How to split string with newline ('\n') in Node?

Question

Within Node, how do I split a string using newline ('\n') ? I have a simple string like var a = "test.js\nagain.js" and I need to get ["test.js", "again.js"]. I tried

a.split("\n");
a.split("\\n");
a.split("\r\n");
a.split("\r");

but the above doesn't work.

possible duplicate of [JavaScript string newline character?](http://stackoverflow.com/questions/1155678/javascript-string-newline-character) — Mene, Feb 20 '14 at 00:28
[`split()`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/String/split) doesn't modify the original string. — thgaskell, Feb 20 '14 at 02:19
This looks like a pure JavaScript question, not requiring a node.js tag. — Wyck, Jan 29 '20 at 15:36
@Wyck, it's useful to know that Node is the intended runtime, because the `newline` character in a string is often platform-dependent. In Node.js, it's very common to operate on file texts with `newline` characters, and it's important to know the platform. In a browser context, this this less of an issue. But I would suggest the OP change the question to specify the target environment, or else ask for solutions that work generically on all major platforms. — Austin Davis, Sep 07 '20 at 01:42

score 331 · Answer 1 · answered Feb 20 '14 at 00:03

331

Try splitting on a regex like /\r?\n/ to be usable by both Windows and UNIX systems.

> "a\nb\r\nc".split(/\r?\n/)
[ 'a', 'b', 'c' ]

answered Feb 20 '14 at 00:03

maerics

151,642
46
269
291

4

What about classic Macs? ;) – AshleyF Nov 07 '15 at 00:03
15

classic Macs died along with Steve Jobs... sadness :( – ymz Jun 15 '17 at 19:23
63

to catch \n, \r and \r\n: `split(/[\r\n]+/)` – Julian TF Jul 26 '17 at 10:15
2

MacOSX don't use single \r anymore, that was only for old Macs. I think they have the same \n as other unixes. – jcubic Dec 10 '17 at 09:10
How can .split(/\r?\n/) be changed to support any number of "\r\n" or "\n"? In other words if I have a string "The\r\n\r\nquick\r\n\r\nbrown\r\nfox\r\njumps\r\n\r\nover" how to change regex above to get just [quick, brown, fox, jumps, over]? – Neal Davis Mar 28 '18 at 22:00
I just tried .split(/[\r?\n]+/) and that seems to work for splitting if you have multiple consecutive line breaks such as "\r\n\r\n..." or "\n\n..." - hope I got that correct. – Neal Davis Mar 28 '18 at 22:18
2

You don't need the `?`. `[\r\n]` is a regex *character set*, which means it can match `\r` or `\n`. `split(/[\r\n]+/)` would work. – Jun 10 '18 at 22:32
21

/[\r\n]+/ will filter out empty lines – Spongman Jul 07 '18 at 18:44

score 64 · Answer 2 · answered Feb 20 '14 at 02:09

64

If the file is native to your system (certainly no guarantees of that), then Node can help you out:

var os = require('os');

a.split(os.EOL);

This is usually more useful for constructing output strings from Node though, for platform portability.

answered Feb 20 '14 at 02:09

qubyte

17,558
3
30
32

8

Yeah, you generally shouldn't do this. You should parse newlines regardless of the platform. – 1j01 Nov 05 '16 at 19:27
1

I highlight that this is specifically for the case of files native to your system in the very first sentence. The answer above is appropriate for when this is unknown or mixed. – qubyte Nov 06 '16 at 10:05
Right. Or when it's "known" now but might change in the future. – 1j01 Nov 06 '16 at 23:24
1

only works when loading files on the platform they were created on. – Spongman Jul 07 '18 at 18:44
Good for temp files(?) – nsandersen Jan 22 '21 at 15:26

score 56 · Answer 3 · edited Jan 30 '21 at 13:44

It looks like regex /\r\n|\r|\n/ handles CR, LF, and CRLF line endings, their mixed sequences, and keeps all the empty lines inbetween. Try that!

    function splitLines(t) { return t.split(/\r\n|\r|\n/); }

    // single newlines
    console.log(splitLines("AAA\rBBB\nCCC\r\nDDD"));
    // double newlines
    console.log(splitLines("EEE\r\rFFF\n\nGGG\r\n\r\nHHH"));
    // mixed sequences
    console.log(splitLines("III\n\r\nJJJ\r\r\nKKK\r\n\nLLL\r\n\rMMM"));

You should get these arrays as a result:

[ "AAA", "BBB", "CCC", "DDD" ]
[ "EEE", "", "FFF", "", "GGG", "", "HHH" ]
[ "III", "", "JJJ", "", "KKK", "", "LLL", "", "MMM" ]

You can also teach that regex to recognize other legit Unicode line terminators by adding |\xHH or |\uHHHH parts, where H's are hexadecimal digits of the additional terminator character codepoint (as seen in Wikipedia article as U+HHHH).

score 41 · Answer 4 · edited Jul 21 '15 at 11:58

41

a = a.split("\n");

Note that splitting returns the new array, rather than just assigning it to the original string. You need to explicitly store it in a variable.

edited Jul 21 '15 at 11:58

Charlie Harding

653
8
22

answered Feb 19 '14 at 23:59

alex

11,935
3
30
42

Doesn't work on windows. i'm seeing 'r' on every end of a line. – Tal Kohavy Mar 23 '22 at 16:21

Alejadro Xalabarder · Answer 5 · 2018-07-09T11:09:56.693

A solution that works with all possible line endings including mixed ones and keeping empty lines as well can be achieved using two replaces and one split as follows

text.replace(/\r\n/g, "\r").replace(/\n/g, "\r").split(/\r/);

some code to test it

  var CR = "\x0D";  //   \r
  var LF = "\x0A";  //   \n

  var mixedfile = "00" + CR + LF +            // 1 x win
                  "01" + LF +                 // 1 x linux
                  "02" + CR +                 // 1 x old mac
                  "03" + CR + CR +            // 2 x old mac
                  "05" + LF + LF +            // 2 x linux
                  "07" + CR + LF + CR + LF +  // 2 x win
                  "09";

  function showarr (desc, arr)
  {
     console.log ("// ----- " + desc);
     for (var ii in arr)
        console.log (ii + ") [" + arr[ii] +  "] (len = " + arr[ii].length + ")");
  }

  showarr ("using 2 replace + 1 split", 
           mixedfile.replace(/\r\n/g, "\r").replace(/\n/g, "\r").split(/\r/));

and the output

  // ----- using 2 replace + 1 split
  0) [00] (len = 2)
  1) [01] (len = 2)
  2) [02] (len = 2)
  3) [03] (len = 2)
  4) [] (len = 0)
  5) [05] (len = 2)
  6) [] (len = 0)
  7) [07] (len = 2)
  8) [] (len = 0)
  9) [09] (len = 2)

Actually the solution given by blakkwater: text.split(/\r\n|\n|\r/); do the same and it is shorter and faster — Alejadro Xalabarder, Jan 11 '19 at 16:23

score 10 · Answer 6 · answered Feb 19 '14 at 23:56

10

The first one should work:

> "a\nb".split("\n");
[ 'a', 'b' ]
> var a = "test.js\nagain.js"
undefined
> a.split("\n");
[ 'test.js', 'again.js' ]

answered Feb 19 '14 at 23:56

TimWolla

31,849
8
63
96

score 4 · Answer 7 · answered Feb 18 '17 at 08:37

4

I made an eol module for working with line endings in node or browsers. It has a split method like

var lines = eol.split(text)

answered Feb 18 '17 at 08:37

ryanve

50,076
30
102
137

How to split string with newline ('\n') in Node?

7 Answers7

Linked

Related