This document is a guide to the various types of parameter widgets available, and the syntax for creating them.
String types
Parameters that have a string value should be of the phylum string
Default values
Default values for strings can be specified by either an attribute:
default="hello"
or for selection (see below), by specifying default="true" for a specific value:
<value default="true">choice 1</value>
Freeform strings
Freeform strings do not need to specify any other specific format or type information.
<parameter phylum="string" cardinality="single" key="result"
friendly-name="Result Name" />
String selection
To select a particular string from a list, the format should be specified as select, and a set of values should be included.
<parameter phylum="string" format="select"
cardinality="single" key="test" friendly-name="Test" >
<values>
<value>choice1</value>
<value>choice2</value>
<value>choice3</value>
</values>
</parameter>
Extension point string selection
To select an extension point to be returned as a string (selected by name, value returned by tag), use format extension-point, and an extension element must be specified:
<parameter phylum="string" format="extension-point" cardinality="single" key="extpt" friendly-name="Ext pt"> <extension point="ncsa.gis.ui.initFileActions"/> </parameter>
Multiple String selection
String lists can be defined using cardinality="multiple"
Choices can be set in the same method as string selection.
<parameter group="Required" phylum="string" format="select" cardinality="multiple" key="list" friendly-name="Field Select">
<dependent-param>bridgeDamage</dependent-param>
<values source="parameter" sourcekey="bridgeDamage" sourcetype="fields"/>
</parameter>
Fields from a dataset string selection
You can also use the string selection to choose between fields/columns of a dataset:
<parameter phylum="string" format="select" cardinality="single" key="dislocationZoneFieldId" friendly-name="Field to use as preferred zone Id" default="tract"> <dependent-param>buildingPopulationDislocation</dependent-param> <values source="parameter" sourcekey="buildingPopulationDislocation" sourcetype="fields" /> </parameter>
The dependent-param element indicates which parameter will hold the dataset that you want to choose fields from. The value element must also specify that parameter in the sourcekey attribute. Setting sourcetype to fields indicates that you want to use the fields from the dataset.
String maps
Any sort of data that, at it's lowest level, is a map of Strings to Strings should use the stringmap phylum.
Keys based on a separate dataset parameter
To create a string map widget that has keys based off of certain records or fields from a separate dataset parameter, use format parameter-keys.
Also, the dependent-param element must be specified to indicate a field that this field is dependent on.
Keys based on fields of a dataset, values one of a set of hardcoded choices.
<parameter phylum="dataset" format="shapefile"
cardinality="single" key="bridges" friendly-name="Bridges">
<parameter phylum="stringmap" format="parameter-keys" cardinality="single" key="map" friendly-name="Agg Type">
<dependant-param>bridges</dependant-param>
<keys source="parameter" sourcekey="bridges" sourcetype="fields"/>
<values source="values">
<value default="true">min</value>
<value>max</value>
<value>sum</value>
</values>
</parameter>
Boolean values
All boolean values should be specific by phylum boolean.
Checkboxes
Checkboxes are the default widget for boolean values.
<parameter phylum="boolean" cardinality="single" key="choice"
friendly-name="Check Me" />
Dataset values
Parameters should use the phylum dataset for dataset values.
Specified by schema type
To select a dataset specified by one or more schema type, you can omit a format,
and just specify the types desired.
<parameter phylum="dataset"
cardinality="single" key="bridges" friendly-name="Bridges">
<types>
<type>bridge</type>
</types>
</parameter>
By format of dataset (feature dataset, fragility dataset, etc)
In this case, types should be omitted, and the format should be specified.
The format should match the string returned by getDatasetFormat() in the various dataset types.
Important note: Feature datasets use the key shapefile, whether or not they are backed by a shapefile.
<parameter phylum="dataset" format="mapping"
cardinality="single" key="mapping" friendly-name="Fragility Mapping">
</parameter>
By feature dataset's geometry type
In this case, types should be omitted, and the format should be specified as shapefile.
Instead of types, geometry-types should be specified.
<parameter phylum="dataset" format="shapefile"
cardinality="single" key="polygon" friendly-name="Polygon ">
<geometry-types>
<geometry-type>Polygon</geometry-type>
</geometry-types>
</parameter>
Multiple datasets
Any of these dataset types can be set to use multiple datasets by setting cardinality to multiple.
<parameter cardinality="multiple"
phylum="dataset" key="hazard" friendly-name="Hazard">
<types>
<type>hazardRaster</type>
<type>deterministicHazardRaster</type>
</types>
</parameter>
View Options
Groups
Parameters can be combined into related groups by adding a group attribute.
<parameter group="Extra Info" phylum="string" key="myExtraParam" friendly-name="Extra stuff"/>
Hiding groups
Groups can be set to only appear when a certain field has a certain value:
<group-show-when group="Socioeconomic Disruption Metrics" node="objectiveOne" value="Include"/>
Where node is the parameter to depend on, and value is the value of the field that indicates this parameter should be shown. You can also be shown on every value EXCEPT a specific value, by adding a ! in front:
value="!Not Used"
Hiding parameters based on selection
Any parameter can by dynamically hidden or shown based on the selection of another parameter. Just add the show-when tags to the parameter definition:
<parameter group="Optimization Method Parameters" phylum="string" key="dmRetrofitMethod" friendly-name="Direct Method Retrofit Method"> <show-when parameter='optimizationMethod' value='Direct Method'/> </parameter>
This will show the dmRetrofitMethod parameter only when the parameter optimizationMethod has been set to Direct Method
Flexible map-based parameters
To enable more flexibility where tasks don't have to have getters and setters for each parameter, you can specify that a parameter be added "as a map" using as-map="true". This will add it to a special parameter called parameterInputMap, which is a map of the keys of any parameters specified as as-map, mapped to their values. This way tasks can have one setter (setParameterInputMap(Map<String,Object>), which is defined in SimpleFeatureTask for convenience), and any number of previously unspecified parameters can be added to this.
Example:
<parameter group="Liquefaction" as-map="true" phylum="dataset" key="liquefaction.groundwatermap" friendly-name="Groundwater map" >
<types>
<type>groundWaterDepthMap</type>
</types>
</parameter>