Everything within the C3 AI Suite is managed through or exists as a C3 Type. Types allow users to describe, process, and interact with data and analytics. Broadly speaking, C3 Types are like Java classes, and include 'fields', 'methods', and an inheritance structure. Once you understand the structure of C3 Types you will be able to write Types to load data, run analytics, build and deploy machine learning models, and configure application logic for your research project. 

All C3 Types are defined in a .c3typ file, stored in 'src' directory of a package. A .c3typ file can only define a single C3 Type. The .c3typ file name must match the name of the C3 Type being defined. 

In this section we will describe the package to download for this tutorial then we will discuss the syntax, special keywords, fields, methods, and inheritance structure of C3 Types. Finally, we will review some examples.

lightbulbAD Package

In this tutorial we will use the lightbulbAD package. To download the source code for this package, please follow the instructions available here: Guide to download C3 lightbulbAD Package.

Syntax

 To help familiarize yourself with the syntax for a C3 Type, let's look at how the 'SmartBulb' Type is defined in the lightbulbAD package:

/*
 * Copyright 2009-2020 C3 (www.c3.ai). All Rights Reserved.
 * This material, including without limitation any software, is the confidential trade secret and proprietary
 * information of C3 and its licensors. Reproduction, use and/or distribution of this material in any form is
 * strictly prohibited except as set forth in a written license agreement with C3 and/or its authorized distributors.
 * This material may be covered by one or more patents or pending patent applications.
 */

/**
 * A single light bulb capable of measuring various properties, including power consumption, light output, etc.
 */
entity type SmartBulb extends LightBulb mixes MetricEvaluatable, NLEvaluatable type key "SMRT_BLB" {

  /**
   * This bulb's historical measurements.
   */
  bulbMeasurements: [SmartBulbMeasurementSeries](smartBulb)

   /**
    * This bulb's historical predictions.
    */
   @db(order='descending(timestamp)')
   bulbPredictions: [SmartBulbPrediction](smartBulb)

   /**
    * This bulb's latest prediction.
    */
   currentPrediction: SmartBulbPrediction stored calc "bulbPredictions[0]"

   /**
    * This bulb's historical events.
    */
   bulbEvents: [SmartBulbEvent](smartBulb)

   /**
    * The latitude of this bulb.
    */
   latitude: double

   /**
    * The longitude of this bulb.
    */
   longitude: double

   /**
    * The unit of measurement used for this bulb's light output measurements.
    */
   lumensUOM: Unit

   /**
    * The unit of measurement used for this bulb's power consumption measurements.
    */
   powerUOM: Unit

   /**
    * The unit of measurement used for this bulb's temperature measurements.
    */
   temperatureUOM: Unit

   /**
    * The unit of measurement used for this bulb's voltage measurements.
    */
   voltageUOM: Unit

   /**
    * A SmartBulb is associated to a {@link Fixture} through a SmartBulbToFixtureRelation.
    */
   @db(order='descending(start), descending(end)')
   fixtureHistory: [SmartBulbToFixtureRelation](from)

   /**
    * The current Fixture to which this bulb is attached.
    */
   currentFixture: Fixture stored calc 'fixtureHistory[0].(end == null).to'

   /**
    * Method to determine the expected lumens of a light bulb
    */
   expectedLumens: function(wattage: !decimal, bulbType: !string): double js server

   /**
    * Returns the life span of this smartBulb
    */
   lifeSpanInYears: function(bulbId: string): double js server

   /**
    * Returns the average life span of all smartBulbs.
    */
   averageLifeSpan: function(): double js server

   /**
    * Returns the id of the smart bulb with the shortest recorded life span to date.
    */
   shortestLifeSpanBulb: function(): string js server

   /**
    * Returns the id of the smart bulb with the longest recorded life span to date.
    */
   longestLifeSpanBulb: function(): string js server

}

At a glance, a .c3typ file has the following components: 

At a high level, the basic syntax to define C3 Types is as follows:

[remix, extendable] [entity] type TypeName extends, mixes AnotherType {
    /* comments */
    [field declaration]
    [method declaration]
}

Note: Everything within square brackets '[]' is optional.

Additional Resources

Keywords

A C3 Type definition is introduced with a series of keywords. These keywords tell the C3 AI Suite how to construct this Type, store it internally, and whether it inherits fields and methods from other already defined C3 Types.

Additional Resources

C3 Types 

There are many categories of C3 Types. 

We will cover the most commonly used categories in this tutorial:

Primitive Types

Like many programming languages, the C3 AI Suite has primitives. Primitives define the units (or data types) of fields on a C3 Type (e.g., "int", "double", "float"). The C3 AI Suite includes a number of primitives, listed below for reference. Adding new primitives will require support from the C3 AI engineering team, however, most DTI researchers should be able to progress their research projects with the C3 AI Suite's existing primitives. 

C3 Also supports certain composites of primitive types. For example, the C3 Type Mapp  is a map between values and keys. We can create a map directly in python in two ways:

  1. Let C3 Figure out the type for us: c3.MappObj(value=python_obj) The new Mapp object can be accessed through the .value  property of the MappObj
  2. Tell C3 exactly which types are supported with c3.Mapp(c3.MappType.fromString("map<string,any>"), {'a': 5, 'b': [1,2,3]})

Similarly, C3 has the Arry  type which supports plain arrays. We can specify them in the same ways.

  1. Let C3 Figure out the type for us: c3.ArryObj(value=python_list)  The new Arry  object can be accessed through the .value  property of the ArryObj
  2. Tell C3 exactly which types are supported with c3.Arry(c3.ArryType.fromString('[int]'), [1,2,3,4]) 

In both cases, these primitive types aren't persisted. They can be used anywhere a C3 Type specifies it must have an Obj  type. (Especially useful for DynBatchJobs  or DynMapReduce jobs for contexts).

Additional Resources

Persistable Types

The most common C3 Type you will define is a 'Persistable' Type. Persistable Types store data in a C3 AI Suite database. By default, all Persistable Types are stored in Postgres though they can also be stored in file systems (e.g., Azure Blob) or key-value stores (e.g., Cassandra), by adding an annotation to your .c3typ file (discussed below). 

When defining a Persistable type, you MUST add the following two key terms to your .c3typ file:

  1. entity: The keyword 'entity' keyword tells the C3 AI Suite this is a 'Persistable' Type.
  2. schema name: The schema names is the name of the database table where the C3 Type's data are stored. When defining a new Persistable Type, you need to add the keywords 'schema name' followed by your chosen table name, in your .c3typ file. (e.g., in the C3 Type LightBulb, we have 'schema name "LGHT_BLB"').
    Note: Extended C3 Types DO NOT NEED a scheme name and Schema names CANNOT exceed 30 characters. 

Additional Resources

Generic (Parametric) Types

Like a Java or C++ Class, C3 Types can be parameterized (or genericized). In fact, the C3 AI Suite uses the exact same syntax as Java and C++ to define a Type's parameters (i.e., angle brackets '<>' ). When defining a Generic (Parametric) Type, your Type name will be followed by angle brackets and a comma-separated list of parameters (usually other C3 Types).  For example:

type TheType<V,U> {
	fieldA: V
	fieldB: [U]
}

Then, when using your Parametric Type in other places, you must specify the arguments for each parameter in the angle brackets. As example, you may define a new C3 Type with the following field:

type NewType {
	newField: TheType<TypeA, TypeB>
}

If your C3 Types will be heavily re-used by other developers you should consider using Parametric (Generic) Types.

Additional Resources

Fields

As discussed above, a C3 Type mainly consists of two components: fields and methods.

Fields are attributes or properties of a C3 Type. To define a field on C3 Type, use the following syntax: field name, followed by a colon ':', followed by a ValueType (e.g., "lumens: int"). By convention, field names are camelCase.

A C3 Type can have many different kinds of fields. Here are the four most commonly used fields:

Primitive Fields

Like many other programming languages the C3 AI Suite has primitive fields (e.g., int, boolean, byte, double, datetime, string, longstring). For example:

doubleField: double
intField: int

Primitive fields are stored in a particular C3 Type's database table.

Reference Fields

Reference fields point (or refer) to other C3 Types. Reference fields link (or join) two C3 Types together. Under the covers, the Reference field stores a pointer (i.e., foreign key reference or ID) to record of another C3 Type. To define a Reference field, use the following syntax: field name, followed by colon ':', followed by Type Name (e.g., "building: Building"). For example:

refField: AnotherType

Collection Fields

Collection fields contain a list of values or records. There are four categories of collections fields:

Collection fields can also be used to model one-to-many and many-to-many relationships between two C3 Types. To do so, we use the Foreign Key Collection Notation, which specifies the foreign key by which two C3 Types (i.e., ThisType and AnotherType) are joined.

To define this annotation use the following syntax: "fieldName :   `[AnotherType]` (fkey, key)" where:

For example:

fieldName: [AnotherType] (fkey[, key])

Shown in the example code and diagram below, the field bulbMeasurements, contains a list of all the SmartBulbMeasurementSeries records, where SmartBulb.id == SmartBulbMeasurements.smartBulb (e.g., a list of various measurements relevant to a SmartBulb, like temperature or power). 

Please see the 'Foreign Key Collection and Schema Names" section in the following C3 AI Develop Documentation for more details: https://developer.c3.ai/docs/7.12.25/topic/mda-fields

Calculated Fields

Calculated fields are derived from other fields on a C3 Type. Calculated fields typically included either JavaScript code or C3 AI ExpressionEngineFunctions and are specified as follows:

fieldName: FieldType [stored] calc "field_formula"

There are two types of calculated fields:

  1. Simple calculated fields: The values of these fields are calculated at runtime and not stored. To define a simple calculated field use the following syntax: keyword `calc`, followed by an expression formula.
  2. Stored calculated fields: The values of these fields are stored in the C3 Type's database table. To define a stored calculated field use the following syntax: keywords `stored calc`, followed by an expression formula (e.g, in the SmartBulb example above, 'stored calc fixtureHistory[0].(end == null).to').

Additional Resources

Methods

At a high-level, methods are pieces of code that take in arguments or parameters and return new values. To define a method on a C3 Type, use the following syntax, method name followed by a colon ':' followed by a function signature, (e.g., function() or function(wattage: !decimal, bulbType: !string)), followed by a colon ':', followed by a return parameter, along with an implementation language and execution location. 

Function Signatures

Here's the syntax for a method's function signature:

function(parName1: [!] parType1, parName2: [!] parType2):

We see the keyword 'function' followed by a comma-separated list of input parameters. To define an input parameter, use the following syntax: argument name followed by a colon ':', followed by argument type (e.g., bulbID: string). Optionally, you can also add an exclamation point '!' before the argument type to indicate the argument is required.

The function itself is NOT implemented in the .c3typ file, rather in a separate .js (JavaScript), .py (Python) or .r (R) files stored in the same directory as the .c3typ file. 

Return Parameter

Following the function signature is a return parameter. Return parameters must be a C3 Type. 

Implementation Language/Execution Location

Finally, let's discuss the implementation language and execution location.

Language Specifications

Methods can be implemented in JavaScript, Python, or R:

Currently, these are the only three programming languages natively supported by the C3 AI Suite to define methods. 

Execution Location

Methods can be executed in three locations:        

Please note, R and Python methods can only be executed on the server. Javascript methods can be executed in either the server or client.

Additionally, for Python, we need to specify a valid 'ActionRuntime' for python function to be executed in. ActionRuntimes are defined in C3 AI Suite and new ActionRuntimes can be defined in a package and are essentially conda environments. For example, 'server' will use the 'py-server' ActionRuntime.

Function Definitions

As mentioned above, the function itself is defined in a different file from the .c3typ file. The name of this file should be the same as the .c3typ file where the method is defined (e.g., SmartBulb.js and SmartBulb.c3typ), but have a different extension, per the function's implementation language (e.g., 'py' for Python, 'js' for JavaScript, 'r' for R). The function names in this file must exactly match the method names in the .c3typ file. For example, we define a C3 Type as such: 

entity type NewType {
	field: double
	funcA: function(num: !int) : float js server
    funcB: function(num: !int, name: string) : double py server
}

This C3 Type will be defined in a file called 'NewType.c3typ'. We also need two other files, 'NewType.js' and 'NewType.py', in the same directory as 'NewType.c3typ'. Here's a peek inside:

'NewType.js'

function funcA(num) {
    ...
}

'NewType.py'

function funcB(num, name='default') {
    ...
}

Additional Resources

Inheritance

Like a Java or C++ Class, a C3 Type can inherit fields and methods from other C3 Types. The keywords`mixes` and `extends` signal inheritance. These keywords follow the new C3 Type's name in your .c3typ file.

These two keywords signal different kinds of Type inheritance.

Mix-ins

Mix-ins are the most common type of inheritance and signaled by the 'mixes' keyword. When a C3 Type 'mixes' another C3 Type, it inherits all the fields and methods of the original C3 Type. In general, mix-ins are used to incorporate non-persistable Types. The new C3 Type is stored in a new table and has its own schema. 

type ParentType {
    fieldA: int
    funcA: function(): ReturnType
}

type ChildType mixes ParentType {
    fieldB: double
    funcB: function(): ReturnType
}

In this example, ChildType mixes ParentType, and inherits 'fieldA' and 'funcA' defined on ParentType.

Persistable Inheritance

Persistable inheritance is signaled by the 'extends' keyword. When a C3 Type 'extends' another C3 Type, it inherits all the original C3 Type's fields and methods. The extension and original C3 Types must be marked with 'entity' keyword and share the same database table. To distinguish data from the original and extension C3 Types, the C3 AI Suite adds a 'type key' field to this database table. Additionally, only C3 Types marked with the `extendable` keyword can be extended. 

extendable entity type ParentType schema name "PRT_TYP" {
    fieldA: int
    funcA: function(): ReturnType
}

entity type ChildType extends ParentType type key "CHLD" {
    fieldB: double
    funcB: function(): ReturnType
}

In this example, ParentType is an extendable Persistable type stored in the table 'PRT_TYP'. ChildType extends ParentType and is also stored in the table 'PRT_TYP'.  Records associated with ChildType have 'type key' == 'CHLD' to distinguish them from ParentType's data.

Additional Resources

Annotations

With annotations you can further configure a C3 Type. Annotations can appear before Type definitions, field declarations, and method declarations. As example, annotations can be used to specify the database technology used to store a C3 Type (e.g., Cassandra, Azure Blob), mark the ActionRuntime in which to run a particular method, deprecate a field or method, specify how time-series data should be treated, and specify how to trigger analytics or DFEs (DataFlow Events).

Here are a few examples of defining annotations: 

  /**
   * This bulb's historical predictions.
   */
  @db(order='descending(timestamp)')
  bulbPredictions: [SmartBulbPrediction](smartBulb)

With the `@db(order = 'descending(timestamp)'` annotation, the bulbPrediction field's records are stored in descending order by timestamp.

@ts(treatment='avg')
lumens: double

With the `@ts(treatment = 'avg') annotation, the value of the lumens field for a given SmartBulb is the average of all that Smartbulb's lumens data. 

Additional Resources

Final Types

Types, methods, and fields can be made final using the 'final' keyword. The fields and methods of a Final Type cannot be modified or overridden (via 'mixes' or 'exends').

type FinalType {
    final method: function(): string
}

Additional Resources

Abstract Types

The C3 AI Suite also supports abstract types. To define an Abstract Type, add the keyword 'abstract' before the Type definition. Abstract Types cannot be instantiated and must mix-in to another concrete Type. This provides a great way to standardize interfaces. Here's an example:

abstract type Interface {
	field1: int
    method1: function(double): int
}

Additional Resources

Examples

To see more examples of C3 Types, check out the .c3typ files in the lightbulbAD package: Guide to download C3 lightbulbAD Package.