Make shell output consistent. (QwenLM#4469)

Author: jacob314

packages/cli/src/ui/components/messages/ToolGroupMessage.tsx

@@ -11,6 +11,7 @@ import { ToolMessage } from './ToolMessage.js';
 import { ToolConfirmationMessage } from './ToolConfirmationMessage.js';
 import { Colors } from '../../colors.js';
 import { Config } from '@google/gemini-cli-core';
+import { SHELL_COMMAND_NAME } from '../../constants.js';
 
 interface ToolGroupMessageProps {
   groupId: number;
@@ -32,7 +33,9 @@ export const ToolGroupMessage: React.FC<ToolGroupMessageProps> = ({
   const hasPending = !toolCalls.every(
     (t) => t.status === ToolCallStatus.Success,
   );
-  const borderColor = hasPending ? Colors.AccentYellow : Colors.Gray;
+  const isShellCommand = toolCalls.some((t) => t.name === SHELL_COMMAND_NAME);
+  const borderColor =
+    hasPending || isShellCommand ? Colors.AccentYellow : Colors.Gray;
 
   const staticHeight = /* border */ 2 + /* marginBottom */ 1;
   // This is a bit of a magic number, but it accounts for the border and

packages/cli/src/ui/constants.ts

@@ -13,3 +13,5 @@ export const UI_WIDTH =
   EstimatedArtWidth + BOX_PADDING_X * 2 + BoxBorderWidth * 2; // ~63
 
 export const STREAM_DEBOUNCE_MS = 100;
+
+export const SHELL_COMMAND_NAME = 'Shell Command';

packages/cli/src/ui/hooks/shellCommandProcessor.test.ts

@@ -10,6 +10,7 @@ import { useShellCommandProcessor } from './shellCommandProcessor';
 import { Config, GeminiClient } from '@google/gemini-cli-core';
 import * as fs from 'fs';
 import EventEmitter from 'events';
+import { ToolCallStatus } from '../types';
 
 // Mock dependencies
 vi.mock('child_process');
@@ -104,8 +105,15 @@ describe('useShellCommandProcessor', () => {
 
     expect(addItemToHistoryMock).toHaveBeenCalledTimes(2);
     expect(addItemToHistoryMock.mock.calls[1][0]).toEqual({
-      type: 'info',
-      text: 'file1.txt\nfile2.txt',
+      type: 'tool_group',
+      tools: [
+        expect.objectContaining({
+          name: 'Shell Command',
+          description: 'ls -l',
+          status: ToolCallStatus.Success,
+          resultDisplay: 'file1.txt\nfile2.txt',
+        }),
+      ],
     });
     expect(geminiClientMock.addHistory).toHaveBeenCalledTimes(1);
   });
@@ -140,8 +148,16 @@ describe('useShellCommandProcessor', () => {
 
     expect(addItemToHistoryMock).toHaveBeenCalledTimes(2);
     expect(addItemToHistoryMock.mock.calls[1][0]).toEqual({
-      type: 'info',
-      text: '[Command produced binary output, which is not shown.]',
+      type: 'tool_group',
+      tools: [
+        expect.objectContaining({
+          name: 'Shell Command',
+          description: 'cat myimage.png',
+          status: ToolCallStatus.Success,
+          resultDisplay:
+            '[Command produced binary output, which is not shown.]',
+        }),
+      ],
     });
   });
 
@@ -172,8 +188,15 @@ describe('useShellCommandProcessor', () => {
 
     expect(addItemToHistoryMock).toHaveBeenCalledTimes(2);
     expect(addItemToHistoryMock.mock.calls[1][0]).toEqual({
-      type: 'error',
-      text: 'Command exited with code 127.\ncommand not found',
+      type: 'tool_group',
+      tools: [
+        expect.objectContaining({
+          name: 'Shell Command',
+          description: 'a-bad-command',
+          status: ToolCallStatus.Error,
+          resultDisplay: 'Command exited with code 127.\ncommand not found',
+        }),
+      ],
     });
   });
 });

packages/cli/src/ui/hooks/shellCommandProcessor.ts

@@ -6,13 +6,18 @@
 
 import { spawn } from 'child_process';
 import { StringDecoder } from 'string_decoder';
-import type { HistoryItemWithoutId } from '../types.js';
+import {
+  HistoryItemWithoutId,
+  IndividualToolCallDisplay,
+  ToolCallStatus,
+} from '../types.js';
 import { useCallback } from 'react';
 import { Config, GeminiClient } from '@google/gemini-cli-core';
 import { type PartListUnion } from '@google/genai';
 import { formatMemoryUsage } from '../utils/formatters.js';
 import { isBinary } from '../utils/textUtils.js';
 import { UseHistoryManagerReturn } from './useHistoryManager.js';
+import { SHELL_COMMAND_NAME } from '../constants.js';
 import crypto from 'crypto';
 import path from 'path';
 import os from 'os';
@@ -204,6 +209,7 @@ ${modelContent}
  * Hook to process shell commands.
  * Orchestrates command execution and updates history and agent context.
  */
+
 export const useShellCommandProcessor = (
   addItemToHistory: UseHistoryManagerReturn['addItem'],
   setPendingHistoryItem: React.Dispatch<
@@ -221,6 +227,7 @@ export const useShellCommandProcessor = (
       }
 
       const userMessageTimestamp = Date.now();
+      const callId = `shell-${userMessageTimestamp}`;
       addItemToHistory(
         { type: 'user_shell', text: rawQuery },
         userMessageTimestamp,
@@ -246,6 +253,20 @@ export const useShellCommandProcessor = (
       const execPromise = new Promise<void>((resolve) => {
         let lastUpdateTime = 0;
 
+        const initialToolDisplay: IndividualToolCallDisplay = {
+          callId,
+          name: SHELL_COMMAND_NAME,
+          description: rawQuery,
+          status: ToolCallStatus.Executing,
+          resultDisplay: '',
+          confirmationDetails: undefined,
+        };
+
+        setPendingHistoryItem({
+          type: 'tool_group',
+          tools: [initialToolDisplay],
+        });
+
         onDebugMessage(`Executing in ${targetDir}: ${commandToExecute}`);
         executeShellCommand(
           commandToExecute,
@@ -254,23 +275,22 @@ export const useShellCommandProcessor = (
           (streamedOutput) => {
             // Throttle pending UI updates to avoid excessive re-renders.
             if (Date.now() - lastUpdateTime > OUTPUT_UPDATE_INTERVAL_MS) {
-              setPendingHistoryItem({ type: 'info', text: streamedOutput });
+              setPendingHistoryItem({
+                type: 'tool_group',
+                tools: [
+                  { ...initialToolDisplay, resultDisplay: streamedOutput },
+                ],
+              });
               lastUpdateTime = Date.now();
             }
           },
           onDebugMessage,
         )
           .then((result) => {
-            // TODO(abhipatel12) - Consider updating pending item and using timeout to ensure
-            // there is no jump where intermediate output is skipped.
             setPendingHistoryItem(null);
 
-            let historyItemType: HistoryItemWithoutId['type'] = 'info';
             let mainContent: string;
 
-            // The context sent to the model utilizes a text tokenizer which means raw binary data is
-            // cannot be parsed and understood and thus would only pollute the context window and waste
-            // tokens.
             if (isBinary(result.rawOutput)) {
               mainContent =
                 '[Command produced binary output, which is not shown.]';
@@ -280,17 +300,19 @@ export const useShellCommandProcessor = (
             }
 
             let finalOutput = mainContent;
+            let finalStatus = ToolCallStatus.Success;
 
             if (result.error) {
-              historyItemType = 'error';
+              finalStatus = ToolCallStatus.Error;
               finalOutput = `${result.error.message}\n${finalOutput}`;
             } else if (result.aborted) {
+              finalStatus = ToolCallStatus.Canceled;
               finalOutput = `Command was cancelled.\n${finalOutput}`;
             } else if (result.signal) {
-              historyItemType = 'error';
+              finalStatus = ToolCallStatus.Error;
               finalOutput = `Command terminated by signal: ${result.signal}.\n${finalOutput}`;
             } else if (result.exitCode !== 0) {
-              historyItemType = 'error';
+              finalStatus = ToolCallStatus.Error;
               finalOutput = `Command exited with code ${result.exitCode}.\n${finalOutput}`;
             }
 
@@ -302,9 +324,18 @@ export const useShellCommandProcessor = (
               }
             }
 
+            const finalToolDisplay: IndividualToolCallDisplay = {
+              ...initialToolDisplay,
+              status: finalStatus,
+              resultDisplay: finalOutput,
+            };
+
             // Add the complete, contextual result to the local UI history.
             addItemToHistory(
-              { type: historyItemType, text: finalOutput },
+              {
+                type: 'tool_group',
+                tools: [finalToolDisplay],
+              } as HistoryItemWithoutId,
               userMessageTimestamp,
             );