# JSONata - $substringBefore

**URL:** <https://discourse.nodered.org/t/jsonata-substringbefore/11920>\
**Category:** General\
**Created:** [6 June 2019 20:14 UTC](https://discourse.nodered.org/t/jsonata-substringbefore/11920 "2019-06-06T20:14:48Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Andrei](https://sea2.discourse-cdn.com/flex026/user_avatar/discourse.nodered.org/andrei/32/10446_2.png) [@Andrei](https://discourse.nodered.org/u/Andrei)\
**Post date:** [6 June 2019 20:14 UTC](https://discourse.nodered.org/t/jsonata-substringbefore/11920/1 "2019-06-06T20:14:48Z")

</div>

It has been asked in [this post](https://discourse.nodered.org/t/ideas-to-remove-an-entry-starting-with-a-period-from-an-array-of-filenames/11854) how to remove from an array the strings that start with a period character.

The first solution proposed is simply brilliant. It uses the array.filter() method, ES6 arrow function and a ternary operator to solve the problem perfectly, in a single line.

A second solution proposed to use a JSONata expression as follows:

```auto
[payload[$substringBefore(".")]]

```

I had troubles in the past to understand this jsonata function and to be honest I never got it right. I decided to submit it to testing.

I will input this array of string and check if it filters all the strings that start with `dot` and only them.

```auto
["aaa", ".bbb", "ccc", " ", "dd.d", "", "eee.", ".", "fff", ". "]

```

and the result is below

 ![s-01](https://us1.discourse-cdn.com/flex026/uploads/nodered/original/2X/7/7e13d1b46fbb8c8adfdf8840086a06a9efbef071.png)

It works perfectly. So another brilliant solution but hard to understand or at least not very intuitive.

I believe it works because of the [sequence flattening rules](https://docs.jsonata.org/processing) that says:

> An **empty sequence** is a sequence with no values and is considered to be 'nothing' or 'no match'. It won't appear in the output of any expression

Edit: Flow

```auto
[{"id":"f990066f.6b1e58","type":"tab","label":"Jsonata - subtringBefore","disabled":false,"info":""},{"id":"fa4c268e.83ff98","type":"inject","z":"f990066f.6b1e58","name":"Array","topic":"","payload":"[\"aaa\",\".bbb\",\"ccc\",\" \",\"dd.d\",\"\",\"eee.\",\".\",\"fff\",\". \"]","payloadType":"json","repeat":"","crontab":"","once":false,"onceDelay":0.1,"x":130,"y":80,"wires":[["c306e321.2f67f"]]},{"id":"a933db56.36c0e8","type":"debug","z":"f990066f.6b1e58","name":"","active":true,"tosidebar":true,"console":false,"tostatus":false,"complete":"true","targetType":"full","x":510,"y":80,"wires":[]},{"id":"c306e321.2f67f","type":"change","z":"f990066f.6b1e58","name":"$substringBefore","rules":[{"t":"set","p":"output","pt":"msg","to":"[payload[$substringBefore(\".\")]]","tot":"jsonata"}],"action":"","property":"","from":"","to":"","reg":false,"x":330,"y":80,"wires":[["a933db56.36c0e8"]]}]

```

.

---

<div class="post-metadata">

**Author:** ![Andrei](https://sea2.discourse-cdn.com/flex026/user_avatar/discourse.nodered.org/andrei/32/10446_2.png) [@Andrei](https://discourse.nodered.org/u/Andrei)\
**Post date:** [6 June 2019 21:07 UTC](https://discourse.nodered.org/t/jsonata-substringbefore/11920/2 "2019-06-06T21:07:53Z")

</div>

I wanted to post an old flow (using the same JSONata function) that confused me but I could not find that flow.

As far as I can remember from that flow..... using `$substringBefore` and `$substringAfter` was resulting in the very same output for the very same dataset.

However, reading the documentation I finally have a clue.

Most likely I was using as the second argument (`chars`) a character that was not present in the dataset passed to the functions. The documentation says that in such case the function will return the first argument untouched. 🤦‍♂️

> ## `$substringBefore()`
> 
> **Signature:** `$substringBefore(str, chars)`
> 
> Returns the substring before the first occurrence of the character sequence `chars` in `str` . If `str` is not specified (i.e. this function is invoked with only one argument), then the context value is used as the value of `str` . If `str` does not contain `chars` , then it returns `str` . An error is thrown if `str` and `chars` are not strings.