Files
llama.cpp/tools/ui/tests/unit/partial-tool-call-cleanup.test.ts
T
Aleksander GrygierandGitHub 32beb244f5 ui: Agentic Content UX improvements (#25450)
* feat: Add shimmer text animation for processing state indicators

* feat: Redesign CollapsibleContentBlock component with improved UX

* feat: Add conditional setting display support with dependsOn field

* feat: Add showAgenticTurnStats setting for per-turn statistics

* feat: Update ChatMessageAgenticContent with improved UI and new features

* feat: Enhance file read tool UI/UX

* feat: Refine styling of collapsible content and code preview blocks

* feat: add terminal variant to CollapsibleContentBlock

* feat: add built-in tools UI registry

* feat: extract ChatMessageReasoningBlock and ChatMessageToolCallBlock

* refactor: simplify ChatMessageAgenticContent to use extracted blocks

* fix: correct markdown content block margin spacing

* fix: reorganize SettingsChatFields layout and reset button positioning

* fix: use direct map access in agentic store session methods

* refactor: remove reasoning preview/throttle system from CollapsibleContentBlock

* feat: add auto-scroll to reasoning block and remove showThoughtInProgress

* feat: add ChatMessageToolCallDateTime component and support for new tool types

* feat: improve auto-scroll reliability in reasoning block with RAF coalescing and MutationObserver

* feat: show MCP server favicon for tools without a built-in icon

* feat: add search-results parsing utilities and tests

* feat: add ChatMessageToolCallSearchResults component

* feat: integrate search results rendering into ChatMessageAgenticContent

* feat: display tool call input alongside output in ChatMessageToolCallBlock

* style: use muted foreground color in reasoning block content

* chore: Format

* feat: Refine reasoning block layout and make pending thoughts display configurable

* feat: Stream tool call code blocks with auto-scroll and handle partial JSON

* feat: add streaming permission gate infrastructure

* feat: wire permission gate into the agentic loop

* fix: bail out on abort and skip already-approved tool calls

* fix: clear partial tool calls on abort and savePartialResponse

* test: cover partial tool call cleanup end-to-end

* refactor: Remove streaming permission gate logic

* fix: Correct autoscroll and streaming gates for tool calls and reasoning blocks

* refactor: Chat Message Assistant componentization

* fix: Show health metadata for disabled MCP servers and promote connections on enable

* fix: Inherit global enabled state for missing MCP per-chat overrides

* refactor: Cleanup

* refactor: Split ChatMessageToolCallBlock into dedicated components

* feat: Add live streaming and auto-scroll for tool execution output

* feat: Add line numbers and change markers to file edit diffs

* chore: Formatting

* feat: Add type definitions and utilities for recommended MCP servers

* feat: Add recommended MCP servers configuration and storage key

* feat: Add McpServerCardCompact component for recommended servers

* feat: Add recommended servers section to Add New Server dialog

* feat: Update McpServerForm to support authorization requirements

* feat: Add select-none classes for text selection prevention

* feat: Add recommended MCP server icon assets

* refactor: Store dismissed MCP recommendations as a boolean flag

* feat: Render tool results as JSON or Markdown based on detected content type

* feat: UI improvement

* feat: Render search block early and update heading to show execution state

* fix: Prevent non-web-search tools from triggering the search UI block

* refactor: Cleanup

* refactor: Extract hardcoded icon size classes into shared constants

* refactor: Extract hardcoded tool result separator into a shared constant

* refactor: Tool Calls UI/logic

* refactor: Cleanup

* refactor: Cleanup

* refactor: Cleanup
2026-07-15 20:31:45 +02:00

96 lines
3.5 KiB
TypeScript

import { describe, it, expect } from 'vitest';
import { MessageRole } from '$lib/enums';
import { deriveAgenticSections } from '$lib/utils/agentic';
import type { DatabaseMessage } from '$lib/types/database';
function makeAssistant(overrides: Partial<DatabaseMessage> = {}): DatabaseMessage {
return {
id: overrides.id ?? 'ast-1',
convId: 'conv-1',
type: 'text',
timestamp: Date.now(),
role: MessageRole.ASSISTANT,
content: overrides.content ?? '',
parent: null,
children: [],
...overrides
} as DatabaseMessage;
}
// Mirrors the filter inside ChatService.convertDbMessageToApiChatMessageData:
// a partial tool call captured mid-stream must not survive into the next request
// payload. The fix in chatStore.savePartialResponseIfNeeded clears toolCalls to ''
// on Stop/Send immediately, mirroring what the agentic flow already does in
// onAssistantTurnComplete(...undefined).
function buildApiToolCalls(message: DatabaseMessage): unknown[] | undefined {
if (!message.toolCalls) return undefined;
try {
const parsed = JSON.parse(message.toolCalls);
return Array.isArray(parsed) && parsed.length > 0 ? parsed : undefined;
} catch {
return undefined;
}
}
describe('partial tool call cleanup', () => {
// Reproduces the broken payload from the user's screenshot: model was
// streaming a tool call whose arguments JSON was cut mid-string. The outer
// envelope still parses, but the arguments themselves are invalid JSON and
// the server rejects the request.
it('marks a partial tool call payload as unsafe to re-send', () => {
const message = makeAssistant({
content: 'partial reasoning',
toolCalls: JSON.stringify([
{
id: 'call_1',
type: 'function',
function: {
name: 'exec_shell_command',
arguments: '{"command":`grep -n \\"read_to\\" ` /Users'
}
}
])
});
const apiToolCalls = buildApiToolCalls(message);
// The bug: even though arguments are invalid, the outer array parses and
// the request gets sent. Function arguments must be parseable JSON on their
// own for the server to execute the tool.
expect(apiToolCalls).toBeDefined();
const args = (apiToolCalls![0] as { function: { arguments: string } }).function.arguments;
expect(() => JSON.parse(args)).toThrow();
});
// After Stop, savePartialResponseIfNeeded clears toolCalls and the agentic
// flow does the same in its silent-return detection. The next request reads
// toolCalls = '' and the conversion drops the field entirely so the server
// never sees the half-streamed call.
it('drops tool_calls from the API request after toolCalls is cleared', () => {
const clearedMessage = makeAssistant({
content: 'partial reasoning',
toolCalls: ''
});
const apiToolCalls = buildApiToolCalls(clearedMessage);
expect(apiToolCalls).toBeUndefined();
});
// The cleanup path keeps the partial reasoning content visible in the UI;
// only the tool_calls field is reset. deriveAgenticSections should still
// surface the reasoning as interrupted (no content / no tool calls behind
// it) without resurrecting the dead tool call block.
it('keeps reasoning content visible after cleanup, without a tool call block', () => {
const cleared = makeAssistant({
content: '',
reasoningContent: 'thinking about read_to',
toolCalls: ''
});
const sections = deriveAgenticSections(cleared);
expect(sections).toHaveLength(1);
expect(sections[0].type).toBe('reasoning');
expect(sections.some((s) => s.type.includes('tool_call'))).toBe(false);
});
});