rag-lite-ts 2.1.1 → 2.3.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +88 -5
- package/dist/{cli → cjs/cli}/indexer.js +73 -15
- package/dist/cjs/cli/ui-server.d.ts +5 -0
- package/dist/cjs/cli/ui-server.js +152 -0
- package/dist/{cli.js → cjs/cli.js} +25 -6
- package/dist/{core → cjs/core}/binary-index-format.js +6 -3
- package/dist/{core → cjs/core}/db.d.ts +56 -0
- package/dist/{core → cjs/core}/db.js +105 -0
- package/dist/{core → cjs/core}/ingestion.js +3 -0
- package/dist/cjs/core/knowledge-base-manager.d.ts +109 -0
- package/dist/cjs/core/knowledge-base-manager.js +256 -0
- package/dist/{core → cjs/core}/model-validator.js +1 -1
- package/dist/{core → cjs/core}/search-pipeline.js +1 -1
- package/dist/{core → cjs/core}/search.js +1 -1
- package/dist/cjs/core/vector-index-messages.d.ts +52 -0
- package/dist/cjs/core/vector-index-messages.js +5 -0
- package/dist/cjs/core/vector-index-worker.d.ts +6 -0
- package/dist/cjs/core/vector-index-worker.js +304 -0
- package/dist/cjs/core/vector-index.d.ts +107 -0
- package/dist/cjs/core/vector-index.js +344 -0
- package/dist/{factories → cjs/factories}/ingestion-factory.js +3 -7
- package/dist/{factories → cjs/factories}/search-factory.js +11 -0
- package/dist/{index-manager.d.ts → cjs/index-manager.d.ts} +23 -3
- package/dist/{index-manager.js → cjs/index-manager.js} +84 -15
- package/dist/{index.d.ts → cjs/index.d.ts} +2 -1
- package/dist/{index.js → cjs/index.js} +3 -1
- package/dist/esm/api-errors.d.ts +90 -0
- package/dist/esm/api-errors.js +320 -0
- package/dist/esm/cli/indexer.d.ts +11 -0
- package/dist/esm/cli/indexer.js +529 -0
- package/dist/esm/cli/search.d.ts +7 -0
- package/dist/esm/cli/search.js +332 -0
- package/dist/esm/cli/ui-server.d.ts +5 -0
- package/dist/esm/cli/ui-server.js +152 -0
- package/dist/esm/cli.d.ts +3 -0
- package/dist/esm/cli.js +548 -0
- package/dist/esm/config.d.ts +51 -0
- package/dist/esm/config.js +79 -0
- package/dist/esm/core/abstract-embedder.d.ts +125 -0
- package/dist/esm/core/abstract-embedder.js +264 -0
- package/dist/esm/core/actionable-error-messages.d.ts +60 -0
- package/dist/esm/core/actionable-error-messages.js +397 -0
- package/dist/esm/core/adapters.d.ts +93 -0
- package/dist/esm/core/adapters.js +139 -0
- package/dist/esm/core/batch-processing-optimizer.d.ts +155 -0
- package/dist/esm/core/batch-processing-optimizer.js +536 -0
- package/dist/esm/core/binary-index-format.d.ts +78 -0
- package/dist/esm/core/binary-index-format.js +294 -0
- package/dist/esm/core/chunker.d.ts +119 -0
- package/dist/esm/core/chunker.js +73 -0
- package/dist/esm/core/cli-database-utils.d.ts +53 -0
- package/dist/esm/core/cli-database-utils.js +239 -0
- package/dist/esm/core/config.d.ts +102 -0
- package/dist/esm/core/config.js +247 -0
- package/dist/esm/core/content-errors.d.ts +111 -0
- package/dist/esm/core/content-errors.js +362 -0
- package/dist/esm/core/content-manager.d.ts +335 -0
- package/dist/esm/core/content-manager.js +1476 -0
- package/dist/esm/core/content-performance-optimizer.d.ts +150 -0
- package/dist/esm/core/content-performance-optimizer.js +516 -0
- package/dist/esm/core/content-resolver.d.ts +104 -0
- package/dist/esm/core/content-resolver.js +285 -0
- package/dist/esm/core/cross-modal-search.d.ts +164 -0
- package/dist/esm/core/cross-modal-search.js +342 -0
- package/dist/esm/core/database-connection-manager.d.ts +109 -0
- package/dist/esm/core/database-connection-manager.js +310 -0
- package/dist/esm/core/db.d.ts +269 -0
- package/dist/esm/core/db.js +1000 -0
- package/dist/esm/core/embedder-factory.d.ts +154 -0
- package/dist/esm/core/embedder-factory.js +311 -0
- package/dist/esm/core/error-handler.d.ts +112 -0
- package/dist/esm/core/error-handler.js +239 -0
- package/dist/esm/core/index.d.ts +59 -0
- package/dist/esm/core/index.js +69 -0
- package/dist/esm/core/ingestion.d.ts +202 -0
- package/dist/esm/core/ingestion.js +904 -0
- package/dist/esm/core/interfaces.d.ts +408 -0
- package/dist/esm/core/interfaces.js +106 -0
- package/dist/esm/core/knowledge-base-manager.d.ts +109 -0
- package/dist/esm/core/knowledge-base-manager.js +256 -0
- package/dist/esm/core/lazy-dependency-loader.d.ts +147 -0
- package/dist/esm/core/lazy-dependency-loader.js +435 -0
- package/dist/esm/core/mode-detection-service.d.ts +150 -0
- package/dist/esm/core/mode-detection-service.js +565 -0
- package/dist/esm/core/mode-model-validator.d.ts +92 -0
- package/dist/esm/core/mode-model-validator.js +203 -0
- package/dist/esm/core/model-registry.d.ts +116 -0
- package/dist/esm/core/model-registry.js +411 -0
- package/dist/esm/core/model-validator.d.ts +217 -0
- package/dist/esm/core/model-validator.js +782 -0
- package/dist/esm/core/path-manager.d.ts +47 -0
- package/dist/esm/core/path-manager.js +71 -0
- package/dist/esm/core/raglite-paths.d.ts +121 -0
- package/dist/esm/core/raglite-paths.js +145 -0
- package/dist/esm/core/reranking-config.d.ts +42 -0
- package/dist/esm/core/reranking-config.js +147 -0
- package/dist/esm/core/reranking-factory.d.ts +92 -0
- package/dist/esm/core/reranking-factory.js +410 -0
- package/dist/esm/core/reranking-strategies.d.ts +310 -0
- package/dist/esm/core/reranking-strategies.js +650 -0
- package/dist/esm/core/resource-cleanup.d.ts +163 -0
- package/dist/esm/core/resource-cleanup.js +371 -0
- package/dist/esm/core/resource-manager.d.ts +212 -0
- package/dist/esm/core/resource-manager.js +564 -0
- package/dist/esm/core/search-pipeline.d.ts +111 -0
- package/dist/esm/core/search-pipeline.js +287 -0
- package/dist/esm/core/search.d.ts +141 -0
- package/dist/esm/core/search.js +320 -0
- package/dist/esm/core/streaming-operations.d.ts +145 -0
- package/dist/esm/core/streaming-operations.js +409 -0
- package/dist/esm/core/types.d.ts +66 -0
- package/dist/esm/core/types.js +6 -0
- package/dist/esm/core/universal-embedder.d.ts +177 -0
- package/dist/esm/core/universal-embedder.js +139 -0
- package/dist/esm/core/validation-messages.d.ts +99 -0
- package/dist/esm/core/validation-messages.js +334 -0
- package/dist/esm/core/vector-index-messages.d.ts +52 -0
- package/dist/esm/core/vector-index-messages.js +5 -0
- package/dist/esm/core/vector-index-worker.d.ts +6 -0
- package/dist/esm/core/vector-index-worker.js +304 -0
- package/dist/esm/core/vector-index.d.ts +107 -0
- package/dist/esm/core/vector-index.js +344 -0
- package/dist/esm/dom-polyfills.d.ts +6 -0
- package/dist/esm/dom-polyfills.js +37 -0
- package/dist/esm/factories/index.d.ts +27 -0
- package/dist/esm/factories/index.js +29 -0
- package/dist/esm/factories/ingestion-factory.d.ts +200 -0
- package/dist/esm/factories/ingestion-factory.js +473 -0
- package/dist/esm/factories/search-factory.d.ts +154 -0
- package/dist/esm/factories/search-factory.js +355 -0
- package/dist/esm/file-processor.d.ts +147 -0
- package/dist/esm/file-processor.js +963 -0
- package/dist/esm/index-manager.d.ts +136 -0
- package/dist/esm/index-manager.js +667 -0
- package/dist/esm/index.d.ts +76 -0
- package/dist/esm/index.js +112 -0
- package/dist/esm/indexer.d.ts +7 -0
- package/dist/esm/indexer.js +54 -0
- package/dist/esm/ingestion.d.ts +63 -0
- package/dist/esm/ingestion.js +124 -0
- package/dist/esm/mcp-server.d.ts +46 -0
- package/dist/esm/mcp-server.js +1820 -0
- package/dist/esm/multimodal/clip-embedder.d.ts +327 -0
- package/dist/esm/multimodal/clip-embedder.js +996 -0
- package/dist/esm/multimodal/index.d.ts +6 -0
- package/dist/esm/multimodal/index.js +6 -0
- package/dist/esm/preprocess.d.ts +19 -0
- package/dist/esm/preprocess.js +203 -0
- package/dist/esm/preprocessors/index.d.ts +17 -0
- package/dist/esm/preprocessors/index.js +38 -0
- package/dist/esm/preprocessors/mdx.d.ts +25 -0
- package/dist/esm/preprocessors/mdx.js +101 -0
- package/dist/esm/preprocessors/mermaid.d.ts +68 -0
- package/dist/esm/preprocessors/mermaid.js +329 -0
- package/dist/esm/preprocessors/registry.d.ts +56 -0
- package/dist/esm/preprocessors/registry.js +179 -0
- package/dist/esm/run-error-recovery-tests.d.ts +7 -0
- package/dist/esm/run-error-recovery-tests.js +101 -0
- package/dist/esm/search-standalone.d.ts +7 -0
- package/dist/esm/search-standalone.js +117 -0
- package/dist/esm/search.d.ts +99 -0
- package/dist/esm/search.js +177 -0
- package/dist/esm/test-utils.d.ts +18 -0
- package/dist/esm/test-utils.js +27 -0
- package/dist/esm/text/chunker.d.ts +33 -0
- package/dist/esm/text/chunker.js +279 -0
- package/dist/esm/text/embedder.d.ts +111 -0
- package/dist/esm/text/embedder.js +386 -0
- package/dist/esm/text/index.d.ts +8 -0
- package/dist/esm/text/index.js +9 -0
- package/dist/esm/text/preprocessors/index.d.ts +17 -0
- package/dist/esm/text/preprocessors/index.js +38 -0
- package/dist/esm/text/preprocessors/mdx.d.ts +25 -0
- package/dist/esm/text/preprocessors/mdx.js +101 -0
- package/dist/esm/text/preprocessors/mermaid.d.ts +68 -0
- package/dist/esm/text/preprocessors/mermaid.js +330 -0
- package/dist/esm/text/preprocessors/registry.d.ts +56 -0
- package/dist/esm/text/preprocessors/registry.js +180 -0
- package/dist/esm/text/reranker.d.ts +49 -0
- package/dist/esm/text/reranker.js +274 -0
- package/dist/esm/text/sentence-transformer-embedder.d.ts +96 -0
- package/dist/esm/text/sentence-transformer-embedder.js +340 -0
- package/dist/esm/text/tokenizer.d.ts +22 -0
- package/dist/esm/text/tokenizer.js +64 -0
- package/dist/esm/types.d.ts +83 -0
- package/dist/esm/types.js +3 -0
- package/dist/esm/utils/vector-math.d.ts +31 -0
- package/dist/esm/utils/vector-math.js +70 -0
- package/package.json +39 -14
- package/dist/core/vector-index.d.ts +0 -72
- package/dist/core/vector-index.js +0 -331
- /package/dist/{api-errors.d.ts → cjs/api-errors.d.ts} +0 -0
- /package/dist/{api-errors.js → cjs/api-errors.js} +0 -0
- /package/dist/{cli → cjs/cli}/indexer.d.ts +0 -0
- /package/dist/{cli → cjs/cli}/search.d.ts +0 -0
- /package/dist/{cli → cjs/cli}/search.js +0 -0
- /package/dist/{cli.d.ts → cjs/cli.d.ts} +0 -0
- /package/dist/{config.d.ts → cjs/config.d.ts} +0 -0
- /package/dist/{config.js → cjs/config.js} +0 -0
- /package/dist/{core → cjs/core}/abstract-embedder.d.ts +0 -0
- /package/dist/{core → cjs/core}/abstract-embedder.js +0 -0
- /package/dist/{core → cjs/core}/actionable-error-messages.d.ts +0 -0
- /package/dist/{core → cjs/core}/actionable-error-messages.js +0 -0
- /package/dist/{core → cjs/core}/adapters.d.ts +0 -0
- /package/dist/{core → cjs/core}/adapters.js +0 -0
- /package/dist/{core → cjs/core}/batch-processing-optimizer.d.ts +0 -0
- /package/dist/{core → cjs/core}/batch-processing-optimizer.js +0 -0
- /package/dist/{core → cjs/core}/binary-index-format.d.ts +0 -0
- /package/dist/{core → cjs/core}/chunker.d.ts +0 -0
- /package/dist/{core → cjs/core}/chunker.js +0 -0
- /package/dist/{core → cjs/core}/cli-database-utils.d.ts +0 -0
- /package/dist/{core → cjs/core}/cli-database-utils.js +0 -0
- /package/dist/{core → cjs/core}/config.d.ts +0 -0
- /package/dist/{core → cjs/core}/config.js +0 -0
- /package/dist/{core → cjs/core}/content-errors.d.ts +0 -0
- /package/dist/{core → cjs/core}/content-errors.js +0 -0
- /package/dist/{core → cjs/core}/content-manager.d.ts +0 -0
- /package/dist/{core → cjs/core}/content-manager.js +0 -0
- /package/dist/{core → cjs/core}/content-performance-optimizer.d.ts +0 -0
- /package/dist/{core → cjs/core}/content-performance-optimizer.js +0 -0
- /package/dist/{core → cjs/core}/content-resolver.d.ts +0 -0
- /package/dist/{core → cjs/core}/content-resolver.js +0 -0
- /package/dist/{core → cjs/core}/cross-modal-search.d.ts +0 -0
- /package/dist/{core → cjs/core}/cross-modal-search.js +0 -0
- /package/dist/{core → cjs/core}/database-connection-manager.d.ts +0 -0
- /package/dist/{core → cjs/core}/database-connection-manager.js +0 -0
- /package/dist/{core → cjs/core}/embedder-factory.d.ts +0 -0
- /package/dist/{core → cjs/core}/embedder-factory.js +0 -0
- /package/dist/{core → cjs/core}/error-handler.d.ts +0 -0
- /package/dist/{core → cjs/core}/error-handler.js +0 -0
- /package/dist/{core → cjs/core}/index.d.ts +0 -0
- /package/dist/{core → cjs/core}/index.js +0 -0
- /package/dist/{core → cjs/core}/ingestion.d.ts +0 -0
- /package/dist/{core → cjs/core}/interfaces.d.ts +0 -0
- /package/dist/{core → cjs/core}/interfaces.js +0 -0
- /package/dist/{core → cjs/core}/lazy-dependency-loader.d.ts +0 -0
- /package/dist/{core → cjs/core}/lazy-dependency-loader.js +0 -0
- /package/dist/{core → cjs/core}/mode-detection-service.d.ts +0 -0
- /package/dist/{core → cjs/core}/mode-detection-service.js +0 -0
- /package/dist/{core → cjs/core}/mode-model-validator.d.ts +0 -0
- /package/dist/{core → cjs/core}/mode-model-validator.js +0 -0
- /package/dist/{core → cjs/core}/model-registry.d.ts +0 -0
- /package/dist/{core → cjs/core}/model-registry.js +0 -0
- /package/dist/{core → cjs/core}/model-validator.d.ts +0 -0
- /package/dist/{core → cjs/core}/path-manager.d.ts +0 -0
- /package/dist/{core → cjs/core}/path-manager.js +0 -0
- /package/dist/{core → cjs/core}/raglite-paths.d.ts +0 -0
- /package/dist/{core → cjs/core}/raglite-paths.js +0 -0
- /package/dist/{core → cjs/core}/reranking-config.d.ts +0 -0
- /package/dist/{core → cjs/core}/reranking-config.js +0 -0
- /package/dist/{core → cjs/core}/reranking-factory.d.ts +0 -0
- /package/dist/{core → cjs/core}/reranking-factory.js +0 -0
- /package/dist/{core → cjs/core}/reranking-strategies.d.ts +0 -0
- /package/dist/{core → cjs/core}/reranking-strategies.js +0 -0
- /package/dist/{core → cjs/core}/resource-cleanup.d.ts +0 -0
- /package/dist/{core → cjs/core}/resource-cleanup.js +0 -0
- /package/dist/{core → cjs/core}/resource-manager.d.ts +0 -0
- /package/dist/{core → cjs/core}/resource-manager.js +0 -0
- /package/dist/{core → cjs/core}/search-pipeline.d.ts +0 -0
- /package/dist/{core → cjs/core}/search.d.ts +0 -0
- /package/dist/{core → cjs/core}/streaming-operations.d.ts +0 -0
- /package/dist/{core → cjs/core}/streaming-operations.js +0 -0
- /package/dist/{core → cjs/core}/types.d.ts +0 -0
- /package/dist/{core → cjs/core}/types.js +0 -0
- /package/dist/{core → cjs/core}/universal-embedder.d.ts +0 -0
- /package/dist/{core → cjs/core}/universal-embedder.js +0 -0
- /package/dist/{core → cjs/core}/validation-messages.d.ts +0 -0
- /package/dist/{core → cjs/core}/validation-messages.js +0 -0
- /package/dist/{dom-polyfills.d.ts → cjs/dom-polyfills.d.ts} +0 -0
- /package/dist/{dom-polyfills.js → cjs/dom-polyfills.js} +0 -0
- /package/dist/{factories → cjs/factories}/index.d.ts +0 -0
- /package/dist/{factories → cjs/factories}/index.js +0 -0
- /package/dist/{factories → cjs/factories}/ingestion-factory.d.ts +0 -0
- /package/dist/{factories → cjs/factories}/search-factory.d.ts +0 -0
- /package/dist/{file-processor.d.ts → cjs/file-processor.d.ts} +0 -0
- /package/dist/{file-processor.js → cjs/file-processor.js} +0 -0
- /package/dist/{indexer.d.ts → cjs/indexer.d.ts} +0 -0
- /package/dist/{indexer.js → cjs/indexer.js} +0 -0
- /package/dist/{ingestion.d.ts → cjs/ingestion.d.ts} +0 -0
- /package/dist/{ingestion.js → cjs/ingestion.js} +0 -0
- /package/dist/{mcp-server.d.ts → cjs/mcp-server.d.ts} +0 -0
- /package/dist/{mcp-server.js → cjs/mcp-server.js} +0 -0
- /package/dist/{multimodal → cjs/multimodal}/clip-embedder.d.ts +0 -0
- /package/dist/{multimodal → cjs/multimodal}/clip-embedder.js +0 -0
- /package/dist/{multimodal → cjs/multimodal}/index.d.ts +0 -0
- /package/dist/{multimodal → cjs/multimodal}/index.js +0 -0
- /package/dist/{preprocess.d.ts → cjs/preprocess.d.ts} +0 -0
- /package/dist/{preprocess.js → cjs/preprocess.js} +0 -0
- /package/dist/{preprocessors → cjs/preprocessors}/index.d.ts +0 -0
- /package/dist/{preprocessors → cjs/preprocessors}/index.js +0 -0
- /package/dist/{preprocessors → cjs/preprocessors}/mdx.d.ts +0 -0
- /package/dist/{preprocessors → cjs/preprocessors}/mdx.js +0 -0
- /package/dist/{preprocessors → cjs/preprocessors}/mermaid.d.ts +0 -0
- /package/dist/{preprocessors → cjs/preprocessors}/mermaid.js +0 -0
- /package/dist/{preprocessors → cjs/preprocessors}/registry.d.ts +0 -0
- /package/dist/{preprocessors → cjs/preprocessors}/registry.js +0 -0
- /package/dist/{run-error-recovery-tests.d.ts → cjs/run-error-recovery-tests.d.ts} +0 -0
- /package/dist/{run-error-recovery-tests.js → cjs/run-error-recovery-tests.js} +0 -0
- /package/dist/{search-standalone.d.ts → cjs/search-standalone.d.ts} +0 -0
- /package/dist/{search-standalone.js → cjs/search-standalone.js} +0 -0
- /package/dist/{search.d.ts → cjs/search.d.ts} +0 -0
- /package/dist/{search.js → cjs/search.js} +0 -0
- /package/dist/{test-utils.d.ts → cjs/test-utils.d.ts} +0 -0
- /package/dist/{test-utils.js → cjs/test-utils.js} +0 -0
- /package/dist/{text → cjs/text}/chunker.d.ts +0 -0
- /package/dist/{text → cjs/text}/chunker.js +0 -0
- /package/dist/{text → cjs/text}/embedder.d.ts +0 -0
- /package/dist/{text → cjs/text}/embedder.js +0 -0
- /package/dist/{text → cjs/text}/index.d.ts +0 -0
- /package/dist/{text → cjs/text}/index.js +0 -0
- /package/dist/{text → cjs/text}/preprocessors/index.d.ts +0 -0
- /package/dist/{text → cjs/text}/preprocessors/index.js +0 -0
- /package/dist/{text → cjs/text}/preprocessors/mdx.d.ts +0 -0
- /package/dist/{text → cjs/text}/preprocessors/mdx.js +0 -0
- /package/dist/{text → cjs/text}/preprocessors/mermaid.d.ts +0 -0
- /package/dist/{text → cjs/text}/preprocessors/mermaid.js +0 -0
- /package/dist/{text → cjs/text}/preprocessors/registry.d.ts +0 -0
- /package/dist/{text → cjs/text}/preprocessors/registry.js +0 -0
- /package/dist/{text → cjs/text}/reranker.d.ts +0 -0
- /package/dist/{text → cjs/text}/reranker.js +0 -0
- /package/dist/{text → cjs/text}/sentence-transformer-embedder.d.ts +0 -0
- /package/dist/{text → cjs/text}/sentence-transformer-embedder.js +0 -0
- /package/dist/{text → cjs/text}/tokenizer.d.ts +0 -0
- /package/dist/{text → cjs/text}/tokenizer.js +0 -0
- /package/dist/{types.d.ts → cjs/types.d.ts} +0 -0
- /package/dist/{types.js → cjs/types.js} +0 -0
- /package/dist/{utils → cjs/utils}/vector-math.d.ts +0 -0
- /package/dist/{utils → cjs/utils}/vector-math.js +0 -0
|
@@ -0,0 +1,239 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* CLI Database Utilities - Database access helpers for CLI commands
|
|
3
|
+
* Provides database locking detection and retry mechanisms for CLI operations
|
|
4
|
+
* Prevents conflicts between CLI commands and long-running processes like MCP server
|
|
5
|
+
*/
|
|
6
|
+
import { DatabaseConnectionManager } from './database-connection-manager.js';
|
|
7
|
+
import { existsSync } from 'fs';
|
|
8
|
+
/**
|
|
9
|
+
* Default options for CLI database access
|
|
10
|
+
*/
|
|
11
|
+
const DEFAULT_CLI_OPTIONS = {
|
|
12
|
+
maxWaitMs: 10000, // 10 seconds
|
|
13
|
+
retryIntervalMs: 500, // 0.5 seconds
|
|
14
|
+
showProgress: true,
|
|
15
|
+
commandName: 'CLI command'
|
|
16
|
+
};
|
|
17
|
+
/**
|
|
18
|
+
* Wait for database to become available for CLI operations
|
|
19
|
+
* Provides user-friendly progress messages and error handling
|
|
20
|
+
*/
|
|
21
|
+
export async function waitForCLIDatabaseAccess(dbPath, options = {}) {
|
|
22
|
+
const opts = { ...DEFAULT_CLI_OPTIONS, ...options };
|
|
23
|
+
// Check if database file exists
|
|
24
|
+
if (!existsSync(dbPath)) {
|
|
25
|
+
throw new Error(`Database file not found: ${dbPath}\n` +
|
|
26
|
+
`Please run 'raglite ingest <path>' first to create the database.`);
|
|
27
|
+
}
|
|
28
|
+
const startTime = Date.now();
|
|
29
|
+
let attempts = 0;
|
|
30
|
+
let lastError = null;
|
|
31
|
+
while (Date.now() - startTime < opts.maxWaitMs) {
|
|
32
|
+
attempts++;
|
|
33
|
+
try {
|
|
34
|
+
// Try to get database access
|
|
35
|
+
await DatabaseConnectionManager.waitForDatabaseAccess(dbPath, 1000);
|
|
36
|
+
if (opts.showProgress && attempts > 1) {
|
|
37
|
+
console.log(`✅ Database is now available (after ${attempts} attempts)`);
|
|
38
|
+
}
|
|
39
|
+
return; // Success!
|
|
40
|
+
}
|
|
41
|
+
catch (error) {
|
|
42
|
+
lastError = error;
|
|
43
|
+
if (lastError.message.includes('SQLITE_BUSY') || lastError.message.includes('busy')) {
|
|
44
|
+
// Database is busy - show progress and retry
|
|
45
|
+
if (opts.showProgress) {
|
|
46
|
+
if (attempts === 1) {
|
|
47
|
+
console.log(`⏳ Database is busy, waiting for access...`);
|
|
48
|
+
console.log(` This usually happens when another process is using the database.`);
|
|
49
|
+
console.log(` Common causes:`);
|
|
50
|
+
console.log(` • MCP server is running`);
|
|
51
|
+
console.log(` • Another CLI command is in progress`);
|
|
52
|
+
console.log(` • Long-running ingestion process`);
|
|
53
|
+
console.log('');
|
|
54
|
+
}
|
|
55
|
+
else if (attempts % 4 === 0) {
|
|
56
|
+
const elapsed = ((Date.now() - startTime) / 1000).toFixed(1);
|
|
57
|
+
console.log(` Still waiting... (${elapsed}s elapsed, attempt ${attempts})`);
|
|
58
|
+
}
|
|
59
|
+
}
|
|
60
|
+
// Wait before retrying
|
|
61
|
+
await new Promise(resolve => setTimeout(resolve, opts.retryIntervalMs));
|
|
62
|
+
continue;
|
|
63
|
+
}
|
|
64
|
+
else {
|
|
65
|
+
// Other error - don't retry
|
|
66
|
+
throw new Error(`Failed to access database: ${lastError.message}\n` +
|
|
67
|
+
`Please check that the database file is not corrupted and you have proper permissions.`);
|
|
68
|
+
}
|
|
69
|
+
}
|
|
70
|
+
}
|
|
71
|
+
// Timeout reached
|
|
72
|
+
const elapsed = ((Date.now() - startTime) / 1000).toFixed(1);
|
|
73
|
+
throw new Error(`Database is still busy after ${elapsed} seconds.\n` +
|
|
74
|
+
`\n` +
|
|
75
|
+
`This might be because:\n` +
|
|
76
|
+
`• Another process is using the database (MCP server, long ingestion, etc.)\n` +
|
|
77
|
+
`• The database is locked due to an interrupted operation\n` +
|
|
78
|
+
`\n` +
|
|
79
|
+
`Solutions:\n` +
|
|
80
|
+
`• Wait for other operations to complete\n` +
|
|
81
|
+
`• Stop the MCP server if running\n` +
|
|
82
|
+
`• Restart your terminal/process\n` +
|
|
83
|
+
`• As a last resort, restart your computer\n` +
|
|
84
|
+
`\n` +
|
|
85
|
+
`Last error: ${lastError?.message || 'Unknown error'}`);
|
|
86
|
+
}
|
|
87
|
+
/**
|
|
88
|
+
* Execute a CLI operation with database access protection
|
|
89
|
+
* Automatically handles database locking and provides user feedback
|
|
90
|
+
*/
|
|
91
|
+
export async function withCLIDatabaseAccess(dbPath, operation, options = {}) {
|
|
92
|
+
const opts = { ...DEFAULT_CLI_OPTIONS, ...options };
|
|
93
|
+
try {
|
|
94
|
+
// Wait for database access
|
|
95
|
+
await waitForCLIDatabaseAccess(dbPath, opts);
|
|
96
|
+
// Execute the operation
|
|
97
|
+
return await operation();
|
|
98
|
+
}
|
|
99
|
+
catch (error) {
|
|
100
|
+
if (error instanceof Error) {
|
|
101
|
+
// Enhance error message with CLI context
|
|
102
|
+
const enhancedMessage = `${opts.commandName} failed: ${error.message}\n` +
|
|
103
|
+
`\n` +
|
|
104
|
+
`If this error persists:\n` +
|
|
105
|
+
`• Check that no other RAG-lite processes are running\n` +
|
|
106
|
+
`• Verify database file permissions\n` +
|
|
107
|
+
`• Try running the command again\n`;
|
|
108
|
+
throw new Error(enhancedMessage);
|
|
109
|
+
}
|
|
110
|
+
throw error;
|
|
111
|
+
}
|
|
112
|
+
}
|
|
113
|
+
/**
|
|
114
|
+
* Check if database is currently busy (non-blocking)
|
|
115
|
+
* Useful for showing warnings or status information
|
|
116
|
+
*/
|
|
117
|
+
export async function isDatabaseBusy(dbPath) {
|
|
118
|
+
try {
|
|
119
|
+
await DatabaseConnectionManager.waitForDatabaseAccess(dbPath, 100);
|
|
120
|
+
return { isBusy: false };
|
|
121
|
+
}
|
|
122
|
+
catch (error) {
|
|
123
|
+
if (error instanceof Error && (error.message.includes('SQLITE_BUSY') ||
|
|
124
|
+
error.message.includes('busy'))) {
|
|
125
|
+
return {
|
|
126
|
+
isBusy: true,
|
|
127
|
+
reason: 'Database is currently in use by another process',
|
|
128
|
+
suggestions: [
|
|
129
|
+
'Wait for other operations to complete',
|
|
130
|
+
'Stop MCP server if running',
|
|
131
|
+
'Check for other CLI commands in progress'
|
|
132
|
+
]
|
|
133
|
+
};
|
|
134
|
+
}
|
|
135
|
+
return {
|
|
136
|
+
isBusy: true,
|
|
137
|
+
reason: `Database access error: ${error instanceof Error ? error.message : String(error)}`,
|
|
138
|
+
suggestions: [
|
|
139
|
+
'Check database file permissions',
|
|
140
|
+
'Verify database file is not corrupted',
|
|
141
|
+
'Ensure you have read/write access to the database directory'
|
|
142
|
+
]
|
|
143
|
+
};
|
|
144
|
+
}
|
|
145
|
+
}
|
|
146
|
+
/**
|
|
147
|
+
* Show database status information for debugging
|
|
148
|
+
* Useful for troubleshooting CLI issues
|
|
149
|
+
*/
|
|
150
|
+
export async function showDatabaseStatus(dbPath) {
|
|
151
|
+
console.log(`📊 Database Status: ${dbPath}`);
|
|
152
|
+
console.log('');
|
|
153
|
+
// Check file existence
|
|
154
|
+
if (!existsSync(dbPath)) {
|
|
155
|
+
console.log('❌ Database file does not exist');
|
|
156
|
+
console.log(' Run "raglite ingest <path>" to create the database');
|
|
157
|
+
return;
|
|
158
|
+
}
|
|
159
|
+
// Check file stats
|
|
160
|
+
try {
|
|
161
|
+
const fs = await import('fs');
|
|
162
|
+
const stats = fs.statSync(dbPath);
|
|
163
|
+
console.log(`📁 File size: ${(stats.size / 1024).toFixed(1)} KB`);
|
|
164
|
+
console.log(`📅 Last modified: ${stats.mtime.toLocaleString()}`);
|
|
165
|
+
}
|
|
166
|
+
catch (error) {
|
|
167
|
+
console.log(`⚠️ Cannot read file stats: ${error}`);
|
|
168
|
+
}
|
|
169
|
+
// Check database access
|
|
170
|
+
const busyStatus = await isDatabaseBusy(dbPath);
|
|
171
|
+
if (busyStatus.isBusy) {
|
|
172
|
+
console.log(`🔒 Status: BUSY`);
|
|
173
|
+
console.log(` Reason: ${busyStatus.reason}`);
|
|
174
|
+
if (busyStatus.suggestions) {
|
|
175
|
+
console.log(' Suggestions:');
|
|
176
|
+
busyStatus.suggestions.forEach(suggestion => {
|
|
177
|
+
console.log(` • ${suggestion}`);
|
|
178
|
+
});
|
|
179
|
+
}
|
|
180
|
+
}
|
|
181
|
+
else {
|
|
182
|
+
console.log(`✅ Status: AVAILABLE`);
|
|
183
|
+
}
|
|
184
|
+
// Show connection manager stats
|
|
185
|
+
const connectionStats = DatabaseConnectionManager.getConnectionStats();
|
|
186
|
+
if (connectionStats.totalConnections > 0) {
|
|
187
|
+
console.log('');
|
|
188
|
+
console.log(`🔗 Active connections: ${connectionStats.totalConnections}`);
|
|
189
|
+
connectionStats.connections.forEach((conn, index) => {
|
|
190
|
+
console.log(` ${index + 1}. ${conn.path}`);
|
|
191
|
+
console.log(` References: ${conn.refCount}`);
|
|
192
|
+
console.log(` Last accessed: ${conn.lastAccessed.toLocaleString()}`);
|
|
193
|
+
console.log(` Idle time: ${(conn.idleTime / 1000).toFixed(1)}s`);
|
|
194
|
+
});
|
|
195
|
+
}
|
|
196
|
+
console.log('');
|
|
197
|
+
}
|
|
198
|
+
/**
|
|
199
|
+
* Force cleanup of database connections (emergency use only)
|
|
200
|
+
* Use with caution - only for recovery from stuck states
|
|
201
|
+
*/
|
|
202
|
+
export async function forceCleanupDatabase(dbPath) {
|
|
203
|
+
console.log(`🚨 Force cleaning up database connections: ${dbPath}`);
|
|
204
|
+
try {
|
|
205
|
+
await DatabaseConnectionManager.forceCloseConnection(dbPath);
|
|
206
|
+
console.log('✅ Force cleanup completed');
|
|
207
|
+
}
|
|
208
|
+
catch (error) {
|
|
209
|
+
console.log(`⚠️ Force cleanup failed: ${error}`);
|
|
210
|
+
console.log('You may need to restart the process or reboot your system');
|
|
211
|
+
}
|
|
212
|
+
}
|
|
213
|
+
/**
|
|
214
|
+
* Graceful shutdown helper for CLI commands
|
|
215
|
+
* Ensures proper cleanup when CLI commands are interrupted
|
|
216
|
+
*/
|
|
217
|
+
export function setupCLICleanup(dbPath) {
|
|
218
|
+
const cleanup = async () => {
|
|
219
|
+
console.log('\n🛑 Shutting down gracefully...');
|
|
220
|
+
if (dbPath) {
|
|
221
|
+
try {
|
|
222
|
+
await DatabaseConnectionManager.releaseConnection(dbPath);
|
|
223
|
+
}
|
|
224
|
+
catch (error) {
|
|
225
|
+
// Ignore cleanup errors during shutdown
|
|
226
|
+
}
|
|
227
|
+
}
|
|
228
|
+
await DatabaseConnectionManager.closeAllConnections();
|
|
229
|
+
process.exit(0);
|
|
230
|
+
};
|
|
231
|
+
// Only set up handlers if they haven't been set up already
|
|
232
|
+
if (!process.listenerCount('SIGINT')) {
|
|
233
|
+
process.on('SIGINT', cleanup);
|
|
234
|
+
}
|
|
235
|
+
if (!process.listenerCount('SIGTERM')) {
|
|
236
|
+
process.on('SIGTERM', cleanup);
|
|
237
|
+
}
|
|
238
|
+
}
|
|
239
|
+
//# sourceMappingURL=cli-database-utils.js.map
|
|
@@ -0,0 +1,102 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* CORE MODULE — Shared between text-only (rag-lite-ts) and future multimodal (rag-lite-mm)
|
|
3
|
+
* Model-agnostic. No transformer or modality-specific logic.
|
|
4
|
+
*/
|
|
5
|
+
/**
|
|
6
|
+
* Core configuration interface - model-agnostic settings
|
|
7
|
+
*/
|
|
8
|
+
export interface CoreConfig {
|
|
9
|
+
chunk_size: number;
|
|
10
|
+
chunk_overlap: number;
|
|
11
|
+
batch_size: number;
|
|
12
|
+
top_k: number;
|
|
13
|
+
db_file: string;
|
|
14
|
+
index_file: string;
|
|
15
|
+
model_cache_path?: string;
|
|
16
|
+
path_storage_strategy: 'absolute' | 'relative';
|
|
17
|
+
embedding_model: string;
|
|
18
|
+
rerank_enabled: boolean;
|
|
19
|
+
preprocessing?: any;
|
|
20
|
+
}
|
|
21
|
+
/**
|
|
22
|
+
* Extensible configuration interface that can be extended by implementations
|
|
23
|
+
*/
|
|
24
|
+
export interface ExtensibleConfig<T = {}> extends CoreConfig {
|
|
25
|
+
implementation: T;
|
|
26
|
+
}
|
|
27
|
+
/**
|
|
28
|
+
* Standard exit codes for different error conditions
|
|
29
|
+
*/
|
|
30
|
+
export declare const EXIT_CODES: {
|
|
31
|
+
readonly SUCCESS: 0;
|
|
32
|
+
readonly GENERAL_ERROR: 1;
|
|
33
|
+
readonly INVALID_ARGUMENTS: 2;
|
|
34
|
+
readonly CONFIGURATION_ERROR: 3;
|
|
35
|
+
readonly FILE_NOT_FOUND: 4;
|
|
36
|
+
readonly DATABASE_ERROR: 5;
|
|
37
|
+
readonly MODEL_ERROR: 6;
|
|
38
|
+
readonly INDEX_ERROR: 7;
|
|
39
|
+
readonly PERMISSION_ERROR: 8;
|
|
40
|
+
};
|
|
41
|
+
/**
|
|
42
|
+
* Configuration validation error with specific exit code
|
|
43
|
+
*/
|
|
44
|
+
export declare class ConfigurationError extends Error {
|
|
45
|
+
exitCode: number;
|
|
46
|
+
constructor(message: string, exitCode?: number);
|
|
47
|
+
}
|
|
48
|
+
/**
|
|
49
|
+
* Get the default model cache path as specified in the requirements
|
|
50
|
+
* @returns Default cache path (~/.raglite/models/)
|
|
51
|
+
*/
|
|
52
|
+
export declare function getDefaultModelCachePath(): string;
|
|
53
|
+
/**
|
|
54
|
+
* Validates core configuration fields
|
|
55
|
+
* @param config - Configuration object to validate
|
|
56
|
+
* @throws {ConfigurationError} If configuration is invalid
|
|
57
|
+
*/
|
|
58
|
+
export declare function validateCoreConfig(config: any): asserts config is CoreConfig;
|
|
59
|
+
/**
|
|
60
|
+
* Model defaults interface for different embedding models
|
|
61
|
+
*/
|
|
62
|
+
export interface ModelDefaults {
|
|
63
|
+
dimensions: number;
|
|
64
|
+
chunk_size: number;
|
|
65
|
+
chunk_overlap: number;
|
|
66
|
+
batch_size: number;
|
|
67
|
+
}
|
|
68
|
+
/**
|
|
69
|
+
* Get default configuration for different embedding models
|
|
70
|
+
* @param modelName - Name of the embedding model
|
|
71
|
+
* @returns Model-specific defaults
|
|
72
|
+
*/
|
|
73
|
+
export declare function getModelDefaults(modelName?: string): ModelDefaults;
|
|
74
|
+
/**
|
|
75
|
+
* Default core configuration object
|
|
76
|
+
* Model-agnostic settings that can be used by core modules
|
|
77
|
+
*/
|
|
78
|
+
export declare const config: CoreConfig;
|
|
79
|
+
/**
|
|
80
|
+
* Validate preprocessing configuration
|
|
81
|
+
*/
|
|
82
|
+
export declare function validatePreprocessingConfig(config: any): asserts config is any;
|
|
83
|
+
/**
|
|
84
|
+
* Merge preprocessing configurations with mode defaults
|
|
85
|
+
*/
|
|
86
|
+
export declare function mergePreprocessingConfig(config: any): any;
|
|
87
|
+
/**
|
|
88
|
+
* Utility function to handle unrecoverable errors with descriptive messages
|
|
89
|
+
* Logs error and exits immediately with appropriate exit code
|
|
90
|
+
* @param error - Error object or message
|
|
91
|
+
* @param context - Context where the error occurred
|
|
92
|
+
* @param exitCode - Exit code to use (defaults to GENERAL_ERROR)
|
|
93
|
+
*/
|
|
94
|
+
export declare function handleUnrecoverableError(error: Error | string, context: string, exitCode?: number): never;
|
|
95
|
+
/**
|
|
96
|
+
* Utility function for safe error logging with context
|
|
97
|
+
* @param error - Error to log
|
|
98
|
+
* @param context - Context where error occurred
|
|
99
|
+
* @param skipError - Whether to skip this error and continue (default: false)
|
|
100
|
+
*/
|
|
101
|
+
export declare function logError(error: Error | string, context: string, skipError?: boolean): void;
|
|
102
|
+
//# sourceMappingURL=config.d.ts.map
|
|
@@ -0,0 +1,247 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* CORE MODULE — Shared between text-only (rag-lite-ts) and future multimodal (rag-lite-mm)
|
|
3
|
+
* Model-agnostic. No transformer or modality-specific logic.
|
|
4
|
+
*/
|
|
5
|
+
import { homedir } from 'os';
|
|
6
|
+
import { join } from 'path';
|
|
7
|
+
/**
|
|
8
|
+
* Standard exit codes for different error conditions
|
|
9
|
+
*/
|
|
10
|
+
export const EXIT_CODES = {
|
|
11
|
+
SUCCESS: 0,
|
|
12
|
+
GENERAL_ERROR: 1,
|
|
13
|
+
INVALID_ARGUMENTS: 2,
|
|
14
|
+
CONFIGURATION_ERROR: 3,
|
|
15
|
+
FILE_NOT_FOUND: 4,
|
|
16
|
+
DATABASE_ERROR: 5,
|
|
17
|
+
MODEL_ERROR: 6,
|
|
18
|
+
INDEX_ERROR: 7,
|
|
19
|
+
PERMISSION_ERROR: 8
|
|
20
|
+
};
|
|
21
|
+
/**
|
|
22
|
+
* Configuration validation error with specific exit code
|
|
23
|
+
*/
|
|
24
|
+
export class ConfigurationError extends Error {
|
|
25
|
+
exitCode;
|
|
26
|
+
constructor(message, exitCode = EXIT_CODES.CONFIGURATION_ERROR) {
|
|
27
|
+
super(message);
|
|
28
|
+
this.exitCode = exitCode;
|
|
29
|
+
this.name = 'ConfigurationError';
|
|
30
|
+
}
|
|
31
|
+
}
|
|
32
|
+
/**
|
|
33
|
+
* Get the default model cache path as specified in the requirements
|
|
34
|
+
* @returns Default cache path (~/.raglite/models/)
|
|
35
|
+
*/
|
|
36
|
+
export function getDefaultModelCachePath() {
|
|
37
|
+
return join(homedir(), '.raglite', 'models');
|
|
38
|
+
}
|
|
39
|
+
/**
|
|
40
|
+
* Validates core configuration fields
|
|
41
|
+
* @param config - Configuration object to validate
|
|
42
|
+
* @throws {ConfigurationError} If configuration is invalid
|
|
43
|
+
*/
|
|
44
|
+
export function validateCoreConfig(config) {
|
|
45
|
+
if (!config || typeof config !== 'object') {
|
|
46
|
+
throw new ConfigurationError('Configuration must be an object');
|
|
47
|
+
}
|
|
48
|
+
// Check required string fields
|
|
49
|
+
const requiredStrings = ['db_file', 'index_file'];
|
|
50
|
+
for (const field of requiredStrings) {
|
|
51
|
+
if (!config[field] || typeof config[field] !== 'string') {
|
|
52
|
+
throw new ConfigurationError(`Configuration error: '${field}' must be a non-empty string.\n` +
|
|
53
|
+
`Current value: ${JSON.stringify(config[field])}\n` +
|
|
54
|
+
`Please check your configuration file.`);
|
|
55
|
+
}
|
|
56
|
+
}
|
|
57
|
+
// Validate path_storage_strategy
|
|
58
|
+
if (!['absolute', 'relative'].includes(config.path_storage_strategy)) {
|
|
59
|
+
throw new ConfigurationError(`Configuration error: 'path_storage_strategy' must be either 'absolute' or 'relative'.\n` +
|
|
60
|
+
`Current value: ${JSON.stringify(config.path_storage_strategy)}\n` +
|
|
61
|
+
`Please set it to 'absolute' or 'relative'.`);
|
|
62
|
+
}
|
|
63
|
+
// Check required numeric fields are positive
|
|
64
|
+
const requiredNumbers = ['chunk_size', 'chunk_overlap', 'batch_size', 'top_k'];
|
|
65
|
+
for (const field of requiredNumbers) {
|
|
66
|
+
if (typeof config[field] !== 'number' || config[field] <= 0) {
|
|
67
|
+
throw new ConfigurationError(`Configuration error: '${field}' must be a positive number.\n` +
|
|
68
|
+
`Current value: ${JSON.stringify(config[field])}\n` +
|
|
69
|
+
`Please ensure all numeric values are greater than 0.`);
|
|
70
|
+
}
|
|
71
|
+
}
|
|
72
|
+
// Validate optional model_cache_path field
|
|
73
|
+
if (config.model_cache_path !== undefined && (typeof config.model_cache_path !== 'string' || config.model_cache_path.trim() === '')) {
|
|
74
|
+
throw new ConfigurationError(`Configuration error: 'model_cache_path' must be a non-empty string when provided.\n` +
|
|
75
|
+
`Current value: ${JSON.stringify(config.model_cache_path)}\n` +
|
|
76
|
+
`Please provide a valid directory path or remove the field to use default caching.`);
|
|
77
|
+
}
|
|
78
|
+
// Validate chunk_overlap is less than chunk_size
|
|
79
|
+
if (config.chunk_overlap >= config.chunk_size) {
|
|
80
|
+
throw new ConfigurationError(`Configuration error: chunk_overlap (${config.chunk_overlap}) must be less than chunk_size (${config.chunk_size}).\n` +
|
|
81
|
+
`Recommended: Set chunk_overlap to about 20% of chunk_size (e.g., chunk_size: 250, chunk_overlap: 50).`);
|
|
82
|
+
}
|
|
83
|
+
// Validate reasonable ranges for performance
|
|
84
|
+
if (config.chunk_size > 1000) {
|
|
85
|
+
console.warn(`Warning: Large chunk_size (${config.chunk_size}) may impact performance. Recommended range: 200-400 tokens.`);
|
|
86
|
+
}
|
|
87
|
+
if (config.batch_size > 64) {
|
|
88
|
+
console.warn(`Warning: Large batch_size (${config.batch_size}) may cause memory issues. Recommended range: 8-32.`);
|
|
89
|
+
}
|
|
90
|
+
}
|
|
91
|
+
/**
|
|
92
|
+
* Get default configuration for different embedding models
|
|
93
|
+
* @param modelName - Name of the embedding model
|
|
94
|
+
* @returns Model-specific defaults
|
|
95
|
+
*/
|
|
96
|
+
export function getModelDefaults(modelName) {
|
|
97
|
+
// Default configuration for most models
|
|
98
|
+
const defaults = {
|
|
99
|
+
dimensions: 384, // Most common dimension for sentence transformers
|
|
100
|
+
chunk_size: 250,
|
|
101
|
+
chunk_overlap: 50,
|
|
102
|
+
batch_size: 16
|
|
103
|
+
};
|
|
104
|
+
// Model-specific overrides based on model name heuristics
|
|
105
|
+
if (modelName) {
|
|
106
|
+
const normalizedName = modelName.toLowerCase();
|
|
107
|
+
// CLIP models - 512 dimensions
|
|
108
|
+
if (normalizedName.includes('clip')) {
|
|
109
|
+
defaults.dimensions = 512;
|
|
110
|
+
defaults.batch_size = 8;
|
|
111
|
+
}
|
|
112
|
+
// MPNet models - 768 dimensions
|
|
113
|
+
else if (normalizedName.includes('all-mpnet-base-v2')) {
|
|
114
|
+
defaults.dimensions = 768;
|
|
115
|
+
defaults.chunk_size = 400;
|
|
116
|
+
defaults.chunk_overlap = 80;
|
|
117
|
+
defaults.batch_size = 8;
|
|
118
|
+
}
|
|
119
|
+
else if (normalizedName.includes('mpnet') || normalizedName.includes('768')) {
|
|
120
|
+
defaults.dimensions = 768;
|
|
121
|
+
}
|
|
122
|
+
// Models with 512 in the name
|
|
123
|
+
else if (normalizedName.includes('512')) {
|
|
124
|
+
defaults.dimensions = 512;
|
|
125
|
+
}
|
|
126
|
+
// MiniLM and other 384-dimensional models (default)
|
|
127
|
+
else if (normalizedName.includes('384') || normalizedName.includes('minilm')) {
|
|
128
|
+
defaults.dimensions = 384;
|
|
129
|
+
}
|
|
130
|
+
}
|
|
131
|
+
return defaults;
|
|
132
|
+
}
|
|
133
|
+
/**
|
|
134
|
+
* Default core configuration object
|
|
135
|
+
* Model-agnostic settings that can be used by core modules
|
|
136
|
+
*/
|
|
137
|
+
export const config = {
|
|
138
|
+
chunk_size: parseInt(process.env.RAG_CHUNK_SIZE || '250', 10),
|
|
139
|
+
chunk_overlap: parseInt(process.env.RAG_CHUNK_OVERLAP || '50', 10),
|
|
140
|
+
batch_size: parseInt(process.env.RAG_BATCH_SIZE || '16', 10),
|
|
141
|
+
top_k: parseInt(process.env.RAG_TOP_K || '10', 10),
|
|
142
|
+
db_file: process.env.RAG_DB_FILE || 'db.sqlite',
|
|
143
|
+
index_file: process.env.RAG_INDEX_FILE || 'vector-index.bin',
|
|
144
|
+
model_cache_path: process.env.RAG_MODEL_CACHE_PATH || getDefaultModelCachePath(),
|
|
145
|
+
path_storage_strategy: process.env.RAG_PATH_STORAGE_STRATEGY || 'relative',
|
|
146
|
+
embedding_model: process.env.RAG_EMBEDDING_MODEL || 'sentence-transformers/all-MiniLM-L6-v2',
|
|
147
|
+
rerank_enabled: process.env.RAG_RERANK_ENABLED === 'true',
|
|
148
|
+
preprocessing: { mode: 'balanced' }
|
|
149
|
+
};
|
|
150
|
+
/**
|
|
151
|
+
* Validate preprocessing configuration
|
|
152
|
+
*/
|
|
153
|
+
export function validatePreprocessingConfig(config) {
|
|
154
|
+
if (!config || typeof config !== 'object') {
|
|
155
|
+
throw new ConfigurationError('Preprocessing configuration must be an object');
|
|
156
|
+
}
|
|
157
|
+
if (!config.mode || !['strict', 'balanced', 'rich'].includes(config.mode)) {
|
|
158
|
+
throw new ConfigurationError('preprocessing.mode must be one of: strict, balanced, rich');
|
|
159
|
+
}
|
|
160
|
+
// Validate overrides if present
|
|
161
|
+
if (config.overrides !== undefined) {
|
|
162
|
+
if (!config.overrides || typeof config.overrides !== 'object') {
|
|
163
|
+
throw new ConfigurationError('preprocessing.overrides must be an object when provided');
|
|
164
|
+
}
|
|
165
|
+
const validValues = ['strip', 'keep', 'placeholder', 'extract'];
|
|
166
|
+
// Validate MDX override
|
|
167
|
+
if (config.overrides.mdx !== undefined && !validValues.includes(config.overrides.mdx)) {
|
|
168
|
+
throw new ConfigurationError('preprocessing.overrides.mdx must be one of: strip, keep, placeholder');
|
|
169
|
+
}
|
|
170
|
+
// Validate Mermaid override
|
|
171
|
+
if (config.overrides.mermaid !== undefined && !validValues.includes(config.overrides.mermaid)) {
|
|
172
|
+
throw new ConfigurationError('preprocessing.overrides.mermaid must be one of: strip, keep, placeholder, extract');
|
|
173
|
+
}
|
|
174
|
+
// Validate code override
|
|
175
|
+
if (config.overrides.code !== undefined && !validValues.includes(config.overrides.code)) {
|
|
176
|
+
throw new ConfigurationError('preprocessing.overrides.code must be one of: strip, keep, placeholder');
|
|
177
|
+
}
|
|
178
|
+
}
|
|
179
|
+
}
|
|
180
|
+
/**
|
|
181
|
+
* Merge preprocessing configurations with mode defaults
|
|
182
|
+
*/
|
|
183
|
+
export function mergePreprocessingConfig(config) {
|
|
184
|
+
const modeDefaults = {
|
|
185
|
+
strict: { mdx: 'strip', mermaid: 'strip', code: 'strip' },
|
|
186
|
+
balanced: { mdx: 'placeholder', mermaid: 'placeholder', code: 'keep' },
|
|
187
|
+
rich: { mdx: 'keep', mermaid: 'extract', code: 'keep' }
|
|
188
|
+
};
|
|
189
|
+
const defaults = modeDefaults[config.mode] || modeDefaults.balanced;
|
|
190
|
+
return {
|
|
191
|
+
...defaults,
|
|
192
|
+
...config.overrides
|
|
193
|
+
};
|
|
194
|
+
}
|
|
195
|
+
/**
|
|
196
|
+
* Utility function to handle unrecoverable errors with descriptive messages
|
|
197
|
+
* Logs error and exits immediately with appropriate exit code
|
|
198
|
+
* @param error - Error object or message
|
|
199
|
+
* @param context - Context where the error occurred
|
|
200
|
+
* @param exitCode - Exit code to use (defaults to GENERAL_ERROR)
|
|
201
|
+
*/
|
|
202
|
+
export function handleUnrecoverableError(error, context, exitCode = EXIT_CODES.GENERAL_ERROR) {
|
|
203
|
+
const errorMessage = error instanceof Error ? error.message : String(error);
|
|
204
|
+
console.error(`\nFatal Error in ${context}:`);
|
|
205
|
+
console.error(errorMessage);
|
|
206
|
+
console.error('\nThe system cannot continue and will exit immediately.');
|
|
207
|
+
// Provide context-specific guidance
|
|
208
|
+
switch (exitCode) {
|
|
209
|
+
case EXIT_CODES.CONFIGURATION_ERROR:
|
|
210
|
+
console.error('\nPlease check your configuration and try again.');
|
|
211
|
+
break;
|
|
212
|
+
case EXIT_CODES.DATABASE_ERROR:
|
|
213
|
+
console.error('\nTry running "raglite rebuild" to fix database issues.');
|
|
214
|
+
break;
|
|
215
|
+
case EXIT_CODES.MODEL_ERROR:
|
|
216
|
+
console.error('\nEnsure you have internet connection for model download and sufficient disk space.');
|
|
217
|
+
break;
|
|
218
|
+
case EXIT_CODES.INDEX_ERROR:
|
|
219
|
+
console.error('\nTry running "raglite rebuild" to recreate the vector index.');
|
|
220
|
+
break;
|
|
221
|
+
case EXIT_CODES.FILE_NOT_FOUND:
|
|
222
|
+
console.error('\nPlease check that the specified files or directories exist and are accessible.');
|
|
223
|
+
break;
|
|
224
|
+
case EXIT_CODES.PERMISSION_ERROR:
|
|
225
|
+
console.error('\nPlease check file and directory permissions.');
|
|
226
|
+
break;
|
|
227
|
+
default:
|
|
228
|
+
console.error('\nIf this problem persists, please report it as a bug.');
|
|
229
|
+
}
|
|
230
|
+
process.exit(exitCode);
|
|
231
|
+
}
|
|
232
|
+
/**
|
|
233
|
+
* Utility function for safe error logging with context
|
|
234
|
+
* @param error - Error to log
|
|
235
|
+
* @param context - Context where error occurred
|
|
236
|
+
* @param skipError - Whether to skip this error and continue (default: false)
|
|
237
|
+
*/
|
|
238
|
+
export function logError(error, context, skipError = false) {
|
|
239
|
+
const errorMessage = error instanceof Error ? error.message : String(error);
|
|
240
|
+
if (skipError) {
|
|
241
|
+
console.error(`Warning in ${context}: ${errorMessage} (skipping and continuing)`);
|
|
242
|
+
}
|
|
243
|
+
else {
|
|
244
|
+
console.error(`Error in ${context}: ${errorMessage}`);
|
|
245
|
+
}
|
|
246
|
+
}
|
|
247
|
+
//# sourceMappingURL=config.js.map
|