import * as path from "path"
import { VectorStoreSearchResult, SearchFilter, IReranker, RerankerCandidate } from "./interfaces"
import { IEmbedder } from "./interfaces/embedder"
import { IVectorStore } from "./interfaces/vector-store"
import { CodeIndexConfigManager } from "./config-manager"
import { CodeIndexStateManager } from "./state-manager"
import { applyQueryPrefill } from "./search/query-prefill"
import { getDefaultModelId } from "../shared/embeddingModels"
import { validateLimit, validateMinScore } from "./validate-search-params"
/**
* Service responsible for searching the code index.
*/
export class CodeIndexSearchService {
constructor(
private readonly configManager: CodeIndexConfigManager,
private readonly stateManager: CodeIndexStateManager,
private readonly embedder: IEmbedder,
private readonly vectorStore: IVectorStore,
private readonly reranker?: IReranker,
) {}
/**
* Searches the code index for relevant content.
* @param query The search query
* @param filter Search filter options
* @returns Array of search results
* @throws Error if the service is not properly configured or ready
*/
public async searchIndex(query: string, filter?: SearchFilter): Promise<VectorStoreSearchResult[]> {
if (!this.configManager.isFeatureEnabled || !this.configManager.isFeatureConfigured) {
throw new Error("Code index feature is disabled or not configured.")
}
// Get configuration values
const configMinScore = this.configManager.currentSearchMinScore
const configMaxResults = this.configManager.currentSearchMaxResults
const currentState = this.stateManager.getCurrentStatus().systemStatus
if (currentState !== "Indexed" && currentState !== "Indexing") {
// Allow search during Indexing too
throw new Error(`Code index is not ready for search. Current state: ${currentState}`)
}
// 使用统一的filter对象，不再单独处理directoryPrefix
// 所有过滤条件都通过pathFilters参数传递
try {
// Apply query prefill for embedding generation
const embedderProvider = this.configManager.currentEmbedderProvider
// Get modelId with fallback to default if not configured
const modelId = this.configManager.currentModelId ?? getDefaultModelId(embedderProvider)
const prefillQuery = applyQueryPrefill(query, embedderProvider, modelId)
// Generate embedding for query (with prefill if applicable)
const embeddingResponse = await this.embedder.createEmbeddings([prefillQuery])
const vector = embeddingResponse?.embeddings[0]
if (!vector) {
throw new Error("Failed to generate embedding for query.")
}
// Perform search - 防止调用方传入未验证的参数
const finalLimit = validateLimit(filter?.limit ?? configMaxResults)
const finalMinScore = validateMinScore(filter?.minScore ?? configMinScore)
let results = await this.vectorStore.search(vector, {
...filter,
minScore: finalMinScore,
limit: finalLimit
})
// 确保结果按分数降序排序
results.sort((a, b) => b.score - a.score)
// If reranker is enabled, rerank the results
if (this.reranker && results.length > 0) {
const candidates = results.map(r => ({
id: r.id,
content: r.payload?.codeChunk || '',
score: r.score,
payload: r.payload
}))
const reranked = await this.reranker.rerank(query, candidates)
// Convert back to VectorStoreSearchResult format, preserving original payload
results = reranked.map(r => ({
id: r.id,
score: r.score, // Use LLM score
payload: r.payload
}))
// Optional: Filter low-score results
const rerankerMinScore = this.configManager.rerankerConfig?.minScore
if (rerankerMinScore !== undefined) {
results = results.filter(r => r.score >= rerankerMinScore)
}
}
return results
} catch (error) {
console.error("[CodeIndexSearchService] Error during search:", error)
this.stateManager.setSystemState("Error", `Search failed: ${(error as Error).message}`)
throw error // Re-throw the error after setting state
}
}
}
