These filters are available in bot scripts and snippets:

• abs
• alphanum
• append
• array_sum
• base_convert
• base64_decode
• base64_encode
• base64url_decode
• base64url_encode
• batch
• bytes_pretty
• capitalize
• cerb_translate
• column
• context_name
• convert_encoding
• csv
• date
• date_modify
• date_pretty
• default
• escape
• filter
• first
• format
• hash_hmac
• image_info
• indent
• join
• json_encode
• json_pretty
• kata_encode
• keys
• last
• length
• lower
• map
• markdown_to_html
• md5
• merge
• nl2br
• number_format
• parse_emails
• parse_url
• permalink
• quote
• reduce
• regexp
• repeat
• replace
• reverse
• round
• secs_pretty
• sha1
• slice
• sort
• split
• split_crlf
• split_csv
• str_pos
• str_sub
• strip_lines
• striptags
• title
• tokenize
• trim
• truncate
• unescape
• upper
• url_decode
• url_encode
• values

abs

Return the absolute value of a number:

{{-5|abs}}

5

alphanum

Remove non-alphanumeric characters from a string:

{{"* Ignore spaces and non-alphanumeric characters+1$2%3!"|alphanum}}

Ignorespacesandnonalphanumericcharacters123

Also allow specific characters:

{{"* Ignore non-alphanumeric but allow spaces$%#!"|alphanum(' !')}}

Ignore nonalphanumeric but allow spaces!

append

Append a suffix to the current text.

(Introduced in 10.0.3)

|append(suffix, delimiter, trim)

{% set emails = "customer@cerb.example" %}
{{emails|append('vendor@cerb.example', delimiter=', ')}}

customer@cerb.example, vendor@cerb.example

{% set emails = null %}
{{emails|append('vendor@cerb.example', delimiter=', ')}}

vendor@cerb.example

array_sum

Sum the numeric elements of an array.

{{array_sum([1,2,3,4,5])}}

15

base_convert

Convert between number system bases.

(Introduced in 9.0.8)

{% set int = 123456789 %}
{{int|base_convert(10,16)}}

{% set hex = '75bcd15' %}
{{hex|base_convert(16,10)}}

75bcd15

123456789

base64_decode

Decode a base64-encoded string:

{% set b64 = "VGhpcyB3YXMgYmFzZTY0LWVuY29kZWQ=" %}
{{b64|base64_decode}}

This was base64-encoded

base64_encode

Encode a string in base64:

{{"This was base64-encoded"|base64_encode}}

VGhpcyB3YXMgYmFzZTY0LWVuY29kZWQ=

base64url_decode

(Added in 9.1.8)

Decode a base64url-encoded string:

{% set b64 = "VGhpcyB3YXMgYmFzZTY0dXJsLWVuY29kZWQ" %}
{{b64|base64url_decode}}

This was base64url-encoded

base64url_encode

(Added in 9.1.8)

Encode a string in base64url:

{{"This was base64url-encoded"|base64url_encode}}

VGhpcyB3YXMgYmFzZTY0dXJsLWVuY29kZWQ

batch

Break a list into smaller chunks with batch:

{% set items = ['red','blue','green'] %}
{{items|batch(2, '(empty)')|json_encode|json_pretty}}

[
    [
        "red",
        "blue"
    ],
    [
        "green",
        "(empty)"
    ]
]

bytes_pretty

Convert a number into a human readable number of bytes:

{{"123456789"|bytes_pretty(2)}}

123.46 MB

The optional argument determines the number of digits of precision.

capitalize

Capitalize the first character of a string (and lowercase the rest):

{% set first_name = "kina" %}
{{first_name|capitalize}}

Kina

cerb_translate

(Added in 9.0)

Converts string IDs (like status.open) into text in the current worker’s language.

The ticket is {{'status.open'|cerb_translate}}

The ticket is open.

column

(Added in 10.1.1)

Extract a key from each item in an array as a new array. This has the same effect as the array_column() function.

{% set people = [
  {'name':'Kina Halpue', 'email':'kina@cerb.example'},
  {'name':'Milo Dade', 'email': 'milo@cerb.example'}
] %}
{{people|column('email')|join(', ')}}

kina@cerb.example, milo@cerb.example

context_name

Convert a Cerb context ID into a human readable label.

|context_name(type)

{{'cerberusweb.contexts.ticket'|context_name('singular')}}
{{'cerberusweb.contexts.task'|context_name('plural')}}
{{'worker'|context_name('id')}}

tickets
task

convert_encoding

Convert character encodings to the first argument from the second. If the second argument is blank then Cerb will attempt to auto-detect the current encoding.

{{"This has 😂 emoji"|convert_encoding('iso-8859-1', 'utf-8')}}

This has ? emoji

csv

(Added in 9.6.4)

Format an array as a comma-separated values list. This is useful for exporting reports for Excel from bots.

{% set records = [
	{
		id: 1,
		subject: "Help with the API",
	},
	{
		id: 2,
		subject: "Automating email replies", 
	}
] %}
ID,Subject
{{records|csv}}

ID,Subject
1,"Help with the API"
2,"Automating email replies"

date

Use the date filter to format a string or variable as a date:

{{'now'|date('F d, Y h:ia T')}}
{{'tomorrow 5pm'|date('D, d F Y H:i T')}}
{{'+2 weeks 08:00'|date('Y-m-d h:ia T')}}

December 12, 2017 11:50am PST
Wed, 13 December 2017 17:00 PST
2017-12-26 08:00am PST

You can use any of the formatting options from PHP DateTime::format.

The second parameter to the date filter can specify a timezone to use:

{% set ts_now = 'now' -%}

Bangalore: {{ts_now|date(time_format, 'Asia/Calcutta')}}
Berlin: {{ts_now|date(time_format, 'Europe/Berlin')}}
New York: {{ts_now|date(time_format, 'America/New_York')}}

Bangalore: December 13, 2017 01:27
Berlin: December 12, 2017 20:57
New York: December 12, 2017 14:57

You can get a Unix timestamp (seconds since 1-Jan-1970 00:00:00 UTC) from a date value with the |date('U') filter:

It has been {{'now'|date('U')}} seconds since {{'0'|date(null, 'UTC')}}

It has been 1513108417 seconds since January 1, 1970 00:00

date_modify

If you need to manipulate a date, create a date object with the date function and use the date_modify filter:

{% set format = 'D, d M Y T' %}
{% set timestamp = date('now') %}
Now: {{timestamp|date(format)}}
+2 days: {{timestamp|date_modify('+2 days')|date(format)}}

Now: Tue, 12 Dec 2017 PST
+2 days: Thu, 14 Dec 2017 PST

date_pretty

Convert a Unix timestamp into a human-readable, relative date:

{% set timestamp = date("Jan 9 2002 10am", "America/Los_Angeles") %}
{{timestamp|date('U')|date_pretty}}

18 years ago

default

You can use the default filter to give a default value to empty variables:

{% set name = '' %}
Hi {{name|default('there')}}

Hi there

escape

Escape strings and variables with the following modes:

• html
• js
• css
• url
• html_attr

{{'This is "escaped" for Javascript'|escape('js')}}
{{'This is "escaped" for <b>HTML</b>'|e('html')}}

This\x20is\x20\x22escaped\x22\x20for\x20Javascript
This is "escaped" for <b>HTML</b>

filter

(Added in 10.1.1)

Exclude items from an array using an arrow function.

|filter(func)

{% set arr = [1,2,3,4,5,6,7,8] %}
{{arr|filter((v,k) => v is even)|values|join(',')}}

2,4,6,8

first

Return the first item of an array, object, or string:

{% set items = [1,2,3] %}
{{items|first}}

1

format

Insert variables into a string:

{% set who = "Kina" %}
{% set quantity = 120 %}
{{"%s closed %d tickets today!"|format(who, quantity)}}

Kina closed 120 tickets today!

hash_hmac

Generate a hash-based message authentication code (HMAC¹) using a secret key.

|hash_hmac(secret_key, algorithm, binary)

For instance, you can use this to sign parameters in a survey URL to verify that the recipient didn’t modify them.

{% set data = {'email': 'kina@cerb.example', 'survey_id': 123} %}
{{data|json_encode|hash_hmac("THIS IS SECRET","sha256")}}

5514f8aed3b39159d455f9a8f74b5d23d4f96391fa4a27d1bea6f940cb7d410f

image_info

(Added in 9.6.7)

Returns information about an image. The image may be provided as bytes or in data URI format.

|image_info()

{% set image_string %}
data:image/png;base64,iVBORw0KGgoAAAA....
{% endset %}
{{image_string|image_info|json_encode|json_pretty}}

{
    "width": 100,
    "height": 100,
    "channels": 3,
    "bits": 8,
    "type": "image/png"
}

indent

(Added in 9.6.4)

Prefix the start of each line with a given marker in a block of text.

|indent(marker, start_line)

{% set text = "Lorem ipsum dolor sit amet, consectetur adipiscing elit. Mauris eget diam 
eu orci hendrerit elementum. Suspendisse egestas, dolor at efficitur sollicitudin, magna eros 
scelerisque risus, at tincidunt massa augue a eros. Nullam scelerisque luctus suscipit. Sed 
dui metus, rhoncus sed diam non, pretium maximus augue. Phasellus feugiat justo mi, in 
tristique quam euismod pellentesque. Curabitur ut libero sagittis sem semper ultrices. Nullam 
et mi id arcu vulputate fringilla ut quis nibh. Fusce lobortis magna eu quam porta scelerisque.
Suspendisse maximus fringilla tellus, a pellentesque sem tincidunt sit amet." -%}

{{text|indent('> ')}}

> Lorem ipsum dolor sit amet, consectetur adipiscing elit. Mauris eget diam 
> eu orci hendrerit elementum. Suspendisse egestas, dolor at efficitur sollicitudin, magna eros 
> scelerisque risus, at tincidunt massa augue a eros. Nullam scelerisque luctus suscipit. Sed 
> dui metus, rhoncus sed diam non, pretium maximus augue. Phasellus feugiat justo mi, in 
> tristique quam euismod pellentesque. Curabitur ut libero sagittis sem semper ultrices. Nullam 
> et mi id arcu vulputate fringilla ut quis nibh. Fusce lobortis magna eu quam porta scelerisque.
> Suspendisse maximus fringilla tellus, a pellentesque sem tincidunt sit amet.

join

Convert an array to a string with delimiters:

{% set items = [1,2,3] %}
{{items|join(',')}}
{{items|join(' ')}}

1,2,3
1 2 3

json_encode

You can encode any variable as a JSON string with the json_encode filter:

{% set json = {'name': 'Joe Customer'} %}
{% set json = dict_set(json, 'order_id', 54321) %}
{% set json = dict_set(json, 'status.text', 'shipped') %}
{% set json = dict_set(json, 'status.tracking_id', 'Z1F238') %}
{{json|json_encode}}	

{"name":"Joe Customer","order_id":54321,"status":{"text":"shipped","tracking_id":"Z1F238"}}	

json_pretty

You can “prettify” a JSON string with the json_pretty filter:

{% set json = {'name': 'Joe Customer'} %}
{% set json = dict_set(json, 'order_id', 54321) %}
{% set json = dict_set(json, 'status.text', 'shipped') %}
{% set json = dict_set(json, 'status.tracking_id', 'Z1F238') %}
{{json|json_encode|json_pretty}}

{
  "name": "Joe Customer",
  "order_id": 54321,
  "status": {
    "text": "shipped",
    "tracking_id": "Z1F238"
  }
}

kata_encode

Emit an object/array as a KATA text block:

{% set object = {
	colors: ["red","green","blue"],
	size: 100,
} %}
{{object|kata_encode}}

colors@list:
  red
  green
  blue
size: 100

keys

Return the keys of an array or object:

{% set list = ['red','green','blue'] %}
{% set obj = { 'name': 'Kina', 'age': 35, 'title': 'Customer Support Supervisor'} %}

{{list|keys|join(',')}}
{{obj|keys|json_encode}}

0,1,2

["name","age","title"]

last

Return the last item of an array, object, or string:

{% set items = [1,2,3] %}
{{items|last}}

3

length

Return the length of a string or array:

{{"This is a string"|length}}
{{[1,2,3,4,5]|length}}

16
5

lower

Convert a string to lowercase:

{{"WHY ARE YOU YELLING?"|lower}}

why are you yelling?

map

(Added in 10.1.1)

Apply a function to each item in an array to create a new array.

|map(func)

{% set samples = [
	[1,2,3,4,5],
	[6,7,8,9,10],
	[1,3,5,7,9],
	[2,4,6,8,10],
] %}
Averages:
{{samples|map((v,k) => array_sum(v)/(samples[k]|length))|join(', ')}}

Averages:
3, 8, 5, 6

markdown_to_html

(Added in 9.5.4)

Convert Markdown² formatting to HTML:

{% set markdown %}
This is **bold** text with a [link](https://cerb.ai/)
{% endset %}
{{markdown|markdown_to_html}}

<p>This is <strong>bold</strong> text with a <a href="https://cerb.ai/">link</a></p>

md5

Generate an MD5³ hash for a string:

{{"You can verify this hash"|md5}}

1c20552e3bae1c4711cf697137002581

merge

Combine two arrays or objects:

{% set mfgs = ['Tesla','Ford'] %}
{% set mfgs = mfgs|merge(['Toyota','GM']) %}
{{mfgs|json_encode}}

["Tesla","Ford","Toyota","GM"]

nl2br

Convert newline characters (\n) to HTML breaks (<br />):

{% set text = "This has
line feeds
in the text
"%}
{{text|nl2br}}

This has<br />
line feeds<br />
in the text<br />

number_format

Format a number with thousand separators and decimal places:

{% set cost = 16858 %}
That will be ${{cost|number_format(2,'.',',')}}

That will be $16,858.00

parse_emails

Parse a delimited string of email addresses into an object. This also assists with email validation.

{% set emails = "kina@cerb.example, milo@cerb.example, karl" %}
{{emails|parse_emails|json_encode|json_pretty}}

{
    "kina@cerb.example": {
        "full_email": "kina@cerb.example",
        "email": "kina@cerb.example",
        "mailbox": "kina",
        "host": "cerb.example",
        "personal": null
    },
    "milo@cerb.example": {
        "full_email": "milo@cerb.example",
        "email": "milo@cerb.example",
        "mailbox": "milo",
        "host": "cerb.example",
        "personal": null
    },
    "karl@localhost": {
        "full_email": "karl@localhost",
        "email": "karl@localhost",
        "mailbox": "karl",
        "host": "localhost",
        "personal": null
    }
}

parse_url

Parse a URL string into an object for validation.

{% set url = "https://cerb.ai/search?q=oauth2#fragment" %}
{{url|parse_url|json_encode|json_pretty}}

{
    "scheme": "https",
    "host": "cerb.ai",
    "path": "/search",
    "query": "q=oauth2",
    "fragment": "fragment"
}

permalink

(Added in 9.2.3)

{% set text = "This is the title of a record!" %}
{{text|permalink|lower}}

this-is-the-title-of-a-record

quote

{% set text = "This is a message you are replying to.

You should quote it.
" %}
{{text|quote}}

> This is a message you are replying to.
>
> You should quote it.

reduce

(Added in 10.1.1)

Reduce an array of items into a single output value.

|reduce(func,initial)

{% set samples = [
	[1,2,3,4,5],
	[6,7,8,9,10],
	[1,3,5,7,9],
	[2,4,6,8,10],
] %}
Sum:
{{samples|reduce((carry,v) => carry + array_sum(v))}}

Sum:
110

regexp

You can use regular expressions⁴ with the regexp filter to match or extract patterns.

|regexp(pattern,group)

• pattern The regular expression pattern to match.
• group: The matching group () from the pattern to extract as a string.

Example:

{% set text = "Your Amazon Order #Z-1234-5678-9 has shipped!" %}
{% set order_id = text|regexp("/Amazon Order #([A-Z0-9\-]+)/", 1) %}
Amazon Order #: {{order_id}}

Amazon Order #: Z-1234-5678-9

If you need to escape characters in your regexp pattern, you should use a set block rather than a string:

{% set pattern %}
#\[.*?\] (.*)#
{% endset %}
{% set bracketed_text = "[ABC-123-45678] Order Processing - 7 Days" %}
{{bracketed_text|regexp(pattern, 1)}}

Order Processing - 7 Days

repeat

(Added in 10.2.3)

Repeat a string a given number of times.

|repeat(times)

{{"*"|repeat(5)}}

*****

replace

{{"I really like %food%"|replace({'%food%':'ice cream'})}}

I really like ice cream

reverse

Reverse a string or array:

{{"Leonardo da Vinci"|reverse}}
{{[1,2,3,4,5]|reverse|join}}

icniV ad odranoeL
54321

round

Round a number with desired precision.

|round(precision,method)

• precision The number of floating point digits.
• method:
  • common
  • ceil
  • floor

{% set pi = 3.141592653589793238462643383279502884197169399375105820974944592307816406286 %}
{{pi|round}}
{{pi|round(5)}}
{{pi|round(5,'ceil')}}

3
3.14159
3.1416

secs_pretty

{{"300"|secs_pretty}}
{{"86400"|secs_pretty}}
{{"604800"|secs_pretty()}}

5 mins
1 day
1 week

sha1

Generate an SHA-1⁵ hash for a string:

{{"You can verify this hash"|sha1}}

50ae61a375994fd178cd47fc7d29f7ec5724dda3

slice

Extract part of a string, array, or object.

|slice(start, length, preserve_keys)

{{[1,2,3,4,5]|slice(2,2)|json_encode}}
{{"This is some text"|slice(0,4)}}

[3,4]
This

sort

Sort an array:

{% set x = [9,5,1,6,4,3] %}
{{x|sort|slice(0,6)|json_encode}}

[1,3,4,5,6,9]

split

Convert a string to an array with the given delimiter.

|split(delimiter, limit)

{{"1,2,3,4,5"|split(',')|json_encode}}

["1","2","3","4","5"]

split_crlf

Split a string on any combination of carriage return (\r) and linefeed (\n) delimiters.

{% set rainbow = "red
orange
yellow
green
blue
indigo
violet" %}
{{rainbow|split_crlf|json_encode}}

["red","orange","yellow","green","blue","indigo","violet"]

split_csv

Split a string on comma delimiters. This automatically handles whitespace padding.

{% set coins = "BTC,   ETH   ,LTC" %}
{{coins|split_csv|json_encode}}

["BTC","ETH","LTC"]

str_pos

(Added in 10.1.2)

Return the position of a substring (needle) within a larger text (haystack). This returns -1 if the substring is not found.

|str_pos(needle, offset, ignoreCase)

{% set alphabet %}
ABCDEFGHIJKLMNOPQRSTUVWXYZ
{% endset %}
{{alphabet|str_pos(needle='hi', offset=0, ignoreCase=true)}}

7

str_sub

(Added in 10.1.2)

Extract a substring from a larger string using starting and ending positions. This is an alternative to |slice(from,length).

|str_sub(from, to)

{% set alphabet %}
ABCDEFGHIJKLMNOPQRSTUVWXYZ
{% endset %}
{{alphabet|str_sub(7,9)}}

HI

strip_lines

Remove lines in a text block that begin with one of the given prefixes.

|strip_lines(prefixes)

{% set email_message %}
> This is some quoted text
> on multiple lines

This is the original message
{% endset %}
{{email_message|strip_lines(prefixes='>')}}

This is the original message

striptags

Remove HTML tags from a string.

{% set html = "This <b>string</b> has <b>HTML</b> tags!" %}
{{html|striptags}}

This string has HTML tags!

title

{% set book_title = "the ultimate bot builder handbook" %}
{{book_title|title}}

The Ultimate Bot Builder Handbook

tokenize

Return an array of word tokens from a text block. This ignores punctuation and returns tokens in the order they appear, including duplicates.

|tokenize

{% set message %}
KATA ("Key Annotated Tree of Attributes") is a human-friendly format for modeling structured 
data that is used throughout Cerb to describe configurations, customizations, sheets, and 
automations. KATA was inspired by YAML but avoids many of its pitfalls.
{% endset %}
{{array_count_values(message|tokenize)|sort|reverse|json_encode|json_pretty}}

{
    "kata": 2,
    "of": 2,
    "is": 2,
    "that": 1,
    "data": 1,
    "structured": 1,
    "modeling": 1,
    "for": 1,
    "format": 1,
    "friendly": 1,
    "human": 1,
    "a": 1,
    "attributes": 1,
    "tree": 1,
    "annotated": 1,
    "used": 1,
    "key": 1,
    "throughout": 1,
    "to": 1,
    "its": 1,
    "many": 1,
    "avoids": 1,
    "but": 1,
    "yaml": 1,
    "by": 1,
    "cerb": 1,
    "inspired": 1,
    "automations": 1,
    "and": 1,
    "sheets": 1,
    "customizations": 1,
    "configurations": 1,
    "describe": 1,
    "was": 1,
    "pitfalls": 1
}

trim

Remove leading and/or trailing whitespace from a string.

|trim(character_mask, side)

• character_mask The characters to remove
• side
  • both
  • left
  • right

{% set str = "    whitespace    " %}
{{str|trim}}
{{str|trim(' ', 'left')}}
{{str|trim(' ', side='right')}}

whitespace
whitespace    
    whitespace

truncate

Ensure that a string is no longer than the given limit.

|truncate(limit)

{% set str = "This string is longer than we'd prefer" %}
{{str|truncate(11)}}

This string...

unescape

Decode HTML entities:

{{""iPhone" is © Apple, Inc."|unescape}}

"iPhone" is © Apple, Inc.

upper

Convert a string to uppercase:

{{"I can't hear you!"|upper}}

I CAN'T HEAR YOU!

url_decode

Decode a URL query string into an array:

{% set query = "name=Kina&action=light_on" %}
{{query|url_decode('json')}}

{"name":"Kina","action":"light_on"}

url_encode

Build a URL query string from an array:

{% set args = {"name": "Kina", "action": "light_on" } %}
{{args|url_encode}}

name=Kina&action=light_on

values

(Added in 10.1.1)

Return the values of an array with sequential keys. This is the filter equivalent of the array_values() function.

{% set countries = {
  'CA': 'Canada',
  'CN': 'China',
  'DE': 'Germany',
  'IN': 'India',
  'MX': 'Mexico',
  'US': 'United States',
} %}
{{countries|values|json_encode}}

["Canada","China","Germany","India","Mexico","United States"]

References