managarten/apps/api/src/lib/search.ts
Till JS e82851985b feat(questions): deep-research module — mana-search + mana-llm pipeline
End-to-end deep-research feature for the questions module: a fire-and-
forget orchestrator in apps/api that plans sub-queries with mana-llm,
retrieves sources via mana-search (with optional Readability extraction),
and streams a structured synthesis back to the web app over SSE.

Backend (apps/api/src/modules/research):
- schema.ts: pgSchema('research') with research_results + sources
- orchestrator.ts: three-phase pipeline (plan / retrieve / synthesise)
  with depth-aware config (quick=1×, standard=3×, deep=6× sub-queries)
- pubsub.ts: in-process event bus, single-node, swappable for Redis
- routes.ts: POST /start (202, fire-and-forget), GET /:id/stream (SSE),
  POST /start-sync (test only), GET /:id, GET /:id/sources
- Credit gating via @mana/shared-hono/credits — validate up-front,
  consume best-effort on `done`. Failed runs cost nothing.

Helpers (apps/api/src/lib):
- llm.ts: llmJson() + llmStream() over mana-llm OpenAI-compat API
- search.ts: webSearch() + bulkExtract() over mana-search Go service
- responses.ts: shared errorResponse / listResponse / validationError

Schema deployment:
- drizzle.config.ts (research-scoped) + drizzle/research/0000_init.sql
  hand-authored migration, deployable via psql -f or drizzle-kit push.
- drizzle-kit added as devDep with db:generate / db:push scripts.

Web client (apps/mana/apps/web/src/lib/api/research.ts):
- Typed start() / get() / listSources() / streamProgress(). The stream
  uses fetch + ReadableStream (not EventSource) so we can attach the
  JWT via Authorization header. Special-cases 402 for friendly toast.
- New PUBLIC_MANA_API_URL plumbing in hooks.server.ts + config.ts.

Module store (modules/questions/stores/answers.svelte.ts):
- New write-side store with createManual / startResearch / accept /
  softDelete. startResearch creates an optimistic empty answer, opens
  the SSE stream, debounces token deltas in 100ms batches into the
  encrypted local row, and on `done` replaces the streamed text with
  the parsed { summary, keyPoints, followUps } payload + citations
  resolved against research.sources.id.

Citation rendering (modules/questions/components/AnswerCitations.svelte):
- Tokenises [n] markers in the answer body into clickable pills with
  hover popovers showing title / host / snippet / external link.
- Lazy-loaded via a session-scoped source cache (stores/sources.svelte.ts)
  that deduplicates concurrent fetches.

UI (routes/(app)/questions/[id]/+page.svelte):
- Recherche card with three-state button (start / cancel / re-run),
  animated phase indicator, source counter.
- Confirmation dialog warning about web/LLM transmission since the
  question itself is locally encrypted.
- Toasts for success / error / cancel via @mana/shared-ui/toast.
- Re-run flow soft-deletes prior research-driven answers but keeps
  manual ones intact.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-08 22:15:35 +02:00

120 lines
3 KiB
TypeScript

/**
* Thin client for the mana-search Go service.
*
* Two helpers, scoped to what the research orchestrator needs:
*
* webSearch() — POST /api/v1/search, returns ranked SearXNG results.
* bulkExtract() — POST /api/v1/extract/bulk, returns Readability text per URL.
*
* Internal service-to-service calls — no auth on the wire (private network).
*/
const SEARCH_URL = process.env.MANA_SEARCH_URL || 'http://localhost:3021';
export interface SearchHit {
url: string;
title: string;
snippet: string;
engine: string;
score: number;
publishedDate?: string;
category: string;
}
export interface ExtractedContent {
title: string;
description?: string;
author?: string;
publishedDate?: string;
siteName?: string;
text: string;
wordCount: number;
}
export interface BulkExtractResult {
url: string;
success: boolean;
content?: ExtractedContent;
error?: string;
}
export class SearchError extends Error {
constructor(
message: string,
public readonly status?: number
) {
super(message);
this.name = 'SearchError';
}
}
export interface WebSearchOptions {
query: string;
limit?: number;
categories?: string[]; // 'general' | 'news' | 'science' | 'it'
language?: string;
signal?: AbortSignal;
}
/** Run one SearXNG query via mana-search and return normalised hits. */
export async function webSearch(opts: WebSearchOptions): Promise<SearchHit[]> {
const res = await fetch(`${SEARCH_URL}/api/v1/search`, {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({
query: opts.query,
options: {
limit: opts.limit ?? 10,
categories: opts.categories,
language: opts.language ?? 'de-DE',
},
}),
signal: opts.signal,
});
if (!res.ok) {
throw new SearchError(`mana-search returned ${res.status}`, res.status);
}
const data = (await res.json()) as { results?: SearchHit[] };
return data.results ?? [];
}
/**
* Extract Readability content for a batch of URLs in parallel server-side.
* mana-search caps at 20 URLs per call; we slice if more come in.
*/
export async function bulkExtract(
urls: string[],
opts: { maxLength?: number; concurrency?: number; signal?: AbortSignal } = {}
): Promise<BulkExtractResult[]> {
if (urls.length === 0) return [];
const batches: string[][] = [];
for (let i = 0; i < urls.length; i += 20) batches.push(urls.slice(i, i + 20));
const all: BulkExtractResult[] = [];
for (const batch of batches) {
const res = await fetch(`${SEARCH_URL}/api/v1/extract/bulk`, {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({
urls: batch,
concurrency: opts.concurrency ?? 5,
options: {
maxLength: opts.maxLength ?? 8000,
},
}),
signal: opts.signal,
});
if (!res.ok) {
throw new SearchError(`mana-search bulk extract returned ${res.status}`, res.status);
}
const data = (await res.json()) as { results?: BulkExtractResult[] };
if (data.results) all.push(...data.results);
}
return all;
}