# NR Linter - \[msg\] Unreachable code
**URL:** https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820
**Category:** General
**Created:** [22 April 2026 16:04 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820 "2026-04-22T16:04:36Z")
**Posts on this page:** 20
**Page:** 1
### Author:  [@mudwalker](https://discourse.nodered.org/u/mudwalker)
#### Post date: [22 April 2026 16:04 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/1 "2026-04-22T16:04:37Z")
I am going through my NR code trying to clear the Linter error messages. I get this error with all of some similarly coded Nodes, they all do what is expected of them (control MQTT Outputs). Could someone explain to me why this error is displayed and how to get rid of it (other than ignore). Or do I need to ignore as it is a correct message because Linter cannot trap the error correctly?
TIA
Colin J
Code for the Node...
```json
[{"id":"6e8f1e99bbbf3993","type":"function","z":"30b349c08c491058","name":"Diverter V/V","func":"//let boilerDivV = global.get(\"boilerDivV\");\nlet Sonoff1 = \"cmnd/DivV/CH_on/POWER\";\nlet Sonoff2 = \"cmnd/DivV/HW_OK/POWER\";\n\n/*\n No Power - B - 0\n White - AB - 1\n White & Grey - A - 2\n Grey - Hold - 3\n*/\nlet msg1;\nlet msg2;\n\nnode.status({ text: msg.boilerDivV });\n\nswitch (msg.boilerDivV) {\n case 0: //HW ON (B)\n msg1 = { topic: Sonoff1, payload: \"OFF\" } //White\n msg2 = { topic: Sonoff2, payload: \"OFF\" } //Grey\n global.set('boilerDivV', 0);\n return [msg1, msg2]\n break;\n case 1: //CH ON (AB)\n msg1 = {topic: Sonoff1, payload: \"ON\"} //White\n msg2 = { topic: Sonoff2, payload: \"OFF\" } //Grey\n global.set('boilerDivV', 1);\n return [msg1, msg2]\n break;\n case 2: //HW OK (A)\n msg1 = { topic: Sonoff1, payload: \"ON\" } //White\n msg2 = { topic: Sonoff2, payload: \"ON\" } //Grey\n global.set('boilerDivV')\n return [msg1, msg2]\n break;\n case 3: //HOLD (~~)\n msg1 = { topic: Sonoff1, payload: \"OFF\" } //White\n msg2 = { topic: Sonoff2, payload: \"ON\" } //Grey\n global.set('boilerDivV', 3);\n return [msg1, msg2]\n break; \n case 4: //Boiler OFF (B)\n msg1 = { topic: Sonoff1, payload: \"OFF\" } //White\n msg2 = { topic: Sonoff2, payload: \"OFF\" } //Grey\n global.set('boilerDivV', 4);\n return [msg1, msg2]\n break;\n\n default: //OFF (B)\n msg1 = { topic: Sonoff1, payload: \"OFF\" } //White\n msg2 = { topic: Sonoff2, payload: \"OFF\" } //Grey\n global.set('boilerDivV', 5);\n return [msg1, msg2]\n break;\n}","outputs":2,"timeout":0,"noerr":0,"initialize":"","finalize":"","libs":[],"x":1150,"y":960,"wires":[["ecfac1afe2bed870"],["06cd21e59f837b89"]]}]
```
---
### Author:  [@Colin](https://discourse.nodered.org/u/Colin)
#### Post date: [22 April 2026 16:37 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/2 "2026-04-22T16:37:31Z")
Delete the `break` statements. Execution can never get to them.
---
### Author:  [@Colin](https://discourse.nodered.org/u/Colin)
#### Post date: [22 April 2026 16:43 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/3 "2026-04-22T16:43:15Z")
Actually leaving the `break` statements in and moving the return down to the end would be a lot better I think.
---
### Author:  [@mudwalker](https://discourse.nodered.org/u/mudwalker)
#### Post date: [22 April 2026 16:43 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/4 "2026-04-22T16:43:20Z")
Thank you @Colin.
The break was superfluous anyway, must have got mixed up with some other language coding!!
---
### Author:  [@mudwalker](https://discourse.nodered.org/u/mudwalker)
#### Post date: [22 April 2026 16:52 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/5 "2026-04-22T16:52:22Z")
Tried that and get: `[msg] Expected a 'break' statement before 'case'`.
I agree, you would think that the return at the very end of the code should work, but doesn't, or at least, for the Linter.
---
### Author:  [@Colin](https://discourse.nodered.org/u/Colin)
#### Post date: [22 April 2026 18:01 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/6 "2026-04-22T18:01:20Z")
It is ok for me
```auto
//let boilerDivV = global.get("boilerDivV");
let Sonoff1 = "cmnd/DivV/CH_on/POWER";
let Sonoff2 = "cmnd/DivV/HW_OK/POWER";
/*
No Power - B - 0
White - AB - 1
White & Grey - A - 2
Grey - Hold - 3
*/
let msg1;
let msg2;
node.status({ text: msg.boilerDivV });
switch (msg.boilerDivV) {
case 0: //HW ON (B)
msg1 = { topic: Sonoff1, payload: "OFF" } //White
msg2 = { topic: Sonoff2, payload: "OFF" } //Grey
global.set('boilerDivV', 0);
break;
case 1: //CH ON (AB)
msg1 = {topic: Sonoff1, payload: "ON"} //White
msg2 = { topic: Sonoff2, payload: "OFF" } //Grey
global.set('boilerDivV', 1);
break;
case 2: //HW OK (A)
msg1 = { topic: Sonoff1, payload: "ON" } //White
msg2 = { topic: Sonoff2, payload: "ON" } //Grey
global.set('boilerDivV')
break;
case 3: //HOLD (~~)
msg1 = { topic: Sonoff1, payload: "OFF" } //White
msg2 = { topic: Sonoff2, payload: "ON" } //Grey
global.set('boilerDivV', 3);
break;
case 4: //Boiler OFF (B)
msg1 = { topic: Sonoff1, payload: "OFF" } //White
msg2 = { topic: Sonoff2, payload: "OFF" } //Grey
global.set('boilerDivV', 4);
break;
default: //OFF (B)
msg1 = { topic: Sonoff1, payload: "OFF" } //White
msg2 = { topic: Sonoff2, payload: "OFF" } //Grey
global.set('boilerDivV', 5);
break;
}
return [msg1, msg2]
```
---
### Author:  [@TotallyInformation](https://discourse.nodered.org/u/TotallyInformation)
#### Post date: [22 April 2026 20:33 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/7 "2026-04-22T20:33:18Z")
A long time ago, I got into the habit of enclosing each cases code in braces which forces JavaScript to know that they are code blocks. Doesn't usually make much difference but occasionally will catch structural errors.
```javascript
switch (msg.boilerDivV) {
case 0: { //HW ON (B)
msg1 = { topic: Sonoff1, payload: "OFF" } //White
msg2 = { topic: Sonoff2, payload: "OFF" } //Grey
global.set('boilerDivV', 0);
break;
}
case 1: { //CH ON (AB)
msg1 = {topic: Sonoff1, payload: "ON"} //White
msg2 = { topic: Sonoff2, payload: "OFF" } //Grey
global.set('boilerDivV', 1);
break;
}
case 2: { //HW OK (A)
msg1 = { topic: Sonoff1, payload: "ON" } //White
msg2 = { topic: Sonoff2, payload: "ON" } //Grey
global.set('boilerDivV')
break;
}
case 3: { //HOLD (~~)
msg1 = { topic: Sonoff1, payload: "OFF" } //White
msg2 = { topic: Sonoff2, payload: "ON" } //Grey
global.set('boilerDivV', 3);
break;
}
case 4: { //Boiler OFF (B)
msg1 = { topic: Sonoff1, payload: "OFF" } //White
msg2 = { topic: Sonoff2, payload: "OFF" } //Grey
global.set('boilerDivV', 4);
break;
}
default: { //OFF (B)
msg1 = { topic: Sonoff1, payload: "OFF" } //White
msg2 = { topic: Sonoff2, payload: "OFF" } //Grey
global.set('boilerDivV', 5);
break;
}
}
```
---
### Author:  [@mudwalker](https://discourse.nodered.org/u/mudwalker)
#### Post date: [23 April 2026 04:08 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/8 "2026-04-23T04:08:58Z")
Ah, I took the `breaks` out. I misunderstood what you were saying, and your solution looks good!!
@TotallyInformation I tend to put more Brackets in than most, it helps my brain, even with the counting up and down of left and right brackets. I think I will adopt this also.
---
### Author:  [@Colin](https://discourse.nodered.org/u/Colin)
#### Post date: [23 April 2026 07:51 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/9 "2026-04-23T07:51:29Z")
`break` means jump to the end. They were redundant originally because there was a return before them so the execution never got to the break.
You can simplify your code further by using
```auto
let msg1 = {topic: Sonoff1}
let msg2 = {topic: Sonoff2}
node.status({ text: msg.boilerDivV });
switch (msg.boilerDivV) {
case 0: //HW ON (B)
msg1.payload = "OFF" //White
msg2.payload: "OFF" //Grey
global.set('boilerDivV', 0);
break;
```
Also the definitions of Sonoff1 and 2 should be `const`. Then the interpreter is free to make optimisations based on that knowledge, and it will stop you accidentally reassigning a value to it.
---
### Author:  [@Colin](https://discourse.nodered.org/u/Colin)
#### Post date: [23 April 2026 07:54 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/10 "2026-04-23T07:54:41Z")
> [@TotallyInformation](#):
>
> A long time ago, I got into the habit of enclosing each cases code in braces
Yes, I do that too.
---
### Author:  [@mudwalker](https://discourse.nodered.org/u/mudwalker)
#### Post date: [23 April 2026 07:59 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/11 "2026-04-23T07:59:43Z")
> [@Colin](#):
>
> ```auto
> case 0: //HW ON (B)
> msg1.payload = "OFF" //White
> msg2.payload = "OFF" //Grey
> global.set('boilerDivV', 0);
> break;
>
> ```
I will update to this (I have several Nodes with similar function). Nice to see how my code can be improved upon, thank you!
---
### Author:  [@ralphwetzel](https://discourse.nodered.org/u/ralphwetzel)
#### Post date: [23 April 2026 17:25 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/12 "2026-04-23T17:25:02Z")
> [@Colin](#):
>
> ` msg2.payload: "OFF" //Grey`
Shouldn't this rather be
```
msg2.payload = "OFF" //Grey
```
? 🤔
---
### Author:  [@Colin](https://discourse.nodered.org/u/Colin)
#### Post date: [23 April 2026 17:46 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/13 "2026-04-23T17:46:24Z")
Yes, previous post corrected
---
### Author:  [@mudwalker](https://discourse.nodered.org/u/mudwalker)
#### Post date: [24 April 2026 07:57 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/14 "2026-04-24T07:57:21Z")
Next 'project' with NR is to change `let` to `const` as I come across them. These were written a while ago before `const` was the new `let` in the NR Forum`.`(Where I get most of my JS training from!) 😃
---
### Author:  [@TotallyInformation](https://discourse.nodered.org/u/TotallyInformation)
#### Post date: [24 April 2026 09:58 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/15 "2026-04-24T09:58:23Z")
Do you mean `var` to `let` or `const`? This is something that the Node-RED core codebase also needs to do!
Though moving `let` to `const` where the variable shouldn't be updated is also sensible. Also noting that objects and arrays can be `const` if all you want to do is change an entry rather than replace the whole thing.
---
### Author:  [@mudwalker](https://discourse.nodered.org/u/mudwalker)
#### Post date: [24 April 2026 10:28 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/16 "2026-04-24T10:28:11Z")
So, if a `variable` is defined at the beginning of a Function Node, _and it does not change in the Function Node_, it can be a `const`no matter what it is?
---
### Author:  [@TotallyInformation](https://discourse.nodered.org/u/TotallyInformation)
#### Post date: [24 April 2026 10:43 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/17 "2026-04-24T10:43:41Z")
Yes, though strictly, I suppose you should call that a CONSTANT. 😉
```javascript
const one = {}
const two = 42
one.something = 'hello' // This is fine
one = {something: 'hello'} // This will fail
two = 24 // this also fails
```
---
### Author:  [@mudwalker](https://discourse.nodered.org/u/mudwalker)
#### Post date: [24 April 2026 10:48 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/18 "2026-04-24T10:48:55Z")
Ah, I thought they meant CONSTANT as in a Number, not a letter/word/anything containing Alphabetic characters
---
### Author:  [@TotallyInformation](https://discourse.nodered.org/u/TotallyInformation)
#### Post date: [24 April 2026 12:50 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/19 "2026-04-24T12:50:46Z")
Nope, it means that the named variable is a "constant" - sort-of. Actually means that you cannot completely replace the content of the named variable. That you can change the properties of a `const` object or array is one of the many strange parts of JavaScript.
---
### Author:  [@mudwalker](https://discourse.nodered.org/u/mudwalker)
#### Post date: [24 April 2026 13:36 UTC](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820/20 "2026-04-24T13:36:56Z")
...and so reigneth confusion again!!🤣
[Next page](https://discourse.nodered.org/t/nr-linter-msg-unreachable-code/100820.md?page=2)