spectra-cli 4.1.1 → 4.3.0
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 +12 -5
- package/dist/.spectra-build-meta.json +9 -0
- package/dist/adapters/go-adapter.d.ts.map +1 -1
- package/dist/adapters/go-adapter.js +2 -0
- package/dist/adapters/go-adapter.js.map +1 -1
- package/dist/adapters/index.d.ts +1 -1
- package/dist/adapters/index.d.ts.map +1 -1
- package/dist/adapters/java-adapter.d.ts.map +1 -1
- package/dist/adapters/java-adapter.js +2 -0
- package/dist/adapters/java-adapter.js.map +1 -1
- package/dist/adapters/language-adapter.d.ts +42 -9
- package/dist/adapters/language-adapter.d.ts.map +1 -1
- package/dist/adapters/python-adapter.d.ts +34 -6
- package/dist/adapters/python-adapter.d.ts.map +1 -1
- package/dist/adapters/python-adapter.js +168 -61
- package/dist/adapters/python-adapter.js.map +1 -1
- package/dist/adapters/ts-js-adapter.d.ts +13 -5
- package/dist/adapters/ts-js-adapter.d.ts.map +1 -1
- package/dist/adapters/ts-js-adapter.js +55 -14
- package/dist/adapters/ts-js-adapter.js.map +1 -1
- package/dist/batch/batch-orchestrator.d.ts +74 -5
- package/dist/batch/batch-orchestrator.d.ts.map +1 -1
- package/dist/batch/batch-orchestrator.js +583 -70
- package/dist/batch/batch-orchestrator.js.map +1 -1
- package/dist/batch/delta-regenerator.d.ts +2 -2
- package/dist/batch/delta-regenerator.d.ts.map +1 -1
- package/dist/batch/delta-regenerator.js +21 -28
- package/dist/batch/delta-regenerator.js.map +1 -1
- package/dist/batch/module-grouper.d.ts +9 -2
- package/dist/batch/module-grouper.d.ts.map +1 -1
- package/dist/batch/module-grouper.js +3 -1
- package/dist/batch/module-grouper.js.map +1 -1
- package/dist/batch/regen-plan.d.ts +78 -0
- package/dist/batch/regen-plan.d.ts.map +1 -0
- package/dist/batch/regen-plan.js +93 -0
- package/dist/batch/regen-plan.js.map +1 -0
- package/dist/cli/commands/batch.d.ts.map +1 -1
- package/dist/cli/commands/batch.js +36 -8
- package/dist/cli/commands/batch.js.map +1 -1
- package/dist/cli/commands/graph.js +1 -1
- package/dist/cli/commands/graph.js.map +1 -1
- package/dist/cli/commands/index.d.ts +8 -0
- package/dist/cli/commands/index.d.ts.map +1 -0
- package/dist/cli/commands/index.js +274 -0
- package/dist/cli/commands/index.js.map +1 -0
- package/dist/cli/commands/query.d.ts.map +1 -1
- package/dist/cli/commands/query.js +3 -1
- package/dist/cli/commands/query.js.map +1 -1
- package/dist/cli/commands/scaffold-kb.d.ts +9 -0
- package/dist/cli/commands/scaffold-kb.d.ts.map +1 -0
- package/dist/cli/commands/scaffold-kb.js +179 -0
- package/dist/cli/commands/scaffold-kb.js.map +1 -0
- package/dist/cli/index.js +33 -14
- package/dist/cli/index.js.map +1 -1
- package/dist/cli/utils/error-handler.d.ts +0 -6
- package/dist/cli/utils/error-handler.d.ts.map +1 -1
- package/dist/cli/utils/error-handler.js +0 -13
- package/dist/cli/utils/error-handler.js.map +1 -1
- package/dist/cli/utils/parse-args.d.ts +55 -3
- package/dist/cli/utils/parse-args.d.ts.map +1 -1
- package/dist/cli/utils/parse-args.js +176 -13
- package/dist/cli/utils/parse-args.js.map +1 -1
- package/dist/cli/version-meta.d.ts +16 -0
- package/dist/cli/version-meta.d.ts.map +1 -0
- package/dist/cli/version-meta.js +28 -0
- package/dist/cli/version-meta.js.map +1 -0
- package/dist/core/ast-analyzer.d.ts.map +1 -1
- package/dist/core/ast-analyzer.js +75 -5
- package/dist/core/ast-analyzer.js.map +1 -1
- package/dist/core/import-resolver.d.ts +93 -0
- package/dist/core/import-resolver.d.ts.map +1 -0
- package/dist/core/import-resolver.js +414 -0
- package/dist/core/import-resolver.js.map +1 -0
- package/dist/core/query-mappers/base-mapper.d.ts +13 -0
- package/dist/core/query-mappers/base-mapper.d.ts.map +1 -1
- package/dist/core/query-mappers/go-mapper.d.ts +68 -0
- package/dist/core/query-mappers/go-mapper.d.ts.map +1 -1
- package/dist/core/query-mappers/go-mapper.js +421 -0
- package/dist/core/query-mappers/go-mapper.js.map +1 -1
- package/dist/core/query-mappers/java-mapper.d.ts +196 -0
- package/dist/core/query-mappers/java-mapper.d.ts.map +1 -1
- package/dist/core/query-mappers/java-mapper.js +600 -0
- package/dist/core/query-mappers/java-mapper.js.map +1 -1
- package/dist/core/query-mappers/python-mapper.d.ts +55 -0
- package/dist/core/query-mappers/python-mapper.d.ts.map +1 -1
- package/dist/core/query-mappers/python-mapper.js +273 -0
- package/dist/core/query-mappers/python-mapper.js.map +1 -1
- package/dist/core/query-mappers/typescript-mapper.d.ts +102 -0
- package/dist/core/query-mappers/typescript-mapper.d.ts.map +1 -1
- package/dist/core/query-mappers/typescript-mapper.js +470 -0
- package/dist/core/query-mappers/typescript-mapper.js.map +1 -1
- package/dist/core/single-spec-orchestrator.d.ts +21 -0
- package/dist/core/single-spec-orchestrator.d.ts.map +1 -1
- package/dist/core/single-spec-orchestrator.js +187 -23
- package/dist/core/single-spec-orchestrator.js.map +1 -1
- package/dist/core/skeleton-hash.d.ts +33 -0
- package/dist/core/skeleton-hash.d.ts.map +1 -0
- package/dist/core/skeleton-hash.js +76 -0
- package/dist/core/skeleton-hash.js.map +1 -0
- package/dist/core/tree-sitter-analyzer.d.ts +7 -0
- package/dist/core/tree-sitter-analyzer.d.ts.map +1 -1
- package/dist/core/tree-sitter-analyzer.js +32 -1
- package/dist/core/tree-sitter-analyzer.js.map +1 -1
- package/dist/core/tree-sitter-fallback.d.ts +12 -4
- package/dist/core/tree-sitter-fallback.d.ts.map +1 -1
- package/dist/core/tree-sitter-fallback.js +160 -12
- package/dist/core/tree-sitter-fallback.js.map +1 -1
- package/dist/extraction/extraction-types.d.ts +25 -25
- package/dist/extraction/image-extractor.d.ts.map +1 -1
- package/dist/extraction/image-extractor.js +3 -1
- package/dist/extraction/image-extractor.js.map +1 -1
- package/dist/extraction/markdown-extractor.d.ts.map +1 -1
- package/dist/extraction/markdown-extractor.js +3 -1
- package/dist/extraction/markdown-extractor.js.map +1 -1
- package/dist/generator/frontmatter.d.ts +5 -0
- package/dist/generator/frontmatter.d.ts.map +1 -1
- package/dist/generator/frontmatter.js +4 -0
- package/dist/generator/frontmatter.js.map +1 -1
- package/dist/generator/index-generator.d.ts +3 -3
- package/dist/generator/index-generator.d.ts.map +1 -1
- package/dist/generator/index-generator.js +1 -1
- package/dist/generator/index-generator.js.map +1 -1
- package/dist/graph/directory-graph.d.ts +14 -4
- package/dist/graph/directory-graph.d.ts.map +1 -1
- package/dist/graph/directory-graph.js +71 -128
- package/dist/graph/directory-graph.js.map +1 -1
- package/dist/graph/mermaid-renderer.d.ts +8 -5
- package/dist/graph/mermaid-renderer.d.ts.map +1 -1
- package/dist/graph/mermaid-renderer.js +3 -3
- package/dist/graph/mermaid-renderer.js.map +1 -1
- package/dist/graph/topological-sort.d.ts +13 -6
- package/dist/graph/topological-sort.d.ts.map +1 -1
- package/dist/graph/topological-sort.js +2 -2
- package/dist/graph/topological-sort.js.map +1 -1
- package/dist/kb-mcp/index.d.ts +14 -0
- package/dist/kb-mcp/index.d.ts.map +1 -0
- package/dist/kb-mcp/index.js +24 -0
- package/dist/kb-mcp/index.js.map +1 -0
- package/dist/kb-mcp/lib/kb-error.d.ts +21 -0
- package/dist/kb-mcp/lib/kb-error.d.ts.map +1 -0
- package/dist/kb-mcp/lib/kb-error.js +28 -0
- package/dist/kb-mcp/lib/kb-error.js.map +1 -0
- package/dist/kb-mcp/lib/kb-locator.d.ts +58 -0
- package/dist/kb-mcp/lib/kb-locator.d.ts.map +1 -0
- package/dist/kb-mcp/lib/kb-locator.js +85 -0
- package/dist/kb-mcp/lib/kb-locator.js.map +1 -0
- package/dist/kb-mcp/lib/result-merger.d.ts +35 -0
- package/dist/kb-mcp/lib/result-merger.d.ts.map +1 -0
- package/dist/kb-mcp/lib/result-merger.js +64 -0
- package/dist/kb-mcp/lib/result-merger.js.map +1 -0
- package/dist/kb-mcp/server.d.ts +12 -0
- package/dist/kb-mcp/server.d.ts.map +1 -0
- package/dist/kb-mcp/server.js +29 -0
- package/dist/kb-mcp/server.js.map +1 -0
- package/dist/kb-mcp/tools/kb-api-lookup.d.ts +28 -0
- package/dist/kb-mcp/tools/kb-api-lookup.d.ts.map +1 -0
- package/dist/kb-mcp/tools/kb-api-lookup.js +237 -0
- package/dist/kb-mcp/tools/kb-api-lookup.js.map +1 -0
- package/dist/kb-mcp/tools/kb-doc-lookup.d.ts +20 -0
- package/dist/kb-mcp/tools/kb-doc-lookup.d.ts.map +1 -0
- package/dist/kb-mcp/tools/kb-doc-lookup.js +92 -0
- package/dist/kb-mcp/tools/kb-doc-lookup.js.map +1 -0
- package/dist/kb-mcp/tools/kb-search.d.ts +23 -0
- package/dist/kb-mcp/tools/kb-search.d.ts.map +1 -0
- package/dist/kb-mcp/tools/kb-search.js +126 -0
- package/dist/kb-mcp/tools/kb-search.js.map +1 -0
- package/dist/knowledge-graph/call-resolver.d.ts +64 -0
- package/dist/knowledge-graph/call-resolver.d.ts.map +1 -0
- package/dist/knowledge-graph/call-resolver.js +323 -0
- package/dist/knowledge-graph/call-resolver.js.map +1 -0
- package/dist/knowledge-graph/import-resolver.d.ts +39 -0
- package/dist/knowledge-graph/import-resolver.d.ts.map +1 -0
- package/dist/knowledge-graph/import-resolver.js +214 -0
- package/dist/knowledge-graph/import-resolver.js.map +1 -0
- package/dist/knowledge-graph/incremental.d.ts +131 -0
- package/dist/knowledge-graph/incremental.d.ts.map +1 -0
- package/dist/knowledge-graph/incremental.js +432 -0
- package/dist/knowledge-graph/incremental.js.map +1 -0
- package/dist/knowledge-graph/index.d.ts +85 -0
- package/dist/knowledge-graph/index.d.ts.map +1 -0
- package/dist/knowledge-graph/index.js +198 -0
- package/dist/knowledge-graph/index.js.map +1 -0
- package/dist/knowledge-graph/module-derivation.d.ts +252 -0
- package/dist/knowledge-graph/module-derivation.d.ts.map +1 -0
- package/dist/knowledge-graph/module-derivation.js +418 -0
- package/dist/knowledge-graph/module-derivation.js.map +1 -0
- package/dist/knowledge-graph/persistence.d.ts +310 -0
- package/dist/knowledge-graph/persistence.d.ts.map +1 -0
- package/dist/knowledge-graph/persistence.js +280 -0
- package/dist/knowledge-graph/persistence.js.map +1 -0
- package/dist/knowledge-graph/query-helpers.d.ts +191 -0
- package/dist/knowledge-graph/query-helpers.d.ts.map +1 -0
- package/dist/knowledge-graph/query-helpers.js +559 -0
- package/dist/knowledge-graph/query-helpers.js.map +1 -0
- package/dist/knowledge-graph/relativize.d.ts +51 -0
- package/dist/knowledge-graph/relativize.d.ts.map +1 -0
- package/dist/knowledge-graph/relativize.js +104 -0
- package/dist/knowledge-graph/relativize.js.map +1 -0
- package/dist/knowledge-graph/unified-graph.d.ts +310 -0
- package/dist/knowledge-graph/unified-graph.d.ts.map +1 -0
- package/dist/knowledge-graph/unified-graph.js +182 -0
- package/dist/knowledge-graph/unified-graph.js.map +1 -0
- package/dist/mcp/agent-context-tools.d.ts +53 -0
- package/dist/mcp/agent-context-tools.d.ts.map +1 -0
- package/dist/mcp/agent-context-tools.js +806 -0
- package/dist/mcp/agent-context-tools.js.map +1 -0
- package/dist/mcp/file-nav-tools.d.ts +42 -0
- package/dist/mcp/file-nav-tools.d.ts.map +1 -0
- package/dist/mcp/file-nav-tools.js +318 -0
- package/dist/mcp/file-nav-tools.js.map +1 -0
- package/dist/mcp/graph-tools.d.ts +22 -0
- package/dist/mcp/graph-tools.d.ts.map +1 -1
- package/dist/mcp/graph-tools.js +204 -133
- package/dist/mcp/graph-tools.js.map +1 -1
- package/dist/mcp/lib/file-nav-helpers.d.ts +128 -0
- package/dist/mcp/lib/file-nav-helpers.d.ts.map +1 -0
- package/dist/mcp/lib/file-nav-helpers.js +305 -0
- package/dist/mcp/lib/file-nav-helpers.js.map +1 -0
- package/dist/mcp/lib/response-helpers.d.ts +67 -0
- package/dist/mcp/lib/response-helpers.d.ts.map +1 -0
- package/dist/mcp/lib/response-helpers.js +103 -0
- package/dist/mcp/lib/response-helpers.js.map +1 -0
- package/dist/mcp/lib/telemetry.d.ts +69 -0
- package/dist/mcp/lib/telemetry.d.ts.map +1 -0
- package/dist/mcp/lib/telemetry.js +113 -0
- package/dist/mcp/lib/telemetry.js.map +1 -0
- package/dist/mcp/lib/tool-response.d.ts +46 -0
- package/dist/mcp/lib/tool-response.d.ts.map +1 -0
- package/dist/mcp/lib/tool-response.js +83 -0
- package/dist/mcp/lib/tool-response.js.map +1 -0
- package/dist/mcp/server.d.ts +14 -1
- package/dist/mcp/server.d.ts.map +1 -1
- package/dist/mcp/server.js +216 -158
- package/dist/mcp/server.js.map +1 -1
- package/dist/models/call-site.d.ts +68 -0
- package/dist/models/call-site.d.ts.map +1 -0
- package/dist/models/call-site.js +60 -0
- package/dist/models/call-site.js.map +1 -0
- package/dist/models/code-skeleton.d.ts +80 -6
- package/dist/models/code-skeleton.d.ts.map +1 -1
- package/dist/models/code-skeleton.js +30 -0
- package/dist/models/code-skeleton.js.map +1 -1
- package/dist/models/drift-item.d.ts +4 -4
- package/dist/models/module-spec.d.ts +173 -94
- package/dist/models/module-spec.d.ts.map +1 -1
- package/dist/models/module-spec.js +6 -0
- package/dist/models/module-spec.js.map +1 -1
- package/dist/panoramic/adr-decision-pipeline.d.ts +53 -0
- package/dist/panoramic/adr-decision-pipeline.d.ts.map +1 -0
- package/dist/panoramic/adr-decision-pipeline.js +685 -0
- package/dist/panoramic/adr-decision-pipeline.js.map +1 -0
- package/dist/panoramic/api-surface-generator.d.ts +56 -0
- package/dist/panoramic/api-surface-generator.d.ts.map +1 -0
- package/dist/panoramic/api-surface-generator.js +1671 -0
- package/dist/panoramic/api-surface-generator.js.map +1 -0
- package/dist/panoramic/architecture-ir-builder.d.ts +19 -0
- package/dist/panoramic/architecture-ir-builder.d.ts.map +1 -0
- package/dist/panoramic/architecture-ir-builder.js +561 -0
- package/dist/panoramic/architecture-ir-builder.js.map +1 -0
- package/dist/panoramic/architecture-ir-exporters.d.ts +7 -0
- package/dist/panoramic/architecture-ir-exporters.d.ts.map +1 -0
- package/dist/panoramic/architecture-ir-exporters.js +183 -0
- package/dist/panoramic/architecture-ir-exporters.js.map +1 -0
- package/dist/panoramic/architecture-ir-generator.d.ts +36 -0
- package/dist/panoramic/architecture-ir-generator.d.ts.map +1 -0
- package/dist/panoramic/architecture-ir-generator.js +51 -0
- package/dist/panoramic/architecture-ir-generator.js.map +1 -0
- package/dist/panoramic/architecture-ir-mermaid-adapter.d.ts +6 -0
- package/dist/panoramic/architecture-ir-mermaid-adapter.d.ts.map +1 -0
- package/dist/panoramic/architecture-ir-mermaid-adapter.js +98 -0
- package/dist/panoramic/architecture-ir-mermaid-adapter.js.map +1 -0
- package/dist/panoramic/architecture-ir-model.d.ts +87 -0
- package/dist/panoramic/architecture-ir-model.d.ts.map +1 -0
- package/dist/panoramic/architecture-ir-model.js +20 -0
- package/dist/panoramic/architecture-ir-model.js.map +1 -0
- package/dist/panoramic/architecture-narrative.d.ts +68 -0
- package/dist/panoramic/architecture-narrative.d.ts.map +1 -0
- package/dist/panoramic/architecture-narrative.js +401 -0
- package/dist/panoramic/architecture-narrative.js.map +1 -0
- package/dist/panoramic/architecture-overview-generator.d.ts +31 -0
- package/dist/panoramic/architecture-overview-generator.d.ts.map +1 -0
- package/dist/panoramic/architecture-overview-generator.js +454 -0
- package/dist/panoramic/architecture-overview-generator.js.map +1 -0
- package/dist/panoramic/architecture-overview-model.d.ts +77 -0
- package/dist/panoramic/architecture-overview-model.d.ts.map +1 -0
- package/dist/panoramic/architecture-overview-model.js +31 -0
- package/dist/panoramic/architecture-overview-model.js.map +1 -0
- package/dist/panoramic/batch-project-docs.d.ts.map +1 -1
- package/dist/panoramic/batch-project-docs.js +4 -2
- package/dist/panoramic/batch-project-docs.js.map +1 -1
- package/dist/panoramic/builders/component-view-builder.d.ts +25 -1
- package/dist/panoramic/builders/component-view-builder.d.ts.map +1 -1
- package/dist/panoramic/builders/component-view-builder.js +83 -3
- package/dist/panoramic/builders/component-view-builder.js.map +1 -1
- package/dist/panoramic/builders/doc-graph-builder.d.ts +15 -2
- package/dist/panoramic/builders/doc-graph-builder.d.ts.map +1 -1
- package/dist/panoramic/builders/doc-graph-builder.js +20 -0
- package/dist/panoramic/builders/doc-graph-builder.js.map +1 -1
- package/dist/panoramic/component-view-builder.d.ts +17 -0
- package/dist/panoramic/component-view-builder.d.ts.map +1 -0
- package/dist/panoramic/component-view-builder.js +652 -0
- package/dist/panoramic/component-view-builder.js.map +1 -0
- package/dist/panoramic/component-view-model.d.ts +127 -0
- package/dist/panoramic/component-view-model.d.ts.map +1 -0
- package/dist/panoramic/component-view-model.js +65 -0
- package/dist/panoramic/component-view-model.js.map +1 -0
- package/dist/panoramic/config-reference-generator.d.ts +78 -0
- package/dist/panoramic/config-reference-generator.d.ts.map +1 -0
- package/dist/panoramic/config-reference-generator.js +199 -0
- package/dist/panoramic/config-reference-generator.js.map +1 -0
- package/dist/panoramic/coverage-auditor.d.ts +95 -0
- package/dist/panoramic/coverage-auditor.d.ts.map +1 -0
- package/dist/panoramic/coverage-auditor.js +290 -0
- package/dist/panoramic/coverage-auditor.js.map +1 -0
- package/dist/panoramic/cross-package-analyzer.d.ts +112 -0
- package/dist/panoramic/cross-package-analyzer.d.ts.map +1 -0
- package/dist/panoramic/cross-package-analyzer.js +258 -0
- package/dist/panoramic/cross-package-analyzer.js.map +1 -0
- package/dist/panoramic/data-model-generator.d.ts +528 -0
- package/dist/panoramic/data-model-generator.d.ts.map +1 -0
- package/dist/panoramic/data-model-generator.js +647 -0
- package/dist/panoramic/data-model-generator.js.map +1 -0
- package/dist/panoramic/doc-graph-builder.d.ts +80 -0
- package/dist/panoramic/doc-graph-builder.d.ts.map +1 -0
- package/dist/panoramic/doc-graph-builder.js +352 -0
- package/dist/panoramic/doc-graph-builder.js.map +1 -0
- package/dist/panoramic/docs-bundle-manifest-reader.d.ts +26 -0
- package/dist/panoramic/docs-bundle-manifest-reader.d.ts.map +1 -0
- package/dist/panoramic/docs-bundle-manifest-reader.js +106 -0
- package/dist/panoramic/docs-bundle-manifest-reader.js.map +1 -0
- package/dist/panoramic/docs-bundle-orchestrator.d.ts +14 -0
- package/dist/panoramic/docs-bundle-orchestrator.d.ts.map +1 -0
- package/dist/panoramic/docs-bundle-orchestrator.js +538 -0
- package/dist/panoramic/docs-bundle-orchestrator.js.map +1 -0
- package/dist/panoramic/docs-bundle-profiles.d.ts +4 -0
- package/dist/panoramic/docs-bundle-profiles.d.ts.map +1 -0
- package/dist/panoramic/docs-bundle-profiles.js +82 -0
- package/dist/panoramic/docs-bundle-profiles.js.map +1 -0
- package/dist/panoramic/docs-bundle-types.d.ts +94 -0
- package/dist/panoramic/docs-bundle-types.d.ts.map +1 -0
- package/dist/panoramic/docs-bundle-types.js +10 -0
- package/dist/panoramic/docs-bundle-types.js.map +1 -0
- package/dist/panoramic/docs-quality-evaluator.d.ts +31 -0
- package/dist/panoramic/docs-quality-evaluator.d.ts.map +1 -0
- package/dist/panoramic/docs-quality-evaluator.js +941 -0
- package/dist/panoramic/docs-quality-evaluator.js.map +1 -0
- package/dist/panoramic/docs-quality-model.d.ts +125 -0
- package/dist/panoramic/docs-quality-model.d.ts.map +1 -0
- package/dist/panoramic/docs-quality-model.js +166 -0
- package/dist/panoramic/docs-quality-model.js.map +1 -0
- package/dist/panoramic/dynamic-scenarios-builder.d.ts +18 -0
- package/dist/panoramic/dynamic-scenarios-builder.d.ts.map +1 -0
- package/dist/panoramic/dynamic-scenarios-builder.js +316 -0
- package/dist/panoramic/dynamic-scenarios-builder.js.map +1 -0
- package/dist/panoramic/event-surface-generator.d.ts +52 -0
- package/dist/panoramic/event-surface-generator.d.ts.map +1 -0
- package/dist/panoramic/event-surface-generator.js +507 -0
- package/dist/panoramic/event-surface-generator.js.map +1 -0
- package/dist/panoramic/generator-registry.d.ts +1 -1
- package/dist/panoramic/generator-registry.d.ts.map +1 -1
- package/dist/panoramic/generator-registry.js +5 -2
- package/dist/panoramic/generator-registry.js.map +1 -1
- package/dist/panoramic/generators/cross-package-analyzer.d.ts +5 -5
- package/dist/panoramic/generators/cross-package-analyzer.d.ts.map +1 -1
- package/dist/panoramic/generators/cross-package-analyzer.js +1 -1
- package/dist/panoramic/generators/cross-package-analyzer.js.map +1 -1
- package/dist/panoramic/generators/data-model-generator.d.ts +27 -27
- package/dist/panoramic/graph/confidence-mapper.d.ts +18 -0
- package/dist/panoramic/graph/confidence-mapper.d.ts.map +1 -1
- package/dist/panoramic/graph/confidence-mapper.js +26 -0
- package/dist/panoramic/graph/confidence-mapper.js.map +1 -1
- package/dist/panoramic/graph/graph-builder.d.ts +72 -2
- package/dist/panoramic/graph/graph-builder.d.ts.map +1 -1
- package/dist/panoramic/graph/graph-builder.js +330 -46
- package/dist/panoramic/graph/graph-builder.js.map +1 -1
- package/dist/panoramic/graph/graph-paths.d.ts +8 -0
- package/dist/panoramic/graph/graph-paths.d.ts.map +1 -1
- package/dist/panoramic/graph/graph-paths.js +10 -0
- package/dist/panoramic/graph/graph-paths.js.map +1 -1
- package/dist/panoramic/graph/graph-query.d.ts +49 -2
- package/dist/panoramic/graph/graph-query.d.ts.map +1 -1
- package/dist/panoramic/graph/graph-query.js +117 -19
- package/dist/panoramic/graph/graph-query.js.map +1 -1
- package/dist/panoramic/graph/graph-types.d.ts +32 -3
- package/dist/panoramic/graph/graph-types.d.ts.map +1 -1
- package/dist/panoramic/graph/index.d.ts +2 -1
- package/dist/panoramic/graph/index.d.ts.map +1 -1
- package/dist/panoramic/graph/index.js +1 -1
- package/dist/panoramic/graph/index.js.map +1 -1
- package/dist/panoramic/hyperedges/schema.d.ts +6 -6
- package/dist/panoramic/interface-surface-generator.d.ts +50 -0
- package/dist/panoramic/interface-surface-generator.d.ts.map +1 -0
- package/dist/panoramic/interface-surface-generator.js +388 -0
- package/dist/panoramic/interface-surface-generator.js.map +1 -0
- package/dist/panoramic/interfaces.d.ts +2 -2
- package/dist/panoramic/mock-readme-generator.d.ts +90 -0
- package/dist/panoramic/mock-readme-generator.d.ts.map +1 -0
- package/dist/panoramic/mock-readme-generator.js +111 -0
- package/dist/panoramic/mock-readme-generator.js.map +1 -0
- package/dist/panoramic/models/component-view-model.d.ts +1 -1
- package/dist/panoramic/models/component-view-model.d.ts.map +1 -1
- package/dist/panoramic/models/component-view-model.js.map +1 -1
- package/dist/panoramic/narrative-provenance-adapter.d.ts +8 -0
- package/dist/panoramic/narrative-provenance-adapter.d.ts.map +1 -0
- package/dist/panoramic/narrative-provenance-adapter.js +99 -0
- package/dist/panoramic/narrative-provenance-adapter.js.map +1 -0
- package/dist/panoramic/parser-registry.d.ts +1 -1
- package/dist/panoramic/parser-registry.js +1 -1
- package/dist/panoramic/parsers/types.d.ts +6 -6
- package/dist/panoramic/pattern-hints-generator.d.ts +28 -0
- package/dist/panoramic/pattern-hints-generator.d.ts.map +1 -0
- package/dist/panoramic/pattern-hints-generator.js +218 -0
- package/dist/panoramic/pattern-hints-generator.js.map +1 -0
- package/dist/panoramic/pattern-hints-model.d.ts +89 -0
- package/dist/panoramic/pattern-hints-model.d.ts.map +1 -0
- package/dist/panoramic/pattern-hints-model.js +61 -0
- package/dist/panoramic/pattern-hints-model.js.map +1 -0
- package/dist/panoramic/pattern-knowledge-base.d.ts +31 -0
- package/dist/panoramic/pattern-knowledge-base.d.ts.map +1 -0
- package/dist/panoramic/pattern-knowledge-base.js +407 -0
- package/dist/panoramic/pattern-knowledge-base.js.map +1 -0
- package/dist/panoramic/pipelines/adr-evidence-verifier.d.ts.map +1 -1
- package/dist/panoramic/pipelines/adr-evidence-verifier.js +1 -29
- package/dist/panoramic/pipelines/adr-evidence-verifier.js.map +1 -1
- package/dist/panoramic/pipelines/coverage-auditor.d.ts.map +1 -1
- package/dist/panoramic/pipelines/coverage-auditor.js +4 -2
- package/dist/panoramic/pipelines/coverage-auditor.js.map +1 -1
- package/dist/panoramic/pipelines/docs-bundle-orchestrator.js +13 -6
- package/dist/panoramic/pipelines/docs-bundle-orchestrator.js.map +1 -1
- package/dist/panoramic/pipelines/product-ux-docs.d.ts +5 -0
- package/dist/panoramic/pipelines/product-ux-docs.d.ts.map +1 -1
- package/dist/panoramic/pipelines/product-ux-docs.js +34 -6
- package/dist/panoramic/pipelines/product-ux-docs.js.map +1 -1
- package/dist/panoramic/product-ux-docs.d.ts +104 -0
- package/dist/panoramic/product-ux-docs.d.ts.map +1 -0
- package/dist/panoramic/product-ux-docs.js +817 -0
- package/dist/panoramic/product-ux-docs.js.map +1 -0
- package/dist/panoramic/qa/__tests__/citation.test.d.ts +2 -0
- package/dist/panoramic/qa/__tests__/citation.test.d.ts.map +1 -0
- package/dist/panoramic/qa/__tests__/citation.test.js +193 -0
- package/dist/panoramic/qa/__tests__/citation.test.js.map +1 -0
- package/dist/panoramic/qa/__tests__/debt-context.test.d.ts +2 -0
- package/dist/panoramic/qa/__tests__/debt-context.test.d.ts.map +1 -0
- package/dist/panoramic/qa/__tests__/debt-context.test.js +154 -0
- package/dist/panoramic/qa/__tests__/debt-context.test.js.map +1 -0
- package/dist/panoramic/qa/__tests__/graph-retriever.test.d.ts +2 -0
- package/dist/panoramic/qa/__tests__/graph-retriever.test.d.ts.map +1 -0
- package/dist/panoramic/qa/__tests__/graph-retriever.test.js +148 -0
- package/dist/panoramic/qa/__tests__/graph-retriever.test.js.map +1 -0
- package/dist/panoramic/qa/__tests__/index.test.d.ts +2 -0
- package/dist/panoramic/qa/__tests__/index.test.d.ts.map +1 -0
- package/dist/panoramic/qa/__tests__/index.test.js +246 -0
- package/dist/panoramic/qa/__tests__/index.test.js.map +1 -0
- package/dist/panoramic/qa/__tests__/llm-caller.test.d.ts +2 -0
- package/dist/panoramic/qa/__tests__/llm-caller.test.d.ts.map +1 -0
- package/dist/panoramic/qa/__tests__/llm-caller.test.js +168 -0
- package/dist/panoramic/qa/__tests__/llm-caller.test.js.map +1 -0
- package/dist/panoramic/qa/__tests__/prompt-builder.test.d.ts +2 -0
- package/dist/panoramic/qa/__tests__/prompt-builder.test.d.ts.map +1 -0
- package/dist/panoramic/qa/__tests__/prompt-builder.test.js +150 -0
- package/dist/panoramic/qa/__tests__/prompt-builder.test.js.map +1 -0
- package/dist/panoramic/qa/__tests__/qa-integration.test.d.ts +2 -0
- package/dist/panoramic/qa/__tests__/qa-integration.test.d.ts.map +1 -0
- package/dist/panoramic/qa/__tests__/qa-integration.test.js +296 -0
- package/dist/panoramic/qa/__tests__/qa-integration.test.js.map +1 -0
- package/dist/panoramic/qa/__tests__/rag-reranker.test.d.ts +2 -0
- package/dist/panoramic/qa/__tests__/rag-reranker.test.d.ts.map +1 -0
- package/dist/panoramic/qa/__tests__/rag-reranker.test.js +136 -0
- package/dist/panoramic/qa/__tests__/rag-reranker.test.js.map +1 -0
- package/dist/panoramic/qa/index.d.ts.map +1 -1
- package/dist/panoramic/qa/index.js +10 -3
- package/dist/panoramic/qa/index.js.map +1 -1
- package/dist/panoramic/query.d.ts +1 -0
- package/dist/panoramic/query.d.ts.map +1 -1
- package/dist/panoramic/query.js +5 -0
- package/dist/panoramic/query.js.map +1 -1
- package/dist/panoramic/runtime-topology-generator.d.ts +77 -0
- package/dist/panoramic/runtime-topology-generator.d.ts.map +1 -0
- package/dist/panoramic/runtime-topology-generator.js +676 -0
- package/dist/panoramic/runtime-topology-generator.js.map +1 -0
- package/dist/panoramic/runtime-topology-model.d.ts +145 -0
- package/dist/panoramic/runtime-topology-model.d.ts.map +1 -0
- package/dist/panoramic/runtime-topology-model.js +300 -0
- package/dist/panoramic/runtime-topology-model.js.map +1 -0
- package/dist/panoramic/troubleshooting-generator.d.ts +51 -0
- package/dist/panoramic/troubleshooting-generator.d.ts.map +1 -0
- package/dist/panoramic/troubleshooting-generator.js +451 -0
- package/dist/panoramic/troubleshooting-generator.js.map +1 -0
- package/dist/panoramic/utils/text-segmenter.d.ts +15 -0
- package/dist/panoramic/utils/text-segmenter.d.ts.map +1 -1
- package/dist/panoramic/utils/text-segmenter.js +38 -0
- package/dist/panoramic/utils/text-segmenter.js.map +1 -1
- package/dist/panoramic/workspace-index-generator.d.ts +148 -0
- package/dist/panoramic/workspace-index-generator.d.ts.map +1 -0
- package/dist/panoramic/workspace-index-generator.js +497 -0
- package/dist/panoramic/workspace-index-generator.js.map +1 -0
- package/dist/runtime-bootstrap.d.ts +9 -0
- package/dist/runtime-bootstrap.d.ts.map +1 -0
- package/dist/runtime-bootstrap.js +33 -0
- package/dist/runtime-bootstrap.js.map +1 -0
- package/dist/scaffold-kb/api-entities-serializer.d.ts +12 -0
- package/dist/scaffold-kb/api-entities-serializer.d.ts.map +1 -0
- package/dist/scaffold-kb/api-entities-serializer.js +144 -0
- package/dist/scaffold-kb/api-entities-serializer.js.map +1 -0
- package/dist/scaffold-kb/arbitration.d.ts +44 -0
- package/dist/scaffold-kb/arbitration.d.ts.map +1 -0
- package/dist/scaffold-kb/arbitration.js +144 -0
- package/dist/scaffold-kb/arbitration.js.map +1 -0
- package/dist/scaffold-kb/chunk-splitter.d.ts +20 -0
- package/dist/scaffold-kb/chunk-splitter.d.ts.map +1 -0
- package/dist/scaffold-kb/chunk-splitter.js +309 -0
- package/dist/scaffold-kb/chunk-splitter.js.map +1 -0
- package/dist/scaffold-kb/doc-graph-builder.d.ts +26 -0
- package/dist/scaffold-kb/doc-graph-builder.d.ts.map +1 -0
- package/dist/scaffold-kb/doc-graph-builder.js +72 -0
- package/dist/scaffold-kb/doc-graph-builder.js.map +1 -0
- package/dist/scaffold-kb/entity-extractor.d.ts +40 -0
- package/dist/scaffold-kb/entity-extractor.d.ts.map +1 -0
- package/dist/scaffold-kb/entity-extractor.js +0 -0
- package/dist/scaffold-kb/entity-extractor.js.map +1 -0
- package/dist/scaffold-kb/entity-heuristic.d.ts +14 -0
- package/dist/scaffold-kb/entity-heuristic.d.ts.map +1 -0
- package/dist/scaffold-kb/entity-heuristic.js +107 -0
- package/dist/scaffold-kb/entity-heuristic.js.map +1 -0
- package/dist/scaffold-kb/entity-matcher.d.ts +20 -0
- package/dist/scaffold-kb/entity-matcher.d.ts.map +1 -0
- package/dist/scaffold-kb/entity-matcher.js +53 -0
- package/dist/scaffold-kb/entity-matcher.js.map +1 -0
- package/dist/scaffold-kb/entity-util.d.ts +11 -0
- package/dist/scaffold-kb/entity-util.d.ts.map +1 -0
- package/dist/scaffold-kb/entity-util.js +23 -0
- package/dist/scaffold-kb/entity-util.js.map +1 -0
- package/dist/scaffold-kb/evidence-envelope.d.ts +17 -0
- package/dist/scaffold-kb/evidence-envelope.d.ts.map +1 -0
- package/dist/scaffold-kb/evidence-envelope.js +32 -0
- package/dist/scaffold-kb/evidence-envelope.js.map +1 -0
- package/dist/scaffold-kb/index.d.ts +20 -0
- package/dist/scaffold-kb/index.d.ts.map +1 -0
- package/dist/scaffold-kb/index.js +102 -0
- package/dist/scaffold-kb/index.js.map +1 -0
- package/dist/scaffold-kb/ingest/ingest-core.d.ts +47 -0
- package/dist/scaffold-kb/ingest/ingest-core.d.ts.map +1 -0
- package/dist/scaffold-kb/ingest/ingest-core.js +231 -0
- package/dist/scaffold-kb/ingest/ingest-core.js.map +1 -0
- package/dist/scaffold-kb/ingest/office-parser.d.ts +33 -0
- package/dist/scaffold-kb/ingest/office-parser.d.ts.map +1 -0
- package/dist/scaffold-kb/ingest/office-parser.js +187 -0
- package/dist/scaffold-kb/ingest/office-parser.js.map +1 -0
- package/dist/scaffold-kb/ingest/url-fetcher.d.ts +43 -0
- package/dist/scaffold-kb/ingest/url-fetcher.d.ts.map +1 -0
- package/dist/scaffold-kb/ingest/url-fetcher.js +245 -0
- package/dist/scaffold-kb/ingest/url-fetcher.js.map +1 -0
- package/dist/scaffold-kb/ingester.d.ts +38 -0
- package/dist/scaffold-kb/ingester.d.ts.map +1 -0
- package/dist/scaffold-kb/ingester.js +261 -0
- package/dist/scaffold-kb/ingester.js.map +1 -0
- package/dist/scaffold-kb/injection-format.d.ts +25 -0
- package/dist/scaffold-kb/injection-format.d.ts.map +1 -0
- package/dist/scaffold-kb/injection-format.js +55 -0
- package/dist/scaffold-kb/injection-format.js.map +1 -0
- package/dist/scaffold-kb/kb-writer.d.ts +13 -0
- package/dist/scaffold-kb/kb-writer.d.ts.map +1 -0
- package/dist/scaffold-kb/kb-writer.js +69 -0
- package/dist/scaffold-kb/kb-writer.js.map +1 -0
- package/dist/scaffold-kb/keyword-extract.d.ts +16 -0
- package/dist/scaffold-kb/keyword-extract.d.ts.map +1 -0
- package/dist/scaffold-kb/keyword-extract.js +49 -0
- package/dist/scaffold-kb/keyword-extract.js.map +1 -0
- package/dist/scaffold-kb/query-sanitizer.d.ts +28 -0
- package/dist/scaffold-kb/query-sanitizer.d.ts.map +1 -0
- package/dist/scaffold-kb/query-sanitizer.js +38 -0
- package/dist/scaffold-kb/query-sanitizer.js.map +1 -0
- package/dist/scaffold-kb/recall-eval.d.ts +48 -0
- package/dist/scaffold-kb/recall-eval.d.ts.map +1 -0
- package/dist/scaffold-kb/recall-eval.js +55 -0
- package/dist/scaffold-kb/recall-eval.js.map +1 -0
- package/dist/scaffold-kb/schema-compat.d.ts +22 -0
- package/dist/scaffold-kb/schema-compat.d.ts.map +1 -0
- package/dist/scaffold-kb/schema-compat.js +32 -0
- package/dist/scaffold-kb/schema-compat.js.map +1 -0
- package/dist/scaffold-kb/search-core.d.ts +42 -0
- package/dist/scaffold-kb/search-core.d.ts.map +1 -0
- package/dist/scaffold-kb/search-core.js +89 -0
- package/dist/scaffold-kb/search-core.js.map +1 -0
- package/dist/scaffold-kb/sqlite-engine.d.ts +58 -0
- package/dist/scaffold-kb/sqlite-engine.d.ts.map +1 -0
- package/dist/scaffold-kb/sqlite-engine.js +61 -0
- package/dist/scaffold-kb/sqlite-engine.js.map +1 -0
- package/dist/scaffold-kb/sqlite-writer.d.ts +18 -0
- package/dist/scaffold-kb/sqlite-writer.d.ts.map +1 -0
- package/dist/scaffold-kb/sqlite-writer.js +76 -0
- package/dist/scaffold-kb/sqlite-writer.js.map +1 -0
- package/dist/scaffold-kb/tokenizer.d.ts +26 -0
- package/dist/scaffold-kb/tokenizer.d.ts.map +1 -0
- package/dist/scaffold-kb/tokenizer.js +97 -0
- package/dist/scaffold-kb/tokenizer.js.map +1 -0
- package/dist/scaffold-kb/types.d.ts +157 -0
- package/dist/scaffold-kb/types.d.ts.map +1 -0
- package/dist/scaffold-kb/types.js +5 -0
- package/dist/scaffold-kb/types.js.map +1 -0
- package/dist/utils/file-scanner.d.ts +15 -0
- package/dist/utils/file-scanner.d.ts.map +1 -1
- package/dist/utils/file-scanner.js +18 -1
- package/dist/utils/file-scanner.js.map +1 -1
- package/dist/utils/string-distance.d.ts +20 -0
- package/dist/utils/string-distance.d.ts.map +1 -0
- package/dist/utils/string-distance.js +38 -0
- package/dist/utils/string-distance.js.map +1 -0
- package/package.json +22 -3
- package/plugins/demo-kb-en/.claude-plugin/plugin.json +11 -0
- package/plugins/demo-kb-en/.mcp.json +8 -0
- package/plugins/demo-kb-en/FIXTURE.json +23 -0
- package/plugins/demo-kb-en/FIXTURE.md +53 -0
- package/plugins/demo-kb-en/ingest-samples/meeting-notes.md +13 -0
- package/plugins/demo-kb-en/kb/api-entities.json +3445 -0
- package/plugins/demo-kb-en/kb/chunks.sqlite +0 -0
- package/plugins/demo-kb-en/kb/doc-graph.json +113 -0
- package/plugins/demo-kb-zh/.claude-plugin/plugin.json +11 -0
- package/plugins/demo-kb-zh/.mcp.json +8 -0
- package/plugins/demo-kb-zh/FIXTURE.json +23 -0
- package/plugins/demo-kb-zh/FIXTURE.md +63 -0
- package/plugins/demo-kb-zh/ingest-samples/meeting-notes.md +13 -0
- package/plugins/demo-kb-zh/kb/api-entities.json +990 -0
- package/plugins/demo-kb-zh/kb/chunks.sqlite +0 -0
- package/plugins/demo-kb-zh/kb/doc-graph.json +113 -0
- package/plugins/spec-driver/.claude-plugin/plugin.json +1 -1
- package/plugins/spec-driver/README.md +50 -1
- package/plugins/spec-driver/agents/implement.md +20 -1
- package/plugins/spec-driver/agents/plan.md +20 -1
- package/plugins/spec-driver/agents/quality-review.md +20 -1
- package/plugins/spec-driver/agents/spec-review.md +20 -1
- package/plugins/spec-driver/agents/verify.md +108 -1
- package/plugins/spec-driver/config/orchestration.yaml +39 -31
- package/plugins/spec-driver/contracts/orchestration-overrides-contract.yaml +27 -0
- package/plugins/spec-driver/contracts/orchestration-schema.mjs +272 -227
- package/plugins/spec-driver/contracts/wrapper-source-of-truth.yaml +3 -0
- package/plugins/spec-driver/docs/customization.md +134 -0
- package/plugins/spec-driver/docs/spectra-mcp-integration.md +165 -0
- package/plugins/spec-driver/hooks/hooks.json +9 -0
- package/plugins/spec-driver/hooks/stop-fix-compliance-check.sh +36 -0
- package/plugins/spec-driver/lib/delegation-contract.mjs +84 -0
- package/plugins/spec-driver/lib/orchestration-resolver.mjs +49 -0
- package/plugins/spec-driver/lib/orchestrator.mjs +29 -1
- package/plugins/spec-driver/lib/preference-rules.mjs +96 -0
- package/plugins/spec-driver/scripts/codex-skills.sh +28 -28
- package/plugins/spec-driver/scripts/dev/spike-fix-compliance-e2e.mjs +130 -0
- package/plugins/spec-driver/scripts/fix-compliance-judge.mjs +415 -0
- package/plugins/spec-driver/scripts/generate-adoption-insights.mjs +48 -1
- package/plugins/spec-driver/scripts/goal-loop-cli.mjs +307 -0
- package/plugins/spec-driver/scripts/init-project.sh +56 -0
- package/plugins/spec-driver/scripts/kb-prequery.mjs +125 -0
- package/plugins/spec-driver/scripts/lib/config-schema.mjs +200 -94
- package/plugins/spec-driver/scripts/lib/ensure-gitignore.sh +277 -0
- package/plugins/spec-driver/scripts/lib/extract-wrapper-body.mjs +136 -0
- package/plugins/spec-driver/scripts/lib/fix-compliance-core.mjs +434 -0
- package/plugins/spec-driver/scripts/lib/fix-compliance-io.mjs +335 -0
- package/plugins/spec-driver/scripts/lib/goal-loop-core.mjs +783 -0
- package/plugins/spec-driver/scripts/lib/init-project-output.sh +28 -0
- package/plugins/spec-driver/scripts/lib/load-zod.mjs +62 -0
- package/plugins/spec-driver/scripts/lib/project-profile-resolver.mjs +141 -49
- package/plugins/spec-driver/scripts/lib/project-profile-schema.mjs +88 -54
- package/plugins/spec-driver/scripts/postinstall.sh +13 -1
- package/plugins/spec-driver/scripts/record-workflow-run.mjs +75 -0
- package/plugins/spec-driver/scripts/sync-delegation-contract.mjs +187 -0
- package/plugins/spec-driver/scripts/sync-preference-rules.mjs +148 -0
- package/plugins/spec-driver/scripts/validate-orchestrator-models.mjs +156 -0
- package/plugins/spec-driver/scripts/validate-wrapper-sources.mjs +46 -1
- package/plugins/spec-driver/skills/spec-driver-feature/SKILL.md +302 -0
- package/plugins/spec-driver/skills/spec-driver-fix/SKILL.md +97 -8
- package/plugins/spec-driver/skills/spec-driver-implement/SKILL.md +18 -0
- package/plugins/spec-driver/skills/spec-driver-refactor/SKILL.md +10 -0
- package/plugins/spec-driver/skills/spec-driver-resume/SKILL.md +12 -1
- package/plugins/spec-driver/skills/spec-driver-story/SKILL.md +30 -0
- package/plugins/spec-driver/templates/delegation-contract.md +27 -0
- package/plugins/spec-driver/templates/goal-loop-override-template.yaml +284 -0
- package/plugins/spec-driver/templates/preference-rules.md +39 -0
- package/plugins/spec-driver/templates/specify-base/project-context-template.yaml +10 -0
- package/plugins/spectra/.claude-plugin/plugin.json +1 -1
- package/plugins/spectra/README.md +48 -1
- package/plugins/spectra/hooks/post-commit.sh +40 -0
- package/templates/module-spec.hbs +1 -0
- package/templates/product-overview.hbs +1 -1
|
@@ -0,0 +1,309 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* F190 scaffold-kb — chunk-splitter
|
|
3
|
+
*
|
|
4
|
+
* 将 ParsedDoc 按三级策略切分为 Chunk[]:
|
|
5
|
+
* 1. 标题级(## / ###)— 每个标题节为候选单元
|
|
6
|
+
* 2. 段落级(\n\n)— 超 400 token 候选按空行切段,组合到接近 400 token
|
|
7
|
+
* 3. 句子级兜底(句号 / 换行)— 单段落超 400 token 时切
|
|
8
|
+
*
|
|
9
|
+
* token 计量:约 4 字符 = 1 token(无外部依赖)
|
|
10
|
+
* 单 chunk 上限:500 token(≈ 2000 字符)
|
|
11
|
+
* 最小 chunk:20 token(≈ 80 字符),不足则合并到邻近
|
|
12
|
+
*/
|
|
13
|
+
// ── 常量 ────────────────────────────────────────────────────────────────────
|
|
14
|
+
/** 约 4 字符 = 1 token 的估算系数 */
|
|
15
|
+
const CHARS_PER_TOKEN = 4;
|
|
16
|
+
/** 段落级再切阈值(token)— 超过此值的标题节需要继续切 */
|
|
17
|
+
const PARAGRAPH_SPLIT_THRESHOLD_TOKENS = 400;
|
|
18
|
+
/** 单 chunk 上限(token) */
|
|
19
|
+
const MAX_CHUNK_TOKENS = 500;
|
|
20
|
+
/** 最小 chunk 阈值(token)— 不足此值合并到邻近 */
|
|
21
|
+
const MIN_CHUNK_TOKENS = 20;
|
|
22
|
+
// 阈值对应字符数
|
|
23
|
+
const PARAGRAPH_SPLIT_THRESHOLD_CHARS = PARAGRAPH_SPLIT_THRESHOLD_TOKENS * CHARS_PER_TOKEN; // 1600
|
|
24
|
+
const MAX_CHUNK_CHARS = MAX_CHUNK_TOKENS * CHARS_PER_TOKEN; // 2000
|
|
25
|
+
const MIN_CHUNK_CHARS = MIN_CHUNK_TOKENS * CHARS_PER_TOKEN; // 80
|
|
26
|
+
// ── 工具函数 ─────────────────────────────────────────────────────────────────
|
|
27
|
+
/**
|
|
28
|
+
* 将标题文本转为 URL slug
|
|
29
|
+
* 小写、空格转 `-`、去除非字母数字和连字符以外的字符
|
|
30
|
+
*/
|
|
31
|
+
function toSlug(heading) {
|
|
32
|
+
return heading
|
|
33
|
+
.toLowerCase()
|
|
34
|
+
.replace(/\s+/g, '-')
|
|
35
|
+
.replace(/[^a-z0-9一-鿿㐀-䶿\-]/g, '') // 保留中文字符和英文字母数字连字符
|
|
36
|
+
.replace(/-+/g, '-')
|
|
37
|
+
.replace(/^-|-$/g, '');
|
|
38
|
+
}
|
|
39
|
+
/**
|
|
40
|
+
* 估算文本的 token 数(4 字符 = 1 token)
|
|
41
|
+
*/
|
|
42
|
+
function estimateTokens(text) {
|
|
43
|
+
return Math.ceil(text.length / CHARS_PER_TOKEN);
|
|
44
|
+
}
|
|
45
|
+
/**
|
|
46
|
+
* 从 Markdown 正文中提取标题文本(去掉 ## / ### 前缀)
|
|
47
|
+
*/
|
|
48
|
+
function extractHeadingText(line) {
|
|
49
|
+
return line.replace(/^#{1,6}\s+/, '').trim();
|
|
50
|
+
}
|
|
51
|
+
/**
|
|
52
|
+
* 判断是否为标题行(仅处理 ## 和 ### 级别)
|
|
53
|
+
*/
|
|
54
|
+
function isHeadingLine(line) {
|
|
55
|
+
return /^#{2,3}\s/.test(line);
|
|
56
|
+
}
|
|
57
|
+
/**
|
|
58
|
+
* 按 ## / ### 标题将文档正文切成标题节列表
|
|
59
|
+
*/
|
|
60
|
+
function splitByHeadings(content) {
|
|
61
|
+
const lines = content.split('\n');
|
|
62
|
+
const sections = [];
|
|
63
|
+
let currentHeading = null;
|
|
64
|
+
let bodyLines = [];
|
|
65
|
+
for (const line of lines) {
|
|
66
|
+
if (isHeadingLine(line)) {
|
|
67
|
+
// 遇到新标题:保存上一节(如果有内容)
|
|
68
|
+
const bodyText = bodyLines.join('\n').trim();
|
|
69
|
+
if (bodyText.length > 0 || currentHeading !== null) {
|
|
70
|
+
sections.push({ heading: currentHeading, body: bodyText });
|
|
71
|
+
}
|
|
72
|
+
currentHeading = extractHeadingText(line);
|
|
73
|
+
bodyLines = [];
|
|
74
|
+
}
|
|
75
|
+
else {
|
|
76
|
+
bodyLines.push(line);
|
|
77
|
+
}
|
|
78
|
+
}
|
|
79
|
+
// 最后一节
|
|
80
|
+
const bodyText = bodyLines.join('\n').trim();
|
|
81
|
+
if (bodyText.length > 0 || currentHeading !== null) {
|
|
82
|
+
sections.push({ heading: currentHeading, body: bodyText });
|
|
83
|
+
}
|
|
84
|
+
return sections;
|
|
85
|
+
}
|
|
86
|
+
// ── 段落与句子切分 ────────────────────────────────────────────────────────────
|
|
87
|
+
/**
|
|
88
|
+
* 将文本按空行(\n\n)切成段落列表(过滤空段落)
|
|
89
|
+
*/
|
|
90
|
+
function splitToParagraphs(text) {
|
|
91
|
+
return text
|
|
92
|
+
.split(/\n\n+/)
|
|
93
|
+
.map((p) => p.trim())
|
|
94
|
+
.filter((p) => p.length > 0);
|
|
95
|
+
}
|
|
96
|
+
/**
|
|
97
|
+
* 将段落按句子切分(句号、问号、感叹号、换行后兜底)
|
|
98
|
+
* 尽量不跨语义单元强切
|
|
99
|
+
*/
|
|
100
|
+
function splitToSentences(paragraph) {
|
|
101
|
+
// 按中英文句号、问号、感叹号切分;保留分隔符在前句末尾
|
|
102
|
+
// 同时按换行切(代码块等大段落)
|
|
103
|
+
const parts = [];
|
|
104
|
+
// 先按换行切,再对每行按句子切
|
|
105
|
+
const lines = paragraph.split('\n');
|
|
106
|
+
for (const line of lines) {
|
|
107
|
+
if (line.trim().length === 0)
|
|
108
|
+
continue;
|
|
109
|
+
// 按中英文句号/问号/感叹号切分,保留标点在前面
|
|
110
|
+
const sentenceBreaks = line.split(/(?<=[。!?.!?])\s*/);
|
|
111
|
+
for (const s of sentenceBreaks) {
|
|
112
|
+
const trimmed = s.trim();
|
|
113
|
+
if (trimmed.length > 0) {
|
|
114
|
+
parts.push(trimmed);
|
|
115
|
+
}
|
|
116
|
+
}
|
|
117
|
+
}
|
|
118
|
+
return parts.filter((p) => p.length > 0);
|
|
119
|
+
}
|
|
120
|
+
/**
|
|
121
|
+
* 将多段落组合为不超过 MAX_CHUNK_CHARS 的文本块
|
|
122
|
+
* 每个段落超过 PARAGRAPH_SPLIT_THRESHOLD_CHARS 时先拆成句子再组合
|
|
123
|
+
*
|
|
124
|
+
* 句子/段落组合目标是接近 400 token(PARAGRAPH_SPLIT_THRESHOLD_CHARS),
|
|
125
|
+
* 绝对上限是 500 token(MAX_CHUNK_CHARS)。
|
|
126
|
+
*/
|
|
127
|
+
function combineParagraphs(paragraphs) {
|
|
128
|
+
const chunks = [];
|
|
129
|
+
let current = '';
|
|
130
|
+
/**
|
|
131
|
+
* 将一批"原子单元"(句子)追加合并到 chunks。
|
|
132
|
+
* 目标大小:PARAGRAPH_SPLIT_THRESHOLD_CHARS(1600 字符 ≈ 400 token)
|
|
133
|
+
* 绝对上限:MAX_CHUNK_CHARS(2000 字符 ≈ 500 token)
|
|
134
|
+
*/
|
|
135
|
+
function flushAtoms(atoms) {
|
|
136
|
+
for (const atom of atoms) {
|
|
137
|
+
if (atom.length > MAX_CHUNK_CHARS) {
|
|
138
|
+
// 单个原子超绝对上限:强制截断(兜底,极端情况)
|
|
139
|
+
if (current.length > 0) {
|
|
140
|
+
chunks.push(current);
|
|
141
|
+
current = '';
|
|
142
|
+
}
|
|
143
|
+
let remaining = atom;
|
|
144
|
+
while (remaining.length > MAX_CHUNK_CHARS) {
|
|
145
|
+
chunks.push(remaining.slice(0, MAX_CHUNK_CHARS));
|
|
146
|
+
remaining = remaining.slice(MAX_CHUNK_CHARS);
|
|
147
|
+
}
|
|
148
|
+
if (remaining.length > 0) {
|
|
149
|
+
current = remaining;
|
|
150
|
+
}
|
|
151
|
+
}
|
|
152
|
+
else {
|
|
153
|
+
// 追加后超过目标大小(400 token)时关闭当前块,开始新块
|
|
154
|
+
const separator = current.length > 0 ? '\n' : '';
|
|
155
|
+
const nextLength = current.length + separator.length + atom.length;
|
|
156
|
+
if (nextLength > PARAGRAPH_SPLIT_THRESHOLD_CHARS && current.length > 0) {
|
|
157
|
+
chunks.push(current);
|
|
158
|
+
current = atom;
|
|
159
|
+
}
|
|
160
|
+
else {
|
|
161
|
+
current = current.length > 0 ? current + '\n' + atom : atom;
|
|
162
|
+
}
|
|
163
|
+
}
|
|
164
|
+
}
|
|
165
|
+
}
|
|
166
|
+
for (const para of paragraphs) {
|
|
167
|
+
if (para.length > PARAGRAPH_SPLIT_THRESHOLD_CHARS) {
|
|
168
|
+
// 段落过大:先按句子切,再组合
|
|
169
|
+
const sentences = splitToSentences(para);
|
|
170
|
+
flushAtoms(sentences);
|
|
171
|
+
}
|
|
172
|
+
else if (current.length + (current.length > 0 ? 2 : 0) + para.length > MAX_CHUNK_CHARS) {
|
|
173
|
+
// 追加后超绝对上限:保存当前,开新 chunk
|
|
174
|
+
if (current.length > 0) {
|
|
175
|
+
chunks.push(current);
|
|
176
|
+
current = '';
|
|
177
|
+
}
|
|
178
|
+
current = para;
|
|
179
|
+
}
|
|
180
|
+
else {
|
|
181
|
+
current = current.length > 0 ? current + '\n\n' + para : para;
|
|
182
|
+
}
|
|
183
|
+
}
|
|
184
|
+
if (current.length > 0) {
|
|
185
|
+
chunks.push(current);
|
|
186
|
+
}
|
|
187
|
+
return chunks;
|
|
188
|
+
}
|
|
189
|
+
// ── 最小 chunk 合并 ────────────────────────────────────────────────────────
|
|
190
|
+
/**
|
|
191
|
+
* 合并过短(< MIN_CHUNK_CHARS)的文本块到邻近块
|
|
192
|
+
* 优先合并到前一块;若前一块不存在则合并到后一块
|
|
193
|
+
*/
|
|
194
|
+
function mergeSmallChunks(texts) {
|
|
195
|
+
if (texts.length === 0)
|
|
196
|
+
return [];
|
|
197
|
+
// 先收集合并结果
|
|
198
|
+
const result = [];
|
|
199
|
+
for (let i = 0; i < texts.length; i++) {
|
|
200
|
+
const text = texts[i];
|
|
201
|
+
// 断言:texts[i] 在循环内已通过索引 i 访问,但 noUncheckedIndexedAccess 要求检查
|
|
202
|
+
if (text === undefined)
|
|
203
|
+
continue;
|
|
204
|
+
if (text.length < MIN_CHUNK_CHARS) {
|
|
205
|
+
if (result.length > 0) {
|
|
206
|
+
// 合并到前一块
|
|
207
|
+
const prev = result[result.length - 1];
|
|
208
|
+
if (prev !== undefined) {
|
|
209
|
+
result[result.length - 1] = prev + '\n\n' + text;
|
|
210
|
+
}
|
|
211
|
+
}
|
|
212
|
+
else {
|
|
213
|
+
// 没有前一块:先缓存,后面尝试合并到下一块
|
|
214
|
+
result.push(text);
|
|
215
|
+
}
|
|
216
|
+
}
|
|
217
|
+
else {
|
|
218
|
+
if (result.length > 0 &&
|
|
219
|
+
(result[result.length - 1]?.length ?? 0) < MIN_CHUNK_CHARS) {
|
|
220
|
+
// 前一个太小,合并进来
|
|
221
|
+
const prev = result[result.length - 1];
|
|
222
|
+
if (prev !== undefined) {
|
|
223
|
+
result[result.length - 1] = prev + '\n\n' + text;
|
|
224
|
+
}
|
|
225
|
+
}
|
|
226
|
+
else {
|
|
227
|
+
result.push(text);
|
|
228
|
+
}
|
|
229
|
+
}
|
|
230
|
+
}
|
|
231
|
+
return result.filter((t) => t.trim().length > 0);
|
|
232
|
+
}
|
|
233
|
+
// ── 主函数 ──────────────────────────────────────────────────────────────────
|
|
234
|
+
/**
|
|
235
|
+
* 将 ParsedDoc 按三级策略切分为 Chunk[]
|
|
236
|
+
*
|
|
237
|
+
* 幂等保证:相同输入产出相同 chunkId 集合与顺序
|
|
238
|
+
*/
|
|
239
|
+
export function splitDocument(doc) {
|
|
240
|
+
const sections = splitByHeadings(doc.content);
|
|
241
|
+
// anchor 序号计数器(同一 anchor 下多个 chunk 时追加序号)
|
|
242
|
+
const anchorCounts = new Map();
|
|
243
|
+
/**
|
|
244
|
+
* 为给定 anchor 分配下一个序号(首次分配返回 1,后续递增)
|
|
245
|
+
*/
|
|
246
|
+
function nextAnchorIndex(anchor) {
|
|
247
|
+
const count = (anchorCounts.get(anchor) ?? 0) + 1;
|
|
248
|
+
anchorCounts.set(anchor, count);
|
|
249
|
+
return count;
|
|
250
|
+
}
|
|
251
|
+
const allChunks = [];
|
|
252
|
+
// 无标题文档的全局序号(anchor 为 null 时使用 _N)
|
|
253
|
+
let noHeadingCounter = 0;
|
|
254
|
+
for (const section of sections) {
|
|
255
|
+
const anchor = section.heading !== null ? toSlug(section.heading) : null;
|
|
256
|
+
// 候选文本:标题(如有)+ 正文
|
|
257
|
+
const sectionText = section.heading !== null
|
|
258
|
+
? `## ${section.heading}\n\n${section.body}`.trim()
|
|
259
|
+
: section.body;
|
|
260
|
+
if (sectionText.trim().length === 0)
|
|
261
|
+
continue;
|
|
262
|
+
// 决定是否需要进一步切分
|
|
263
|
+
let rawTexts;
|
|
264
|
+
if (estimateTokens(sectionText) <= PARAGRAPH_SPLIT_THRESHOLD_TOKENS) {
|
|
265
|
+
// 整节 ≤ 400 token:直接作为一个候选块
|
|
266
|
+
rawTexts = [sectionText];
|
|
267
|
+
}
|
|
268
|
+
else {
|
|
269
|
+
// 超过 400 token:按段落+句子切分
|
|
270
|
+
const paragraphs = splitToParagraphs(
|
|
271
|
+
// 正文部分切段落(不含标题行本身)
|
|
272
|
+
section.body);
|
|
273
|
+
const combined = combineParagraphs(paragraphs);
|
|
274
|
+
const merged = mergeSmallChunks(combined);
|
|
275
|
+
// 如果正文被切开,把标题前缀加到第一块
|
|
276
|
+
if (merged.length > 0 && section.heading !== null) {
|
|
277
|
+
merged[0] = `## ${section.heading}\n\n${merged[0] ?? ''}`;
|
|
278
|
+
}
|
|
279
|
+
rawTexts = merged.length > 0 ? merged : [sectionText];
|
|
280
|
+
}
|
|
281
|
+
// 最终过滤 + 生成 Chunk 对象
|
|
282
|
+
for (const text of rawTexts) {
|
|
283
|
+
if (text.trim().length === 0)
|
|
284
|
+
continue;
|
|
285
|
+
// 过短:计入下一次迭代合并(此处已经过 mergeSmallChunks,理论上不应出现)
|
|
286
|
+
// 但仍做最终检查
|
|
287
|
+
if (text.length < MIN_CHUNK_CHARS && rawTexts.length > 1)
|
|
288
|
+
continue;
|
|
289
|
+
// 生成 chunkId
|
|
290
|
+
let chunkId;
|
|
291
|
+
if (anchor !== null) {
|
|
292
|
+
const idx = nextAnchorIndex(anchor);
|
|
293
|
+
chunkId = idx === 1 ? `${doc.id}#${anchor}` : `${doc.id}#${anchor}-${idx}`;
|
|
294
|
+
}
|
|
295
|
+
else {
|
|
296
|
+
noHeadingCounter += 1;
|
|
297
|
+
chunkId = `${doc.id}#_${noHeadingCounter}`;
|
|
298
|
+
}
|
|
299
|
+
allChunks.push({
|
|
300
|
+
chunkId,
|
|
301
|
+
docId: doc.id,
|
|
302
|
+
contentRaw: text,
|
|
303
|
+
anchor,
|
|
304
|
+
});
|
|
305
|
+
}
|
|
306
|
+
}
|
|
307
|
+
return allChunks;
|
|
308
|
+
}
|
|
309
|
+
//# sourceMappingURL=chunk-splitter.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"chunk-splitter.js","sourceRoot":"","sources":["../../src/scaffold-kb/chunk-splitter.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;GAWG;AAIH,6EAA6E;AAE7E,6BAA6B;AAC7B,MAAM,eAAe,GAAG,CAAC,CAAC;AAE1B,oCAAoC;AACpC,MAAM,gCAAgC,GAAG,GAAG,CAAC;AAE7C,wBAAwB;AACxB,MAAM,gBAAgB,GAAG,GAAG,CAAC;AAE7B,oCAAoC;AACpC,MAAM,gBAAgB,GAAG,EAAE,CAAC;AAE5B,UAAU;AACV,MAAM,+BAA+B,GAAG,gCAAgC,GAAG,eAAe,CAAC,CAAC,OAAO;AACnG,MAAM,eAAe,GAAG,gBAAgB,GAAG,eAAe,CAAC,CAAC,OAAO;AACnE,MAAM,eAAe,GAAG,gBAAgB,GAAG,eAAe,CAAC,CAAC,KAAK;AAEjE,4EAA4E;AAE5E;;;GAGG;AACH,SAAS,MAAM,CAAC,OAAe;IAC7B,OAAO,OAAO;SACX,WAAW,EAAE;SACb,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC;SACpB,OAAO,CAAC,oBAAoB,EAAE,EAAE,CAAC,CAAC,mBAAmB;SACrD,OAAO,CAAC,KAAK,EAAE,GAAG,CAAC;SACnB,OAAO,CAAC,QAAQ,EAAE,EAAE,CAAC,CAAC;AAC3B,CAAC;AAED;;GAEG;AACH,SAAS,cAAc,CAAC,IAAY;IAClC,OAAO,IAAI,CAAC,IAAI,CAAC,IAAI,CAAC,MAAM,GAAG,eAAe,CAAC,CAAC;AAClD,CAAC;AAED;;GAEG;AACH,SAAS,kBAAkB,CAAC,IAAY;IACtC,OAAO,IAAI,CAAC,OAAO,CAAC,YAAY,EAAE,EAAE,CAAC,CAAC,IAAI,EAAE,CAAC;AAC/C,CAAC;AAED;;GAEG;AACH,SAAS,aAAa,CAAC,IAAY;IACjC,OAAO,WAAW,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;AAChC,CAAC;AAUD;;GAEG;AACH,SAAS,eAAe,CAAC,OAAe;IACtC,MAAM,KAAK,GAAG,OAAO,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;IAClC,MAAM,QAAQ,GAAqB,EAAE,CAAC;IAEtC,IAAI,cAAc,GAAkB,IAAI,CAAC;IACzC,IAAI,SAAS,GAAa,EAAE,CAAC;IAE7B,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;QACzB,IAAI,aAAa,CAAC,IAAI,CAAC,EAAE,CAAC;YACxB,qBAAqB;YACrB,MAAM,QAAQ,GAAG,SAAS,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,IAAI,EAAE,CAAC;YAC7C,IAAI,QAAQ,CAAC,MAAM,GAAG,CAAC,IAAI,cAAc,KAAK,IAAI,EAAE,CAAC;gBACnD,QAAQ,CAAC,IAAI,CAAC,EAAE,OAAO,EAAE,cAAc,EAAE,IAAI,EAAE,QAAQ,EAAE,CAAC,CAAC;YAC7D,CAAC;YACD,cAAc,GAAG,kBAAkB,CAAC,IAAI,CAAC,CAAC;YAC1C,SAAS,GAAG,EAAE,CAAC;QACjB,CAAC;aAAM,CAAC;YACN,SAAS,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;QACvB,CAAC;IACH,CAAC;IAED,OAAO;IACP,MAAM,QAAQ,GAAG,SAAS,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,IAAI,EAAE,CAAC;IAC7C,IAAI,QAAQ,CAAC,MAAM,GAAG,CAAC,IAAI,cAAc,KAAK,IAAI,EAAE,CAAC;QACnD,QAAQ,CAAC,IAAI,CAAC,EAAE,OAAO,EAAE,cAAc,EAAE,IAAI,EAAE,QAAQ,EAAE,CAAC,CAAC;IAC7D,CAAC;IAED,OAAO,QAAQ,CAAC;AAClB,CAAC;AAED,0EAA0E;AAE1E;;GAEG;AACH,SAAS,iBAAiB,CAAC,IAAY;IACrC,OAAO,IAAI;SACR,KAAK,CAAC,OAAO,CAAC;SACd,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;SACpB,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;AACjC,CAAC;AAED;;;GAGG;AACH,SAAS,gBAAgB,CAAC,SAAiB;IACzC,6BAA6B;IAC7B,kBAAkB;IAClB,MAAM,KAAK,GAAa,EAAE,CAAC;IAC3B,iBAAiB;IACjB,MAAM,KAAK,GAAG,SAAS,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;IAEpC,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;QACzB,IAAI,IAAI,CAAC,IAAI,EAAE,CAAC,MAAM,KAAK,CAAC;YAAE,SAAS;QAEvC,0BAA0B;QAC1B,MAAM,cAAc,GAAG,IAAI,CAAC,KAAK,CAAC,kBAAkB,CAAC,CAAC;QACtD,KAAK,MAAM,CAAC,IAAI,cAAc,EAAE,CAAC;YAC/B,MAAM,OAAO,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC;YACzB,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACvB,KAAK,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC;YACtB,CAAC;QACH,CAAC;IACH,CAAC;IAED,OAAO,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;AAC3C,CAAC;AAED;;;;;;GAMG;AACH,SAAS,iBAAiB,CAAC,UAAoB;IAC7C,MAAM,MAAM,GAAa,EAAE,CAAC;IAC5B,IAAI,OAAO,GAAG,EAAE,CAAC;IAEjB;;;;OAIG;IACH,SAAS,UAAU,CAAC,KAAe;QACjC,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;YACzB,IAAI,IAAI,CAAC,MAAM,GAAG,eAAe,EAAE,CAAC;gBAClC,0BAA0B;gBAC1B,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;oBACvB,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC;oBACrB,OAAO,GAAG,EAAE,CAAC;gBACf,CAAC;gBACD,IAAI,SAAS,GAAG,IAAI,CAAC;gBACrB,OAAO,SAAS,CAAC,MAAM,GAAG,eAAe,EAAE,CAAC;oBAC1C,MAAM,CAAC,IAAI,CAAC,SAAS,CAAC,KAAK,CAAC,CAAC,EAAE,eAAe,CAAC,CAAC,CAAC;oBACjD,SAAS,GAAG,SAAS,CAAC,KAAK,CAAC,eAAe,CAAC,CAAC;gBAC/C,CAAC;gBACD,IAAI,SAAS,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;oBACzB,OAAO,GAAG,SAAS,CAAC;gBACtB,CAAC;YACH,CAAC;iBAAM,CAAC;gBACN,kCAAkC;gBAClC,MAAM,SAAS,GAAG,OAAO,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC;gBACjD,MAAM,UAAU,GAAG,OAAO,CAAC,MAAM,GAAG,SAAS,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAC;gBACnE,IAAI,UAAU,GAAG,+BAA+B,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;oBACvE,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC;oBACrB,OAAO,GAAG,IAAI,CAAC;gBACjB,CAAC;qBAAM,CAAC;oBACN,OAAO,GAAG,OAAO,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,GAAG,IAAI,GAAG,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC;gBAC9D,CAAC;YACH,CAAC;QACH,CAAC;IACH,CAAC;IAED,KAAK,MAAM,IAAI,IAAI,UAAU,EAAE,CAAC;QAC9B,IAAI,IAAI,CAAC,MAAM,GAAG,+BAA+B,EAAE,CAAC;YAClD,iBAAiB;YACjB,MAAM,SAAS,GAAG,gBAAgB,CAAC,IAAI,CAAC,CAAC;YACzC,UAAU,CAAC,SAAS,CAAC,CAAC;QACxB,CAAC;aAAM,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,OAAO,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,GAAG,IAAI,CAAC,MAAM,GAAG,eAAe,EAAE,CAAC;YACzF,yBAAyB;YACzB,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACvB,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC;gBACrB,OAAO,GAAG,EAAE,CAAC;YACf,CAAC;YACD,OAAO,GAAG,IAAI,CAAC;QACjB,CAAC;aAAM,CAAC;YACN,OAAO,GAAG,OAAO,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,GAAG,MAAM,GAAG,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC;QAChE,CAAC;IACH,CAAC;IAED,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACvB,MAAM,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC;IACvB,CAAC;IAED,OAAO,MAAM,CAAC;AAChB,CAAC;AAED,0EAA0E;AAE1E;;;GAGG;AACH,SAAS,gBAAgB,CAAC,KAAe;IACvC,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC;QAAE,OAAO,EAAE,CAAC;IAElC,UAAU;IACV,MAAM,MAAM,GAAa,EAAE,CAAC;IAE5B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACtC,MAAM,IAAI,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC;QACtB,6DAA6D;QAC7D,IAAI,IAAI,KAAK,SAAS;YAAE,SAAS;QAEjC,IAAI,IAAI,CAAC,MAAM,GAAG,eAAe,EAAE,CAAC;YAClC,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACtB,SAAS;gBACT,MAAM,IAAI,GAAG,MAAM,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;gBACvC,IAAI,IAAI,KAAK,SAAS,EAAE,CAAC;oBACvB,MAAM,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,GAAG,IAAI,GAAG,MAAM,GAAG,IAAI,CAAC;gBACnD,CAAC;YACH,CAAC;iBAAM,CAAC;gBACN,uBAAuB;gBACvB,MAAM,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;YACpB,CAAC;QACH,CAAC;aAAM,CAAC;YACN,IACE,MAAM,CAAC,MAAM,GAAG,CAAC;gBACjB,CAAC,MAAM,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,EAAE,MAAM,IAAI,CAAC,CAAC,GAAG,eAAe,EAC1D,CAAC;gBACD,aAAa;gBACb,MAAM,IAAI,GAAG,MAAM,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;gBACvC,IAAI,IAAI,KAAK,SAAS,EAAE,CAAC;oBACvB,MAAM,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,GAAG,IAAI,GAAG,MAAM,GAAG,IAAI,CAAC;gBACnD,CAAC;YACH,CAAC;iBAAM,CAAC;gBACN,MAAM,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;YACpB,CAAC;QACH,CAAC;IACH,CAAC;IAED,OAAO,MAAM,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;AACnD,CAAC;AAED,4EAA4E;AAE5E;;;;GAIG;AACH,MAAM,UAAU,aAAa,CAAC,GAAc;IAC1C,MAAM,QAAQ,GAAG,eAAe,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC;IAE9C,0CAA0C;IAC1C,MAAM,YAAY,GAAG,IAAI,GAAG,EAAkB,CAAC;IAE/C;;OAEG;IACH,SAAS,eAAe,CAAC,MAAc;QACrC,MAAM,KAAK,GAAG,CAAC,YAAY,CAAC,GAAG,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC;QAClD,YAAY,CAAC,GAAG,CAAC,MAAM,EAAE,KAAK,CAAC,CAAC;QAChC,OAAO,KAAK,CAAC;IACf,CAAC;IAED,MAAM,SAAS,GAAY,EAAE,CAAC;IAE9B,mCAAmC;IACnC,IAAI,gBAAgB,GAAG,CAAC,CAAC;IAEzB,KAAK,MAAM,OAAO,IAAI,QAAQ,EAAE,CAAC;QAC/B,MAAM,MAAM,GAAG,OAAO,CAAC,OAAO,KAAK,IAAI,CAAC,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC;QAEzE,kBAAkB;QAClB,MAAM,WAAW,GACf,OAAO,CAAC,OAAO,KAAK,IAAI;YACtB,CAAC,CAAC,MAAM,OAAO,CAAC,OAAO,OAAO,OAAO,CAAC,IAAI,EAAE,CAAC,IAAI,EAAE;YACnD,CAAC,CAAC,OAAO,CAAC,IAAI,CAAC;QAEnB,IAAI,WAAW,CAAC,IAAI,EAAE,CAAC,MAAM,KAAK,CAAC;YAAE,SAAS;QAE9C,cAAc;QACd,IAAI,QAAkB,CAAC;QAEvB,IAAI,cAAc,CAAC,WAAW,CAAC,IAAI,gCAAgC,EAAE,CAAC;YACpE,2BAA2B;YAC3B,QAAQ,GAAG,CAAC,WAAW,CAAC,CAAC;QAC3B,CAAC;aAAM,CAAC;YACN,wBAAwB;YACxB,MAAM,UAAU,GAAG,iBAAiB;YAClC,mBAAmB;YACnB,OAAO,CAAC,IAAI,CACb,CAAC;YACF,MAAM,QAAQ,GAAG,iBAAiB,CAAC,UAAU,CAAC,CAAC;YAC/C,MAAM,MAAM,GAAG,gBAAgB,CAAC,QAAQ,CAAC,CAAC;YAE1C,qBAAqB;YACrB,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,IAAI,OAAO,CAAC,OAAO,KAAK,IAAI,EAAE,CAAC;gBAClD,MAAM,CAAC,CAAC,CAAC,GAAG,MAAM,OAAO,CAAC,OAAO,OAAO,MAAM,CAAC,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC;YAC5D,CAAC;YAED,QAAQ,GAAG,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,WAAW,CAAC,CAAC;QACxD,CAAC;QAED,qBAAqB;QACrB,KAAK,MAAM,IAAI,IAAI,QAAQ,EAAE,CAAC;YAC5B,IAAI,IAAI,CAAC,IAAI,EAAE,CAAC,MAAM,KAAK,CAAC;gBAAE,SAAS;YACvC,+CAA+C;YAC/C,UAAU;YACV,IAAI,IAAI,CAAC,MAAM,GAAG,eAAe,IAAI,QAAQ,CAAC,MAAM,GAAG,CAAC;gBAAE,SAAS;YAEnE,aAAa;YACb,IAAI,OAAe,CAAC;YACpB,IAAI,MAAM,KAAK,IAAI,EAAE,CAAC;gBACpB,MAAM,GAAG,GAAG,eAAe,CAAC,MAAM,CAAC,CAAC;gBACpC,OAAO,GAAG,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,GAAG,GAAG,CAAC,EAAE,IAAI,MAAM,EAAE,CAAC,CAAC,CAAC,GAAG,GAAG,CAAC,EAAE,IAAI,MAAM,IAAI,GAAG,EAAE,CAAC;YAC7E,CAAC;iBAAM,CAAC;gBACN,gBAAgB,IAAI,CAAC,CAAC;gBACtB,OAAO,GAAG,GAAG,GAAG,CAAC,EAAE,KAAK,gBAAgB,EAAE,CAAC;YAC7C,CAAC;YAED,SAAS,CAAC,IAAI,CAAC;gBACb,OAAO;gBACP,KAAK,EAAE,GAAG,CAAC,EAAE;gBACb,UAAU,EAAE,IAAI;gBAChB,MAAM;aACP,CAAC,CAAC;QACL,CAAC;IACH,CAAC;IAED,OAAO,SAAS,CAAC;AACnB,CAAC"}
|
|
@@ -0,0 +1,26 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* F190 scaffold-kb — doc-graph 构建器
|
|
3
|
+
*
|
|
4
|
+
* 将 ParsedDoc 列表转换为 DocGraph(doc-graph.json 结构契约)。
|
|
5
|
+
* 幂等保证:相同输入(同 docs 同 opts)产出在去掉 builtAt 后字节级一致。
|
|
6
|
+
* 排序规则:nodes 按 doc.id 升序,edges 按 (source, target) 升序。
|
|
7
|
+
*/
|
|
8
|
+
import type { DocGraph, ParsedDoc } from './types.js';
|
|
9
|
+
/** buildDocGraph 的选项参数 */
|
|
10
|
+
export interface BuildDocGraphOpts {
|
|
11
|
+
/** 文档来源类型 */
|
|
12
|
+
source: 'llms.txt' | 'directory';
|
|
13
|
+
/** SDK 版本(可选),写入产物元数据 */
|
|
14
|
+
sdkVersion: string | null;
|
|
15
|
+
/** 构建时间戳(ISO 8601),幂等对比时排除 */
|
|
16
|
+
builtAt: string;
|
|
17
|
+
}
|
|
18
|
+
/**
|
|
19
|
+
* 从解析文档列表构建文档图(doc-graph.json)。
|
|
20
|
+
*
|
|
21
|
+
* @param docs ingester 解析出的文档数组
|
|
22
|
+
* @param opts 来源类型、SDK 版本、构建时间戳
|
|
23
|
+
* @returns 满足 spec §3.2 结构契约的 DocGraph
|
|
24
|
+
*/
|
|
25
|
+
export declare function buildDocGraph(docs: ParsedDoc[], opts: BuildDocGraphOpts): DocGraph;
|
|
26
|
+
//# sourceMappingURL=doc-graph-builder.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"doc-graph-builder.d.ts","sourceRoot":"","sources":["../../src/scaffold-kb/doc-graph-builder.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,OAAO,KAAK,EAAE,QAAQ,EAAoB,SAAS,EAAE,MAAM,YAAY,CAAC;AAExE,0BAA0B;AAC1B,MAAM,WAAW,iBAAiB;IAChC,aAAa;IACb,MAAM,EAAE,UAAU,GAAG,WAAW,CAAC;IACjC,yBAAyB;IACzB,UAAU,EAAE,MAAM,GAAG,IAAI,CAAC;IAC1B,8BAA8B;IAC9B,OAAO,EAAE,MAAM,CAAC;CACjB;AAED;;;;;;GAMG;AACH,wBAAgB,aAAa,CAAC,IAAI,EAAE,SAAS,EAAE,EAAE,IAAI,EAAE,iBAAiB,GAAG,QAAQ,CA2DlF"}
|
|
@@ -0,0 +1,72 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* F190 scaffold-kb — doc-graph 构建器
|
|
3
|
+
*
|
|
4
|
+
* 将 ParsedDoc 列表转换为 DocGraph(doc-graph.json 结构契约)。
|
|
5
|
+
* 幂等保证:相同输入(同 docs 同 opts)产出在去掉 builtAt 后字节级一致。
|
|
6
|
+
* 排序规则:nodes 按 doc.id 升序,edges 按 (source, target) 升序。
|
|
7
|
+
*/
|
|
8
|
+
/**
|
|
9
|
+
* 从解析文档列表构建文档图(doc-graph.json)。
|
|
10
|
+
*
|
|
11
|
+
* @param docs ingester 解析出的文档数组
|
|
12
|
+
* @param opts 来源类型、SDK 版本、构建时间戳
|
|
13
|
+
* @returns 满足 spec §3.2 结构契约的 DocGraph
|
|
14
|
+
*/
|
|
15
|
+
export function buildDocGraph(docs, opts) {
|
|
16
|
+
// 第一步:构建 id 集合,用于悬空引用过滤
|
|
17
|
+
const idSet = new Set(docs.map((d) => d.id));
|
|
18
|
+
// 第二步:映射 ParsedDoc → DocNode,按 id 升序排列确保幂等
|
|
19
|
+
const nodes = docs
|
|
20
|
+
.slice()
|
|
21
|
+
.sort((a, b) => a.id.localeCompare(b.id))
|
|
22
|
+
.map((doc) => {
|
|
23
|
+
const node = {
|
|
24
|
+
id: doc.id,
|
|
25
|
+
title: doc.title,
|
|
26
|
+
lang: doc.lang,
|
|
27
|
+
sourceUrl: doc.sourceUrl,
|
|
28
|
+
};
|
|
29
|
+
// summary 和 tags 是可选字段,仅当 ParsedDoc 携带时写入
|
|
30
|
+
// ParsedDoc 接口本身没有 summary/tags,为 future-proof 保留扩展检查
|
|
31
|
+
return node;
|
|
32
|
+
});
|
|
33
|
+
// 第三步:从 doc.references 提取 DocEdge,跳过悬空引用(目标 id 不在集合内)
|
|
34
|
+
const edgeMap = new Map();
|
|
35
|
+
for (const doc of docs) {
|
|
36
|
+
const refs = doc.references;
|
|
37
|
+
if (refs === undefined || refs.length === 0) {
|
|
38
|
+
continue;
|
|
39
|
+
}
|
|
40
|
+
for (const targetId of refs) {
|
|
41
|
+
// 悬空引用跳过(目标文档不存在于 docs 集合)
|
|
42
|
+
if (!idSet.has(targetId)) {
|
|
43
|
+
continue;
|
|
44
|
+
}
|
|
45
|
+
// 使用 "source|target" 作为 key 去重(同一对可能从多处引用)
|
|
46
|
+
const key = `${doc.id}\x00${targetId}`;
|
|
47
|
+
if (!edgeMap.has(key)) {
|
|
48
|
+
edgeMap.set(key, {
|
|
49
|
+
source: doc.id,
|
|
50
|
+
target: targetId,
|
|
51
|
+
relation: 'references',
|
|
52
|
+
});
|
|
53
|
+
}
|
|
54
|
+
}
|
|
55
|
+
}
|
|
56
|
+
// 按 (source, target) 升序排列,保证幂等
|
|
57
|
+
const edges = Array.from(edgeMap.values()).sort((a, b) => {
|
|
58
|
+
const cmp = a.source.localeCompare(b.source);
|
|
59
|
+
if (cmp !== 0)
|
|
60
|
+
return cmp;
|
|
61
|
+
return a.target.localeCompare(b.target);
|
|
62
|
+
});
|
|
63
|
+
return {
|
|
64
|
+
schemaVersion: '1.0',
|
|
65
|
+
source: opts.source,
|
|
66
|
+
builtAt: opts.builtAt,
|
|
67
|
+
sdkVersion: opts.sdkVersion,
|
|
68
|
+
nodes,
|
|
69
|
+
edges,
|
|
70
|
+
};
|
|
71
|
+
}
|
|
72
|
+
//# sourceMappingURL=doc-graph-builder.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"doc-graph-builder.js","sourceRoot":"","sources":["../../src/scaffold-kb/doc-graph-builder.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAcH;;;;;;GAMG;AACH,MAAM,UAAU,aAAa,CAAC,IAAiB,EAAE,IAAuB;IACtE,wBAAwB;IACxB,MAAM,KAAK,GAAG,IAAI,GAAG,CAAS,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC;IAErD,2CAA2C;IAC3C,MAAM,KAAK,GAAc,IAAI;SAC1B,KAAK,EAAE;SACP,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC,aAAa,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC;SACxC,GAAG,CAAC,CAAC,GAAG,EAAE,EAAE;QACX,MAAM,IAAI,GAAY;YACpB,EAAE,EAAE,GAAG,CAAC,EAAE;YACV,KAAK,EAAE,GAAG,CAAC,KAAK;YAChB,IAAI,EAAE,GAAG,CAAC,IAAI;YACd,SAAS,EAAE,GAAG,CAAC,SAAS;SACzB,CAAC;QACF,0CAA0C;QAC1C,sDAAsD;QACtD,OAAO,IAAI,CAAC;IACd,CAAC,CAAC,CAAC;IAEL,sDAAsD;IACtD,MAAM,OAAO,GAAG,IAAI,GAAG,EAAmB,CAAC;IAC3C,KAAK,MAAM,GAAG,IAAI,IAAI,EAAE,CAAC;QACvB,MAAM,IAAI,GAAG,GAAG,CAAC,UAAU,CAAC;QAC5B,IAAI,IAAI,KAAK,SAAS,IAAI,IAAI,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;YAC5C,SAAS;QACX,CAAC;QACD,KAAK,MAAM,QAAQ,IAAI,IAAI,EAAE,CAAC;YAC5B,2BAA2B;YAC3B,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC,QAAQ,CAAC,EAAE,CAAC;gBACzB,SAAS;YACX,CAAC;YACD,2CAA2C;YAC3C,MAAM,GAAG,GAAG,GAAG,GAAG,CAAC,EAAE,OAAO,QAAQ,EAAE,CAAC;YACvC,IAAI,CAAC,OAAO,CAAC,GAAG,CAAC,GAAG,CAAC,EAAE,CAAC;gBACtB,OAAO,CAAC,GAAG,CAAC,GAAG,EAAE;oBACf,MAAM,EAAE,GAAG,CAAC,EAAE;oBACd,MAAM,EAAE,QAAQ;oBAChB,QAAQ,EAAE,YAAY;iBACvB,CAAC,CAAC;YACL,CAAC;QACH,CAAC;IACH,CAAC;IAED,+BAA+B;IAC/B,MAAM,KAAK,GAAc,KAAK,CAAC,IAAI,CAAC,OAAO,CAAC,MAAM,EAAE,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE;QAClE,MAAM,GAAG,GAAG,CAAC,CAAC,MAAM,CAAC,aAAa,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC;QAC7C,IAAI,GAAG,KAAK,CAAC;YAAE,OAAO,GAAG,CAAC;QAC1B,OAAO,CAAC,CAAC,MAAM,CAAC,aAAa,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC;IAC1C,CAAC,CAAC,CAAC;IAEH,OAAO;QACL,aAAa,EAAE,KAAK;QACpB,MAAM,EAAE,IAAI,CAAC,MAAM;QACnB,OAAO,EAAE,IAAI,CAAC,OAAO;QACrB,UAAU,EAAE,IAAI,CAAC,UAAU;QAC3B,KAAK;QACL,KAAK;KACN,CAAC;AACJ,CAAC"}
|
|
@@ -0,0 +1,40 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* F192 T004 — API 实体抽取器(LLM 提质 + heuristic 兜底,FR-001/002/003)
|
|
3
|
+
*
|
|
4
|
+
* 流程:chunks → section 窗口聚合(W-4,防签名跨 chunk 切碎)→ 逐 section
|
|
5
|
+
* LLM 抽取(无认证/--no-llm/返回 null/解析失败 → heuristic 兜底,EC-002)。
|
|
6
|
+
* 成本护栏:maxSections 截断 + coverage 元字段(FR-001)。
|
|
7
|
+
* 诚实:LLM prompt 严禁编造;无 LLM 不阻断(FR-003)。
|
|
8
|
+
*
|
|
9
|
+
* callLLM/llmAvailable 可注入(仿 ingester fetchImpl),保证测试确定性。
|
|
10
|
+
*/
|
|
11
|
+
import type { ApiEntity, Chunk, ExtractionSection } from './types.js';
|
|
12
|
+
import { type LLMCallOptions } from '../panoramic/utils/llm-facade.js';
|
|
13
|
+
export type CallLLMFn = (prompt: string, options?: LLMCallOptions) => Promise<string | null>;
|
|
14
|
+
export interface ExtractOptions {
|
|
15
|
+
/** 强制跳过 LLM,只走 heuristic(CLI --no-llm,FR-001) */
|
|
16
|
+
noLlm?: boolean;
|
|
17
|
+
/** 成本护栏:最多处理 section 数(超出截断并记 coverage)。缺省全量 */
|
|
18
|
+
maxSections?: number;
|
|
19
|
+
/** 成本护栏:累计输入字符上限(token 代理;超出停止后续 section)。缺省无限 */
|
|
20
|
+
maxInputChars?: number;
|
|
21
|
+
/** 进度回调(可观测,FR-001) */
|
|
22
|
+
onProgress?: (done: number, total: number) => void;
|
|
23
|
+
/** 注入 LLM 调用(测试用),缺省走 facade */
|
|
24
|
+
callLLM?: CallLLMFn;
|
|
25
|
+
/** 注入可用性检测(测试用),缺省走 facade */
|
|
26
|
+
llmAvailable?: () => boolean;
|
|
27
|
+
}
|
|
28
|
+
export interface ExtractResult {
|
|
29
|
+
entities: ApiEntity[];
|
|
30
|
+
coverage: {
|
|
31
|
+
totalSections: number;
|
|
32
|
+
extractedSections: number;
|
|
33
|
+
};
|
|
34
|
+
method: 'llm' | 'heuristic' | 'mixed';
|
|
35
|
+
}
|
|
36
|
+
/** chunks → section 窗口(同 docId+anchor 相邻 chunk 聚合,W-4) */
|
|
37
|
+
export declare function aggregateSections(chunks: Chunk[], docLang: Map<string, string>): ExtractionSection[];
|
|
38
|
+
/** 抽取实体:LLM 优先、heuristic 兜底,带成本护栏 + coverage */
|
|
39
|
+
export declare function extractEntities(sections: ExtractionSection[], opts?: ExtractOptions): Promise<ExtractResult>;
|
|
40
|
+
//# sourceMappingURL=entity-extractor.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"entity-extractor.d.ts","sourceRoot":"","sources":["../../src/scaffold-kb/entity-extractor.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAEH,OAAO,KAAK,EAAE,SAAS,EAAE,KAAK,EAAE,iBAAiB,EAAE,MAAM,YAAY,CAAC;AAGtE,OAAO,EAIL,KAAK,cAAc,EACpB,MAAM,kCAAkC,CAAC;AAE1C,MAAM,MAAM,SAAS,GAAG,CAAC,MAAM,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE,cAAc,KAAK,OAAO,CAAC,MAAM,GAAG,IAAI,CAAC,CAAC;AAE7F,MAAM,WAAW,cAAc;IAC7B,iDAAiD;IACjD,KAAK,CAAC,EAAE,OAAO,CAAC;IAChB,gDAAgD;IAChD,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,kDAAkD;IAClD,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,uBAAuB;IACvB,UAAU,CAAC,EAAE,CAAC,IAAI,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,KAAK,IAAI,CAAC;IACnD,gCAAgC;IAChC,OAAO,CAAC,EAAE,SAAS,CAAC;IACpB,8BAA8B;IAC9B,YAAY,CAAC,EAAE,MAAM,OAAO,CAAC;CAC9B;AAED,MAAM,WAAW,aAAa;IAC5B,QAAQ,EAAE,SAAS,EAAE,CAAC;IACtB,QAAQ,EAAE;QAAE,aAAa,EAAE,MAAM,CAAC;QAAC,iBAAiB,EAAE,MAAM,CAAA;KAAE,CAAC;IAC/D,MAAM,EAAE,KAAK,GAAG,WAAW,GAAG,OAAO,CAAC;CACvC;AAeD,0DAA0D;AAC1D,wBAAgB,iBAAiB,CAAC,MAAM,EAAE,KAAK,EAAE,EAAE,OAAO,EAAE,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC,GAAG,iBAAiB,EAAE,CAwBpG;AA0ED,gDAAgD;AAChD,wBAAsB,eAAe,CACnC,QAAQ,EAAE,iBAAiB,EAAE,EAC7B,IAAI,GAAE,cAAmB,GACxB,OAAO,CAAC,aAAa,CAAC,CA0DxB"}
|
|
Binary file
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"entity-extractor.js","sourceRoot":"","sources":["../../src/scaffold-kb/entity-extractor.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAGH,OAAO,EAAE,gBAAgB,EAAE,MAAM,uBAAuB,CAAC;AACzD,OAAO,EAAE,YAAY,EAAE,eAAe,EAAE,MAAM,kBAAkB,CAAC;AACjE,OAAO,EACL,OAAO,IAAI,cAAc,EACzB,gBAAgB,EAChB,cAAc,IAAI,qBAAqB,GAExC,MAAM,kCAAkC,CAAC;AAyB1C,MAAM,cAAc,GAAG;IACrB,6DAA6D;IAC7D,0DAA0D;IAC1D,mCAAmC;IACnC,wEAAwE;IACxE,2FAA2F;IAC3F,8DAA8D;CAC/D,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;AAEX,MAAM,WAAW,GAAG,IAAI,GAAG,CAAoB;IAC7C,UAAU,EAAE,QAAQ,EAAE,OAAO,EAAE,UAAU,EAAE,MAAM,EAAE,UAAU,EAAE,YAAY,EAAE,OAAO;CACrF,CAAC,CAAC;AAEH,0DAA0D;AAC1D,MAAM,UAAU,iBAAiB,CAAC,MAAe,EAAE,OAA4B;IAC7E,MAAM,KAAK,GAAa,EAAE,CAAC;IAC3B,MAAM,MAAM,GAAG,IAAI,GAAG,EAAyF,CAAC;IAChH,KAAK,MAAM,CAAC,IAAI,MAAM,EAAE,CAAC;QACvB,MAAM,GAAG,GAAG,GAAG,CAAC,CAAC,KAAK,IAAI,CAAC,CAAC,MAAM,IAAI,EAAE,EAAE,CAAC;QAC3C,IAAI,CAAC,GAAG,MAAM,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC;QACxB,IAAI,CAAC,CAAC,EAAE,CAAC;YACP,CAAC,GAAG,EAAE,KAAK,EAAE,CAAC,CAAC,KAAK,EAAE,MAAM,EAAE,CAAC,CAAC,MAAM,EAAE,QAAQ,EAAE,EAAE,EAAE,KAAK,EAAE,EAAE,EAAE,CAAC;YAClE,MAAM,CAAC,GAAG,CAAC,GAAG,EAAE,CAAC,CAAC,CAAC;YACnB,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;QAClB,CAAC;QACD,CAAC,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC,CAAC,OAAO,CAAC,CAAC;QAC3B,CAAC,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC,UAAU,CAAC,CAAC;IAC7B,CAAC;IACD,OAAO,KAAK,CAAC,GAAG,CAAC,CAAC,GAAG,EAAE,EAAE;QACvB,MAAM,CAAC,GAAG,MAAM,CAAC,GAAG,CAAC,GAAG,CAAE,CAAC;QAC3B,OAAO;YACL,KAAK,EAAE,CAAC,CAAC,KAAK;YACd,MAAM,EAAE,CAAC,CAAC,MAAM;YAChB,IAAI,EAAE,OAAO,CAAC,GAAG,CAAC,CAAC,CAAC,KAAK,CAAC,IAAI,IAAI;YAClC,QAAQ,EAAE,CAAC,CAAC,QAAQ;YACpB,IAAI,EAAE,CAAC,CAAC,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC;SAC3B,CAAC;IACJ,CAAC,CAAC,CAAC;AACL,CAAC;AAED,oEAAoE;AACpE,SAAS,gBAAgB,CAAC,IAAY,EAAE,OAA0B;IAChE,IAAI,GAAY,CAAC;IACjB,IAAI,CAAC;QACH,GAAG,GAAG,gBAAgB,CAAC,IAAI,CAAC,CAAC;IAC/B,CAAC;IAAC,MAAM,CAAC;QACP,OAAO,IAAI,CAAC;IACd,CAAC;IACD,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,GAAG,CAAC;QAAE,OAAO,IAAI,CAAC;IACrC,MAAM,YAAY,GAAG,OAAO,CAAC,QAAQ,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;IAC/C,MAAM,GAAG,GAAgB,EAAE,CAAC;IAC5B,MAAM,IAAI,GAAG,IAAI,GAAG,EAAU,CAAC;IAC/B,KAAK,MAAM,GAAG,IAAI,GAAG,EAAE,CAAC;QACtB,IAAI,OAAO,GAAG,KAAK,QAAQ,IAAI,GAAG,KAAK,IAAI;YAAE,SAAS;QACtD,MAAM,CAAC,GAAG,GAA8B,CAAC;QACzC,MAAM,IAAI,GAAG,OAAO,CAAC,CAAC,MAAM,CAAC,KAAK,QAAQ,CAAC,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;QACnE,IAAI,CAAC,IAAI;YAAE,SAAS;QACpB,MAAM,OAAO,GAAG,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC,IAAI,UAAU,CAAC,CAAC;QAChD,MAAM,IAAI,GAAG,CAAC,WAAW,CAAC,GAAG,CAAC,OAA4B,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,UAAU,CAAsB,CAAC;QACzG,MAAM,aAAa,GAAG,OAAO,CAAC,CAAC,eAAe,CAAC,KAAK,QAAQ,IAAI,CAAC,CAAC,eAAe,CAAC,CAAC,IAAI,EAAE;YACvF,CAAC,CAAC,CAAC,CAAC,eAAe,CAAC,CAAC,IAAI,EAAE;YAC3B,CAAC,CAAC,IAAI,CAAC;QACT,MAAM,WAAW,GAAG,OAAO,CAAC,CAAC,aAAa,CAAC,KAAK,QAAQ,CAAC,CAAC,CAAC,CAAC,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC;QACnF,MAAM,EAAE,GAAG,YAAY,CAAC,aAAa,EAAE,IAAI,EAAE,WAAW,CAAC,CAAC;QAC1D,IAAI,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC;YAAE,SAAS;QAC3B,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;QAEb,MAAM,CAAC,GAAc;YACnB,EAAE;YACF,IAAI;YACJ,aAAa;YACb,IAAI;YACJ,WAAW,EAAE,OAAO,CAAC,KAAK;YAC1B,aAAa,EAAE,YAAY;YAC3B,IAAI,EAAE,OAAO,CAAC,IAAI;YAClB,UAAU,EAAE,eAAe,CAAC,CAAC,CAAC,YAAY,CAAC,EAAE,GAAG,CAAC;YACjD,gBAAgB,EAAE,KAAK;SACxB,CAAC;QACF,IAAI,OAAO,CAAC,QAAQ,CAAC,MAAM,GAAG,CAAC;YAAE,CAAC,CAAC,cAAc,GAAG,OAAO,CAAC,QAAQ,CAAC;QACrE,IAAI,OAAO,CAAC,MAAM;YAAE,CAAC,CAAC,YAAY,GAAG,OAAO,CAAC,MAAM,CAAC;QACpD,IAAI,WAAW;YAAE,CAAC,CAAC,WAAW,GAAG,WAAW,CAAC;QAC7C,IAAI,OAAO,CAAC,CAAC,WAAW,CAAC,KAAK,QAAQ,IAAI,CAAC,CAAC,WAAW,CAAC;YAAE,CAAC,CAAC,SAAS,GAAG,CAAC,CAAC,WAAW,CAAC,CAAC;QACvF,IAAI,OAAO,CAAC,CAAC,WAAW,CAAC,KAAK,QAAQ;YAAE,CAAC,CAAC,SAAS,GAAG,CAAC,CAAC,WAAW,CAAC,CAAC;QACrE,IAAI,OAAO,CAAC,CAAC,SAAS,CAAC,KAAK,QAAQ;YAAE,CAAC,CAAC,OAAO,GAAG,CAAC,CAAC,SAAS,CAAC,CAAC;QAC/D,IAAI,OAAO,CAAC,CAAC,cAAc,CAAC,KAAK,QAAQ;YAAE,CAAC,CAAC,YAAY,GAAG,CAAC,CAAC,cAAc,CAAC,CAAC;QAC9E,IAAI,OAAO,CAAC,CAAC,eAAe,CAAC,KAAK,QAAQ;YAAE,CAAC,CAAC,aAAa,GAAG,CAAC,CAAC,eAAe,CAAC,CAAC;QACjF,IAAI,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,EAAE,CAAC;YAC/B,CAAC,CAAC,MAAM,GAAI,CAAC,CAAC,QAAQ,CAAe;iBAClC,MAAM,CAAC,CAAC,CAAC,EAAgC,EAAE,CAAC,OAAO,CAAC,KAAK,QAAQ,IAAI,CAAC,KAAK,IAAI,CAAC;iBAChF,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE;gBACT,MAAM,KAAK,GAA6C,EAAE,IAAI,EAAE,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC,IAAI,EAAE,CAAC,EAAE,CAAC;gBAC1F,IAAI,OAAO,CAAC,CAAC,MAAM,CAAC,KAAK,QAAQ;oBAAE,KAAK,CAAC,IAAI,GAAG,CAAC,CAAC,MAAM,CAAC,CAAC;gBAC1D,IAAI,OAAO,CAAC,CAAC,UAAU,CAAC,KAAK,SAAS;oBAAE,KAAK,CAAC,QAAQ,GAAG,CAAC,CAAC,UAAU,CAAC,CAAC;gBACvE,IAAI,OAAO,CAAC,CAAC,KAAK,CAAC,KAAK,QAAQ;oBAAE,KAAK,CAAC,GAAG,GAAG,CAAC,CAAC,KAAK,CAAC,CAAC;gBACvD,OAAO,KAAK,CAAC;YACf,CAAC,CAAC;iBACD,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;QACtC,CAAC;QACD,MAAM,GAAG,GAAG,CAAC,CAAC,YAAY,CAAC,CAAC;QAC5B,IAAI,OAAO,GAAG,KAAK,QAAQ,IAAI,GAAG,KAAK,IAAI,EAAE,CAAC;YAC5C,MAAM,CAAC,GAAG,GAA8B,CAAC;YACzC,IAAI,CAAC,CAAC,cAAc,CAAC,KAAK,IAAI,EAAE,CAAC;gBAC/B,CAAC,CAAC,UAAU,GAAG,EAAE,YAAY,EAAE,IAAI,EAAE,CAAC;gBACtC,IAAI,OAAO,CAAC,CAAC,OAAO,CAAC,KAAK,QAAQ;oBAAE,CAAC,CAAC,UAAU,CAAC,KAAK,GAAG,CAAC,CAAC,OAAO,CAAC,CAAC;gBACpE,IAAI,OAAO,CAAC,CAAC,aAAa,CAAC,KAAK,QAAQ;oBAAE,CAAC,CAAC,UAAU,CAAC,WAAW,GAAG,CAAC,CAAC,aAAa,CAAC,CAAC;YACxF,CAAC;QACH,CAAC;QACD,GAAG,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;IACd,CAAC;IACD,OAAO,GAAG,CAAC;AACb,CAAC;AAED,gDAAgD;AAChD,MAAM,CAAC,KAAK,UAAU,eAAe,CACnC,QAA6B,EAC7B,OAAuB,EAAE;IAEzB,MAAM,SAAS,GAAG,IAAI,CAAC,OAAO,IAAI,cAAc,CAAC;IACjD,MAAM,SAAS,GAAG,CAAC,IAAI,CAAC,YAAY,IAAI,qBAAqB,CAAC,EAAE,CAAC;IACjE,MAAM,MAAM,GAAG,IAAI,CAAC,KAAK,KAAK,IAAI,IAAI,SAAS,CAAC;IAChD,MAAM,WAAW,GAAG,IAAI,CAAC,WAAW,IAAI,QAAQ,CAAC,MAAM,CAAC;IACxD,MAAM,SAAS,GAAG,QAAQ,CAAC,KAAK,CAAC,CAAC,EAAE,WAAW,CAAC,CAAC;IAEjD,MAAM,QAAQ,GAAgB,EAAE,CAAC;IACjC,MAAM,IAAI,GAAG,IAAI,GAAG,EAAU,CAAC;IAC/B,IAAI,QAAQ,GAAG,CAAC,CAAC;IACjB,IAAI,SAAS,GAAG,CAAC,CAAC;IAClB,IAAI,SAAS,GAAG,CAAC,CAAC;IAClB,IAAI,UAAU,GAAG,CAAC,CAAC;IAEnB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,SAAS,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QAC1C,MAAM,OAAO,GAAG,SAAS,CAAC,CAAC,CAAE,CAAC;QAC9B,uCAAuC;QACvC,IAAI,IAAI,CAAC,aAAa,KAAK,SAAS,IAAI,SAAS,GAAG,CAAC,IAAI,UAAU,GAAG,OAAO,CAAC,IAAI,CAAC,MAAM,GAAG,IAAI,CAAC,aAAa,EAAE,CAAC;YAC/G,MAAM;QACR,CAAC;QACD,UAAU,IAAI,OAAO,CAAC,IAAI,CAAC,MAAM,CAAC;QAClC,IAAI,WAAW,GAAuB,IAAI,CAAC;QAC3C,IAAI,MAAM,EAAE,CAAC;YACX,IAAI,IAAI,GAAkB,IAAI,CAAC;YAC/B,IAAI,CAAC;gBACH,IAAI,GAAG,MAAM,SAAS,CAAC,OAAO,CAAC,IAAI,EAAE,EAAE,YAAY,EAAE,cAAc,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC;YAC1F,CAAC;YAAC,MAAM,CAAC;gBACP,IAAI,GAAG,IAAI,CAAC,CAAC,4BAA4B;YAC3C,CAAC;YACD,IAAI,IAAI,KAAK,IAAI,EAAE,CAAC;gBAClB,MAAM,MAAM,GAAG,gBAAgB,CAAC,IAAI,EAAE,OAAO,CAAC,CAAC;gBAC/C,IAAI,MAAM,KAAK,IAAI,EAAE,CAAC;oBACpB,WAAW,GAAG,MAAM,CAAC;oBACrB,QAAQ,EAAE,CAAC;gBACb,CAAC;YACH,CAAC;QACH,CAAC;QACD,IAAI,WAAW,KAAK,IAAI,EAAE,CAAC;YACzB,WAAW,GAAG,gBAAgB,CAAC,OAAO,CAAC,CAAC;YACxC,SAAS,EAAE,CAAC;QACd,CAAC;QACD,KAAK,MAAM,CAAC,IAAI,WAAW,EAAE,CAAC;YAC5B,IAAI,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,CAAC;gBACpB,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC;gBACf,QAAQ,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;YACnB,CAAC;QACH,CAAC;QACD,SAAS,EAAE,CAAC;QACZ,IAAI,CAAC,UAAU,EAAE,CAAC,SAAS,EAAE,SAAS,CAAC,MAAM,CAAC,CAAC;IACjD,CAAC;IAED,MAAM,MAAM,GACV,QAAQ,GAAG,CAAC,IAAI,SAAS,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,QAAQ,GAAG,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,WAAW,CAAC;IAC/E,OAAO;QACL,QAAQ;QACR,QAAQ,EAAE,EAAE,aAAa,EAAE,QAAQ,CAAC,MAAM,EAAE,iBAAiB,EAAE,SAAS,EAAE;QAC1E,MAAM;KACP,CAAC;AACJ,CAAC"}
|
|
@@ -0,0 +1,14 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* F192 T003 — 无 LLM 的确定性 API 实体抽取(heuristic fallback,EC-002)
|
|
3
|
+
*
|
|
4
|
+
* 纯函数、零外部依赖、可重复。规则保守(宁可漏抽不可编造,FR-003):
|
|
5
|
+
* - 围栏代码块 + 内联里的函数/方法签名 `name(params)`
|
|
6
|
+
* - 错误码模式(E\d+ / ALL_CAPS_ERROR)
|
|
7
|
+
* - 段落级 deprecated 标记(@deprecated / 已废弃 / deprecated since X)
|
|
8
|
+
*
|
|
9
|
+
* 反过拟合(FR-018):规则 MUST NOT 匹配任何 eval manifest 的具体 API 名字面量。
|
|
10
|
+
*/
|
|
11
|
+
import type { ApiEntity, ExtractionSection } from './types.js';
|
|
12
|
+
/** 对单个 section 做确定性抽取,返回去重后的实体列表 */
|
|
13
|
+
export declare function extractHeuristic(section: ExtractionSection): ApiEntity[];
|
|
14
|
+
//# sourceMappingURL=entity-heuristic.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"entity-heuristic.d.ts","sourceRoot":"","sources":["../../src/scaffold-kb/entity-heuristic.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAEH,OAAO,KAAK,EAAE,SAAS,EAAE,iBAAiB,EAAE,MAAM,YAAY,CAAC;AAsD/D,oCAAoC;AACpC,wBAAgB,gBAAgB,CAAC,OAAO,EAAE,iBAAiB,GAAG,SAAS,EAAE,CA+CxE"}
|
|
@@ -0,0 +1,107 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* F192 T003 — 无 LLM 的确定性 API 实体抽取(heuristic fallback,EC-002)
|
|
3
|
+
*
|
|
4
|
+
* 纯函数、零外部依赖、可重复。规则保守(宁可漏抽不可编造,FR-003):
|
|
5
|
+
* - 围栏代码块 + 内联里的函数/方法签名 `name(params)`
|
|
6
|
+
* - 错误码模式(E\d+ / ALL_CAPS_ERROR)
|
|
7
|
+
* - 段落级 deprecated 标记(@deprecated / 已废弃 / deprecated since X)
|
|
8
|
+
*
|
|
9
|
+
* 反过拟合(FR-018):规则 MUST NOT 匹配任何 eval manifest 的具体 API 名字面量。
|
|
10
|
+
*/
|
|
11
|
+
import { makeEntityId } from './entity-util.js';
|
|
12
|
+
const HEURISTIC_CONFIDENCE = 0.5;
|
|
13
|
+
// 函数/方法签名:可选前缀(export/async/function/def/fn/func/public...) + 名称 + (参数)
|
|
14
|
+
const SIG_RE = /(?:export\s+|public\s+|async\s+|function\s+|def\s+|fn\s+|func\s+)*([A-Za-z_$][\w$.]{1,})\s*\(([^)]{0,400})\)/g;
|
|
15
|
+
// 错误码:E + ≥2 位数字,或 全大写下划线 + _ERROR/_CODE/_EXCEPTION
|
|
16
|
+
const ERRCODE_RE = /\b(E\d{2,}|[A-Z][A-Z0-9]*(?:_[A-Z0-9]+)*_(?:ERROR|CODE|EXCEPTION))\b/g;
|
|
17
|
+
// deprecated 标记 + 可选 since 版本
|
|
18
|
+
const DEPRECATED_RE = /@deprecated|已废弃|deprecated\s+since\s+([\w.\-]+)|废弃于\s*([\w.\-]+)/i;
|
|
19
|
+
// 噪声名(语言关键字 / 控制流),不作为 API 名
|
|
20
|
+
const NOISE_NAMES = new Set([
|
|
21
|
+
'if', 'for', 'while', 'switch', 'return', 'function', 'class', 'catch', 'await',
|
|
22
|
+
'async', 'def', 'fn', 'func', 'import', 'export', 'const', 'let', 'var', 'new',
|
|
23
|
+
'typeof', 'instanceof', 'in', 'of', 'do', 'else', 'try', 'throw', 'with', 'super',
|
|
24
|
+
]);
|
|
25
|
+
function isNoiseName(name) {
|
|
26
|
+
if (name.length < 2)
|
|
27
|
+
return true;
|
|
28
|
+
const last = name.includes('.') ? (name.split('.').pop() ?? name) : name;
|
|
29
|
+
return NOISE_NAMES.has(last.toLowerCase());
|
|
30
|
+
}
|
|
31
|
+
/** 解析 `a, b: number, c = 1` → params(name + 可选 type) */
|
|
32
|
+
function parseParams(paramStr) {
|
|
33
|
+
const trimmed = paramStr.trim();
|
|
34
|
+
if (!trimmed)
|
|
35
|
+
return [];
|
|
36
|
+
return trimmed
|
|
37
|
+
.split(',')
|
|
38
|
+
.map((p) => p.trim())
|
|
39
|
+
.filter((p) => p.length > 0)
|
|
40
|
+
.map((p) => {
|
|
41
|
+
// 支持 `name: type` / `name=default` / `type name`(go/java 粗略)
|
|
42
|
+
const colon = p.split(':');
|
|
43
|
+
const rawName = (colon[0] ?? p).split('=')[0]?.trim() ?? p;
|
|
44
|
+
const name = rawName.replace(/^\.\.\./, '').replace(/\?$/, '').trim();
|
|
45
|
+
const type = colon.length > 1 ? colon.slice(1).join(':').trim() : null;
|
|
46
|
+
const entry = { name };
|
|
47
|
+
if (type)
|
|
48
|
+
entry.type = type;
|
|
49
|
+
return entry;
|
|
50
|
+
})
|
|
51
|
+
.filter((e) => e.name.length > 0 && /^[A-Za-z_$]/.test(e.name));
|
|
52
|
+
}
|
|
53
|
+
function parseDeprecated(text) {
|
|
54
|
+
const m = DEPRECATED_RE.exec(text);
|
|
55
|
+
if (!m)
|
|
56
|
+
return { isDeprecated: false };
|
|
57
|
+
const since = m[1] ?? m[2] ?? null;
|
|
58
|
+
return since ? { isDeprecated: true, since } : { isDeprecated: true };
|
|
59
|
+
}
|
|
60
|
+
/** 对单个 section 做确定性抽取,返回去重后的实体列表 */
|
|
61
|
+
export function extractHeuristic(section) {
|
|
62
|
+
const out = [];
|
|
63
|
+
const seen = new Set();
|
|
64
|
+
const primaryChunk = section.chunkIds[0] ?? '';
|
|
65
|
+
const dep = parseDeprecated(section.text);
|
|
66
|
+
const push = (name, kind, signature, params) => {
|
|
67
|
+
const id = makeEntityId(name, kind);
|
|
68
|
+
if (seen.has(id))
|
|
69
|
+
return;
|
|
70
|
+
seen.add(id);
|
|
71
|
+
const e = {
|
|
72
|
+
id,
|
|
73
|
+
name,
|
|
74
|
+
qualifiedName: name,
|
|
75
|
+
kind,
|
|
76
|
+
signature,
|
|
77
|
+
sourceDocId: section.docId,
|
|
78
|
+
sourceChunkId: primaryChunk,
|
|
79
|
+
sourceChunkIds: section.chunkIds,
|
|
80
|
+
sourceAnchor: section.anchor,
|
|
81
|
+
lang: section.lang,
|
|
82
|
+
confidence: HEURISTIC_CONFIDENCE,
|
|
83
|
+
extractionMethod: 'heuristic',
|
|
84
|
+
};
|
|
85
|
+
if (params !== undefined)
|
|
86
|
+
e.params = params;
|
|
87
|
+
if (dep.isDeprecated)
|
|
88
|
+
e.deprecated = dep;
|
|
89
|
+
out.push(e);
|
|
90
|
+
};
|
|
91
|
+
// 1. 错误码
|
|
92
|
+
for (const m of section.text.matchAll(ERRCODE_RE)) {
|
|
93
|
+
const code = m[1];
|
|
94
|
+
if (code)
|
|
95
|
+
push(code, 'error_code', null, undefined);
|
|
96
|
+
}
|
|
97
|
+
// 2. 函数/方法签名
|
|
98
|
+
for (const m of section.text.matchAll(SIG_RE)) {
|
|
99
|
+
const name = m[1];
|
|
100
|
+
const paramStr = m[2] ?? '';
|
|
101
|
+
if (!name || isNoiseName(name))
|
|
102
|
+
continue;
|
|
103
|
+
push(name, 'function', `${name}(${paramStr.trim()})`, parseParams(paramStr));
|
|
104
|
+
}
|
|
105
|
+
return out;
|
|
106
|
+
}
|
|
107
|
+
//# sourceMappingURL=entity-heuristic.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"entity-heuristic.js","sourceRoot":"","sources":["../../src/scaffold-kb/entity-heuristic.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAGH,OAAO,EAAE,YAAY,EAAE,MAAM,kBAAkB,CAAC;AAEhD,MAAM,oBAAoB,GAAG,GAAG,CAAC;AAEjC,wEAAwE;AACxE,MAAM,MAAM,GACV,+GAA+G,CAAC;AAClH,oDAAoD;AACpD,MAAM,UAAU,GAAG,uEAAuE,CAAC;AAC3F,8BAA8B;AAC9B,MAAM,aAAa,GAAG,mEAAmE,CAAC;AAE1F,6BAA6B;AAC7B,MAAM,WAAW,GAAG,IAAI,GAAG,CAAC;IAC1B,IAAI,EAAE,KAAK,EAAE,OAAO,EAAE,QAAQ,EAAE,QAAQ,EAAE,UAAU,EAAE,OAAO,EAAE,OAAO,EAAE,OAAO;IAC/E,OAAO,EAAE,KAAK,EAAE,IAAI,EAAE,MAAM,EAAE,QAAQ,EAAE,QAAQ,EAAE,OAAO,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK;IAC9E,QAAQ,EAAE,YAAY,EAAE,IAAI,EAAE,IAAI,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,EAAE,OAAO,EAAE,MAAM,EAAE,OAAO;CAClF,CAAC,CAAC;AAEH,SAAS,WAAW,CAAC,IAAY;IAC/B,IAAI,IAAI,CAAC,MAAM,GAAG,CAAC;QAAE,OAAO,IAAI,CAAC;IACjC,MAAM,IAAI,GAAG,IAAI,CAAC,QAAQ,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,GAAG,EAAE,IAAI,IAAI,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC;IACzE,OAAO,WAAW,CAAC,GAAG,CAAC,IAAI,CAAC,WAAW,EAAE,CAAC,CAAC;AAC7C,CAAC;AAED,wDAAwD;AACxD,SAAS,WAAW,CAAC,QAAgB;IACnC,MAAM,OAAO,GAAG,QAAQ,CAAC,IAAI,EAAE,CAAC;IAChC,IAAI,CAAC,OAAO;QAAE,OAAO,EAAE,CAAC;IACxB,OAAO,OAAO;SACX,KAAK,CAAC,GAAG,CAAC;SACV,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;SACpB,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,GAAG,CAAC,CAAC;SAC3B,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE;QACT,6DAA6D;QAC7D,MAAM,KAAK,GAAG,CAAC,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;QAC3B,MAAM,OAAO,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,EAAE,IAAI,EAAE,IAAI,CAAC,CAAC;QAC3D,MAAM,IAAI,GAAG,OAAO,CAAC,OAAO,CAAC,SAAS,EAAE,EAAE,CAAC,CAAC,OAAO,CAAC,KAAK,EAAE,EAAE,CAAC,CAAC,IAAI,EAAE,CAAC;QACtE,MAAM,IAAI,GAAG,KAAK,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC;QACvE,MAAM,KAAK,GAA6C,EAAE,IAAI,EAAE,CAAC;QACjE,IAAI,IAAI;YAAE,KAAK,CAAC,IAAI,GAAG,IAAI,CAAC;QAC5B,OAAO,KAAK,CAAC;IACf,CAAC,CAAC;SACD,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,GAAG,CAAC,IAAI,aAAa,CAAC,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC;AACpE,CAAC;AAED,SAAS,eAAe,CAAC,IAAY;IACnC,MAAM,CAAC,GAAG,aAAa,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IACnC,IAAI,CAAC,CAAC;QAAE,OAAO,EAAE,YAAY,EAAE,KAAK,EAAE,CAAC;IACvC,MAAM,KAAK,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,IAAI,IAAI,CAAC;IACnC,OAAO,KAAK,CAAC,CAAC,CAAC,EAAE,YAAY,EAAE,IAAI,EAAE,KAAK,EAAE,CAAC,CAAC,CAAC,EAAE,YAAY,EAAE,IAAI,EAAE,CAAC;AACxE,CAAC;AAED,oCAAoC;AACpC,MAAM,UAAU,gBAAgB,CAAC,OAA0B;IACzD,MAAM,GAAG,GAAgB,EAAE,CAAC;IAC5B,MAAM,IAAI,GAAG,IAAI,GAAG,EAAU,CAAC;IAC/B,MAAM,YAAY,GAAG,OAAO,CAAC,QAAQ,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;IAC/C,MAAM,GAAG,GAAG,eAAe,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC;IAE1C,MAAM,IAAI,GAAG,CACX,IAAY,EACZ,IAAuB,EACvB,SAAwB,EACxB,MAA2B,EACrB,EAAE;QACR,MAAM,EAAE,GAAG,YAAY,CAAC,IAAI,EAAE,IAAI,CAAC,CAAC;QACpC,IAAI,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC;YAAE,OAAO;QACzB,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,CAAC;QACb,MAAM,CAAC,GAAc;YACnB,EAAE;YACF,IAAI;YACJ,aAAa,EAAE,IAAI;YACnB,IAAI;YACJ,SAAS;YACT,WAAW,EAAE,OAAO,CAAC,KAAK;YAC1B,aAAa,EAAE,YAAY;YAC3B,cAAc,EAAE,OAAO,CAAC,QAAQ;YAChC,YAAY,EAAE,OAAO,CAAC,MAAM;YAC5B,IAAI,EAAE,OAAO,CAAC,IAAI;YAClB,UAAU,EAAE,oBAAoB;YAChC,gBAAgB,EAAE,WAAW;SAC9B,CAAC;QACF,IAAI,MAAM,KAAK,SAAS;YAAE,CAAC,CAAC,MAAM,GAAG,MAAM,CAAC;QAC5C,IAAI,GAAG,CAAC,YAAY;YAAE,CAAC,CAAC,UAAU,GAAG,GAAG,CAAC;QACzC,GAAG,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;IACd,CAAC,CAAC;IAEF,SAAS;IACT,KAAK,MAAM,CAAC,IAAI,OAAO,CAAC,IAAI,CAAC,QAAQ,CAAC,UAAU,CAAC,EAAE,CAAC;QAClD,MAAM,IAAI,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QAClB,IAAI,IAAI;YAAE,IAAI,CAAC,IAAI,EAAE,YAAY,EAAE,IAAI,EAAE,SAAS,CAAC,CAAC;IACtD,CAAC;IACD,aAAa;IACb,KAAK,MAAM,CAAC,IAAI,OAAO,CAAC,IAAI,CAAC,QAAQ,CAAC,MAAM,CAAC,EAAE,CAAC;QAC9C,MAAM,IAAI,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QAClB,MAAM,QAAQ,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;QAC5B,IAAI,CAAC,IAAI,IAAI,WAAW,CAAC,IAAI,CAAC;YAAE,SAAS;QACzC,IAAI,CAAC,IAAI,EAAE,UAAU,EAAE,GAAG,IAAI,IAAI,QAAQ,CAAC,IAAI,EAAE,GAAG,EAAE,WAAW,CAAC,QAAQ,CAAC,CAAC,CAAC;IAC/E,CAAC;IACD,OAAO,GAAG,CAAC;AACb,CAAC"}
|