KNOWLEDGE BASE
Log In    |    Knowledge Base    |    4D Home
Tech Tip: Introduction to Error Handling in 4D AIKit
PRODUCT: 4D | VERSION: 20 R | PLATFORM: Mac & Win
Published On: December 9, 2025
When interacting with the OpenAI API through 4D AIKit, a request may fail due to invalid parameters, authentication issues, rate limits, or server-side problems.

AIKit returns a structured OpenAIError object whenever an API request does not succeed.

Detecting an Error

Every AIKit method returns an object containing a success flag:
  • success = True → the request completed successfully

  • success = False → an error occurred and the error property contains an OpenAIError instance

Example :
var $client:=cs.AIKit.OpenAI.new("API Key")
var $messages : Collection
var $options : Object
var $result : Object

$messages:=[{role: "user"; content: "Explain quantum physics"}]

$options:={model: "gpt-4o-mini"; max_tokens: 100}

$result:=$client.chat.completions.create($messages; $options)

If (Not($result.success))
   $error:=$result.errors
   ALERT("Error: "+$error[0].message)
End if


Behavior on Failure

When success = False:

  • No partial data is returned in the result

  • The error property always contains a valid OpenAIError instance

  • Error details can be extracted safely

  • This provides a predictable pattern for error inspection.


Through this unified structure, error inspection remains consistent regardless of the failure type.