@retrivora-ai/rag-engine 0.2.8 → 0.3.0

package/dist/{RagConfig--ibz0b3W.d.mts → RagConfig-D3Inaf9N.d.mts}

@@ -47,7 +47,7 @@ interface VectorDBConfig {
      *
      * For multi-table PostgreSQL search, the following options are also supported:
      * - tables?: string[] | string
-     * - searchFields?: string[] // optional override for which columns receive exact-match boosts
+     * - searchFields?: string[] | string // optional override for which columns receive exact-match boosts
      */
     options: Record<string, unknown>;
 }

package/dist/{RagConfig--ibz0b3W.d.ts → RagConfig-D3Inaf9N.d.ts}

@@ -47,7 +47,7 @@ interface VectorDBConfig {
      *
      * For multi-table PostgreSQL search, the following options are also supported:
      * - tables?: string[] | string
-     * - searchFields?: string[] // optional override for which columns receive exact-match boosts
+     * - searchFields?: string[] | string // optional override for which columns receive exact-match boosts
      */
     options: Record<string, unknown>;
 }

package/dist/{chunk-CYVPACA7.mjs → chunk-GT72OIOD.mjs}

@@ -316,6 +316,20 @@ var ConfigValidator = class {
         severity: "error"
       });
     }
+    if (opts.tables && typeof opts.tables !== "string" && !Array.isArray(opts.tables)) {
+      errors.push({
+        field: "vectorDb.options.tables",
+        message: "PostgreSQL tables must be a string or a string array",
+        severity: "error"
+      });
+    }
+    if (opts.searchFields && typeof opts.searchFields !== "string" && !Array.isArray(opts.searchFields)) {
+      errors.push({
+        field: "vectorDb.options.searchFields",
+        message: "PostgreSQL searchFields must be a string or a string array",
+        severity: "error"
+      });
+    }
     return errors;
   }
   /**
@@ -1556,7 +1570,11 @@ var EmbeddingStrategyResolver = class {
 function normalizeHintValue(value) {
   return value.replace(/\s+/g, " ").trim();
 }
+function isLikelyPromptPhrase(value) {
+  return /^(what|which|who|where|when|why|how)\b/i.test(value.trim());
+}
 function extractQueryFieldHints(question) {
+  var _a, _b, _c, _d;
   if (!question.trim()) return [];
   const hints = /* @__PURE__ */ new Map();
   const addHint = (value, field) => {
@@ -1573,6 +1591,46 @@ function extractQueryFieldHints(question) {
   for (const match of question.matchAll(/["']([^"']{2,100})["']/g)) {
     addHint(match[1]);
   }
+  const naturalQuestionPatterns = [
+    /\b(?:what|which)\s+(?:is|are|was|were)\s+(?:the\s+)?([^?.!,]{1,60}?)\s+of\s+["']?([^"'\n?.!,]{2,120})["']?(?=[?.!,]|$)/gi,
+    /\b(?:who|what)\s+(?:is|are|was|were)\s+["']?([^"'\n?.!,]{2,120})["']?(?=[?.!,]|$)/gi,
+    /\b(?:about|for|regarding)\s+["']?([^"'\n?.!,]{2,120})["']?(?=[?.!,]|$)/gi
+  ];
+  const personCompanyPatterns = [
+    /\bcompany(?:\s+name)?\s+(?:of|for)\s+["']?([^"'\n?.!,]{2,120})["']?(?=[?.!,]|$)/gi,
+    /\b(?:which|what)\s+company\s+does\s+["']?([^"'\n?.!,]{2,120})["']?\s+work(?:\s+for|\s+at)?(?=[?.!,]|$)/gi,
+    /\bwhere\s+does\s+["']?([^"'\n?.!,]{2,120})["']?\s+work(?:\s+for|\s+at)?(?=[?.!,]|$)/gi
+  ];
+  for (const pattern of personCompanyPatterns) {
+    for (const match of question.matchAll(pattern)) {
+      const name = match[1];
+      if (name) addHint(name, "name");
+    }
+  }
+  const universalPatterns = [
+    { regex: /([a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,})/gi, field: "email", group: 1 },
+    { regex: /(\+?\d[\d\-\.\s\(\)]{6,}\d)/g, field: "phone", group: 1 },
+    { regex: /(\b\d{4}-\d{2}-\d{2}\b|\b\d{1,2}\/\d{1,2}\/\d{2,4}\b)/g, field: "date", group: 1 },
+    { regex: /(\$\s?\d{1,3}(?:,\d{3})*(?:\.\d+)?)/g, field: "amount", group: 1 },
+    { regex: /\b(ID|id|identifier)[: ]\s*([A-Za-z0-9\-]{3,})\b/gi, field: "id", group: 2 },
+    // Generic quoted phrase / proper-noun sequences as keywords (already partially handled above)
+    { regex: /"([^"]{2,120})"/g, group: 1 },
+    { regex: /'([^']{2,120})'/g, group: 1 }
+  ];
+  for (const p of universalPatterns) {
+    for (const match of question.matchAll(p.regex)) {
+      const val = p.group ? (_a = match[p.group]) != null ? _a : match[0] : match[0];
+      if (!val) continue;
+      if (p.field) addHint(val, p.field);
+      else addHint(val);
+    }
+  }
+  for (const pattern of naturalQuestionPatterns) {
+    for (const match of question.matchAll(pattern)) {
+      const value = (_b = match[2]) != null ? _b : match[1];
+      if (value) addHint(value);
+    }
+  }
   const fieldPattern = `([^\\n:=?.!,]{1,60}?)`;
   const valuePattern = `([^\\n?.!,]{1,120}?)`;
   const fieldValuePatterns = [
@@ -1582,11 +1640,37 @@ function extractQueryFieldHints(question) {
   ];
   for (const pattern of fieldValuePatterns) {
     for (const match of question.matchAll(pattern)) {
-      addHint(match[2], match[1]);
+      const field = normalizeHintValue((_c = match[1]) != null ? _c : "");
+      const value = (_d = match[2]) != null ? _d : "";
+      if (field && !isLikelyPromptPhrase(field)) {
+        addHint(value, field);
+      } else {
+        addHint(value);
+      }
     }
   }
+  for (const match of question.matchAll(/\b[A-Z][a-z]+(?:\s+[A-Z][a-z]+){0,3}\b/g)) {
+    addHint(match[0]);
+  }
   return [...hints.values()];
 }
+function buildQueryFilter(question, hints) {
+  const filter = { metadata: {}, keywords: [], queryText: question };
+  for (const hint of hints) {
+    if (hint.field) {
+      filter.metadata[hint.field] = hint.value;
+    } else {
+      filter.keywords.push(hint.value);
+    }
+  }
+  for (const match of question.matchAll(/\b[A-Z][a-z]+(?:\s+[A-Z][a-z]+){0,3}\b/g)) {
+    const term = normalizeHintValue(match[0]);
+    if (term && !filter.keywords.includes(term)) filter.keywords.push(term);
+  }
+  if (Object.keys(filter.metadata || {}).length === 0) delete filter.metadata;
+  if (filter.keywords && filter.keywords.length === 0) delete filter.keywords;
+  return filter;
+}
 var Pipeline = class {
   constructor(config) {
     this.initialised = false;
@@ -1675,10 +1759,9 @@ var Pipeline = class {
     try {
       const queryVector = await this.embeddingProvider.embed(question, { taskType: "query" });
       const fieldHints = extractQueryFieldHints(question);
-      const rawMatches = await this.vectorDB.query(queryVector, topK, ns, {
-        __queryText: question,
-        __fieldHints: fieldHints
-      });
+      const filter = buildQueryFilter(question, fieldHints);
+      filter.__entityHints = fieldHints;
+      const rawMatches = await this.vectorDB.query(queryVector, topK, ns, filter);
       const sources = rawMatches.filter((m) => m.score >= scoreThreshold);
       const context = sources.length ? sources.map((m, i) => `[Source ${i + 1}]
 ${m.content}`).join("\n\n---\n\n") : "No relevant context found.";

package/dist/handlers/index.d.mts

@@ -1,3 +1,3 @@
-import '../RagConfig--ibz0b3W.mjs';
-export { c as createChatHandler, a as createHealthHandler, b as createIngestHandler, d as createUploadHandler } from '../index-w8qIEFvi.mjs';
+import '../RagConfig-D3Inaf9N.mjs';
+export { c as createChatHandler, a as createHealthHandler, b as createIngestHandler, d as createUploadHandler } from '../index-Ymwm-_OR.mjs';
 import 'next/server';

package/dist/handlers/index.d.ts

@@ -1,3 +1,3 @@
-import '../RagConfig--ibz0b3W.js';
-export { c as createChatHandler, a as createHealthHandler, b as createIngestHandler, d as createUploadHandler } from '../index-Dr1HN0se.js';
+import '../RagConfig-D3Inaf9N.js';
+export { c as createChatHandler, a as createHealthHandler, b as createIngestHandler, d as createUploadHandler } from '../index-BhNJQ2SS.js';
 import 'next/server';

package/dist/handlers/index.js

@@ -1484,6 +1484,20 @@ var ConfigValidator = class {
         severity: "error"
       });
     }
+    if (opts.tables && typeof opts.tables !== "string" && !Array.isArray(opts.tables)) {
+      errors.push({
+        field: "vectorDb.options.tables",
+        message: "PostgreSQL tables must be a string or a string array",
+        severity: "error"
+      });
+    }
+    if (opts.searchFields && typeof opts.searchFields !== "string" && !Array.isArray(opts.searchFields)) {
+      errors.push({
+        field: "vectorDb.options.searchFields",
+        message: "PostgreSQL searchFields must be a string or a string array",
+        severity: "error"
+      });
+    }
     return errors;
   }
   /**
@@ -2680,7 +2694,11 @@ var EmbeddingStrategyResolver = class {
 function normalizeHintValue(value) {
   return value.replace(/\s+/g, " ").trim();
 }
+function isLikelyPromptPhrase(value) {
+  return /^(what|which|who|where|when|why|how)\b/i.test(value.trim());
+}
 function extractQueryFieldHints(question) {
+  var _a, _b, _c, _d;
   if (!question.trim()) return [];
   const hints = /* @__PURE__ */ new Map();
   const addHint = (value, field) => {
@@ -2697,6 +2715,46 @@ function extractQueryFieldHints(question) {
   for (const match of question.matchAll(/["']([^"']{2,100})["']/g)) {
     addHint(match[1]);
   }
+  const naturalQuestionPatterns = [
+    /\b(?:what|which)\s+(?:is|are|was|were)\s+(?:the\s+)?([^?.!,]{1,60}?)\s+of\s+["']?([^"'\n?.!,]{2,120})["']?(?=[?.!,]|$)/gi,
+    /\b(?:who|what)\s+(?:is|are|was|were)\s+["']?([^"'\n?.!,]{2,120})["']?(?=[?.!,]|$)/gi,
+    /\b(?:about|for|regarding)\s+["']?([^"'\n?.!,]{2,120})["']?(?=[?.!,]|$)/gi
+  ];
+  const personCompanyPatterns = [
+    /\bcompany(?:\s+name)?\s+(?:of|for)\s+["']?([^"'\n?.!,]{2,120})["']?(?=[?.!,]|$)/gi,
+    /\b(?:which|what)\s+company\s+does\s+["']?([^"'\n?.!,]{2,120})["']?\s+work(?:\s+for|\s+at)?(?=[?.!,]|$)/gi,
+    /\bwhere\s+does\s+["']?([^"'\n?.!,]{2,120})["']?\s+work(?:\s+for|\s+at)?(?=[?.!,]|$)/gi
+  ];
+  for (const pattern of personCompanyPatterns) {
+    for (const match of question.matchAll(pattern)) {
+      const name = match[1];
+      if (name) addHint(name, "name");
+    }
+  }
+  const universalPatterns = [
+    { regex: /([a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,})/gi, field: "email", group: 1 },
+    { regex: /(\+?\d[\d\-\.\s\(\)]{6,}\d)/g, field: "phone", group: 1 },
+    { regex: /(\b\d{4}-\d{2}-\d{2}\b|\b\d{1,2}\/\d{1,2}\/\d{2,4}\b)/g, field: "date", group: 1 },
+    { regex: /(\$\s?\d{1,3}(?:,\d{3})*(?:\.\d+)?)/g, field: "amount", group: 1 },
+    { regex: /\b(ID|id|identifier)[: ]\s*([A-Za-z0-9\-]{3,})\b/gi, field: "id", group: 2 },
+    // Generic quoted phrase / proper-noun sequences as keywords (already partially handled above)
+    { regex: /"([^"]{2,120})"/g, group: 1 },
+    { regex: /'([^']{2,120})'/g, group: 1 }
+  ];
+  for (const p of universalPatterns) {
+    for (const match of question.matchAll(p.regex)) {
+      const val = p.group ? (_a = match[p.group]) != null ? _a : match[0] : match[0];
+      if (!val) continue;
+      if (p.field) addHint(val, p.field);
+      else addHint(val);
+    }
+  }
+  for (const pattern of naturalQuestionPatterns) {
+    for (const match of question.matchAll(pattern)) {
+      const value = (_b = match[2]) != null ? _b : match[1];
+      if (value) addHint(value);
+    }
+  }
   const fieldPattern = `([^\\n:=?.!,]{1,60}?)`;
   const valuePattern = `([^\\n?.!,]{1,120}?)`;
   const fieldValuePatterns = [
@@ -2706,11 +2764,37 @@ function extractQueryFieldHints(question) {
   ];
   for (const pattern of fieldValuePatterns) {
     for (const match of question.matchAll(pattern)) {
-      addHint(match[2], match[1]);
+      const field = normalizeHintValue((_c = match[1]) != null ? _c : "");
+      const value = (_d = match[2]) != null ? _d : "";
+      if (field && !isLikelyPromptPhrase(field)) {
+        addHint(value, field);
+      } else {
+        addHint(value);
+      }
     }
   }
+  for (const match of question.matchAll(/\b[A-Z][a-z]+(?:\s+[A-Z][a-z]+){0,3}\b/g)) {
+    addHint(match[0]);
+  }
   return [...hints.values()];
 }
+function buildQueryFilter(question, hints) {
+  const filter = { metadata: {}, keywords: [], queryText: question };
+  for (const hint of hints) {
+    if (hint.field) {
+      filter.metadata[hint.field] = hint.value;
+    } else {
+      filter.keywords.push(hint.value);
+    }
+  }
+  for (const match of question.matchAll(/\b[A-Z][a-z]+(?:\s+[A-Z][a-z]+){0,3}\b/g)) {
+    const term = normalizeHintValue(match[0]);
+    if (term && !filter.keywords.includes(term)) filter.keywords.push(term);
+  }
+  if (Object.keys(filter.metadata || {}).length === 0) delete filter.metadata;
+  if (filter.keywords && filter.keywords.length === 0) delete filter.keywords;
+  return filter;
+}
 var Pipeline = class {
   constructor(config) {
     this.initialised = false;
@@ -2799,10 +2883,9 @@ var Pipeline = class {
     try {
       const queryVector = await this.embeddingProvider.embed(question, { taskType: "query" });
       const fieldHints = extractQueryFieldHints(question);
-      const rawMatches = await this.vectorDB.query(queryVector, topK, ns, {
-        __queryText: question,
-        __fieldHints: fieldHints
-      });
+      const filter = buildQueryFilter(question, fieldHints);
+      filter.__entityHints = fieldHints;
+      const rawMatches = await this.vectorDB.query(queryVector, topK, ns, filter);
       const sources = rawMatches.filter((m) => m.score >= scoreThreshold);
       const context = sources.length ? sources.map((m, i) => `[Source ${i + 1}]
 ${m.content}`).join("\n\n---\n\n") : "No relevant context found.";

package/dist/handlers/index.mjs

@@ -3,7 +3,7 @@ import {
   createHealthHandler,
   createIngestHandler,
   createUploadHandler
-} from "../chunk-CYVPACA7.mjs";
+} from "../chunk-GT72OIOD.mjs";
 import "../chunk-EDLTMSNY.mjs";
 import "../chunk-FWCSY2DS.mjs";
 export {

package/dist/{index-Dr1HN0se.d.ts → index-BhNJQ2SS.d.ts}

@@ -1,4 +1,4 @@
-import { c as VectorDBConfig, L as LLMConfig, E as EmbeddingConfig, a as RagConfig, C as ChatResponse } from './RagConfig--ibz0b3W.js';
+import { c as VectorDBConfig, L as LLMConfig, E as EmbeddingConfig, a as RagConfig, C as ChatResponse } from './RagConfig-D3Inaf9N.js';
 import { NextRequest, NextResponse } from 'next/server';
 
 /**

package/dist/{index-w8qIEFvi.d.mts → index-Ymwm-_OR.d.mts}

@@ -1,4 +1,4 @@
-import { c as VectorDBConfig, L as LLMConfig, E as EmbeddingConfig, a as RagConfig, C as ChatResponse } from './RagConfig--ibz0b3W.mjs';
+import { c as VectorDBConfig, L as LLMConfig, E as EmbeddingConfig, a as RagConfig, C as ChatResponse } from './RagConfig-D3Inaf9N.mjs';
 import { NextRequest, NextResponse } from 'next/server';
 
 /**

package/dist/index.d.mts

@@ -2,8 +2,8 @@ import * as react_jsx_runtime from 'react/jsx-runtime';
 import React$1, { ReactNode } from 'react';
 import { C as ChatMessage } from './DocumentChunker-BICIjSuG.mjs';
 export { a as ChatOptions, b as Chunk, c as ChunkOptions, E as EmbedOptions, I as ILLMProvider } from './DocumentChunker-BICIjSuG.mjs';
-import { V as VectorMatch, U as UIConfig } from './RagConfig--ibz0b3W.mjs';
-export { C as ChatResponse, E as EmbeddingConfig, I as IngestDocument, L as LLMConfig, R as RAGConfig, a as RagConfig, b as UpsertDocument, c as VectorDBConfig } from './RagConfig--ibz0b3W.mjs';
+import { V as VectorMatch, U as UIConfig } from './RagConfig-D3Inaf9N.mjs';
+export { C as ChatResponse, E as EmbeddingConfig, I as IngestDocument, L as LLMConfig, R as RAGConfig, a as RagConfig, b as UpsertDocument, c as VectorDBConfig } from './RagConfig-D3Inaf9N.mjs';
 
 interface ChatWidgetProps {
     /** Position of the floating button. Defaults to bottom-right. */

package/dist/index.d.ts

@@ -2,8 +2,8 @@ import * as react_jsx_runtime from 'react/jsx-runtime';
 import React$1, { ReactNode } from 'react';
 import { C as ChatMessage } from './DocumentChunker-BICIjSuG.js';
 export { a as ChatOptions, b as Chunk, c as ChunkOptions, E as EmbedOptions, I as ILLMProvider } from './DocumentChunker-BICIjSuG.js';
-import { V as VectorMatch, U as UIConfig } from './RagConfig--ibz0b3W.js';
-export { C as ChatResponse, E as EmbeddingConfig, I as IngestDocument, L as LLMConfig, R as RAGConfig, a as RagConfig, b as UpsertDocument, c as VectorDBConfig } from './RagConfig--ibz0b3W.js';
+import { V as VectorMatch, U as UIConfig } from './RagConfig-D3Inaf9N.js';
+export { C as ChatResponse, E as EmbeddingConfig, I as IngestDocument, L as LLMConfig, R as RAGConfig, a as RagConfig, b as UpsertDocument, c as VectorDBConfig } from './RagConfig-D3Inaf9N.js';
 
 interface ChatWidgetProps {
     /** Position of the floating button. Defaults to bottom-right. */

package/dist/server.d.mts

@@ -1,9 +1,9 @@
-import { a as RagConfig, C as ChatResponse, I as IngestDocument, c as VectorDBConfig, b as UpsertDocument, V as VectorMatch, L as LLMConfig, E as EmbeddingConfig, d as VectorDBProvider, e as LLMProvider, f as EmbeddingProvider } from './RagConfig--ibz0b3W.mjs';
-export { R as RAGConfig, U as UIConfig } from './RagConfig--ibz0b3W.mjs';
+import { a as RagConfig, C as ChatResponse, I as IngestDocument, c as VectorDBConfig, b as UpsertDocument, V as VectorMatch, L as LLMConfig, E as EmbeddingConfig, d as VectorDBProvider, e as LLMProvider, f as EmbeddingProvider } from './RagConfig-D3Inaf9N.mjs';
+export { R as RAGConfig, U as UIConfig } from './RagConfig-D3Inaf9N.mjs';
 import { C as ChatMessage, I as ILLMProvider, a as ChatOptions, E as EmbedOptions } from './DocumentChunker-BICIjSuG.mjs';
 export { b as Chunk, c as ChunkOptions, D as DocumentChunker } from './DocumentChunker-BICIjSuG.mjs';
-import { H as HealthCheckResult } from './index-w8qIEFvi.mjs';
-export { P as ProviderHealthCheck, c as createChatHandler, a as createHealthHandler, b as createIngestHandler, d as createUploadHandler } from './index-w8qIEFvi.mjs';
+import { H as HealthCheckResult } from './index-Ymwm-_OR.mjs';
+export { P as ProviderHealthCheck, c as createChatHandler, a as createHealthHandler, b as createIngestHandler, d as createUploadHandler } from './index-Ymwm-_OR.mjs';
 import 'next/server';
 
 /**
@@ -706,7 +706,7 @@ declare class MultiTablePostgresProvider extends BaseVectorProvider {
     private readonly dimensions;
     private readonly connectionString;
     private tables;
-    private tableSearchConfig;
+    private searchFields;
     constructor(config: VectorDBConfig);
     initialize(): Promise<void>;
     /**
@@ -726,8 +726,6 @@ declare class MultiTablePostgresProvider extends BaseVectorProvider {
     deleteNamespace(_namespace: string): Promise<void>;
     ping(): Promise<boolean>;
     disconnect(): Promise<void>;
-    private loadTableSearchConfig;
-    private parseConfiguredSearchFields;
 }
 
 /**

package/dist/server.d.ts

@@ -1,9 +1,9 @@
-import { a as RagConfig, C as ChatResponse, I as IngestDocument, c as VectorDBConfig, b as UpsertDocument, V as VectorMatch, L as LLMConfig, E as EmbeddingConfig, d as VectorDBProvider, e as LLMProvider, f as EmbeddingProvider } from './RagConfig--ibz0b3W.js';
-export { R as RAGConfig, U as UIConfig } from './RagConfig--ibz0b3W.js';
+import { a as RagConfig, C as ChatResponse, I as IngestDocument, c as VectorDBConfig, b as UpsertDocument, V as VectorMatch, L as LLMConfig, E as EmbeddingConfig, d as VectorDBProvider, e as LLMProvider, f as EmbeddingProvider } from './RagConfig-D3Inaf9N.js';
+export { R as RAGConfig, U as UIConfig } from './RagConfig-D3Inaf9N.js';
 import { C as ChatMessage, I as ILLMProvider, a as ChatOptions, E as EmbedOptions } from './DocumentChunker-BICIjSuG.js';
 export { b as Chunk, c as ChunkOptions, D as DocumentChunker } from './DocumentChunker-BICIjSuG.js';
-import { H as HealthCheckResult } from './index-Dr1HN0se.js';
-export { P as ProviderHealthCheck, c as createChatHandler, a as createHealthHandler, b as createIngestHandler, d as createUploadHandler } from './index-Dr1HN0se.js';
+import { H as HealthCheckResult } from './index-BhNJQ2SS.js';
+export { P as ProviderHealthCheck, c as createChatHandler, a as createHealthHandler, b as createIngestHandler, d as createUploadHandler } from './index-BhNJQ2SS.js';
 import 'next/server';
 
 /**
@@ -706,7 +706,7 @@ declare class MultiTablePostgresProvider extends BaseVectorProvider {
     private readonly dimensions;
     private readonly connectionString;
     private tables;
-    private tableSearchConfig;
+    private searchFields;
     constructor(config: VectorDBConfig);
     initialize(): Promise<void>;
     /**
@@ -726,8 +726,6 @@ declare class MultiTablePostgresProvider extends BaseVectorProvider {
     deleteNamespace(_namespace: string): Promise<void>;
     ping(): Promise<boolean>;
     disconnect(): Promise<void>;
-    private loadTableSearchConfig;
-    private parseConfiguredSearchFields;
 }
 
 /**