API Creator executes your business logic in response to REST POST/PUT/DELETE requests. This article provides information about how the server operates so you can debug your API. Shown conceptually in the Reactive Logic video, this article shows how to use rules, how the rules work, and how rules integrate with events.
For more information about the Reactive Logic video, see Videos.
In this article:
Verify the Prerequisites
Before reviewing the information in this article, ensure that you have completed the following prerequisites:
API Creator uses your business rules (reactive logic) to process RESTful updates (POST, PUT, DELETE).
Connect and Declare Rules
When you create an API and you connect API Creator to your database,
For more information about the row object, see Row Objects.
Validation: return row.CreditLimit >= row.BalanceDerive Balance asSum(Orders.AmountTotal where ShippedDate === null)
RESTful Update Processing
The following subsections explain how API Creator processes update requests. The following diagram shows the workflow:
For more information about request events, see Event Handlers.
You can define multiple custom resources on the same underlying base table. The resources can represent projects and aliases of table columns. Integrity demands that the base table logic is enforced regardless of the resource used.
Resource/object mapping is required to map the resource objects onto their respective row objects. This includes column name de-aliasing and materialization of columns not projected (sometimes called "hydrated"). The declarative and event logic deals with a full row. API Creator shares that logic over all custom resources. To do this, API Creator must read the database, with concurrency control supplied by the DBMS.
So that you can define logic based on data changes, API Creator builds the following row objects:
- row. This object reflects client changes from JSON, and all logic processing.
- oldRow. This object reflects the row before the transaction started. This object is from the database read.
API Server ensures that updated rows do not overlay the updates of other users by performing optimistic locking checks. The check is on a time-stamp field, if provided. Otherwise it is the hash of all attributes in the resource row.
For more information:
DB Key Generation
Databases support system-generated primary keys. There are special requirements when processing JSON POSTs. These requirements include child rows for such tables, such as the items for an order. API Creator must "stamp" the generated order# into each item. API Creator does this before logic execution and when it performs managed parent.
When using a resource with nested children (Customer contains Orders, contains Items), you can POST (insert) a new Customer, Order, and Items in one transaction. You can do this only when API Creator can propagate the primary key of the parent (Customer or Orders) down to the child (for example, Orders or Items).
For more information:
Row Logic Cycle
API Creator processes each submitted row in the order it receives them, as follows:
- Calls early row events, supplied with therow,oldRow, andlogicContextvariables. Your event can inspect/alter the row before rules fire. For example, you can compute network-unique primary keys.
- Executes (only) those rules whose dependent row data has changed (based on comparingrowwitholdRow). It computes the rule execution order based on rule dependencies, discovered by automatic rule parsing. The rule updates the row (state), so it is visible to ensuing rules.
- Executes row events so that you can do whatever is not expressed in declarative logic. For example, send email, post data to other systems, and credit card checks. You are passed row, oldRow, and logicContext. The effects of rule changes are visible in your row objects.You can alter the row, but you must save your changes.For more information about logic events, see Logic Event Rule Types.
- If the logic has altered data referenced by rules in related objects, API Creator instantiates rows for them and invokes their logic. For example, altering
updates the related customer's balance, whose rules would verify theOrderTotal
. This is an efficient one-row update, not an aggregate query.CreditLimit
A transaction is used for all updates of a given request, both rows from the client, and chained updates (step 4). Rows are buffered into the write-cache so that multiple updates to the same row (for example, many line items might adjust the
Row Commit Cycle
The write-cache is flushed to the database at the end of the transaction, after all rows have been processed. In addition to the flush, there is an important logic provision.
Your logic specifications for validations and events can stipulate that they run during normal per-row execution or can be deferred until commit. If you elect your validations and events run during normal per-row execution,
CA Live API Creatorexecutes the logic only prior to transaction flush. The logic phase has been completed, so all the logic for all the rows are visible in your row objects.
For example, you want to ensure Purchase Orders have at least one line item. You can define a
Purchaseorder.item_count, with a
item_count > 0.
While a good approach, this would fail. Why? API Creator processes the
Purchaseorderinsert first before line items. At this point, the count is zero (0), so the validation fails. Instead, you can commit validations and events. These validations and events must operate on the end-result of logic processing for all the rows in the entire transaction.
For more information:
Request Events - Response
API Creator raises request post events. You can alter the response message or can perform other functions such as logging.
With forward chaining, if you change a referenced value, API Creator recomputes the derived referencing attributes. The term chaining correctly infers that a derived attribute (for example,
Purchaseorder.amount_total) is itself referenced in another derivation (
Customer.balance). API Creator tracks these references and performs the forward chaining, automatically.
For formulas (for example, price * quantity), forward chaining entails evaluating the expression (though see ordering, in the following sections). Forward chaining is more complicated for dependencies and multi-table derivations.
Columns dependent on changed columns can themselves have interdependencies. For example:
a = b + x
b = x + 2
It is clear that
b, so if
xis altered, API Creator must recompute
bbefore it recomputes
a. You can state these rules in any order. You can change the rules during maintenance, without concern for ordering.
Customer.balanceexample, imagine a simple update where a
Purchaseorderis marked paid. We need to recompute the new balance. A dreadful approach is to issue a SQL Sum query to add all the existing orders. In general, there could be thousands! And worse, this could be chained, where the summed attributes depend on further summed attributes. That is, in fact, just the case here: the
Purchaseorder.amount_totalis itself a sum of the
Lineitem.amount. This is a significant performance factor. ORM products are often blamed for poor performance due to excessive use of chained aggregate queries.
API Creator adjusts the parent sum by the change in the child. The result is a one-row update (unless it was pruned).
Analogous considerations are where the client alters a parent attribute referred to in child logic (for example, a formula). When this occurs, API Creator visits each related child to run the dependent logic. Running the dependent logic might update the child, and might trigger further adjustment / cascade processing to other related data.
For more information about formulas, see Formula Rule Type.
Adjustment and cascade-processing make updates to related data. API Creator often issues SQL updates for data beyond that originally sent from the client. This is a natural consequence of your logic and exactly what business logic is supposed to do. These triggered updates are subjected to the full logic analysis/chaining process, so will often result in still other updates. For example, consider a simple update to a
- Lineitem.amountis derived as price*quantity, so is recomputed.
- Purchaseorder.amount_totalis derived as Sum (Lineitem.amount), so it is recomputed (adjusted).
- Customer.balance is derived as Sum (Purchaseorder.amount_totalwhere Paid = false), so is is adjusted.
The customer logic re-evaluates the credit limit check - if not passed, the entire transaction is rolled back, and an exception is returned to the client.
Chaining means that API Creator can execute your logic more than once on the same row multiple times within a transaction. Consider a transaction comprised of a purchase order with multiple Line Items. The purchase order logic is clearly executed on insertion. Now consider that each Line Item would adjust the Purchase Order's amount_total. This re-executes the purchase order logic, now as an update. Your logic can determine
initialVerbby way of the
For more information about the LogicContext object, see The logicContext Object.
Reactive Programming vs Conventional Procedural Programming
The following are key observations about some fundamental characteristics that distinguish reactive programming from conventional procedural (imperative) programming:
- No Control flow.API Creator invokes the rules and only in reaction to actual changes. You do not order their execution. Rules are bound to the data, not a specific use case, so they apply to all in-coming transactions. In other words, the logic automatically processes the following transactions:
- Order inserted - balance increased
- Order deleted - balance decreased (if not paid)
- Order unshipped - balance decreased
- Order shipped - balance decreased
- Order amountTotal changed - balance adjusted
- Order reassigned to different customer - balance increased for new customer, decreased for old
- OrderDetail inserted - obtain price, adjust Order and Customer (and check credit)
- OrderDetail Deleted - reduce Order and customer totals
- OrderDetail Quantity increased - adjust Order and Customer (and check credit)
- OrderDetail Product Changed - obtain price, adjust Order and Customer (and check credit)
- OrderDetail Quantity and Product Changed - obtain price, adjust Order and Customer (and check credit)
- Customer CreditLimit changed - check credit
- Elimination of Boilerplate code.Reactive programming automates detection, change propagation, and persistence handling (SQL commands). The logic is executable:
- Change detection.Most of the alternative code is determining when to propagate updates by detecting changes. This is eliminated in the declarative-reactive approach.
- SQL (caching).SQL handling is tedious. Rules automate the SQL, including the underlying services for caching.
Simple Example: Check Credit
In the example, a solution of Check Credit is devised. Building on the previous two rules:
This represents the complete, executable solution, that includes:
- Ordering.You can alter the previous rules in any order because they are automatically ordered per dependency management.
- Re-use.API Creator applies the rules to all incoming transactions, automatically invoking the previous (relevant) logic.
- Automatic Persistence.API Creator processes incoming transactions by providing the SQL. Adjusting a quantity automatically reads/adjusts the Orders and Customer rows. It does so efficiently (a one-row update, not an expensive select sum query).
Common Logic Patterns
This simple "
CreditLimit" example illustrates one of the most common logic patterns, validating a rollup/constraint-derived result. Other examples of the pattern include:
- Rollup employee salaries to department, constrain to budget.
- Rollup departments, constrain to budget.
- Rollup Student Course Credit, constrain to max for student, max for course.
- Compute Product Price from the sum of the Component Parts (nested).
- Compute Event
from sum ofGiftsAmount
A similar pattern is Existence Checks: validations on [qualified] counts, such as an Order must have items and Department must have employees.
For more information about the common logic patterns, including validate a sum (rollup) and existence checks, see Learning Rules.
Business Perspective: Agility, Transparency, and Quality
Declarative logic is more expressive than imperative code. The previous five lines of logic equate to over 200 lines of triggers, or 500 lines of Java. It is also far more readable, in fact understandable, to business users. In an industry where we walk over hot coals for a 30% gain, this is a 40X improvement in expression factor. You can deliver update business logic 10X faster using rules. Removing boilerplate code and automatic re-use drives this compression factor.
For more information about the Add Payment example, see Reactive Logic Tutorial.
- About how to debug your API, see Debug.
- About the row object behavior, including attribute/object accessors, persistence, and event publishing, see Customize your API.
- About update-business logic, see Reactive Logic Operation.