refactor: split ask flow state and view models

This commit is contained in:
Erik Kotsiuba
2026-04-26 02:17:11 +02:00
parent 8ea067dcde
commit 2618d6ea57
21 changed files with 1455 additions and 1173 deletions
+35 -170
View File
@@ -1,187 +1,52 @@
import type { ExtensionAPI } from "@mariozechner/pi-coding-agent";
import { Text, truncateToWidth } from "@mariozechner/pi-tui";
import { renderResultText } from "./result.ts";
import { AskParamsSchema } from "./schema.ts";
import { createInitialState } from "./state/create.ts";
import { collectValidationIssues } from "./state/normalize.ts";
import { summarizeResult, toAskResult } from "./state/result.ts";
import type {
AskParams,
AskQuestionInput,
AskResult,
AskValidationIssue,
} from "./types.ts";
import { UI_DIMENSIONS } from "./ui/constants.ts";
ExtensionAPI,
ExtensionContext,
} from "@mariozechner/pi-coding-agent";
import {
ASK_TOOL_DESCRIPTION,
ASK_TOOL_PROMPT_GUIDELINES,
invalidPayloadResponse,
nonInteractiveResponse,
renderAskToolCall,
renderAskToolResult,
successfulResponse,
validateParams,
} from "./ask-tool-helpers.ts";
import { AskParamsSchema } from "./schema.ts";
import type { AskParams } from "./types.ts";
import { runAskFlow } from "./ui/controller.ts";
export function registerAskTool(pi: ExtensionAPI) {
pi.registerTool({
name: "ask_user",
label: "Ask User",
description:
"Interactive clarification tool for cases where the next step depends on user preferences, missing requirements, or choosing between multiple valid directions. Ask a short structured interview, collect normalized answers, and continue using those answers explicitly instead of guessing. Supports single-select, multi-select, and preview-pane questions. Always include a machine-readable `value` for every option. Use `preview` only when every option includes `preview` text; descriptions alone are not enough.",
description: ASK_TOOL_DESCRIPTION,
promptSnippet:
"Clarify ambiguous or preference-sensitive decisions with a short interactive interview before proceeding",
promptGuidelines: [
"Use this tool before making preference-sensitive decisions about scope, tone, UX, naming, architecture, docs, or implementation direction.",
"When multiple valid directions exist, ask 1-3 concise questions instead of committing to one path on your own.",
"Prefer one focused decision per question and use short labels with 2-4 clear options.",
"Always include a non-empty `value` for every option.",
'Use `type: "single"` by default and `type: "multi"` only when several answers can genuinely apply.',
'Use `type: "preview"` only when every option includes non-empty `preview` text for the dedicated preview pane. Option descriptions do not satisfy this requirement.',
"After clarifying a note or follow-up question, prefer another structured ask_user follow-up if a choice is still needed instead of switching to plain-text multiple choice in chat.",
"When prior answers already narrow the branch, bundle the next 2-3 related unresolved decisions into one follow-up ask instead of issuing a long sequence of single-question asks.",
"Use one-at-a-time follow-up asks only when the next question materially depends on the previous answer.",
],
promptGuidelines: [...ASK_TOOL_PROMPT_GUIDELINES],
parameters: AskParamsSchema,
async execute(_toolCallId, params: AskParams, _signal, _onUpdate, ctx) {
const validation = validateParams(params);
if (!validation.ok) {
return {
content: [
{ type: "text", text: formatValidationError(validation.issues) },
],
details: errorResultDetails(params, validation.issues),
};
}
if (!ctx.hasUI) {
return {
content: [
{
type: "text",
text: formatNonInteractiveMessage(validation.state),
},
],
details: {
...toAskResult(validation.state),
cancelled: true,
},
};
}
const result = await runAskFlow(ctx, params);
return {
content: [{ type: "text", text: summarizeResult(result) }],
details: result,
};
},
renderCall(args, theme) {
const params = args as AskParams;
const labels = Array.isArray(params.questions)
? params.questions
.map(
(question: AskQuestionInput, index) =>
question.label || `Q${index + 1}`
)
.join(", ")
: "";
let text = theme.fg("toolTitle", theme.bold("ask_user "));
text += theme.fg("muted", `${params.questions?.length ?? 0} question(s)`);
if (labels) {
text += theme.fg(
"dim",
` (${truncateToWidth(labels, UI_DIMENSIONS.callLabelTruncateWidth)})`
);
}
return new Text(text, 0, 0);
},
renderResult(result, _options, theme) {
const details = result.details as AskResult | undefined;
if (!details) {
const text = result.content[0];
return new Text(text?.type === "text" ? text.text : "", 0, 0);
}
if (details.error) {
return new Text(theme.fg("warning", "Invalid input"), 0, 0);
}
if (details.cancelled) {
return new Text(theme.fg("warning", "Cancelled"), 0, 0);
}
return new Text(renderResultText(details), 0, 0);
},
execute: executeAskTool,
renderCall: renderAskToolCall,
renderResult: renderAskToolResult,
});
}
function errorResultDetails(
interface ExecuteContext extends Pick<ExtensionContext, "cwd" | "hasUI"> {}
async function executeAskTool(
_toolCallId: string,
params: AskParams,
issues: AskValidationIssue[]
): AskResult {
try {
const state = createInitialState(params);
return {
...toAskResult(state),
cancelled: true,
error: {
kind: "invalid_input",
issues,
},
};
} catch {
return {
title: params.title,
cancelled: true,
mode: "submit",
questions: [],
answers: {},
error: {
kind: "invalid_input",
issues,
},
};
_signal: AbortSignal | undefined,
_onUpdate: unknown,
ctx: ExecuteContext
) {
const validation = validateParams(params);
if (!validation.ok) {
return invalidPayloadResponse(params, validation.issues);
}
}
function validateParams(
params: AskParams
):
| { ok: true; state: ReturnType<typeof createInitialState> }
| { ok: false; issues: AskValidationIssue[] } {
const issues = collectValidationIssues(params);
if (issues.length > 0) {
return {
ok: false,
issues,
};
if (!ctx.hasUI) {
return nonInteractiveResponse(validation.state);
}
return {
ok: true,
state: createInitialState(params),
};
}
function formatValidationError(issues: AskValidationIssue[]): string {
const lines = [
"Invalid ask_user payload:",
...issues.map((issue) => `- ${issue.path}: ${issue.message}`),
];
return lines.join("\n");
}
function formatNonInteractiveMessage(
state: ReturnType<typeof createInitialState>
): string {
const lines = [
"Needs user input: ask_user requires interactive UI.",
"Run same tool call in interactive session, or ask user these questions manually:",
];
for (const [index, question] of state.questions.entries()) {
lines.push(`${index + 1}. ${question.label}: ${question.prompt}`);
for (const option of question.options) {
lines.push(` - ${option.label} [${option.value}]`);
}
lines.push(" - Type your own [custom]");
}
lines.push(
"details.questions contains normalized pending questions. details.answers stays empty until user responds."
);
return lines.join("\n");
const result = await runAskFlow(ctx as never, params);
return successfulResponse(result);
}