/**
* Local copy of the core Roo Code defaults we rely on for this library.
* In Roo Code proper these come from `@roo-code/types` as `CODEBASE_INDEX_DEFAULTS`.
*/
const CODEBASE_INDEX_DEFAULTS = {
DEFAULT_SEARCH_MIN_SCORE: 0.4,
DEFAULT_SEARCH_RESULTS: 50,
} as const
/**
* Default configuration for the code index
*/
import { CodeIndexConfig } from '../interfaces/config'
export const DEFAULT_CONFIG: CodeIndexConfig = {
isEnabled: true,
embedderProvider: "ollama",
embedderModelId: "nomic-embed-text",
embedderModelDimension: 768,
embedderOllamaBaseUrl: "http://localhost:11434",
qdrantUrl: "http://localhost:6333",
vectorSearchMinScore: 0.1,
vectorSearchMaxResults: 20,
rerankerEnabled: false,
rerankerConcurrency: 3,
rerankerMaxRetries: 3,
rerankerRetryDelayMs: 1000,
summarizerProvider: 'ollama',
summarizerOllamaBaseUrl: 'http://localhost:11434',
summarizerOllamaModelId: 'qwen3-vl:4b-instruct',
summarizerOpenAiCompatibleBaseUrl: 'http://localhost:8080/v1',
summarizerOpenAiCompatibleModelId: 'gpt-4',
summarizerOpenAiCompatibleApiKey: '',
summarizerLanguage: 'English',
summarizerBatchSize: 2,
summarizerConcurrency: 2,
summarizerMaxRetries: 3,
summarizerRetryDelayMs: 1000
}
/**Parser */
export const MAX_BLOCK_CHARS = 2000
export const MIN_BLOCK_CHARS = 100
export const MIN_CHUNK_REMAINDER_CHARS = 200 // Minimum characters for the *next* chunk after a split
export const MAX_CHARS_TOLERANCE_FACTOR = 1.15 // 15% tolerance for max chars
/**Search */
/**
* @deprecated Use SEARCH_CONFIG from './search-config' instead
*/
export const DEFAULT_SEARCH_MIN_SCORE = CODEBASE_INDEX_DEFAULTS.DEFAULT_SEARCH_MIN_SCORE
/**
* @deprecated Use SEARCH_CONFIG from './search-config' instead
*/
export const DEFAULT_MAX_SEARCH_RESULTS = CODEBASE_INDEX_DEFAULTS.DEFAULT_SEARCH_RESULTS
/**File Watcher */
export const QDRANT_CODE_BLOCK_NAMESPACE = "f47ac10b-58cc-4372-a567-0e02b2c3d479"
export const MAX_FILE_SIZE_BYTES = 1 * 1024 * 1024 // 1MB
/**Directory Scanner */
export const MAX_LIST_FILES_LIMIT_CODE_INDEX = 50_000
export const BATCH_SEGMENT_THRESHOLD = 60 // Number of code segments to batch for embeddings/upserts (default for OpenAI)
export const MAX_BATCH_RETRIES = 3
export const INITIAL_RETRY_DELAY_MS = 500
// Dynamic batch sizes for different embedder types
export const EMBEDDER_BATCH_SIZES: { [key: string]: number } = {
"openai": 60,
"openai-compatible": 60,
"jina": 30,
"gemini": 40,
"mistral": 30,
"vercel-ai-gateway": 60,
"openrouter": 60,
"ollama": 20, // Smaller batch size for Ollama to prevent timeouts with large local models
}
/**
* Gets the optimal batch size for a specific embedder type or embedder instance
* @param embedderType The embedder provider type, or an embedder instance with optimalBatchSize property
* @returns The optimal batch size for the embedder
*/
export function getBatchSizeForEmbedder(embedder: any): number {
// Check if embedder has an optimalBatchSize property
if (embedder && typeof embedder.optimalBatchSize === 'number') {
return embedder.optimalBatchSize
}
// Check if embedder has an embedderInfo property with name
const embedderType = embedder?.embedderInfo?.name || embedder
return EMBEDDER_BATCH_SIZES[embedderType] || BATCH_SEGMENT_THRESHOLD
}
export const PARSING_CONCURRENCY = 10
export const MAX_PENDING_BATCHES = 20 // Maximum number of batches to accumulate before waiting
/**OpenAI Embedder */
export const MAX_BATCH_TOKENS = 100000
export const MAX_ITEM_TOKENS = 8191
export const BATCH_PROCESSING_CONCURRENCY = 10
/**Gemini Embedder */
export const GEMINI_MAX_ITEM_TOKENS = 2048
/**BatchProcessor Truncation - 截断降级功能用于处理超长文本 */
export const TRUNCATION_INITIAL_THRESHOLD = 800 // 初始截断阈值（chars）
export const TRUNCATION_REDUCTION_FACTOR = 0.7 // 每次降低 30%
export const MIN_TRUNCATION_THRESHOLD = 200 // 最小阈值
export const MAX_TRUNCATION_ATTEMPTS = 3 // 最大重试次数
export const INDIVIDUAL_PROCESSING_TIMEOUT_MS = 60000 // 降级处理超时（1分钟）
/**Feature Flags - 功能开关 */
export const ENABLE_TRUNCATION_FALLBACK = true // 是否启用截断降级功能
