These functions are available in bot scripts and snippets:

• array_column
• array_combine
• array_count_values
• array_diff
• array_extract_keys
• array_fill_keys
• array_intersect
• array_matches
• array_sort_keys
• array_unique
• array_values
• attribute
• cerb_automation
• cerb_avatar_image
• cerb_avatar_url
• cerb_calendar_get_relative_date
• cerb_calendar_time_elapsed
• cerb_current_worker
• cerb_extract_uris
• cerb_file_url
• cerb_has_priv
• cerb_placeholders_list
• cerb_plugin_enabled
• cerb_record_readable
• cerb_record_writeable
• cerb_url
• cerb_workflow_config
• cerb_workflow_resources
• clamp_float
• clamp_int
• cycle
• date
• date_lerp
• dict_set
• dict_unset
• dns_get_record
• dns_host_by_ip
• json_decode
• jsonpath_set
• kata_parse
• max
• min
• random
• random_string
• range
• regexp_match_all
• shuffle
• validate_email
• validate_number
• vobject_parse
• xml_attr
• xml_attrs
• xml_decode
• xml_encode
• xml_xpath
• xml_xpath_ns
• xml_xpath_remove
• References

array_column

(Added in 9.2.1)

The array_column function extracts a column from the elements of an array:

{% set people = [
	{"id": 1, "name": "Kina Halpue", "email": "kina@cerb.example"},
	{"id": 2, "name": "Milo Dade", "email": "milo@cerb.example"},
	{"id": 3, "name": "Janey Youve", "email": "janey@cerb.example"},
] %}
The email addresses are: {{array_column(people,'email')|join(', ')}}

The email addresses are: kina@cerb.example, milo@cerb.example, janey@cerb.example

array_combine

(Added in 9.0)

The array_combine function creates a new array with the given keys and values:

{% set keys = ['name', 'age', 'email'] %}
{% set values = ['Janey Youve', '30-ish', 'janey@cerb.example'] %}
{% set person = array_combine(keys, values) %}
{{person.name}} can be reached at {{person.email}}

Janey Youve can be reached at janey@cerb.example

array_count_values

(Added in 10.0.4)

The array_count_values function takes an array of values as input, and returns an array with distinct values as keys and their count of occurrences. This function only works on arrays of strings or numbers.

{% set values = [1,2,3,1,3,2,3,1,2,1,3,1,3] %}
{{array_count_values(values)|json_encode|json_pretty}}

{
    "1": 5,
    "2": 3,
    "3": 5
}

array_diff

The array_diff function returns the items in the second array that are not present in the first array:

{% set arr1 = ['Apple', 'Google', 'Microsoft'] %}
{% set arr2 = ['Apple', 'Microsoft', 'Cerb'] %}
{% set diff = array_diff(arr2, arr1) %}
These are new: {{diff|join(', ')}}

These are new: Cerb

array_extract_keys

(Added in 9.6.4)

Returns the given keys from all elements of a list.

{% set records = [
	{
		id: 1,
		subject: "Help with the API",
		status: "open",
		sender: "customer@cerb.example",
	},
	{
		id: 2,
		subject: "Automating email replies",
		status: "open",
		sender: "customer@cerb.example",
	}
] %}
Sender,Subject,Status
{{array_extract_keys(records, ['sender','subject','status'])|csv}}

Sender,Subject,Status
customer@cerb.example,"Help with the API",open
customer@cerb.example,"Automating email replies",open

array_fill_keys

(Added in 10.0)

Create an array with the given keys, each set to the default value.

array_fill_keys(keys,value)

{{array_fill_keys(range(1,10),true)|json_encode}}

{"1":true,"2":true,"3":true,"4":true,"5":true,"6":true,"7":true,"8":true,"9":true,"10":true}

array_intersect

(Added in 9.0)

Returns a new array for all the elements in array1 that are also present in array2. This is the opposite of array_diff.

{% set arr1 = ['Apple', 'Google', 'Microsoft'] %}
{% set arr2 = ['Apple', 'Microsoft', 'Cerb'] %}
{% set intersect = array_intersect(arr2, arr1) %}
These are in both: {{intersect|join(', ')}}

These are in both: Apple, Microsoft

array_matches

(Added in 9.6.4)

Compares an array of values to an array of patterns.

{% set recipients = ['support@cerb.example','sales@cerb.example'] %}
{% set patterns = ['sales@*'] %}
{% set results = array_matches(recipients, patterns) %}
Matches: {{results|join(', ')}}

Matches: sales@cerb.example

array_sort_keys

(Added in 9.0)

Sort an associative array by its keys rather than its values.

{% set arr = {"z":"A", "a":"B", "m":"C"} %}
{% set arr = array_sort_keys(arr) %}
{{arr|keys|join(',')}}

a,m,z

array_unique

(Added in 9.0)

Return a new array with only the distinct values from the array argument.

{% set arr = [1,1,2,2,3,3,4,4,5,5,6] %}
Unique values {{array_unique(arr)|join(',')}}

Unique values 1,2,3,4,5,6

array_values

(Added in 9.0)

Return the values from an associative array as a new indexed array. For instance, this can affect the output in JSON encoding by using [] rather than {key:value}.

{% set arr = {"z":"A", "a":"B", "m":"C"} %}
{{array_values(arr)|json_encode}}

["A","B","C"]

attribute

Access the values of an object with a variable key:

{% set person = {"first_name": "Kina", "last_name": "Halpue", "title": "Customer Support Supervisor"} %}
{% set key = 'title' %}
{{attribute(person, key)}}

Customer Support Supervisor

cerb_automation

Invoke a scripting.function automation from any feature that supports scripting.

The function returns keys for exit_state: (exit, return, error) and return: (an arbitrary dictionary).

This brings the full functionality of automations to email signatures, snippets, legacy bot behaviors, automation event bindings, toolbars bindings, etc.

For instance, a snippet could use an automation to dynamically generate content based on the target record or current worker. This solves many feature requests.

cerb_automation(uri, inputs)

{% set ip_data = cerb_automation('wgm.scripting.getLocationByIP', { ip:"1.2.3.4" } ) %}
{% if ip_data.return.data %}
I see you are contacting us from {{ip_data.return.data.country_name}}.
{% endif %}

I see you are contacting us from Australia.

cerb_avatar_image

Retrieve the avatar image for a given record type and ID.

cerb_avatar_image(record_type, id, updated)

{{cerb_avatar_image('worker','1','now'|date('U'))}}

<img src="https:/cerb.example/avatars/worker/1?v=1513212603" style="height:16px;width:16px;border-radius:16px;vertical-align:middle;">

cerb_avatar_url

Retrieve the avatar image URL for a given record type and ID.

cerb_avatar_url(record_type, id, updated)

{{cerb_avatar_url('worker','1','now'|date('U'))}}

https://cerb.example/avatars/worker/1?v=1513212702

cerb_calendar_get_relative_date

(Added in 11.0)

Calculate a future timestamp using calendar availability. For instance, this can be used for SLAs to generate a due date like “+4 business hours”.

cerb_calendar_get_relative_date(calendar,rel_date,now)

Now: {{"now"|date('r')}}
Due: {{cerb_calendar_get_relative_date(123,'+2 hours')|date('r')}}

Now: Fri, 18 Oct 2024 20:02:18 -0700
Due: Mon, 21 Oct 2024 09:00:00 -0700

cerb_calendar_time_elapsed

(Added in 10.1.1)

Calculate the time elapsed (in seconds) between two dates using calendar availability.

cerb_calendar_time_elapsed(calendar,date_from,date_to)

{{cerb_calendar_time_elapsed(123,'last Friday 5pm','now')|secs_pretty}}

18 hours, 13 mins

cerb_current_worker

(Added in 10.4.14)

Return a dictionary for the currently logged in worker. This returns an empty dictionary when used outside a browser session.

cerb_current_worker(expand)

Hello {{cerb_current_worker().first_name}}!

Hello Kina!

cerb_extract_uris

(Added in 9.5.3)

Return an array of URLs found in HTML content, along with metadata (e.g. tag, attributes, URI parts).

In the response, URLs are replaced with tokens in the template which can be modified with the |replace filter.

For instance, this function can be used to rewrite all links in an email template for click tracking.

cerb_extract_uris(html)

{% set html %}
This is some <b>HTML</b> with <a href="https://cerb.ai/">links</a>.
{% endset %}
{% set results = cerb_extract_uris(html) %}
{{results|json_encode|json_pretty}}

{
    "tokens": {
        "#uri-61411f091662a": "https://cerb.ai/"
    },
    "context": {
        "#uri-61411f091662a": {
            "is_tag": true,
            "name": "a",
            "attr": "href",
            "attrs": {
                "href": "https://cerb.ai/"
            },
            "uri_parts": {
                "scheme": "https",
                "userinfo": null,
                "host": "cerb.ai",
                "port": null,
                "path": "/",
                "query": null,
                "fragment": null
            }
        }
    },
    "template": "This is some <b>HTML</b> with <a href=\"#uri-61411f091662a\">links</a>.\n"
}

To rewrite links:

{% set html %}
This is some <b>HTML</b> with <a href="https://cerb.ai/">links</a>.
{% endset %}
{% set results = cerb_extract_uris(html) %}
{% set new_urls = results.tokens|map(
  (url,token) => "https://proxy.example/click?url=" ~ url|url_encode
)%}
{{results.template|replace(new_urls)}}

This is some <b>HTML</b> with <a href="https://proxy.example/click?url=https%3A%2F%2Fcerb.ai%2F">links</a>.

cerb_file_url

Retrieve the download link for a given attachment ID.

This automatically adapts to use within Cerb and community portals (e.g. SSL, proxies).

cerb_file_url(id)

{{cerb_file_url('1')}}

https://cerb.example/files/1/original_message.html

cerb_has_priv

(Added in 9.0)

Returns a boolean depending on whether the given actor has the given privilege among their roles. If no actor is given, the current worker is assumed. This allows bot functionality, snippets, and widgets, to adapt based on worker permissions. This is particularly useful in HTML-based profile widgets.

{% if cerb_has_priv('contexts.cerberusweb.context.ticket.create', 'worker', 1) %}
Worker #1 has permission to create tickets.
{% endif %}

Worker #1 has permission to create tickets.

cerb_placeholders_list

Return an object with every placeholder in the current behavior.

cerb_placeholders_list(extract, prefix)

{{cerb_placeholders_list()|json_encode|json_pretty}}

{
  "worker__context": "cerberusweb.contexts.worker",
  "worker__loaded": true,
  "worker__label": "Kina Halpue",
  "worker__image_url": "https://cerb.example/avatars/worker/1?v=1512582324",
  "worker_at_mention_name": "Kina",
  "worker_calendar_id": 7,
  "worker_dob": null,
  "worker_id": 1,
  "worker_first_name": "Kina",
  "worker_full_name": "Kina Halpue",
  "worker_gender": "F",
  "worker_is_disabled": 0,
  "worker_is_superuser": 1,
  "worker_language": "en_US",
  "worker_last_name": "Halpue",
  "worker_location": "",
  "worker_mobile": "15555555555",
  "worker_phone": "",
  "worker_time_format": "D, d M Y h:i a",
  "worker_timezone": "America/Los_Angeles",
  "worker_title": "Customer Support",
  "worker_updated": 1512582324,
  "worker_record_url": "https://cerb.example/profiles/worker/1-Kina-Halpue",
  ...
}

cerb_plugin_enabled

(Added in 11.0)

Test if a Cerb plugin is installed and enabled.

For instance, this can be used to make dashboard tabs or widgets conditional on a particular plugin being enabled (e.g. project boards).

cerb_plugin_enabled(plugin_id)

{{cerb_plugin_enabled('cerb.classifiers')}}

1

cerb_record_readable

(Added in 9.0)

Returns a boolean if the given actor has read access to the given record. If no actor is provided then the current worker is assumed. This allows bots and widgets to adapt based on record permissions. For instance, an HTML widget on a profile dashboard could only show a button to workers who can modify the record.

{% if cerb_record_readable('ticket', 123, 'worker', 1) %}
Worker #1 can read ticket #123.
{% endif %}

Worker #1 can read ticket #123.

cerb_record_writeable

(Added in 9.0)

Returns a boolean if the given actor has write access to the given record. If no actor is provided then the current worker is assumed. This allows bots and widgets to adapt based on record permissions. For instance, an HTML widget on a profile dashboard could only show a button to workers who can modify the record.

{% if cerb_record_writeable('ticket', 123, 'worker', 1) %}
Worker #1 can modify ticket #123.
{% endif %}

Worker #1 can modify ticket #123.

cerb_url

Retrieve a full URL to a page or resource in Cerb.

This automatically adapts to use within Cerb and community portals (e.g. SSL, proxies).

{{cerb_url("c=profiles&type=ticket&id=5")}}

https://cerb.example/profiles/ticket/5

cerb_workflow_config

(Added in 11.0)

Perform runtime configuration lookups from any feature that supports automation scripting (e.g. automations, workflows, snippets). For instance, you can create a workflow just for sharing values (e.g. API keys) between multiple workflows.

cerb_workflow_config(name_or_id,key,default)

{{cerb_workflow_config('example.workflow','secretCode',null)}}

sup3rs3cr3t

cerb_workflow_resources

(Added in 11.0)

Perform runtime resource lookups and return a map of workflow resources and their local record IDs. This is useful from automations, event listeners, and toolbars.

cerb_workflow_resources(name_or_id)

{{cerb_workflow_resources('example.workflow'|json_encode}}

{"records":{"automation/example":123}}

clamp_float

Set the range boundaries for a decimal value.

{{clamp_float(-105.19,0,100)}}

0

clamp_int

Set the range boundaries for an integer value.

{{clamp_int(110,-90,90)}}

90

cycle

Round-robin through a sequence.

{% set options = ['odd','even'] %}
{% for n in 1..10 %}
* {{cycle(options, n)}}
{% endfor %}

* even
* odd
* even
* odd
* even
* odd
* even
* odd
* even
* odd

date

Create a date object for use with the date_modify filter.

{% set d = date('1-Jan-2018 10:00am') %}
{{d|date_modify('+2 hours')|date('F d, Y g:ia')}}

January 01, 2018 12:00pm

date_lerp

Interpolate the timestamps between two dates with the given unit and step.

date_lerp(date_range,unit,step,limit)

Arguments:

Returns: An array of Unix timestamps.

{{date_lerp('this month',unit='day',step=5)|map((v) => v|date('r'))|json_encode|json_pretty}}

[
    "Sat, 01 Oct 2022 00:00:00 -0700",
    "Thu, 06 Oct 2022 00:00:00 -0700",
    "Tue, 11 Oct 2022 00:00:00 -0700",
    "Sun, 16 Oct 2022 00:00:00 -0700",
    "Fri, 21 Oct 2022 00:00:00 -0700",
    "Wed, 26 Oct 2022 00:00:00 -0700",
    "Mon, 31 Oct 2022 00:00:00 -0700"
]

dict_set

You can use the dict_set function to quickly add, modify, or append items in an array or object.

dict_set(object,path,value,delimiter) : object

Arguments:

Returns: The function returns a modified version of object.

You can set deeply nested keys in a single line using dot-notation:

{% set var = {"group": {}} %}
{% set var = dict_set(var, 'group.name', 'Support') %}
{% set var = dict_set(var, 'group.manager.name.first', 'Kina') %}
{% set var = dict_set(var, 'group.manager.name.last', 'Halpue') %}
{{var|json_encode|json_pretty}}

{
  "group": {
    "name": "Support",
    "manager": {
      "name": {
        "first": "Kina",
        "last": "Halpue"
      }
    }
  }
}

Append items to an array by adding .[] to the key:

{% set var = {"group": {}} %}
{% set var = dict_set(var, 'group.name', 'Support') %}
{% set var = dict_set(var, 'group.members.[]', 'Kina Halpue') %}
{% set var = dict_set(var, 'group.members.[]', 'William Portcullis') %}
{% set var = dict_set(var, 'group.members.[]', 'Steven Emplois') %}
{{var|json_encode|json_pretty}}

{
  "group": {
    "name": "Support",
    "members": [
      "Kina Halpue",
      "William Portcullis",
      "Steven Emplois"
    ]
  }
}

Append to nested arrays:

{% set var = [1,2,[3,4,[5,6]]] %}
{% set var = dict_set(var, '2.2.[]', 7) %}
{% set var = dict_set(var, '2.2.[]', 8) %}
{% set var = dict_set(var, '2.3', 9) %}
{{var|json_encode|json_pretty}}

[
  1,
  2,
  [
    3,
    4,
    [
      5,
      6,
      7,
      8
    ],
    9
  ]
]

dict_unset

You can use the dict_unset function to remove items by key from an array or object.

You can unset deeply nested keys in a single line using dot-notation:

{% set person = {"person":{"name":{"first":"Jane","last":"Tester"},"age":28,"location":"Secret"}} %}
{% set person = dict_unset(person, ['person.name.last','person.age','person.location']) %}
{{person|json_encode|json_pretty}}

{
    "person": {
        "name": {
            "first": "Jane"
        }
    }
}

dns_get_record

(Added in 10.2.6)

Resolve DNS records by hostname and type. This enables workflows like verifying domain ownership via TXT records, validating SPF/DKIM, verifying MX servers, etc.

dns_get_record(hostname,type)

• hostname: The lookup hostname.
• type: The record type (a, aaaa, caa, cname, mx, ns, ptr, soa, srv, txt)

{{dns_get_record('cerb.ai','a')|json_encode|json_pretty}}

[
    {
        "host": "cerb.ai",
        "class": "IN",
        "ttl": 77,
        "type": "A",
        "ip": "54.192.81.51"
    },
    {
        "host": "cerb.ai",
        "class": "IN",
        "ttl": 77,
        "type": "A",
        "ip": "54.192.81.69"
    }
]

dns_host_by_ip

(Added in 10.2.6)

Resolve a hostname from an IP. If a name can’t be resolved for a valid IP, the IP is returned. If an invalid IP is provided, the result is an empty string.

dns_host_by_ip(ip)

• ip: The IP address to reverse lookup a hostname.

{{dns_host_by_ip('54.148.127.4')}}

cerb.email

json_decode

You can decode a JSON-encoded string with the json_decode function:

{% set json_string = "{\"name\":\"Joe Customer\",\"order_id\":12345}" %}
{% set json = json_decode(json_string) %}
Customer: {{json.name}}
Order #: {{json.order_id}}	

Customer: Joe Customer
Order #: 12345

This returns an object.

jsonpath_set

This is nearly identical to dict_set.

{% set json_string = "{\"name\":\"Joe Customer\",\"order_id\":12345}" %}
{% set json = json_decode(json_string) %}
{% set json = jsonpath_set(json, 'order_id', '67890') %}
{{json.order_id}}

67890

You can specify an array by appending [] without a leading dot (.):

{% set json_string = "{\"team\":{\"groups\":[]}}" %}
{% set json = json_decode(json_string) %}
{% set json = jsonpath_set(json, 'team.groups[]', 'Support') %}
{% set json = jsonpath_set(json, 'team.groups[]', 'Sales') %}
{% set json = jsonpath_set(json, 'team.groups[]', 'Development') %}
{{json|json_encode|json_pretty}}

{
  "team": {
    "groups": [
      "Support",
      "Sales",
      "Development"
    ]
  }
}

kata_parse

Parses a KATA text block into an object.

{% set kata %}
colors@list:
  red
  green
  blue
size@int: 100
{% endset %}
{{kata_parse(kata)|json_encode|json_pretty}}

{
    "colors@list": "red\ngreen\nblue",
    "size@int": "100"
}

max

Return the largest value in an array or object.

{% set numbers = [1,9,8,4,2] %}
{{max(numbers)}}

9

min

Return the smallest value in an array or object.

{% set numbers = [1,9,8,4,2] %}
{{min(numbers)}}

1

random

Return a random item from a string or array, or a random number between 0 and the given number (inclusive).

{{random([1,2,3,4,5,6,7,8,9,0])}}
{{random("abcdefghijklmnopqrstuvwxyz")}}
{{random(20)}}

9
o
17

random_string

Generate a random string of the given length. This is useful for generating confirmation codes or temporary passwords.

{{random_string(16)}}

61AE3XG3ZMW8QDTM

range

Return an array with values between from and to (inclusive).

range(from,to,step)

{{range(5,15)|json_encode}}
{{range(5,15,2)|json_encode}}

[5,6,7,8,9,10,11,12,13,14,15]
[5,7,9,11,13,15]

regexp_match_all

regexp_match_all(pattern, string, group)

{% set headers = 
"X-Mailer: Cerb
From: customer@cerb.example
To: support@cerb.example
"
%}
{% set results = regexp_match_all("#^(.*?): (.*?)$#m", headers) %}
{{results|json_encode|json_pretty}}

[
  [
    "X-Mailer: Cerb",
    "From: customer@cerb.example",
    "To: support@cerb.example"
  ],
  [
    "X-Mailer",
    "From",
    "To"
  ],
  [
    "Cerb",
    "customer@cerb.example",
    "support@cerb.example"
  ]
]

shuffle

Randomize an array:

{{shuffle([1,2,3,4,5])|json_encode}}

[2,4,5,1,3]

validate_email

Validate an email address:

{{validate_email('kina')|json_encode}}
{{validate_email('kina#cerb.example')|json_encode}}
{{validate_email('kina@cerb.example')|json_encode}}

false
false
true

validate_number

Validate a number:

{{validate_number('abcde')|json_encode}}
{{validate_number('20.f')|json_encode}}
{{validate_number(10)|json_encode}}
{{validate_number('123.45')|json_encode}}

false
false
true
true

vobject_parse

Parse a block of text in VObject format (e.g. vCard, iCal).

vobject_parse(text)

Arguments:

Returns: An object with properties and parameters.

{% set vcard %}
begin:vcard
source:ldap://cn=Meister%20Berger,o=Universitaet%20Goerlitz,c=DE
name:Meister Berger
fn:Meister Berger
n:Berger;Meister
bday;value=date:1963-09-21
o:Universit=E6t G=F6rlitz
title:Mayor
title;language=de;value=text:Burgermeister
note:The Mayor of the great city of
  Goerlitz in the great country of Germany.
email;internet:mb@goerlitz.de
home.tel;type=fax,voice,msg:+49 3581 123456
home.label:Hufenshlagel 1234\n
 02828 Goerlitz\n
 Deutschland
end:vcard
{% endset %}
{{vobject_parse(vcard)|json_encode|json_pretty}}

{
    "VCARD": [
        {
            "props": {
                "SOURCE": [
                    {
                        "params": [],
                        "value": "ldap://cn=Meister%20Berger,o=Universitaet%20Goerlitz,c=DE"
                    }
                ],
                "NAME": [
                    {
                        "params": [],
                        "value": "Meister Berger"
                    }
                ],
                "FN": [
                    {
                        "params": [],
                        "value": "Meister Berger"
                    }
                ],
                "N": [
                    {
                        "params": [],
                        "value": "Berger;Meister"
                    }
                ],
                "BDAY": [
                    {
                        "params": {
                            "value": "date"
                        },
                        "value": "1963-09-21"
                    }
                ],
                "O": [
                    {
                        "params": [],
                        "value": "Universit=E6t G=F6rlitz"
                    }
                ],
                "TITLE": [
                    {
                        "params": [],
                        "value": "Mayor"
                    },
                    {
                        "params": {
                            "language": "de",
                            "value": "text"
                        },
                        "value": "Burgermeister"
                    }
                ],
                "NOTE": [
                    {
                        "params": [],
                        "value": "The Mayor of the great city of Goerlitz in the great country of Germany."
                    }
                ],
                "EMAIL": [
                    {
                        "params": {
                            "internet": ""
                        },
                        "value": "mb@goerlitz.de"
                    }
                ],
                "HOME.TEL": [
                    {
                        "params": {
                            "type": "fax,voice,msg"
                        },
                        "value": "+49 3581 123456"
                    }
                ],
                "HOME.LABEL": [
                    {
                        "params": [],
                        "value": "Hufenshlagel 1234\n02828 Goerlitz\nDeutschland"
                    }
                ]
            }
        }
    ]
}

xml_attr

Return a single attribute from an XML node.

xml_attr(xml_node, attr)

Arguments:

Returns: A string from the given XML attribute, or false.

{% set xml_string %}
<?xml version = "1.0" encoding = "UTF-8"?>
<Movies>
    <Movie rating="R">
        <Title runtime="142">The Shawshank Redemption</Title>
        <Genre>Drama</Genre>
        <Director>
            <Name highratedmovie="The Mist">
                <First>Frank</First>
                <Last>Darabont</Last>
            </Name>
        </Director>
        <Studio>Columbia Pictures</Studio>
        <Year>1994</Year>
    </Movie>
</Movies>
{% endset %}
{% set xml = xml_decode(xml_string) %}
{% set movie = xml_xpath(xml, '//Movie')|first %}
{% set runtime = xml_attr(movie.Title,'runtime') %}
The runtime of {{movie.Title}} is {{runtime ? (60*runtime)|secs_pretty : 'unknown'}}.

The runtime of The Shawshank Redemption is 2 hours, 22 mins.

xml_attrs

Return all attributes from an XML node.

xml_attrs(xml_node)

Arguments:

Returns: An array of attribute keys and values.

{% set xml_string %}
<?xml version = "1.0" encoding = "UTF-8"?>
<Movies>
    <Movie rating="R">
        <Title runtime="177">The Godfather</Title>
        <Genre> Crime Drama </Genre>
        <Director>
            <Name>
                <First>Francis Ford</First>
                <Last>Coppola</Last>
            </Name>
        </Director>
        <Studio>Paramount Pictures</Studio>
        <Year>1972</Year>
    </Movie>
    <Movie rating= "R">
        <Title runtime="142">The Shawshank Redemption</Title>
        <Genre>Drama</Genre>
        <Director>
            <Name highratedmovie="The Mist">
                <First>Frank</First>
                <Last>Darabont</Last>
            </Name>
        </Director>
        <Studio>Columbia Pictures</Studio>
        <Year>1994</Year>
    </Movie>
</Movies>
{% endset %}
{% set xml = xml_decode(xml_string) %}
{% set movies = xml_xpath(xml, '//Movie') %}
{{xml_attrs(movies[1])|json_encode|json_pretty}}

{
    "rating": "R"
}

xml_decode

You can decode an XML¹ string into an XML object with the xml_decode function.

Use the xml_xpath function to extract values with XPath² queries.

xml_decode(xml_string,namespaces,mode)

• xml_string: The string of XML to convert into an object.
• namespaces: An optional array of namespaces.
• mode: Use html to convert an HTML DOM into an XML document.

{% set string_of_xml = 
"<response>
  <client_id>1</client_id>
  <invoice_id>123</invoice_id>
</response>"
-%}
{% set xml = xml_decode(string_of_xml) %}
{{xml_encode(xml)}}

<?xml version="1.0"?>
<response>
  <client_id>1</client_id>
  <invoice_id>123</invoice_id>
</response>

xml_encode

You can encode an object as XML with the xml_encode function:

{% set string_of_xml = 
"<response xmlns=\"http://www.example.com/api/\">
  <client_id>1</client_id>
  <invoice_id>123</invoice_id>
</response>"
-%}
{% set xml = xml_decode(string_of_xml) %}
{{xml_encode(xml.client_id)}}	

<client_id>1</client_id>

xml_xpath

Use the xml_xpath function to extract values with XPath² queries:

{% set string_of_xml = 
"<response>
  <client_id>1</client_id>
  <invoice_id>123</invoice_id>
</response>"
-%}
{% set xml = xml_decode(string_of_xml) %}
{% set client_id = xml_xpath(xml, '//client_id')|first %}
{% set invoice_id = xml_xpath(xml, '//invoice_id')|first %}
Client ID: {{client_id}}
Invoice ID: {{invoice_id}}

Client ID: 1
Invoice ID: 123

xml_xpath_ns

You can define an XML namespace with the xml_xpath_ns function:

{% set string_of_xml = 
"<response xmlns=\"http://www.example.com/api/\">
  <client_id>1</client_id>
  <invoice_id>123</invoice_id>
</response>"
-%}
{% set xml = xml_decode(string_of_xml) %}
{% set xml = xml_xpath_ns(xml, 'ns', 'http://www.example.com/api/') %}
{% set client_id = xml_xpath(xml, '//ns:client_id')|first %}
{% set invoice_id = xml_xpath(xml, '//ns:invoice_id')|first %}
Client ID: {{client_id}}
Invoice ID: {{invoice_id}}

Client ID: 1
Invoice ID: 123

xml_xpath_remove

Remove elements from an XML document with an XPath query.

xml_xpath_remove(xml,path)

• xml: An XML object created by xml_decode.
• path: The XPath query to match elements for removal.

{% set string_of_xml =
"<response>
  <client_id>1</client_id>
  <invoice_id>123</invoice_id>
</response>"
-%}
{% set xml = xml_decode(string_of_xml) %}
{% set xml = xml_xpath_remove(xml, '//invoice_id') %}
{{xml_encode(xml)}}

<?xml version="1.0"?>
<response>
  <client_id>1</client_id>
</response>

References