Skip to content

Add support for updated Web Search tooling #213

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Add support for updated Web Search tooling #213

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
53 changes: 51 additions & 2 deletions Sources/AIProxy/OpenAI/OpenAICreateResponseRequestBody.swift
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,9 @@ public struct OpenAICreateResponseRequestBody: Encodable {
/// Text, image, or file inputs to the model, used to generate a response.
public let input: OpenAIResponse.Input?

/// Specify additional output data to include in the model response.
public let include: [Include]?

/// Model ID used to generate the response, like gpt-4o or o1.
/// OpenAI offers a wide range of models with different capabilities, performance characteristics, and price points.
/// Refer to the model guide to browse and compare available models: https://platform.openai.com/docs/models
Expand Down Expand Up @@ -81,6 +84,7 @@ public struct OpenAICreateResponseRequestBody: Encodable {

private enum CodingKeys: String, CodingKey {
case input
case include
case model
case tools
case toolChoice = "tool_choice"
Expand All @@ -102,6 +106,7 @@ public struct OpenAICreateResponseRequestBody: Encodable {
// To format, place the cursor in the initializer's parameter list and use `ctrl-m`
public init(
input: OpenAIResponse.Input? = nil,
include: [Include]? = nil,
model: String? = nil,
parallelToolCalls: Bool? = nil,
previousResponseId: String? = nil,
Expand All @@ -118,6 +123,7 @@ public struct OpenAICreateResponseRequestBody: Encodable {
user: String? = nil
) {
self.input = input
self.include = include
self.model = model
self.parallelToolCalls = parallelToolCalls
self.previousResponseId = previousResponseId
Expand All @@ -140,6 +146,30 @@ public struct OpenAICreateResponseRequestBody: Encodable {

extension OpenAICreateResponseRequestBody {

/// Specify additional output data to include in the model response.
public enum Include: String, Codable {
/// Include the outputs of python code execution in code interpreter tool call items.
case codeInterpreterCallOutputs = "code_interpreter_call.outputs"

/// Include image urls from the computer call output.
case computerCallOutputImageUrl = "computer_call_output.output.image_url"

/// Include the search results of the file search tool call.
case fileSearchCallResults = "file_search_call.results"

/// Include image urls from the input message.
case messageInputImageImageUrl = "message.input_image.image_url"

/// Include logprobs with assistant messages.
case messageOutputTextLogprobs = "message.output_text.logprobs"

/// Includes an encrypted version of reasoning tokens in reasoning item outputs.
case reasoningEncryptedContent = "reasoning.encrypted_content"

/// Include the sources of the web search tool call.
case webSearchCallActionSources = "web_search_call.action.sources"
}

/// The truncation strategy to use for the model response.
public enum Truncation: String, Encodable {
/// If the context of this response and previous ones exceeds the model's context window size, the model will truncate the response to fit the context window by dropping input items in the middle of the conversation.
Expand Down Expand Up @@ -205,6 +235,11 @@ extension OpenAICreateResponseRequestBody {
/// https://platform.openai.com/docs/guides/tools-web-search?api-mode=responses
case webSearch(WebSearchTool)

/// Allow models to search the web for the latest information before generating a response (preview version).
/// https://platform.openai.com/docs/guides/tools-web-search?api-mode=responses
@available(*, deprecated, message: "Use webSearch if using a modern GTP model. webSearchPreview will be removed in a future version.")
case webSearchPreview(WebSearchPreviewTool)

private enum CodingKeys: String, CodingKey {
case description
case displayHeight = "display_height"
Expand Down Expand Up @@ -234,6 +269,11 @@ extension OpenAICreateResponseRequestBody {
try container.encodeIfPresent(tool.rankingOptions, forKey: .rankingOptions)

case .webSearch(let tool):
try container.encode("web_search", forKey: .type)
try container.encodeIfPresent(tool.searchContextSize, forKey: .searchContextSize)
try container.encodeIfPresent(tool.userLocation, forKey: .userLocation)

case .webSearchPreview(let tool):
try container.encode("web_search_preview", forKey: .type)
try container.encodeIfPresent(tool.searchContextSize, forKey: .searchContextSize)
try container.encodeIfPresent(tool.userLocation, forKey: .userLocation)
Expand Down Expand Up @@ -265,11 +305,16 @@ extension OpenAICreateResponseRequestBody {
let rankingOptions = try container.decodeIfPresent(FileSearchTool.RankingOptions.self, forKey: .rankingOptions)
self = .fileSearch(FileSearchTool(vectorStoreIDs: vectorStoreIDs, filters: filters, maxNumResults: maxNumResults, rankingOptions: rankingOptions))

case "web_search_preview":
case "web_search":
let searchContextSize = try container.decodeIfPresent(WebSearchTool.SearchContextSize.self, forKey: .searchContextSize)
let userLocation = try container.decodeIfPresent(WebSearchTool.UserLocation.self, forKey: .userLocation)
self = .webSearch(WebSearchTool(searchContextSize: searchContextSize, userLocation: userLocation))

case "web_search_preview":
let searchContextSize = try container.decodeIfPresent(WebSearchTool.SearchContextSize.self, forKey: .searchContextSize)
let userLocation = try container.decodeIfPresent(WebSearchTool.UserLocation.self, forKey: .userLocation)
self = .webSearchPreview(WebSearchTool(searchContextSize: searchContextSize, userLocation: userLocation))

case "computer_use_preview":
let displayWidth = try container.decode(Int.self, forKey: .displayWidth)
let displayHeight = try container.decode(Int.self, forKey: .displayHeight)
Expand Down Expand Up @@ -421,7 +466,7 @@ extension OpenAICreateResponseRequestBody {
case userLocation = "user_location"
}

public let type = "web_search_preview"
public let type = "web_search"
public let searchContextSize: SearchContextSize?
public let userLocation: UserLocation?

Expand Down Expand Up @@ -460,6 +505,10 @@ extension OpenAICreateResponseRequestBody {
}
}

// MARK: - Web Search Tool (Preview)
@available(*, deprecated, message: "Use WebSearchTool instead. WebSearchPreviewTool will be removed in a future version.")
public typealias WebSearchPreviewTool = WebSearchTool

// MARK: - Computer Use Tool
public struct ComputerUseTool: Codable {
private enum CodingKeys: String, CodingKey {
Expand Down
12 changes: 12 additions & 0 deletions Sources/AIProxy/OpenAI/OpenAIResponse.swift
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -339,6 +339,18 @@ extension OpenAIResponse {
public var type = "web_search_call"
public let id: String
public let status: String
public let action: WebSearchAction?

public struct WebSearchAction: Decodable {
public let type: String
public let query: String?
public let sources: [WebSearchSource]?
}

public struct WebSearchSource: Decodable {
public let type: String
public let url: String
}
}

// MARK: - File Search Call
Expand Down
34 changes: 34 additions & 0 deletions Tests/AIProxyTests/OpenAIResponseStreamingEventTests.swift
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -102,6 +102,40 @@ class OpenAIResponseStreamingEventTests: XCTestCase {
XCTAssertEqual(webSearchCall.status, "completed")
}

func testWebSearchCallWithSourcesIsDecodable() throws {
let line = #"data: {"type":"response.output_item.done","sequence_number":8,"output_index":1,"item":{"id":"ws_68bf1fcb40ec819b815add7f2df9dcff0a2285864df44dd5","type":"web_search_call","status":"completed","action":{"type":"search","query":"Bills game score live Buffalo Bills score","sources":[{"type":"url","url":"https://www.wsj.com/sports/football/bills-ravens-josh-allen-lamar-jackson-derrick-henry-f8fc4d38"},{"type":"url","url":"https://timesofindia.indiatimes.com/sports/nfl/news/nfl-game-results-today-bills-vs-ravens-match-highlights-top-scorers-and-more/articleshow/123759411.cms"},{"type":"url","url":"https://www.bild.de/sport/mehr-sport/buffalo-bills-vs-baltimore-ravens-dieses-nfl-spiel-hatte-alles-68be463d72019652f128d780"},{"type":"url","url":"https://nypost.com/2025/09/08/sports/lamar-jackson-speaks-out-after-shoving-fan-in-snf-loss-to-bills/"},{"type":"url","url":"https://www.capstone-companies.com/summary/bills-game-today"}]}}}"#
let event = OpenAIResponseStreamingEvent.deserialize(fromLine: line)

guard case .outputItemDone(let outputItemDone) = event else {
return XCTFail("Expected response.output_item.done")
}
XCTAssertEqual(outputItemDone.sequenceNumber, 8)

guard case .webSearchCall(let webSearchCall) = outputItemDone.item else {
return XCTFail("Expected web search call")
}

XCTAssertEqual(webSearchCall.id, "ws_68bf1fcb40ec819b815add7f2df9dcff0a2285864df44dd5")
XCTAssertEqual(webSearchCall.status, "completed")

// Verify the action is properly decoded
XCTAssertNotNil(webSearchCall.action)
XCTAssertEqual(webSearchCall.action?.type, "search")
XCTAssertEqual(webSearchCall.action?.query, "Bills game score live Buffalo Bills score")

// Verify the sources are properly decoded
XCTAssertNotNil(webSearchCall.action?.sources)
XCTAssertEqual(webSearchCall.action?.sources?.count, 5)

let firstSource = webSearchCall.action?.sources?.first
XCTAssertEqual(firstSource?.type, "url")
XCTAssertEqual(firstSource?.url, "https://www.wsj.com/sports/football/bills-ravens-josh-allen-lamar-jackson-derrick-henry-f8fc4d38")

let lastSource = webSearchCall.action?.sources?.last
XCTAssertEqual(lastSource?.type, "url")
XCTAssertEqual(lastSource?.url, "https://www.capstone-companies.com/summary/bills-game-today")
}

func testOutputItemAddedForContentIsDecodable() throws {
let line = #"data: {"type":"response.output_item.added","sequence_number":7,"output_index":1,"item":{"id":"msg_123","type":"message","status":"in_progress","content":[],"role":"assistant"}}"#
let event = OpenAIResponseStreamingEvent.deserialize(fromLine: line)
Expand Down