up to date verion 1.2.0

Author: MaxiDonkey@hotmail.com

Changelog.md

@@ -1,3 +1,42 @@
+#### 2025, August 12 version 1.2.0
+**Modifications to ensure full use of the gtp-5 model**
+
+- JSON Normalization Before Deserialization
+  - New `GenAI.API.Normalizer` module (`TJSONNormalizer`, `TWrapKind`, `TNormalizationRule`) to unify polymorphic fields (e.g., string vs. object).
+  - Direct integration in the HTTP layer: new `Get(..., Path)` overloads allow targeted normalization of a JSON subtree before object mapping.
+
+- Canceling Background Requests
+  - New `Responses.AsyncAwaitCancel(response_id)` method to cancel an asynchronous (`background = true`) response, with full callback support (`OnStart`, `OnSuccess`, `OnError`).
+
+- Streaming Enhancements
+  - Extended typed coverage for streaming events and outputs (MCP, Code Interpreter, Image Generation, etc.) via new `Responses.OutputParams` classes (`TResponseOutput*`, `TResponseImageGenerationTool`, `TResponseCodeInterpreter`, etc.).
+
+- New Types and Parameters
+  - InputParams: full coverage for computer interactions, local shell, MCP, web search, code, image generation, reasoning, text/JSON formats, tool choice/hosted tool, and file search filters.
+  - OutputParams: states (`Created`, `InProgress`, etc.), events (`Added`, `Delta`), usage metrics, and statistics.
+  - New enums (`TOutputIncluding`, `TReasoningGenerateSummary`, `TFidelityType`, etc.).
+
+- API `v1/chat/completions`
+  - New parameters:
+    - `prompt_cache_key` (prompt caching)
+    - `safety_identifier` (stable ID for safety monitoring)
+    - `verbosity` (low/medium/high)
+
+- API `v1/responses`
+  - New parameters:
+    - `max_tool_calls`
+    - `prompt` (template reference via `TPromptParams`)
+    - `prompt_cache_key`, `safety_identifier`
+    - `stream_options`, `top_logprobs`, `verbosity`
+
+- Structured System and Developer Messages
+  - New overloads:
+    - `TMessagePayload.Developer(const Content: TArray; const Name: string = '')`
+    - `TMessagePayload.System(const Content: TArray; const Name: string = '')`
+  - Improves parity between plain text and structured content flows.
+
+<br>
+
 ### 2025, June 14 version 1.1.0 (**Getit version**)
 - Given the project’s rapid progress, it’s now essential to embed versioning directly into the GenAI wrapper’s source code. For any client implementing the IGenAI interface, the version number can be retrieved via the Version property, for example:
 ```Delphi

ChatCompletion.md

@@ -112,7 +112,7 @@ The Chat API can be used for both single-turn requests and multi-turn, stateless
 //      procedure (E: Exception)
 //      begin
 //        Display(TutorialHub, E.Message);
-//      end
+//      end);
 ```
 <br>
 

README.md

@@ -5,13 +5,13 @@ ___
 ![GitHub](https://img.shields.io/badge/IDE%20Version-Delphi%2010.4/11/12-ffffba)
 [![GetIt – Available](https://img.shields.io/badge/GetIt-Available-baffc9?logo=delphi&logoColor=white)](https://getitnow.embarcadero.com/genai-optimized-openai-integration-wrapper/)
 ![GitHub](https://img.shields.io/badge/platform-all%20platforms-baffc9)
-![GitHub](https://img.shields.io/badge/Updated%20on%20June%2013,%202025-blue)
+![GitHub](https://img.shields.io/badge/Updated%20on%20August%2012,%202025-blue)
 
 <br>
 
 NEW: 
 - Getit current version: 1.1.0
-- [Changelog v1.1.0](https://github.com/MaxiDonkey/DelphiGenAI/blob/main/Changelog.md)
+- [Changelog v1.2.0](https://github.com/MaxiDonkey/DelphiGenAI/blob/main/Changelog.md)
 - [Mini-lab to experiment with the v1/responses endpoint `File2knowledge`](https://github.com/MaxiDonkey/file2knowledge)  
 - [Responses endpoint](https://github.com/MaxiDonkey/DelphiGenAI/blob/main/Responses.md)
 ___
@@ -75,7 +75,7 @@ ___
 **GenAI** is a powerful Delphi library that brings the latest OpenAI APIs to your desktop, mobile, and server apps.
 
 **Core capabilities**  
-- Unified access to text, vision, speech, and audio endpoints  
+- Unified access to text, vision, speech and audio endpoints  
 - Agentic workflows via the `v1/responses` endpoint, with built-in tools `file_search`, `web_search`, `Code Interpreter`, and `remote MCP`  
 - Supports state-of-the-art models, including ***gpt-4o***, ***gpt-4.1***, ***gpt-4.5*** and the reasoning-centric *o1 · o3 · o4* series  
 

Responses.md

@@ -13,6 +13,7 @@
         - [Get a model response](#get-a-model-response)
         - [Delete a model response](#delete-a-model-response)
         - [List input items](#list-input-items)
+    - [Canceling Background Tasks](#canceling-background-tasks)
 - [Vision](#vision)
     - [Analyze single source](#analyze-single-source)
     - [Analyze multi-source](#analyze-multi-source)
@@ -621,6 +622,62 @@ Returns a list of input items for a given response. Refer to the [official docum
 
 ___
 
+### Canceling Background Tasks
+
+Terminates an ongoing model response identified by its ID. This action is only available for responses that were initiated with the `background` parameter set to `true`.
+
+Some reasoning models—such as Codex and Deep Research—can take several minutes to solve complex problems. Background mode allows you to run long-running operations on models like o3 and o1-pro more reliably, avoiding timeouts and network interruptions.
+
+When background mode is enabled, the task starts asynchronously. You can then periodically poll the response object to monitor its progress. To start a background process, send an API request with `background` set to `true`.
+
+And, of course, you can cancel a background task, as shown below.
+
+<br>
+
+```pascal
+//uses GenAI, GenAI.Types, GenAI.Tutorial.VCL;
+
+  TutorialHub.JSONRequestClear;
+
+  //Asynchronous promise example
+  var Promise := Client.Responses.AsyncAwaitCancel('Response_ID');
+    
+  Promise
+    .&Then<TResponse>(
+      function (Value: TResponse): TResponse
+      begin
+        Result := Value;
+        Display(TutorialHub, Value);
+      end)
+    .&Catch(
+      procedure (E: Exception)
+      begin
+        Display(TutorialHub, E.Message);
+      end);
+
+  //Asynchronous example
+//  Client.Responses.AsynCancel('Response_ID',
+//    function : TAsynResponse
+//    begin
+//      Result.Sender := TutorialHub;
+//      Result.OnStart := Start;
+//      Result.OnSuccess := Display;
+//      Result.OnError := Display;
+//    end);
+
+  //Synchronous example
+//  var Value := Client.Responses.Cancel('Response_ID');
+//  try
+//    Display(TutorialHub, Value);
+//  finally
+//    Value.Free;
+//  end;
+```
+
+<br>
+
+___
+
 ## Vision
 
 Refer to the [official documentation](https://platform.openai.com/docs/guides/vision).