, if you are only interested in the first capture group, you can only add those to the matches array:
var matches = [],
match,
regex = /#\{(.*?)\}/g,
oURL = "https://graph.facebook.com/#{username}/posts?access_token=#{token}";
while (match = regex.exec(oURL)) {
matches.push(match[1])
}
console.log(matches)

Try this instead:
oURL.match(/#\{(.*?)\}/g)

The answer you accepted is perfect, but I thought I'd also add that it's pretty easy to create a little helper function like this:
function getMatches(str, expr) {
var matches = [];
var match;
while (match = expr.exec(str)) {
matches.push(match[1]);
}
return matches;
}
Then you can use it a little more intuitively.
var oURL = "https://graph.facebook.com/#{username}/posts?access_token=#{token}";
var expr = /#\{([^\{]*)?\}/g;
var result = getMatches(oURL, expr);
console.log(result);
http://codepen.io/Chevex/pen/VLyaeG

Try this:
var match = (/#\{(.*?)\}.*?#\{(.*?)\}/g.exec(oURL));

How to extract string in regex

I have string in this format:
var a="input_[2][invoiceNO]";
I want to extract "invoiceNo" string. I've tried:
var a="input_[2][invoiceNO]";
var patt = new RegExp('\[(.*?)\]');
var res = patt.exec(a);
However, I get the following output:
Array [ "[2]", "2" ]
I want to extract only invoiceNo from the string.
Note: Input start can be any string and in place of number 2 it can be any number.

I would check if the [...] before the necessary [InvoiceNo] contains digits and is preceded with _ with this regex:
/_\[\d+\]\s*\[([^\]]+)\]/g
Explanation:
_ - Match underscore
\[\d+\] - Match [1234]-like substring
\s* - Optional spaces
\[([^\]]+)\] - The [some_invoice_123]-like substring
You can even use this regex to find invoice numbers inside larger texts.
The value is in capture group 1 (see m[1] below).
Sample code:
var re = /_\[\d+\]\s*\[([^\]]+)\]/g;
var str = 'input_[2][invoiceNO]';
while ((m = re.exec(str)) !== null) {
alert(m[1]);
}

You can use this regex:
/\[(\w{2,})\]/
and grab captured group #1 from resulting array of String.match function.
var str = 'input_[2][invoiceNO]'
var m = str.match(/\[(\w{2,})\]/);
//=> ["[invoiceNO]", "invoiceNO"]
PS: You can also use negative lookahead to grab same string:
var m = str.match(/\[(\w+)\](?!\[)/);

var a="input_[2][invoiceNO]";
var patt = new RegExp('\[(.*?)\]$');
var res = patt.exec(a);

Try this:
var a="input_[2][invoiceNO]";
var patt = new RegExp(/\]\[(.*)\]/);
var res = patt.exec(a)[1];
console.log(res);
Output:
invoiceNO

You could use something like so: \[([^[]+)\]$. This will extract the content within the last set of brackets. Example available here.

Use the greediness of .*
var a="input_[2][invoiceNO]";
var patt = new RegExp('.*\[(.*?)\]');
var res = patt.exec(a);

Extract Numbers between text in Javascript?

I have a string as a1234b5.
I am trying to get 1234 (in between a and b5). i tried the following way
number.replace(/[^0-9\.]/g, '');
But it's giving me like 12345. But I need 1234. how to achieve this in Javascript ?

You can use:
var m = 'a1234b5'.match(/\d+/);
if (m)
console.log(m[0]);
//=> "1234"

slighty different approach
var a = "a1234b5243,523kmw3254n293f9823i32lia3un2al542n5j5j6j7k7j565h5h2ghb3bg43";
var b;
if ( typeof a != "undefined" )
{
b = a.match( /[0-9]{2,}/g );
console.log( b );
}
no output if a isn't set.
if a is empty => null
if somethings found => ["1234", "5243", "523", "3254", "293", "9823", "32", "542", "565", "43"]

Assuming that there are always letters around the numbers you want and that you only care about the very first group of numbers that are surrounded by letters, you can use this:
("abc123456def1234ghi123".match(/[^\d](\d+)[^\d]/) || []).pop()
// "123456"

var number = 'a1234b5';
var firstMatch = number.match(/[0-9]+/);
var matches = number.match(/[0-9]+/g);
var without = matches.join('');
var withoutNum = Number(without);
console.log(firstMatch); // ["1234"]
console.log(matches); // ["1234","5"]
console.log(without); // "12345"
console.log(withoutNum); // 12345
I have a feeling that number is actually a hexadecimal. I urge you to update the question with more information (i.e. context) than you're providing.

It's not clear if a and b are always part of the strings you are working with; but if you want to 'extract' the number out, you could use:
var s = "a1234b5",
res = s.match(/[^\d](\d+)[^\d]/);
// res => ["a1234b", "1234"]
then, you could reassign or do whatever. It's not clear what your intention is based on your use of replace. But if you are using replace to convert that string to just the number inside the [a-z] characters, this would work:
s.replace(/[^\d](\d+)[^\d](.*)$/, "$1")
But, that's assuming the first non-digit character of the match has nothing before it.

Regex to grab strings between square brackets

I have the following string: pass[1][2011-08-21][total_passes]
How would I extract the items between the square brackets into an array? I tried
match(/\[(.*?)\]/);
var s = 'pass[1][2011-08-21][total_passes]';
var result = s.match(/\[(.*?)\]/);
console.log(result);
but this only returns [1].
Not sure how to do this.. Thanks in advance.

You are almost there, you just need a global match (note the /g flag):
match(/\[(.*?)\]/g);
Example: http://jsfiddle.net/kobi/Rbdj4/
If you want something that only captures the group (from MDN):
var s = "pass[1][2011-08-21][total_passes]";
var matches = [];
var pattern = /\[(.*?)\]/g;
var match;
while ((match = pattern.exec(s)) != null)
{
matches.push(match[1]);
}
Example: http://jsfiddle.net/kobi/6a7XN/
Another option (which I usually prefer), is abusing the replace callback:
var matches = [];
s.replace(/\[(.*?)\]/g, function(g0,g1){matches.push(g1);})
Example: http://jsfiddle.net/kobi/6CEzP/

var s = 'pass[1][2011-08-21][total_passes]';
r = s.match(/\[([^\]]*)\]/g);
r ; //# => [ '[1]', '[2011-08-21]', '[total_passes]' ]
example proving the edge case of unbalanced [];
var s = 'pass[1]]][2011-08-21][total_passes]';
r = s.match(/\[([^\]]*)\]/g);
r; //# => [ '[1]', '[2011-08-21]', '[total_passes]' ]

add the global flag to your regex , and iterate the array returned .
match(/\[(.*?)\]/g)

I'm not sure if you can get this directly into an array. But the following code should work to find all occurences and then process them:
var string = "pass[1][2011-08-21][total_passes]";
var regex = /\[([^\]]*)\]/g;
while (match = regex.exec(string)) {
alert(match[1]);
}
Please note: i really think you need the character class [^\]] here. Otherwise in my test the expression would match the hole string because ] is also matches by .*.

'pass[1][2011-08-21][total_passes]'.match(/\[.+?\]/g); // ["[1]","[2011-08-21]","[total_passes]"]
Explanation
\[ # match the opening [
Note: \ before [ tells that do NOT consider as a grouping symbol.
.+? # Accept one or more character but NOT greedy
\] # match the closing ] and again do NOT consider as a grouping symbol
/g # do NOT stop after the first match. Do it for the whole input string.
You can play with other combinations of the regular expression
https://regex101.com/r/IYDkNi/1

[C#]
string str1 = " pass[1][2011-08-21][total_passes]";
string matching = #"\[(.*?)\]";
Regex reg = new Regex(matching);
MatchCollection matches = reg.Matches(str1);
you can use foreach for matched strings.