Merge remote-tracking branch 'origin/main' into support-roo-directory

Author: mrubens

package.json

@@ -174,6 +174,11 @@
 				"command": "roo-cline.setCustomStoragePath",
 				"title": "Set Custom Storage Path",
 				"category": "Roo Code"
+			},
+			{
+				"command": "roo-cline.focusInput",
+				"title": "Focus Input Field",
+				"category": "Roo Code"
 			}
 		],
 		"menus": {

src/activate/registerCommands.ts

@@ -114,6 +114,9 @@ const getCommandsMap = ({ context, outputChannel, provider }: RegisterCommandOpt
 			const { promptForCustomStoragePath } = await import("../shared/storagePathManager")
 			await promptForCustomStoragePath()
 		},
+		"roo-cline.focusInput": () => {
+			provider.postMessageToWebview({ type: "action", action: "focusInput" })
+		},
 	}
 }
 

src/core/assistant-message/index.ts

@@ -60,6 +60,7 @@ export const toolParamNames = [
 	"cwd",
 	"follow_up",
 	"task",
+	"size",
 ] as const
 
 export type ToolParamName = (typeof toolParamNames)[number]
@@ -115,7 +116,7 @@ export interface ListCodeDefinitionNamesToolUse extends ToolUse {
 
 export interface BrowserActionToolUse extends ToolUse {
 	name: "browser_action"
-	params: Partial<Pick<Record<ToolParamName, string>, "action" | "url" | "coordinate" | "text">>
+	params: Partial<Pick<Record<ToolParamName, string>, "action" | "url" | "coordinate" | "text" | "size">>
 }
 
 export interface UseMcpToolToolUse extends ToolUse {

src/core/prompts/__tests__/__snapshots__/system.test.ts.snap

@@ -2714,19 +2714,26 @@ Parameters:
     * launch: Launch a new Puppeteer-controlled browser instance at the specified URL. This **must always be the first action**.
         - Use with the \`url\` parameter to provide the URL.
         - Ensure the URL is valid and includes the appropriate protocol (e.g. http://localhost:3000/page, file:///path/to/file.html, etc.)
+    * hover: Move the cursor to a specific x,y coordinate.
+        - Use with the \`coordinate\` parameter to specify the location.
+        - Always move to the center of an element (icon, button, link, etc.) based on coordinates derived from a screenshot.
     * click: Click at a specific x,y coordinate.
         - Use with the \`coordinate\` parameter to specify the location.
         - Always click in the center of an element (icon, button, link, etc.) based on coordinates derived from a screenshot.
     * type: Type a string of text on the keyboard. You might use this after clicking on a text field to input text.
         - Use with the \`text\` parameter to provide the string to type.
+    * resize: Resize the viewport to a specific w,h size.
+        - Use with the \`size\` parameter to specify the new size.
     * scroll_down: Scroll down the page by one page height.
     * scroll_up: Scroll up the page by one page height.
     * close: Close the Puppeteer-controlled browser instance. This **must always be the final browser action**.
         - Example: \`<action>close</action>\`
 - url: (optional) Use this for providing the URL for the \`launch\` action.
     * Example: <url>https://example.com</url>
-- coordinate: (optional) The X and Y coordinates for the \`click\` action. Coordinates should be within the **900x600** resolution.
+- coordinate: (optional) The X and Y coordinates for the \`click\` and \`hover\` actions. Coordinates should be within the **900x600** resolution.
     * Example: <coordinate>450,300</coordinate>
+- size: (optional) The width and height for the \`resize\` action.
+    * Example: <size>1280,720</size>
 - text: (optional) Use this for providing the text for the \`type\` action.
     * Example: <text>Hello, world!</text>
 Usage:
@@ -3625,19 +3632,26 @@ Parameters:
     * launch: Launch a new Puppeteer-controlled browser instance at the specified URL. This **must always be the first action**.
         - Use with the \`url\` parameter to provide the URL.
         - Ensure the URL is valid and includes the appropriate protocol (e.g. http://localhost:3000/page, file:///path/to/file.html, etc.)
+    * hover: Move the cursor to a specific x,y coordinate.
+        - Use with the \`coordinate\` parameter to specify the location.
+        - Always move to the center of an element (icon, button, link, etc.) based on coordinates derived from a screenshot.
     * click: Click at a specific x,y coordinate.
         - Use with the \`coordinate\` parameter to specify the location.
         - Always click in the center of an element (icon, button, link, etc.) based on coordinates derived from a screenshot.
     * type: Type a string of text on the keyboard. You might use this after clicking on a text field to input text.
         - Use with the \`text\` parameter to provide the string to type.
+    * resize: Resize the viewport to a specific w,h size.
+        - Use with the \`size\` parameter to specify the new size.
     * scroll_down: Scroll down the page by one page height.
     * scroll_up: Scroll up the page by one page height.
     * close: Close the Puppeteer-controlled browser instance. This **must always be the final browser action**.
         - Example: \`<action>close</action>\`
 - url: (optional) Use this for providing the URL for the \`launch\` action.
     * Example: <url>https://example.com</url>
-- coordinate: (optional) The X and Y coordinates for the \`click\` action. Coordinates should be within the **1280x800** resolution.
+- coordinate: (optional) The X and Y coordinates for the \`click\` and \`hover\` actions. Coordinates should be within the **1280x800** resolution.
     * Example: <coordinate>450,300</coordinate>
+- size: (optional) The width and height for the \`resize\` action.
+    * Example: <size>1280,720</size>
 - text: (optional) Use this for providing the text for the \`type\` action.
     * Example: <text>Hello, world!</text>
 Usage:

src/core/prompts/tools/browser-action.ts

@@ -15,19 +15,26 @@ Parameters:
     * launch: Launch a new Puppeteer-controlled browser instance at the specified URL. This **must always be the first action**.
         - Use with the \`url\` parameter to provide the URL.
         - Ensure the URL is valid and includes the appropriate protocol (e.g. http://localhost:3000/page, file:///path/to/file.html, etc.)
+    * hover: Move the cursor to a specific x,y coordinate.
+        - Use with the \`coordinate\` parameter to specify the location.
+        - Always move to the center of an element (icon, button, link, etc.) based on coordinates derived from a screenshot.
     * click: Click at a specific x,y coordinate.
         - Use with the \`coordinate\` parameter to specify the location.
         - Always click in the center of an element (icon, button, link, etc.) based on coordinates derived from a screenshot.
     * type: Type a string of text on the keyboard. You might use this after clicking on a text field to input text.
         - Use with the \`text\` parameter to provide the string to type.
+    * resize: Resize the viewport to a specific w,h size.
+        - Use with the \`size\` parameter to specify the new size.
     * scroll_down: Scroll down the page by one page height.
     * scroll_up: Scroll up the page by one page height.
     * close: Close the Puppeteer-controlled browser instance. This **must always be the final browser action**.
         - Example: \`<action>close</action>\`
 - url: (optional) Use this for providing the URL for the \`launch\` action.
     * Example: <url>https://example.com</url>
-- coordinate: (optional) The X and Y coordinates for the \`click\` action. Coordinates should be within the **${args.browserViewportSize}** resolution.
+- coordinate: (optional) The X and Y coordinates for the \`click\` and \`hover\` actions. Coordinates should be within the **${args.browserViewportSize}** resolution.
     * Example: <coordinate>450,300</coordinate>
+- size: (optional) The width and height for the \`resize\` action.
+    * Example: <size>1280,720</size>
 - text: (optional) Use this for providing the text for the \`type\` action.
     * Example: <text>Hello, world!</text>
 Usage:

src/core/tools/browserActionTool.ts

@@ -21,6 +21,7 @@ export async function browserActionTool(
 	const url: string | undefined = block.params.url
 	const coordinate: string | undefined = block.params.coordinate
 	const text: string | undefined = block.params.text
+	const size: string | undefined = block.params.size
 	if (!action || !browserActions.includes(action)) {
 		// checking for action to ensure it is complete and valid
 		if (!block.partial) {
@@ -72,7 +73,7 @@ export async function browserActionTool(
 				await cline.browserSession.launchBrowser()
 				browserActionResult = await cline.browserSession.navigateToUrl(url)
 			} else {
-				if (action === "click") {
+				if (action === "click" || action === "hover") {
 					if (!coordinate) {
 						cline.consecutiveMistakeCount++
 						pushToolResult(await cline.sayAndCreateMissingParamError("browser_action", "coordinate"))
@@ -88,6 +89,14 @@ export async function browserActionTool(
 						return
 					}
 				}
+				if (action === "resize") {
+					if (!size) {
+						cline.consecutiveMistakeCount++
+						pushToolResult(await cline.sayAndCreateMissingParamError("browser_action", "size"))
+						await cline.browserSession.closeBrowser()
+						return
+					}
+				}
 				cline.consecutiveMistakeCount = 0
 				await cline.say(
 					"browser_action",
@@ -103,6 +112,9 @@ export async function browserActionTool(
 					case "click":
 						browserActionResult = await cline.browserSession.click(coordinate!)
 						break
+					case "hover":
+						browserActionResult = await cline.browserSession.hover(coordinate!)
+						break
 					case "type":
 						browserActionResult = await cline.browserSession.type(text!)
 						break
@@ -112,6 +124,9 @@ export async function browserActionTool(
 					case "scroll_up":
 						browserActionResult = await cline.browserSession.scrollUp()
 						break
+					case "resize":
+						browserActionResult = await cline.browserSession.resize(size!)
+						break
 					case "close":
 						browserActionResult = await cline.browserSession.closeBrowser()
 						break
@@ -121,9 +136,11 @@ export async function browserActionTool(
 			switch (action) {
 				case "launch":
 				case "click":
+				case "hover":
 				case "type":
 				case "scroll_down":
 				case "scroll_up":
+				case "resize":
 					await cline.say("browser_action_result", JSON.stringify(browserActionResult))
 					pushToolResult(
 						formatResponse.toolResult(

src/exports/api.ts

@@ -132,6 +132,21 @@ export class API extends EventEmitter<RooCodeEvents> implements RooCodeAPI {
 		return taskId
 	}
 
+	public async resumeTask(taskId: string): Promise<void> {
+		const { historyItem } = await this.provider.getTaskWithId(taskId)
+		await this.provider.initClineWithHistoryItem(historyItem)
+		await this.provider.postMessageToWebview({ type: "action", action: "chatButtonClicked" })
+	}
+
+	public async isTaskInHistory(taskId: string): Promise<boolean> {
+		try {
+			await this.provider.getTaskWithId(taskId)
+			return true
+		} catch {
+			return false
+		}
+	}
+
 	public getCurrentTaskStack() {
 		return this.sidebarProvider.getCurrentTaskStack()
 	}

src/exports/roo-code.d.ts

@@ -551,6 +551,20 @@ interface RooCodeAPI extends EventEmitter<RooCodeEvents> {
 		images?: string[]
 		newTab?: boolean
 	}): Promise<string>
+	/**
+	 * Resumes a task with the given ID.
+	 * @param taskId The ID of the task to resume.
+	 * @throws Error if the task is not found in the task history.
+	 */
+	resumeTask(taskId: string): Promise<void>
+
+	/**
+	 * Checks if a task with the given ID is in the task history.
+	 * @param taskId The ID of the task to check.
+	 * @returns True if the task is in the task history, false otherwise.
+	 */
+	isTaskInHistory(taskId: string): Promise<boolean>
+
 	/**
 	 * Returns the current task stack.
 	 * @returns An array of task IDs.

src/services/browser/BrowserSession.ts

@@ -538,4 +538,17 @@ export class BrowserSession {
 			})
 		})
 	}
+
+	async resize(size: string): Promise<BrowserActionResult> {
+		return this.doAction(async (page) => {
+			const [width, height] = size.split(",").map(Number)
+			const session = await page.createCDPSession()
+			await page.setViewport({ width, height })
+			const { windowId } = await session.send("Browser.getWindowForTarget")
+			await session.send("Browser.setWindowBounds", {
+				bounds: { width, height },
+				windowId,
+			})
+		})
+	}
 }

src/shared/ExtensionMessage.ts

@@ -77,6 +77,7 @@ export interface ExtensionMessage {
 		| "historyButtonClicked"
 		| "promptsButtonClicked"
 		| "didBecomeVisible"
+		| "focusInput"
 	invoke?: "newChat" | "sendMessage" | "primaryButtonClick" | "secondaryButtonClick" | "setChatBoxMessage"
 	state?: ExtensionState
 	images?: string[]
@@ -232,13 +233,23 @@ export interface ClineSayTool {
 }
 
 // Must keep in sync with system prompt.
-export const browserActions = ["launch", "click", "hover", "type", "scroll_down", "scroll_up", "close"] as const
+export const browserActions = [
+	"launch",
+	"click",
+	"hover",
+	"type",
+	"scroll_down",
+	"scroll_up",
+	"resize",
+	"close",
+] as const
 
 export type BrowserAction = (typeof browserActions)[number]
 
 export interface ClineSayBrowserAction {
 	action: BrowserAction
 	coordinate?: string
+	size?: string
 	text?: string
 }