Node properties¶
A node’s properties are defined by the defaults
object in its html definition. These are the properties that get passed to the node constructor function when an instance of the node is created in the runtime.
In the example from the creating your first node section, the node had a single property called name
. In this section, we’ll add a new property called prefix
to the node:
Add a new entry to the defaults
object:
defaults: {
name: {value:""},
prefix: {value:""}
},
The entry includes the default value
to be used when a new node of this type is dragged onto the workspace.
Add an entry to the edit template for the node:
<div class="form-row">
<label for="node-input-prefix"><i class="fa fa-tag"></i> Prefix</label>
<input type="text" id="node-input-prefix">
</div>
The template should contain an <input>
element with an id
set to node-input-<propertyname>
.
Use the property in the node:
Depending on the programming language the property can either be found in the info Struct: nodeInfo["info"]["prefix"]
(all programming languages except JavaScript). Or accessed directly: config.prefix
(JavaScript).
Property definitions¶
The entries in the defaults
object must be objects and can have the following attributes:
value
: (any type) the default value the property takesrequired
: (boolean) optional whether the property is required. If set to true, the property will be invalid if its value is null or an empty string.validate
: (function) optional a function that can be used to validate the value of the property.type
: (string) optional if this property is a pointer to a configuration node, this identifies the type of the node.
Reserved property names¶
There are some reserved names for properties that must not be used. These are:
- Any single character -
x
,y
,z
,d
,g
,l
are already used. Others are reserved for future use. id
,type
,wires
,inputs
,outputs
If a node wants to allow the number of inputs or outputs it provides to be configurable then inputs
or outputs
may be included in the defaults
array. The Function node is an example of how this works.
Property validation¶
The editor attempts to validate all properties to warn the user if invalid values have been given.
The required
attribute can be used to indicate a property must be non-null and non-blank.
If more specific validation is required, the validate
attribute can be used to provide a function that will check the value is valid. The function is passed the value and should return either true or false. It is called within the context of the node which means this
can be used to access other properties of the node. This allows the validation to depend on other property values. While editing a node the this
object reflects the current configuration of the node and not the current form element value. The validator function should try to access the property configuration element and take the this
object as a fallback to achieve the right user experience.
There is a group of common validation functions provided.
RED.validators.number()
- check the value is a numberRED.validators.regex(re)
- check the value matches the provided regular expression
Both methods - required
attribute and validate
attribute - are reflected by the UI in the same way. The missing configuration marker on the node is triggered and the corresponding input is red surrounded when a value is not valid or missing.
The following example shows how each of these validators can be applied.
defaults: {
minimumLength: { value:0, validate:RED.validators.number() },
lowerCaseOnly: {value:"", validate:RED.validators.regex(/[a-z]+/) },
custom: { value:"", validate:function(v) {
var minimumLength=$("#node-input-minimumLength").length?$("#node-input-minimumLength").val():this.minimumLength;
return v.length > minimumLength
} }
},
Note how the custom
property is only valid if its length is greater than the current value of the minimumLength
property or the value of the minimumLength
form element.
Property edit dialog¶
When the edit dialog is opened, the editor populates the dialog with the edit template for the node.
For each of the properties in the defaults
array, it looks for an <input>
element with an id
set to node-input-<propertyname>
, or node-config-input-<propertyname>
in the case of Configuration nodes. This input is then automatically populated with the current value of the property. When the edit dialog is closed, the property takes whatever value is in the input.
More information about the edit dialog is available here.
Custom edit behavior¶
The default behavior works in many cases, but sometimes it is necessary to define some node-specific behavior. For example, if a property cannot be properly edited as a simple <input>
or <select>
, or if the edit dialog content itself needs to have certain behaviors based on what options are selected.
A node definition can include two functions to customize the edit behavior.
oneditprepare
is called immediately before the dialog is displayed.oneditsave
is called when the edit dialog is okayed.oneditcancel
is called when the edit dialog is canceled.oneditdelete
is called when the delete button in a configuration node’s edit dialog is pressed.oneditresize
is called when the edit dialog is resized.
For example, when the Inject node is configured to repeat, it stores the configuration as a cron-like string: 1,2 * * * *
. The node defines an oneditprepare
function that can parse that string and present a more user-friendly UI. It also has an oneditsave
function that compiles the options chosen by the user back into the corresponding cron string.