Merge branch 'main' into font-aliasing

Author: DonDawsy

CHANGELOG.md

@@ -1,5 +1,29 @@
 # Roo Code Changelog
 
+## [3.12.0] - 2025-04-15
+
+- Add xAI provider and expose reasoning effort options for Grok on OpenRouter (thanks Cline!)
+- Make diff editing config per-profile and improve pre-diff string normalization
+- Make checkpoints faster and more reliable
+- Add a search bar to mode and profile select dropdowns (thanks @samhvw8!)
+- Add telemetry for code action usage, prompt enhancement usage, and consecutive mistake errors
+- Suppress zero cost values in the task header (thanks @do-it!)
+- Make JSON parsing safer to avoid crashing the webview on bad input
+- Allow users to bind a keyboard shortcut for accepting suggestions or input in the chat view (thanks @axkirillov!)
+
+## [3.11.17] - 2025-04-14
+
+- Improvements to OpenAI cache reporting and cost estimates (thanks @monotykamary and Cline!)
+- Visual improvements to the auto-approve toggles (thanks @sachasayan!)
+- Bugfix to diff apply logic (thanks @avtc for the test case!) and telemetry to track errors going forward
+- Fix race condition in capturing short-running terminal commands (thanks @KJ7LNW!)
+- Fix eslint error (thanks @nobu007!)
+
+## [3.11.16] - 2025-04-14
+
+- Add gpt-4.1, gpt-4.1-mini, and gpt-4.1-nano to the OpenAI provider
+- Include model ID in environment details and when exporting tasks (thanks @feifei325!)
+
 ## [3.11.15] - 2025-04-13
 
 - Add ability to filter task history by workspace (thanks @samhvw8!)

README.md

@@ -3,7 +3,7 @@
 	"version": "0.1.0",
 	"private": true,
 	"scripts": {
-		"lint": "eslint src --ext ts",
+		"lint": "eslint src/**/*.ts",
 		"check-types": "tsc --noEmit",
 		"test": "npm run build && npx dotenvx run -f .env.local -- node ./out/runTest.js",
 		"ci": "npm run vscode-test && npm run test",

e2e/package.json

@@ -3,7 +3,7 @@
 	"private": true,
 	"type": "module",
 	"scripts": {
-		"lint": "eslint src --ext ts --max-warnings=0",
+		"lint": "eslint src/**/*.ts --max-warnings=0",
 		"check-types": "tsc --noEmit",
 		"format": "prettier --write src",
 		"dev": "dotenvx run -f ../../.env -- tsx src/index.ts"

evals/apps/cli/package.json

@@ -16,6 +16,7 @@ import {
 	IpcMessageType,
 	TaskCommandName,
 	rooCodeDefaults,
+	EvalEventName,
 } from "@evals/types"
 import {
 	type Run,
@@ -34,7 +35,7 @@ import { IpcServer, IpcClient } from "@evals/ipc"
 import { __dirname, extensionDevelopmentPath, exercisesPath } from "./paths.js"
 import { getExercises } from "./exercises.js"
 
-type TaskResult = { success: boolean; retry: boolean }
+type TaskResult = { success: boolean }
 type TaskPromise = Promise<TaskResult>
 
 const TASK_START_DELAY = 10 * 1_000
@@ -116,24 +117,25 @@ const run = async (toolbox: GluegunToolbox) => {
 
 	const runningPromises: TaskPromise[] = []
 
-	// Retries aren't implemented yet, but the return values are set up to
-	// support them.
 	const processTask = async (task: Task, delay = 0) => {
 		if (task.finishedAt === null) {
 			await new Promise((resolve) => setTimeout(resolve, delay))
-			const { retry } = await runExercise({ run, task, server })
-
-			if (retry) {
-				return { success: false, retry: true }
-			}
+			await runExercise({ run, task, server })
 		}
 
 		if (task.passed === null) {
 			const passed = await runUnitTest({ task })
 			await updateTask(task.id, { passed })
-			return { success: passed, retry: false }
+
+			server.broadcast({
+				type: IpcMessageType.TaskEvent,
+				origin: IpcOrigin.Server,
+				data: { eventName: passed ? EvalEventName.Pass : EvalEventName.Fail, taskId: task.id },
+			})
+
+			return { success: passed }
 		} else {
-			return { success: task.passed, retry: false }
+			return { success: task.passed }
 		}
 	}
 
@@ -200,7 +202,7 @@ const runExercise = async ({ run, task, server }: { run: Run; task: Task; server
 	} catch (error) {
 		console.log(`${Date.now()} [cli#runExercise | ${language} / ${exercise}] unable to connect`)
 		client.disconnect()
-		return { success: false, retry: false }
+		return { success: false }
 	}
 
 	let taskStartedAt = Date.now()
@@ -209,16 +211,15 @@ const runExercise = async ({ run, task, server }: { run: Run; task: Task; server
 	let rooTaskId: string | undefined
 	let isClientDisconnected = false
 
-	const ignoreEvents: RooCodeEventName[] = [
-		RooCodeEventName.Message,
-		RooCodeEventName.TaskTokenUsageUpdated,
-		RooCodeEventName.TaskAskResponded,
-	]
+	const ignoreEvents: Record<"broadcast" | "log", (RooCodeEventName | EvalEventName)[]> = {
+		broadcast: [RooCodeEventName.Message],
+		log: [RooCodeEventName.Message, RooCodeEventName.TaskTokenUsageUpdated, RooCodeEventName.TaskAskResponded],
+	}
 
 	client.on(IpcMessageType.TaskEvent, async (taskEvent) => {
 		const { eventName, payload } = taskEvent
 
-		if (taskEvent.eventName !== RooCodeEventName.Message) {
+		if (!ignoreEvents.broadcast.includes(eventName)) {
 			server.broadcast({
 				type: IpcMessageType.TaskEvent,
 				origin: IpcOrigin.Server,
@@ -227,7 +228,7 @@ const runExercise = async ({ run, task, server }: { run: Run; task: Task; server
 			})
 		}
 
-		if (!ignoreEvents.includes(eventName)) {
+		if (!ignoreEvents.log.includes(eventName)) {
 			console.log(
 				`${Date.now()} [cli#runExercise | ${language} / ${exercise}] taskEvent -> ${eventName}`,
 				payload,
@@ -320,11 +321,10 @@ const runExercise = async ({ run, task, server }: { run: Run; task: Task; server
 				data: { commandName: TaskCommandName.CancelTask, data: rooTaskId },
 			})
 
-			// Give the server some time to cancel the task.
+			// Allow some time for the task to cancel.
 			await new Promise((resolve) => setTimeout(resolve, 5_000))
 		}
 
-		// TODO: Notify clients that the task timed out.
 		await updateTask(task.id, { finishedAt: new Date() })
 	}
 
@@ -336,12 +336,15 @@ const runExercise = async ({ run, task, server }: { run: Run; task: Task; server
 				clientId: client.clientId!,
 				data: { commandName: TaskCommandName.CloseTask, data: rooTaskId },
 			})
+
+			// Allow some time for the window to close.
+			await new Promise((resolve) => setTimeout(resolve, 2_000))
 		}
 
 		client.disconnect()
 	}
 
-	return { success: !!taskFinishedAt, retry: false }
+	return { success: !!taskFinishedAt }
 }
 
 const runUnitTest = async ({ task }: { task: Task }) => {
@@ -372,7 +375,7 @@ const runUnitTest = async ({ task }: { task: Task }) => {
 				})
 
 				console.log(
-					`${Date.now()} [cli#runUnitTest | ${task.language} / ${task.exercise}] "${command.join(" ")}": ${subprocess.pid} -> ${JSON.stringify(descendants)}`,
+					`${Date.now()} [cli#runUnitTest | ${task.language} / ${task.exercise}] "${command.join(" ")}": unit tests timed out, killing ${subprocess.pid} + ${JSON.stringify(descendants)}`,
 				)
 
 				if (descendants.length > 0) {
@@ -384,7 +387,10 @@ const runUnitTest = async ({ task }: { task: Task }) => {
 
 							await execa`kill -9 ${descendant}`
 						} catch (error) {
-							console.error("Error killing descendant processes:", error)
+							console.error(
+								`${Date.now()} [cli#runUnitTest | ${task.language} / ${task.exercise}] Error killing descendant processes:`,
+								error,
+							)
 						}
 					}
 				}
@@ -396,7 +402,10 @@ const runUnitTest = async ({ task }: { task: Task }) => {
 				try {
 					await execa`kill -9 ${subprocess.pid!}`
 				} catch (error) {
-					console.error("Error killing process:", error)
+					console.error(
+						`${Date.now()} [cli#runUnitTest | ${task.language} / ${task.exercise}] Error killing process:`,
+						error,
+					)
 				}
 			}, UNIT_TEST_TIMEOUT)
 

evals/apps/cli/src/index.ts

@@ -1,7 +1,7 @@
 import { useState, useCallback, useRef } from "react"
 import { useQuery, keepPreviousData } from "@tanstack/react-query"
 
-import { RooCodeEventName, taskEventSchema, TokenUsage } from "@evals/types"
+import { TokenUsage, taskEventSchema, RooCodeEventName, EvalEventName } from "@evals/types"
 import { Run } from "@evals/db"
 
 import { getTasks } from "@/lib/server/tasks"
@@ -51,17 +51,17 @@ export const useRunStatus = (run: Run) => {
 			case RooCodeEventName.TaskStarted:
 				startTimes.current.set(taskId, Date.now())
 				break
-			case RooCodeEventName.TaskCompleted:
-			case RooCodeEventName.TaskAborted:
-				setTasksUpdatedAt(Date.now())
-				break
 			case RooCodeEventName.TaskTokenUsageUpdated: {
 				const startTime = startTimes.current.get(taskId)
 				const duration = startTime ? Date.now() - startTime : undefined
 				tokenUsage.current.set(taskId, { ...payload[1], duration })
 				setUsageUpdatedAt(Date.now())
 				break
 			}
+			case EvalEventName.Pass:
+			case EvalEventName.Fail:
+				setTasksUpdatedAt(Date.now())
+				break
 		}
 	}, [])
 

evals/apps/web/src/hooks/use-run-status.ts

@@ -4,7 +4,7 @@
 	"type": "module",
 	"exports": "./src/index.ts",
 	"scripts": {
-		"lint": "eslint src --ext ts --max-warnings=0",
+		"lint": "eslint src/**/*.ts --max-warnings=0",
 		"check-types": "tsc --noEmit",
 		"format": "prettier --write src",
 		"drizzle-kit": "dotenvx run -f ../../.env -- tsx node_modules/drizzle-kit/bin.cjs",
@@ -23,6 +23,7 @@
 		"@libsql/client": "^0.14.0",
 		"drizzle-orm": "^0.40.0",
 		"drizzle-zod": "^0.7.0",
+		"p-map": "^7.0.3",
 		"zod": "^3.24.2"
 	},
 	"devDependencies": {

evals/packages/db/package.json

@@ -1,5 +1,6 @@
 import { drizzle } from "drizzle-orm/libsql"
 import { eq } from "drizzle-orm"
+import pMap from "p-map"
 
 import { db as sourceDb } from "../src/db.js"
 import { schema } from "../src/schema.js"
@@ -52,29 +53,33 @@ const copyRun = async (runId: number) => {
 
 	console.log(`Copying ${tasks.length} tasks`)
 
-	for (const task of tasks) {
-		// eslint-disable-next-line @typescript-eslint/no-unused-vars
-		const { id: _, ...newTaskMetricsValues } = task.taskMetrics!
-		const [newTaskMetrics] = await destDb.insert(schema.taskMetrics).values(newTaskMetricsValues).returning()
-
-		if (!newTaskMetrics) {
-			throw new Error(`Failed to insert taskMetrics for task ${task.id}`)
-		}
-
-		// eslint-disable-next-line @typescript-eslint/no-unused-vars
-		const { id: __, ...newTaskValues } = task
-
-		const [newTask] = await destDb
-			.insert(schema.tasks)
-			.values({ ...newTaskValues, runId: newRun.id, taskMetricsId: newTaskMetrics.id })
-			.returning()
-
-		if (!newTask) {
-			throw new Error(`Failed to insert task ${task.id}`)
-		}
-	}
-
-	console.log(`Successfully copied run ${runId} with ${tasks.length} tasks`)
+	await pMap(
+		tasks,
+		async (task) => {
+			// eslint-disable-next-line @typescript-eslint/no-unused-vars
+			const { id: _, ...newTaskMetricsValues } = task.taskMetrics!
+			const [newTaskMetrics] = await destDb.insert(schema.taskMetrics).values(newTaskMetricsValues).returning()
+
+			if (!newTaskMetrics) {
+				throw new Error(`Failed to insert taskMetrics for task ${task.id}`)
+			}
+
+			// eslint-disable-next-line @typescript-eslint/no-unused-vars
+			const { id: __, ...newTaskValues } = task
+
+			const [newTask] = await destDb
+				.insert(schema.tasks)
+				.values({ ...newTaskValues, runId: newRun.id, taskMetricsId: newTaskMetrics.id })
+				.returning()
+
+			if (!newTask) {
+				throw new Error(`Failed to insert task ${task.id}`)
+			}
+		},
+		{ concurrency: 25 },
+	)
+
+	console.log(`\nSuccessfully copied run ${runId} with ${tasks.length} tasks`)
 }
 
 const main = async () => {

evals/packages/db/scripts/copy-run.mts

@@ -4,7 +4,7 @@
 	"type": "module",
 	"exports": "./src/index.ts",
 	"scripts": {
-		"lint": "eslint src --ext ts --max-warnings=0",
+		"lint": "eslint src/**/*.ts --max-warnings=0",
 		"check-types": "tsc --noEmit",
 		"format": "prettier --write src"
 	},

evals/packages/ipc/package.json

@@ -4,7 +4,7 @@
 	"type": "module",
 	"exports": "./src/index.ts",
 	"scripts": {
-		"lint": "eslint src --ext ts --max-warnings=0",
+		"lint": "eslint src/**/*.ts --max-warnings=0",
 		"check-types": "tsc --noEmit",
 		"test": "vitest --globals --run",
 		"format": "prettier --write src"