The data (process variables, node inputs, node outputs, rule inputs, data store entities, or constants) used by Appian must conform to certain data types.
Appian data types can be one of the system types or a custom data type built from an XML Schema Definition (XSD), Java object, or imported from a WSDL by the Call Web Service smart service.
See Also: Conversion Functions
The Any Type is a generic data type only available for use as an expression input for rules and certain expression functions, such as the if() function. It accepts data of any data type.
Data is stored in variables of this type by mapping an existing variable, rule, constant, or expression to its value.
The map data type is for associative arrays that store data in key value pairs (e.g. a!map(key1: "value1", key2: "value2").
A map can be returned by a number of functions and objects or created using the function a!map(). The map type can also be selected as a variable type in process models.
Inside of a map, each key has an associated value and that value's type is preserved (i.e. the value is not variant-wrapped).
A system data type is a required format for data stored in Appian and includes primitive types and complex types. Each system data type can be used to store either a single value or a list of values.
The following primitive system data types are available.
Values include True / False.
The default value is null, which appears as [Empty Value] in a process variable. The output of an empty boolean value in an expression depends on the function used.
You can populate the value using results from the true() or false() functions.
The toboolean() function can be used to convert true/false and 0/1 values into a Boolean data type value.
See also: Casting
Data in a date format can be created using the date(year, month, date) function. Variables that have a date data type do not accept text string input.
The default value for a date is [Empty Value]. The minimum value is 1/1/1000, and maximum value is 12/31/9999. Dates values are not adjusted to a different time zone when saved or displayed.
Numerical and text values can be converted to a date value using the todate() function.
Variables that hold a Date and Time data type refer to a point in time that is the same for all users. Date and time variables do not accept text strings as input.
A Date and Time value is saved in Greenwich Mean Time (GMT) then converted to the end-user's time zone (accounting for daylight saving time) when displayed.
GMT is the default time zone used when evaluating expressions that include Date and Time values. You can specify a different time zone in your process model properties. The time zone used for display can be the user's preferred time zone, a globally specified time zone, or the time zone context.
If a function expects a Date and Time value, and you pass it a date, the date is automatically converted to a Date and Time value, using 12:00 am for the time component.
The Date and Time value is only converted to the end user's time zone when it is displayed in the following manner. (Use separate date values and time values if you do not want this conversion to take place.)
See also Time Zone Context
Date and Time format data can be created using the datetime(year, month, date, hour, minute, second) function.
Numerical and text values can be converted to a Date and Time value using the todatetime() function.
See also: Appian Functions, Internationalization Settings
This type is used to store an Encrypted Text value. An Encrypted Text value can be created in one of the following ways:
EncryptionService public Java APIAn Encrypted Text value is only decrypted when displayed as the value in an EncryptedTextField or within a plug-in using the EncryptionService public API.
A value of type Encrypted Text cannot be cast to any other type and no other type can be cast to it.
An encrypted value is larger than the corresponding plaintext value. Specifically, the value stored in memory or on disk will be the lowest multiple of 16 bytes that is greater than the size of the corresponding plaintext value in bytes.
An Encrypted Text value remains encrypted when stored on disk. The encryption key is unique to each installation.
The data type does not provide any additional access controls. Encrypted text entered by one user may be decrypted and displayed to another user if that other user has permission to view the interface in which the value is displayed.
See also: Encrypted Text Component
Holds numeric data stored as double precision floating-point decimal numbers. The default value is 0.0.
Decimal numbers can be created from text strings using the todecimal() function. When converted with this function, numbers less than -10,000,000 and greater than 10,000,000 are represented in scientific notation.
For forms, use one of the following Text Functions to display a decimal number as a default value to avoid rounding.
Also for forms, use the a!currency() function to localize a currency value based on a given ISO currency code.
See also: Text Functions
For PVs, if you enter a number that exceeds the maximum number of digits supported by double precision floating-points, the number is truncated down to the maximum number of digits when you save the process model. It does not provide you with a warning message if this occurs.
Integer numbers can range from -2,147,483,647 to 2,147,483,647 (or from -231+1 to 231-1 in scientific notation).
The default value is 0 and the null value is -2\^31.
Integer numbers can be created from text strings using the tointeger() function.
When an arithmetic operation (such as an expression) creates a Number (Integer) value that exceeds the type's limits, the value wraps.
2147483647 + 10 = -2,147,483,6392147483647 + 1 = -2147483648 - interpreted as null. When looking in the user interface at a process variable changed to this value, an [Empty Value] result is displayed.When values that exceed the Number(Integer) range are passed through a user interface, the excessive value is changed.
When a value that exceeds the Number(Integer) range is converted to Number(Integer) from a string in an engine server (such as when you convert text to an integer using the tointeger() function), the excessive value is changed to the maximum value.
On task forms, the Number Form Component prevents entry of values outside of the valid range.
2147483647 displays as 2147483000.For PVs, if you enter a value that exceeds the Number(Integer) range, the integer will be replaced with a null value when you save the model. The Modeler does not provide you with a warning message if this occurs.
This type is used to store any UTF-8 text string. Numerical values can be entered into the text data type; however, data manipulation cannot be performed on the text data type (except for report aggregations). The default value is [Empty Value].
To display text, enclose it within double quotation marks ("").
Time data can be created using the time(hour,minute,second) function. Variables that have a time data type do not accept text string input.
Time values are not adjusted to a different time zone when saved or displayed.
The following complex system types are made available in the system to support smart services.
These types cannot be edited or deleted. Their XML structure is not guaranteed to remain the same from release to release.
The ApplicationTestResult data type is designed to hold test result information for all expression rules in an application. This type also includes execution statistics for an individual application.
See also: ApplicationTestResult and Automated Testing for Expression Rules
The DataSubset data type is designed to hold the data returned by a query configured with a paging parameter.
It contains the following fields:
The EntityData data type lets you define a target data store entity and the values to store in the target entity as an input value for the Write to Multiple Data Store Entities Smart Service.
It contains the following fields:
For example, a value of type EntityData where the entity and data values are stored as process variables could resemble the following:
1
a!entityData(entity: pv!ENTITY_OPPORTUNITIES, data: {pv!RadiationOpp, pv!NewBusinessOpp})
See also: Data Store Entity Data Type and Write to Multiple Data Store Entities Smart Service
The EntityDataIdentifiers data type lets you define a target data store entity and the values to delete from the target entity as an input value for the Delete from Data Store Entities Smart Service.
It contains the following fields:
For example, a value of type EntityDataIdentifiers where the entity and data values are stored as process variables could resemble the following:
1
a!entityDataIdentifiers(entity: pv!ENTITY_OPPORTUNITIES, identifiers: {pv!RadiationOpp.id, pv!NewBusinessOpp.id})
Note: Make sure to use the primary key value IDs for the data rather than CDT values (for example, pv!opportunities.id rather than pv!opportunities). Using a CDT value will result in a casting error.
See also: Data Store Entity Data Type and Delete from Data Store Entities Smart Service
The HealthCheckOutput data type is designed to hold the details of the latest Health Check run.
It contains the following fields:
See also: a!latestHealthCheck()
The IntegrationError data type contains details on errors that occur when calling a web service or other integration via an Integration Rule.
It contains the following fields:
This data type is returned in any integration result if the integration returns an error. The response of the IntegrationError is dependent on information from the web service. For instance, suppose an API request returns a 400 - Bad Request error. The integration error may return a result like this:
1
2
3
title: 400 - Bad Request
message: The server cannot process the request due to an invalid request syntax
details: The request is missing a query parameter for dateRange to identify the start and end date to use in this query
In addition to the default response for an integration, the IntegrationError type can also be returned using the function a!integrationError().
The LabelValue data type is designed to hold event labels and a nested hierarchy of subevent labels for use by the Post Event to Feed Smart Service and the Post System Event to Feed Smart Service.
It contains the following fields:
The LabelValueTable data type references the LabelValue type.
It contains the following field:
Data type used to define the record list view for record types.
It contains the following fields:
See also: a!listViewItem()
Data type produced when defining record user filters using expressions. See FacetOption for list options and configuration details.
See also: a!recordFilterList() and a!recordFilterDateRange()
Defines a list of options that a user can select from in a record user filter.
See also: a!recordFilterListOption()
The ObjectTestResult data type is designed to hold data for each of the expression rules with test cases.
See also: ObjectTestResult and Automated Testing for Expression Rules
The PagingInfo data type is designed to hold the paging configuration passed as a parameter to queries and the Read-Only Grid component.
It's used primarily as an argument for the todatasubset() and a!gridField() functions.
To create a value of type PagingInfo, use the a!pagingInfo() function.
See also: todatasubset() and a!pagingInfo()
The ProcessInfo data type is designed to hold information about a running process. It contains three fields.
See also: Process Model Properties, Process Variables, Start Process smart service.
The Query data type defines the grouping, aggregation, filtering, paging, and sorting configuration to be applied when querying using a!queryEntity(). It also contains several supporting data types for each of these configurations.
Note: Process variables cannot be created as a Query data type.
It contains the following fields:
See also: a!query()
Data type accepted in the selection|aggregation field of the Query data type. It can contain one or more Column data types and should be used instead of an Aggregation data type when you just want to select the columns, rather than group them together or apply an aggregation function.
See also: a!querySelection()
The Column data type is only used in conjunction with the Selection data type.
It contains the following fields:
true.See also: a!queryColumn()
Data type accepted in the selection|aggregation field of the Query data type. It can contain one or more AggregationColumn data types and should be used instead of a Selection data type when you want to perform a function on the selected columns. The following aggregation functions are supported: COUNT, SUM, AVG, MIN, and MAX.
See also: a!queryAggregation()
The AggregationColumn data type is only used in conjunction with the Aggregation data type.
It contains the following fields:
true.false.false.See also: a!queryAggregationColumn()
Data type that determines the filtration to apply.
It contains the following fields:
AND and OR.See also: a!queryLogicalExpression()
This data type is required to configure the filter options for a!pickerFieldRecords, a!query, a!queryRecordType, a!recordData, and a!recordFilterListOption.
It contains the following fields:
=, <>, >, >=, <, <=, between, in, not in, is null, not null, starts with, not starts with, ends with, not ends with, includes, not includes.is null or not null. If the operator value is between, the value must be a list of only two elements with the lower bound as the first element and the upper bound as the second.See also: a!queryFilter(), a!queryRecordType(), a!recordData()
Data type that indicates a user's search term. Used in the logicalExpression|filter|search field of a Query or LogicalExpression data type.
It contains a single field:
The Save data type is designed to be used in conjunction with the a!save() function to create reusable custom components.
See also: a!save()
The SortInfo data type is referenced by the PagingInfo and DataSubset types and determines how data is sorted in a subset.
To create a value of type SortInfo, use the a!sortInfo() function.
See also: a!sortInfo(), PagingInfo, and DataSubset
The TestCaseResult data type is designed to hold data for each of the test cases in an object.
See also: TestCaseResult and Automated Testing for Expression Rules
The TestRunResult data type is designed to hold test statistics for a test run across all applications being tested.
See also: TestRunResult and Automated Testing for Expression Rules
The Writer data type is a special data type returned by expression functions that intend to modify data. The modification of data must not happen during expression evaluation, so these functions return a Writer, which is then handled in a special way during the phase of an interface evaluation where saving into variables takes place. The Writer data type has no impact during expression evaluation - no data is written by the function that returns a writer until a variable created with the bind function is saved into an interface.
It contains the following fields:
See also: bind() and Writer Functions
An Appian Object data type is a required format for objects specific to the Appian system. Similar to primitive and complex system types, each can be used to store either a single value or a list of values.
Appian Object data types are only recognizable within the Appian system.
Holds an integer ID number that represents an Appian application. It can be used as a rule input to expression rules or interfaces; it can also be used as a constant and referenced from interfaces, web APIs, and process models; and finally used as a process variable from the Process Modeler, or as inputs to the Start Rule Tests (Applications) - Smart Service. Custom data types cannot use this data type.
Stores data for each prediction made by the Classify Documents and Classify Emails smart services.
Each prediction result is stored in the Above Threshold, Below Threshold, or Failed outputs in these smart services, based on what the user enters as the confidence threshold in the smart service input. Above Threshold and Below Threshold outputs contain the document ID, class, and confidence score for each prediction. The Failed output contains the document ID and error message for each failed prediction.
See also: AI skill object
Holds an integer ID number that represents a connected system. A connected system represents an external system that is integrated with Appian.
See also: Connected System Objects
Holds an integer ID number that represents a data store entity. Data store entities are named, typed storage units within a data store. They can map to one or more tables in an external database.
It can only be applied to process variables and cannot be used to save form data from a mobile form. Data store entity IDs are not reused if the entity is deleted.
See also: Data Stores
Holds an integer ID number that represents a document in Document Management. It can be used to save form data selected from a dropdown, radio button, or checkbox field input on a mobile form.
If the ID number of a document is known, you can convert it to this data type using the todocument() function. Document IDs are not reused if the document is deleted.
See also: Document Management
Holds an integer number that represents a Document Management Community ID.
If the ID of a Community is known, you can convert it to this data type using the tocommunity() function. Community IDs are not reused if the community is deleted.
It cannot be used to save form data from a mobile form.
Holds an integer ID number that represents a document or a folder that exists within Document Management. Document or folder IDs are not reused if the object is deleted.
It cannot be used to save form data from a mobile form.
Holds data formatted as an email address. Variables that hold it do not accept text strings as direct input.
Email address data can be created from text strings using the toemailaddress() function.
Use the email recipient data type if you are sending email from a process.
It cannot be used to save form data from a mobile form.
Email address data must be converted to email recipient data for use by the Send E-Mail smart service.
This is done with the toemailrecipient() function, which accepts email address data, user data, or group data.
It cannot be used to save form data from a mobile form.
Holds an integer ID number that represents a folder that exists within Document Management. It can be used to save form data selected from a dropdown, radio button, or checkbox field input on a mobile form.
If the ID number of a folder is known, you can convert it to this data type using the tofolder() function. Folder IDs are not reused if the folder is deleted.
Holds an integer ID number that represents a group within the system.
It can be used as a record field type for record types with data sync enabled. It can also be used to save form data selected from a dropdown, radio button, or checkbox field input on a mobile form.
Tip: As a best practice, you should select Group as the record field type for fields with integer values that represent a group. This will not impact how you reference the record field, and it will allow you to use the fields in your record-level security configuration.
If the ID of a group is known, you can convert it to this data type using the togroup() function. A group ID may be reused if the group is deleted.
See also: Creating Groups
Holds an integer ID number that represents a Knowledge Center in Document Management.
If the ID number of a Knowledge Center is known, you can convert it to this data type using the toknowledgecenter() function. Knowledge Center IDs are not reused if it is deleted.
It cannot be used to save form data from a mobile form.
Holds the definition of a portal.
You can use the portal object reference domain, portal!, to call a portal as a parameter value for functions related to portals, such as a!urlForPortal().
You can't use a constant to reference a portal, instead use the portal! domain.
Holds the definition of a portal page. Portal pages are referenced using portal!portalName.pages.portalPageName. See Referencing portal pages for more information.
Holds the integer ID number of an instance of a process model.
Holds the integer ID number of a process model. Each time a process model is launched, it runs as a separate process.
A process model ID may be reused if the process model is deleted.
This data type points to an action configured on a record type, which makes it possible to reference the action's properties like the display name, key, description, icon, process model, and visibility configuration. References to this data type are used as the value of the action parameter in the a!recordActionItem() to display actions as interface components. Record actions are referenced using recordType!recordTypeName.actions.actionName.
The data type returned for the a!recordData() function. This reference is used to define the record type and filters for a grid or chart.
See a!recordData() for more information.
Holds the definition of a record field. Record field references, like those defined in the fields parameter of a!queryRecordType(), use this type to define the data that is returned in a query or displayed in a grid or chart. Record fields are referenced using recordType!recordTypeName.fields.fieldName.
Holds a record definition used to configure a link to the record in an interface such as a news post.
You can only create constants of this data type. Process variables cannot be created of this type, and custom data types cannot be saved as this type.
To create a value of type Record Identifier, use the a!toRecordIdentifier function. To create a value of type Record Identifier for a User record, use the a!userRecordIdentifier function.
See also: a!toRecordIdentifier() and a!userRecordIdentifier()
Values of type Record Identifier are used as inputs for the Post Event to Feed Smart Service and Post System Event to Feed Smart Service to specify the record tags for an event.
See also: Post Event to Feed Smart Service and Post System Event to Feed Smart Service
Holds the definition of a record type relationship. These definitions are used to reference related record types in functions like a!relatedRecordData to sort, limit, or filter a query based on the value of a record's related data. Record relationships are referenced using recordType!recordTypeName.relationships.relationshipName.
Holds the definition of a record type.
You can use the record type object reference domain, recordType!, to call a record type as a parameter value for functions related to records, such as a!queryRecordType(), queryRecordByIdentifier(), and urlforrecord().
Although you can still use a constant to reference your record type, the recordType! domain eliminates the need to create this additional object.
See urlforrecord(), Constants, a!queryRecordByIdentifier(), and a!queryRecordType() for more information.
Holds an integer ID number that represents a report within the system.
Constants of type Report are commonly used as parameter values for report links.
Holds a URI value and enforces security rules during casting.
When casting from Text, the string will fail verification if it contains any of the following:
http, https, ftp, tel, and mailto.mailto value that contains an apostrophe.The string does not change when cast to or from a Text data type.
This data type cannot be used in expressions for events or process reports.
See also: Safe Link: Link type that accepts SafeURI values to create an external link.
Holds the integer ID number of a process task.
It can be used to add a link to the Read-Only Grid component that opens a process task in Tempo.
See also: Read-Only Grid
Holds an integer ID number that represents a task report within the system.
Constants of type Task Report are commonly used as parameter values for report links.
Holds an Appian user account ID number.
It can be used as a record field type for record types with data sync enabled. It can also be used to save form data selected from a dropdown, radio button, or checkbox field input on a mobile form.
Tip: As a best practice, you should select User as the record field type for fields with text values that represent users. This will not impact how you reference the record field, and it will allow you to use the field in your record-level security configuration.
See also: User Management
Holds an Appian user account or group. It is sometimes referred to as a People data type.
It can be used to save form data selected from a dropdown, radio button, or checkbox field input on a mobile form.
Holds the definition of an end-user filter configured for a record type. A user filter can be applied by users viewing the record list or a records-powered grid as a convenient way to limit the list of records. User filters are referenced using recordType!recordTypeName.filters.filterName. Learn how to create user filters.
This data type holds the data definition for all of the fields in a record type (or a subset of fields defined by your record type configuration), as well as the data values for the record.
When you make changes to the record type object, the record data type is also updated and stays current with any configuration changes you make. For example, if you change the name of a record field on a record type, the change is also captured on the record data type.
Note: Renaming record fields is available for record types with data sync enabled.
The record data type allows you to easily pass record data to your interfaces, records-powered components, and expression rules.
Learn more about using records in your apps:
Designers can create or import their own custom data types (CDTs). These organize data into a structure that represents a logical grouping of related data, such as Employee and Contract.
For more information about creating and editing data types, see Custom Data Types (CDTs)
Careful attention must be paid to data types when passing the values from one variable to another (also called mapping data).
The variables used at the node level are called node inputs and node outputs. A node is often referred to as an activity. A node input or output can also be called an Activity Class. These variables can be mapped into process variables, enabling Process Modelers to access the variables in other nodes within the process model and pass the data to other processes and subprocesses. The following properties must be taken into consideration when mapping node inputs/outputs to process variables.
Appian does not support a direct mapping of data from one type into another, regardless of whether the data values are compatible. You can use an expression to cast a variable from one type into another and then save the result into a new variable.
Appian allows you to map variables that only store single values into variables that can store multiple values using a custom output on the output tab, but not when using results listed on the output tab. Results must match according to data type and whether the data type holds multiple values (its cardinality).
Mapping scalars (single values) to vectors (multiple values) is only applicable when mapping process variables. Therefore, vector values cannot be set as default values in a scalar variable. For example, a variable cannot be given the default value {1,6,9,8} if the variable does not support multiple values.
Data types also apply to rule inputs and constants, but data from a node input or a process variable cannot be mapped to a rule input or a constant.
See also: Casting
Appian Data Types
