driftdetect-cortex 0.9.41

package/dist/embeddings/lexical/tokenizer.d.ts

@@ -0,0 +1,71 @@
+/**
+ * Code Tokenizer
+ *
+ * Tokenizes code for lexical analysis, handling:
+ * - CamelCase splitting
+ * - snake_case splitting
+ * - Common token filtering
+ * - Code-specific tokenization
+ *
+ * @module embeddings/lexical/tokenizer
+ */
+/**
+ * Tokenizer configuration
+ */
+export interface TokenizerConfig {
+    /** Minimum token length */
+    minTokenLength: number;
+    /** Maximum token length */
+    maxTokenLength: number;
+    /** Whether to lowercase tokens */
+    lowercase: boolean;
+    /** Whether to split compound words */
+    splitCompounds: boolean;
+    /** Custom stop words to filter */
+    stopWords?: Set<string>;
+}
+/**
+ * Code tokenizer for lexical analysis
+ */
+export declare class CodeTokenizer {
+    private config;
+    private stopWords;
+    constructor(config?: Partial<TokenizerConfig>);
+    /**
+     * Tokenize code into meaningful tokens
+     */
+    tokenize(code: string): string[];
+    /**
+     * Tokenize with frequency information
+     */
+    tokenizeWithFrequency(code: string): Map<string, number>;
+    /**
+     * Remove comments and string literals
+     */
+    private removeCommentsAndStrings;
+    /**
+     * Extract tokens from cleaned code
+     */
+    private extractTokens;
+    /**
+     * Split compound words (camelCase, PascalCase, snake_case)
+     */
+    private splitCompoundWord;
+    /**
+     * Split camelCase and PascalCase
+     */
+    private splitCamelCase;
+    /**
+     * Check if a token is valid
+     */
+    private isValidToken;
+    /**
+     * Get the stop words set
+     */
+    getStopWords(): Set<string>;
+    /**
+     * Add custom stop words
+     */
+    addStopWords(words: string[]): void;
+}
+//# sourceMappingURL=tokenizer.d.ts.map

package/dist/embeddings/lexical/tokenizer.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"tokenizer.d.ts","sourceRoot":"","sources":["../../../src/embeddings/lexical/tokenizer.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AAEH;;GAEG;AACH,MAAM,WAAW,eAAe;IAC9B,2BAA2B;IAC3B,cAAc,EAAE,MAAM,CAAC;IACvB,2BAA2B;IAC3B,cAAc,EAAE,MAAM,CAAC;IACvB,kCAAkC;IAClC,SAAS,EAAE,OAAO,CAAC;IACnB,sCAAsC;IACtC,cAAc,EAAE,OAAO,CAAC;IACxB,kCAAkC;IAClC,SAAS,CAAC,EAAE,GAAG,CAAC,MAAM,CAAC,CAAC;CACzB;AA8BD;;GAEG;AACH,qBAAa,aAAa;IACxB,OAAO,CAAC,MAAM,CAAkB;IAChC,OAAO,CAAC,SAAS,CAAc;gBAEnB,MAAM,CAAC,EAAE,OAAO,CAAC,eAAe,CAAC;IAK7C;;OAEG;IACH,QAAQ,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,EAAE;IAqBhC;;OAEG;IACH,qBAAqB,CAAC,IAAI,EAAE,MAAM,GAAG,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC;IAoBxD;;OAEG;IACH,OAAO,CAAC,wBAAwB;IAehC;;OAEG;IACH,OAAO,CAAC,aAAa;IAMrB;;OAEG;IACH,OAAO,CAAC,iBAAiB;IAezB;;OAEG;IACH,OAAO,CAAC,cAAc;IAUtB;;OAEG;IACH,OAAO,CAAC,YAAY;IAiBpB;;OAEG;IACH,YAAY,IAAI,GAAG,CAAC,MAAM,CAAC;IAI3B;;OAEG;IACH,YAAY,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,IAAI;CAKpC"}

package/dist/embeddings/lexical/tokenizer.js

@@ -0,0 +1,167 @@
+/**
+ * Code Tokenizer
+ *
+ * Tokenizes code for lexical analysis, handling:
+ * - CamelCase splitting
+ * - snake_case splitting
+ * - Common token filtering
+ * - Code-specific tokenization
+ *
+ * @module embeddings/lexical/tokenizer
+ */
+/**
+ * Default configuration
+ */
+const DEFAULT_CONFIG = {
+    minTokenLength: 2,
+    maxTokenLength: 50,
+    lowercase: true,
+    splitCompounds: true,
+};
+/**
+ * Common programming stop words
+ */
+const DEFAULT_STOP_WORDS = new Set([
+    // Language keywords
+    'const', 'let', 'var', 'function', 'class', 'interface', 'type',
+    'import', 'export', 'from', 'return', 'if', 'else', 'for', 'while',
+    'switch', 'case', 'break', 'continue', 'try', 'catch', 'finally',
+    'throw', 'new', 'this', 'super', 'extends', 'implements',
+    'public', 'private', 'protected', 'static', 'async', 'await',
+    'true', 'false', 'null', 'undefined', 'void',
+    // Common short words
+    'the', 'and', 'for', 'not', 'with', 'has', 'get', 'set',
+    // Common code words that are too generic
+    'data', 'value', 'item', 'items', 'list', 'array', 'object',
+    'string', 'number', 'boolean', 'any', 'unknown',
+]);
+/**
+ * Code tokenizer for lexical analysis
+ */
+export class CodeTokenizer {
+    config;
+    stopWords;
+    constructor(config) {
+        this.config = { ...DEFAULT_CONFIG, ...config };
+        this.stopWords = config?.stopWords ?? DEFAULT_STOP_WORDS;
+    }
+    /**
+     * Tokenize code into meaningful tokens
+     */
+    tokenize(code) {
+        // Step 1: Remove comments and strings
+        const cleaned = this.removeCommentsAndStrings(code);
+        // Step 2: Extract identifiers and words
+        const rawTokens = this.extractTokens(cleaned);
+        // Step 3: Split compound words if configured
+        const splitTokens = this.config.splitCompounds
+            ? rawTokens.flatMap(t => this.splitCompoundWord(t))
+            : rawTokens;
+        // Step 4: Normalize and filter
+        const normalized = splitTokens
+            .map(t => this.config.lowercase ? t.toLowerCase() : t)
+            .filter(t => this.isValidToken(t));
+        // Step 5: Remove duplicates while preserving order
+        return [...new Set(normalized)];
+    }
+    /**
+     * Tokenize with frequency information
+     */
+    tokenizeWithFrequency(code) {
+        const cleaned = this.removeCommentsAndStrings(code);
+        const rawTokens = this.extractTokens(cleaned);
+        const splitTokens = this.config.splitCompounds
+            ? rawTokens.flatMap(t => this.splitCompoundWord(t))
+            : rawTokens;
+        const frequency = new Map();
+        for (const token of splitTokens) {
+            const normalized = this.config.lowercase ? token.toLowerCase() : token;
+            if (this.isValidToken(normalized)) {
+                frequency.set(normalized, (frequency.get(normalized) ?? 0) + 1);
+            }
+        }
+        return frequency;
+    }
+    /**
+     * Remove comments and string literals
+     */
+    removeCommentsAndStrings(code) {
+        // Remove single-line comments
+        let result = code.replace(/\/\/.*$/gm, '');
+        // Remove multi-line comments
+        result = result.replace(/\/\*[\s\S]*?\*\//g, '');
+        // Remove string literals (simple approach)
+        result = result.replace(/"(?:[^"\\]|\\.)*"/g, '');
+        result = result.replace(/'(?:[^'\\]|\\.)*'/g, '');
+        result = result.replace(/`(?:[^`\\]|\\.)*`/g, '');
+        return result;
+    }
+    /**
+     * Extract tokens from cleaned code
+     */
+    extractTokens(code) {
+        // Match identifiers and words
+        const pattern = /[a-zA-Z_$][a-zA-Z0-9_$]*/g;
+        return code.match(pattern) ?? [];
+    }
+    /**
+     * Split compound words (camelCase, PascalCase, snake_case)
+     */
+    splitCompoundWord(word) {
+        const parts = [];
+        // Split on underscores first
+        const underscoreParts = word.split('_').filter(Boolean);
+        for (const part of underscoreParts) {
+            // Split camelCase/PascalCase
+            const camelParts = this.splitCamelCase(part);
+            parts.push(...camelParts);
+        }
+        return parts.length > 0 ? parts : [word];
+    }
+    /**
+     * Split camelCase and PascalCase
+     */
+    splitCamelCase(word) {
+        // Insert space before uppercase letters that follow lowercase
+        // Also handle acronyms (e.g., XMLParser -> XML Parser)
+        const spaced = word
+            .replace(/([a-z])([A-Z])/g, '$1 $2')
+            .replace(/([A-Z]+)([A-Z][a-z])/g, '$1 $2');
+        return spaced.split(' ').filter(Boolean);
+    }
+    /**
+     * Check if a token is valid
+     */
+    isValidToken(token) {
+        // Check length
+        if (token.length < this.config.minTokenLength)
+            return false;
+        if (token.length > this.config.maxTokenLength)
+            return false;
+        // Check if it's a stop word
+        if (this.stopWords.has(token.toLowerCase()))
+            return false;
+        // Check if it's all digits
+        if (/^\d+$/.test(token))
+            return false;
+        // Check if it's a single repeated character
+        if (/^(.)\1+$/.test(token))
+            return false;
+        return true;
+    }
+    /**
+     * Get the stop words set
+     */
+    getStopWords() {
+        return new Set(this.stopWords);
+    }
+    /**
+     * Add custom stop words
+     */
+    addStopWords(words) {
+        for (const word of words) {
+            this.stopWords.add(word.toLowerCase());
+        }
+    }
+}
+//# sourceMappingURL=tokenizer.js.map

package/dist/embeddings/lexical/tokenizer.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"tokenizer.js","sourceRoot":"","sources":["../../../src/embeddings/lexical/tokenizer.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AAkBH;;GAEG;AACH,MAAM,cAAc,GAAoB;IACtC,cAAc,EAAE,CAAC;IACjB,cAAc,EAAE,EAAE;IAClB,SAAS,EAAE,IAAI;IACf,cAAc,EAAE,IAAI;CACrB,CAAC;AAEF;;GAEG;AACH,MAAM,kBAAkB,GAAG,IAAI,GAAG,CAAC;IACjC,oBAAoB;IACpB,OAAO,EAAE,KAAK,EAAE,KAAK,EAAE,UAAU,EAAE,OAAO,EAAE,WAAW,EAAE,MAAM;IAC/D,QAAQ,EAAE,QAAQ,EAAE,MAAM,EAAE,QAAQ,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,EAAE,OAAO;IAClE,QAAQ,EAAE,MAAM,EAAE,OAAO,EAAE,UAAU,EAAE,KAAK,EAAE,OAAO,EAAE,SAAS;IAChE,OAAO,EAAE,KAAK,EAAE,MAAM,EAAE,OAAO,EAAE,SAAS,EAAE,YAAY;IACxD,QAAQ,EAAE,SAAS,EAAE,WAAW,EAAE,QAAQ,EAAE,OAAO,EAAE,OAAO;IAC5D,MAAM,EAAE,OAAO,EAAE,MAAM,EAAE,WAAW,EAAE,MAAM;IAC5C,qBAAqB;IACrB,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,MAAM,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK;IACvD,yCAAyC;IACzC,MAAM,EAAE,OAAO,EAAE,MAAM,EAAE,OAAO,EAAE,MAAM,EAAE,OAAO,EAAE,QAAQ;IAC3D,QAAQ,EAAE,QAAQ,EAAE,SAAS,EAAE,KAAK,EAAE,SAAS;CAChD,CAAC,CAAC;AAEH;;GAEG;AACH,MAAM,OAAO,aAAa;IAChB,MAAM,CAAkB;IACxB,SAAS,CAAc;IAE/B,YAAY,MAAiC;QAC3C,IAAI,CAAC,MAAM,GAAG,EAAE,GAAG,cAAc,EAAE,GAAG,MAAM,EAAE,CAAC;QAC/C,IAAI,CAAC,SAAS,GAAG,MAAM,EAAE,SAAS,IAAI,kBAAkB,CAAC;IAC3D,CAAC;IAED;;OAEG;IACH,QAAQ,CAAC,IAAY;QACnB,sCAAsC;QACtC,MAAM,OAAO,GAAG,IAAI,CAAC,wBAAwB,CAAC,IAAI,CAAC,CAAC;QAEpD,wCAAwC;QACxC,MAAM,SAAS,GAAG,IAAI,CAAC,aAAa,CAAC,OAAO,CAAC,CAAC;QAE9C,6CAA6C;QAC7C,MAAM,WAAW,GAAG,IAAI,CAAC,MAAM,CAAC,cAAc;YAC5C,CAAC,CAAC,SAAS,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,iBAAiB,CAAC,CAAC,CAAC,CAAC;YACnD,CAAC,CAAC,SAAS,CAAC;QAEd,+BAA+B;QAC/B,MAAM,UAAU,GAAG,WAAW;aAC3B,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,MAAM,CAAC,SAAS,CAAC,CAAC,CAAC,CAAC,CAAC,WAAW,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC;aACrD,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,YAAY,CAAC,CAAC,CAAC,CAAC,CAAC;QAErC,mDAAmD;QACnD,OAAO,CAAC,GAAG,IAAI,GAAG,CAAC,UAAU,CAAC,CAAC,CAAC;IAClC,CAAC;IAED;;OAEG;IACH,qBAAqB,CAAC,IAAY;QAChC,MAAM,OAAO,GAAG,IAAI,CAAC,wBAAwB,CAAC,IAAI,CAAC,CAAC;QACpD,MAAM,SAAS,GAAG,IAAI,CAAC,aAAa,CAAC,OAAO,CAAC,CAAC;QAE9C,MAAM,WAAW,GAAG,IAAI,CAAC,MAAM,CAAC,cAAc;YAC5C,CAAC,CAAC,SAAS,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,iBAAiB,CAAC,CAAC,CAAC,CAAC;YACnD,CAAC,CAAC,SAAS,CAAC;QAEd,MAAM,SAAS,GAAG,IAAI,GAAG,EAAkB,CAAC;QAE5C,KAAK,MAAM,KAAK,IAAI,WAAW,EAAE,CAAC;YAChC,MAAM,UAAU,GAAG,IAAI,CAAC,MAAM,CAAC,SAAS,CAAC,CAAC,CAAC,KAAK,CAAC,WAAW,EAAE,CAAC,CAAC,CAAC,KAAK,CAAC;YACvE,IAAI,IAAI,CAAC,YAAY,CAAC,UAAU,CAAC,EAAE,CAAC;gBAClC,SAAS,CAAC,GAAG,CAAC,UAAU,EAAE,CAAC,SAAS,CAAC,GAAG,CAAC,UAAU,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;YAClE,CAAC;QACH,CAAC;QAED,OAAO,SAAS,CAAC;IACnB,CAAC;IAED;;OAEG;IACK,wBAAwB,CAAC,IAAY;QAC3C,8BAA8B;QAC9B,IAAI,MAAM,GAAG,IAAI,CAAC,OAAO,CAAC,WAAW,EAAE,EAAE,CAAC,CAAC;QAE3C,6BAA6B;QAC7B,MAAM,GAAG,MAAM,CAAC,OAAO,CAAC,mBAAmB,EAAE,EAAE,CAAC,CAAC;QAEjD,2CAA2C;QAC3C,MAAM,GAAG,MAAM,CAAC,OAAO,CAAC,oBAAoB,EAAE,EAAE,CAAC,CAAC;QAClD,MAAM,GAAG,MAAM,CAAC,OAAO,CAAC,oBAAoB,EAAE,EAAE,CAAC,CAAC;QAClD,MAAM,GAAG,MAAM,CAAC,OAAO,CAAC,oBAAoB,EAAE,EAAE,CAAC,CAAC;QAElD,OAAO,MAAM,CAAC;IAChB,CAAC;IAED;;OAEG;IACK,aAAa,CAAC,IAAY;QAChC,8BAA8B;QAC9B,MAAM,OAAO,GAAG,2BAA2B,CAAC;QAC5C,OAAO,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,IAAI,EAAE,CAAC;IACnC,CAAC;IAED;;OAEG;IACK,iBAAiB,CAAC,IAAY;QACpC,MAAM,KAAK,GAAa,EAAE,CAAC;QAE3B,6BAA6B;QAC7B,MAAM,eAAe,GAAG,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC;QAExD,KAAK,MAAM,IAAI,IAAI,eAAe,EAAE,CAAC;YACnC,6BAA6B;YAC7B,MAAM,UAAU,GAAG,IAAI,CAAC,cAAc,CAAC,IAAI,CAAC,CAAC;YAC7C,KAAK,CAAC,IAAI,CAAC,GAAG,UAAU,CAAC,CAAC;QAC5B,CAAC;QAED,OAAO,KAAK,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC;IAC3C,CAAC;IAED;;OAEG;IACK,cAAc,CAAC,IAAY;QACjC,8DAA8D;QAC9D,uDAAuD;QACvD,MAAM,MAAM,GAAG,IAAI;aAChB,OAAO,CAAC,iBAAiB,EAAE,OAAO,CAAC;aACnC,OAAO,CAAC,uBAAuB,EAAE,OAAO,CAAC,CAAC;QAE7C,OAAO,MAAM,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC;IAC3C,CAAC;IAED;;OAEG;IACK,YAAY,CAAC,KAAa;QAChC,eAAe;QACf,IAAI,KAAK,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAC,cAAc;YAAE,OAAO,KAAK,CAAC;QAC5D,IAAI,KAAK,CAAC,MAAM,GAAG,IAAI,CAAC,MAAM,CAAC,cAAc;YAAE,OAAO,KAAK,CAAC;QAE5D,4BAA4B;QAC5B,IAAI,IAAI,CAAC,SAAS,CAAC,GAAG,CAAC,KAAK,CAAC,WAAW,EAAE,CAAC;YAAE,OAAO,KAAK,CAAC;QAE1D,2BAA2B;QAC3B,IAAI,OAAO,CAAC,IAAI,CAAC,KAAK,CAAC;YAAE,OAAO,KAAK,CAAC;QAEtC,4CAA4C;QAC5C,IAAI,UAAU,CAAC,IAAI,CAAC,KAAK,CAAC;YAAE,OAAO,KAAK,CAAC;QAEzC,OAAO,IAAI,CAAC;IACd,CAAC;IAED;;OAEG;IACH,YAAY;QACV,OAAO,IAAI,GAAG,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC;IACjC,CAAC;IAED;;OAEG;IACH,YAAY,CAAC,KAAe;QAC1B,KAAK,MAAM,IAAI,IAAI,KAAK,EAAE,CAAC;YACzB,IAAI,CAAC,SAAS,CAAC,GAAG,CAAC,IAAI,CAAC,WAAW,EAAE,CAAC,CAAC;QACzC,CAAC;IACH,CAAC;CACF"}

package/dist/embeddings/local.d.ts

@@ -0,0 +1,35 @@
+/**
+ * Local Embedding Provider (Transformers.js)
+ *
+ * Default provider that works offline using Transformers.js.
+ * Uses all-MiniLM-L6-v2 model (384 dimensions).
+ */
+import type { IEmbeddingProvider } from './interface.js';
+/**
+ * Local embedding provider using Transformers.js
+ */
+export declare class LocalEmbeddingProvider implements IEmbeddingProvider {
+    readonly name = "local";
+    readonly dimensions = 384;
+    readonly maxTokens = 512;
+    private extractor;
+    private modelId;
+    private initialized;
+    /**
+     * Initialize the provider
+     */
+    initialize(): Promise<void>;
+    /**
+     * Generate embedding for a single text
+     */
+    embed(text: string): Promise<number[]>;
+    /**
+     * Generate embeddings for multiple texts
+     */
+    embedBatch(texts: string[]): Promise<number[][]>;
+    /**
+     * Check if the provider is available
+     */
+    isAvailable(): Promise<boolean>;
+}
+//# sourceMappingURL=local.d.ts.map

package/dist/embeddings/local.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"local.d.ts","sourceRoot":"","sources":["../../src/embeddings/local.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,KAAK,EAAE,kBAAkB,EAAE,MAAM,gBAAgB,CAAC;AAKzD;;GAEG;AACH,qBAAa,sBAAuB,YAAW,kBAAkB;IAC/D,QAAQ,CAAC,IAAI,WAAW;IACxB,QAAQ,CAAC,UAAU,OAAO;IAC1B,QAAQ,CAAC,SAAS,OAAO;IAEzB,OAAO,CAAC,SAAS,CAA0C;IAC3D,OAAO,CAAC,OAAO,CAA6B;IAC5C,OAAO,CAAC,WAAW,CAAS;IAE5B;;OAEG;IACG,UAAU,IAAI,OAAO,CAAC,IAAI,CAAC;IAejC;;OAEG;IACG,KAAK,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC;IAe5C;;OAEG;IACG,UAAU,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC;IAItD;;OAEG;IACG,WAAW,IAAI,OAAO,CAAC,OAAO,CAAC;CAUtC"}

package/dist/embeddings/local.js

@@ -0,0 +1,71 @@
+/**
+ * Local Embedding Provider (Transformers.js)
+ *
+ * Default provider that works offline using Transformers.js.
+ * Uses all-MiniLM-L6-v2 model (384 dimensions).
+ */
+/**
+ * Local embedding provider using Transformers.js
+ */
+export class LocalEmbeddingProvider {
+    name = 'local';
+    dimensions = 384;
+    maxTokens = 512;
+    extractor = null;
+    modelId = 'Xenova/all-MiniLM-L6-v2';
+    initialized = false;
+    /**
+     * Initialize the provider
+     */
+    async initialize() {
+        if (this.initialized)
+            return;
+        try {
+            const { pipeline } = await import('@xenova/transformers');
+            this.extractor = await pipeline('feature-extraction', this.modelId, {
+                quantized: true, // Smaller, faster
+            });
+            this.initialized = true;
+        }
+        catch (error) {
+            console.error('Failed to initialize local embedding provider:', error);
+            throw new Error('Failed to initialize Transformers.js');
+        }
+    }
+    /**
+     * Generate embedding for a single text
+     */
+    async embed(text) {
+        if (!this.extractor) {
+            throw new Error('Provider not initialized');
+        }
+        // Use type assertion to work around complex Transformers.js types
+        const output = await this.extractor(text, {
+            pooling: 'mean',
+            normalize: true,
+        });
+        // Extract the embedding array
+        return Array.from(output.data);
+    }
+    /**
+     * Generate embeddings for multiple texts
+     */
+    async embedBatch(texts) {
+        return Promise.all(texts.map(t => this.embed(t)));
+    }
+    /**
+     * Check if the provider is available
+     */
+    async isAvailable() {
+        try {
+            if (!this.initialized) {
+                await this.initialize();
+            }
+            return true;
+        }
+        catch {
+            return false;
+        }
+    }
+}
+//# sourceMappingURL=local.js.map

package/dist/embeddings/local.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"local.js","sourceRoot":"","sources":["../../src/embeddings/local.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAOH;;GAEG;AACH,MAAM,OAAO,sBAAsB;IACxB,IAAI,GAAG,OAAO,CAAC;IACf,UAAU,GAAG,GAAG,CAAC;IACjB,SAAS,GAAG,GAAG,CAAC;IAEjB,SAAS,GAAqC,IAAI,CAAC;IACnD,OAAO,GAAG,yBAAyB,CAAC;IACpC,WAAW,GAAG,KAAK,CAAC;IAE5B;;OAEG;IACH,KAAK,CAAC,UAAU;QACd,IAAI,IAAI,CAAC,WAAW;YAAE,OAAO;QAE7B,IAAI,CAAC;YACH,MAAM,EAAE,QAAQ,EAAE,GAAG,MAAM,MAAM,CAAC,sBAAsB,CAAC,CAAC;YAC1D,IAAI,CAAC,SAAS,GAAG,MAAM,QAAQ,CAAC,oBAAoB,EAAE,IAAI,CAAC,OAAO,EAAE;gBAClE,SAAS,EAAE,IAAI,EAAE,kBAAkB;aACpC,CAAC,CAAC;YACH,IAAI,CAAC,WAAW,GAAG,IAAI,CAAC;QAC1B,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,OAAO,CAAC,KAAK,CAAC,gDAAgD,EAAE,KAAK,CAAC,CAAC;YACvE,MAAM,IAAI,KAAK,CAAC,sCAAsC,CAAC,CAAC;QAC1D,CAAC;IACH,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,KAAK,CAAC,IAAY;QACtB,IAAI,CAAC,IAAI,CAAC,SAAS,EAAE,CAAC;YACpB,MAAM,IAAI,KAAK,CAAC,0BAA0B,CAAC,CAAC;QAC9C,CAAC;QAED,kEAAkE;QAClE,MAAM,MAAM,GAAG,MAAO,IAAI,CAAC,SAAiH,CAAC,IAAI,EAAE;YACjJ,OAAO,EAAE,MAAM;YACf,SAAS,EAAE,IAAI;SAChB,CAAC,CAAC;QAEH,8BAA8B;QAC9B,OAAO,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC;IACjC,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,UAAU,CAAC,KAAe;QAC9B,OAAO,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;IACpD,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,WAAW;QACf,IAAI,CAAC;YACH,IAAI,CAAC,IAAI,CAAC,WAAW,EAAE,CAAC;gBACtB,MAAM,IAAI,CAAC,UAAU,EAAE,CAAC;YAC1B,CAAC;YACD,OAAO,IAAI,CAAC;QACd,CAAC;QAAC,MAAM,CAAC;YACP,OAAO,KAAK,CAAC;QACf,CAAC;IACH,CAAC;CACF"}

package/dist/embeddings/ollama.d.ts

@@ -0,0 +1,35 @@
+/**
+ * Ollama Embedding Provider
+ *
+ * Local embeddings using Ollama.
+ * Uses nomic-embed-text model (768 dimensions).
+ */
+import type { IEmbeddingProvider } from './interface.js';
+/**
+ * Ollama embedding provider
+ */
+export declare class OllamaEmbeddingProvider implements IEmbeddingProvider {
+    readonly name = "ollama";
+    readonly dimensions = 768;
+    readonly maxTokens = 8192;
+    private baseUrl;
+    private model;
+    constructor(baseUrl?: string, model?: string);
+    /**
+     * Initialize the provider
+     */
+    initialize(): Promise<void>;
+    /**
+     * Generate embedding for a single text
+     */
+    embed(text: string): Promise<number[]>;
+    /**
+     * Generate embeddings for multiple texts
+     */
+    embedBatch(texts: string[]): Promise<number[][]>;
+    /**
+     * Check if the provider is available
+     */
+    isAvailable(): Promise<boolean>;
+}
+//# sourceMappingURL=ollama.d.ts.map

package/dist/embeddings/ollama.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"ollama.d.ts","sourceRoot":"","sources":["../../src/embeddings/ollama.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,KAAK,EAAE,kBAAkB,EAAE,MAAM,gBAAgB,CAAC;AAEzD;;GAEG;AACH,qBAAa,uBAAwB,YAAW,kBAAkB;IAChE,QAAQ,CAAC,IAAI,YAAY;IACzB,QAAQ,CAAC,UAAU,OAAO;IAC1B,QAAQ,CAAC,SAAS,QAAQ;IAE1B,OAAO,CAAC,OAAO,CAAS;IACxB,OAAO,CAAC,KAAK,CAAS;gBAEV,OAAO,SAA2B,EAAE,KAAK,SAAqB;IAK1E;;OAEG;IACG,UAAU,IAAI,OAAO,CAAC,IAAI,CAAC;IAOjC;;OAEG;IACG,KAAK,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC;IAmB5C;;OAEG;IACG,UAAU,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC;IAItD;;OAEG;IACG,WAAW,IAAI,OAAO,CAAC,OAAO,CAAC;CAQtC"}

package/dist/embeddings/ollama.js

@@ -0,0 +1,67 @@
+/**
+ * Ollama Embedding Provider
+ *
+ * Local embeddings using Ollama.
+ * Uses nomic-embed-text model (768 dimensions).
+ */
+/**
+ * Ollama embedding provider
+ */
+export class OllamaEmbeddingProvider {
+    name = 'ollama';
+    dimensions = 768; // nomic-embed-text
+    maxTokens = 8192;
+    baseUrl;
+    model;
+    constructor(baseUrl = 'http://localhost:11434', model = 'nomic-embed-text') {
+        this.baseUrl = baseUrl;
+        this.model = model;
+    }
+    /**
+     * Initialize the provider
+     */
+    async initialize() {
+        const available = await this.isAvailable();
+        if (!available) {
+            throw new Error('Ollama not available. Is it running?');
+        }
+    }
+    /**
+     * Generate embedding for a single text
+     */
+    async embed(text) {
+        const response = await fetch(`${this.baseUrl}/api/embeddings`, {
+            method: 'POST',
+            headers: { 'Content-Type': 'application/json' },
+            body: JSON.stringify({
+                model: this.model,
+                prompt: text,
+            }),
+        });
+        if (!response.ok) {
+            const error = await response.text();
+            throw new Error(`Ollama error: ${response.status} ${error}`);
+        }
+        const data = await response.json();
+        return data.embedding;
+    }
+    /**
+     * Generate embeddings for multiple texts
+     */
+    async embedBatch(texts) {
+        return Promise.all(texts.map(t => this.embed(t)));
+    }
+    /**
+     * Check if the provider is available
+     */
+    async isAvailable() {
+        try {
+            const response = await fetch(`${this.baseUrl}/api/tags`);
+            return response.ok;
+        }
+        catch {
+            return false;
+        }
+    }
+}
+//# sourceMappingURL=ollama.js.map

package/dist/embeddings/ollama.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"ollama.js","sourceRoot":"","sources":["../../src/embeddings/ollama.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAIH;;GAEG;AACH,MAAM,OAAO,uBAAuB;IACzB,IAAI,GAAG,QAAQ,CAAC;IAChB,UAAU,GAAG,GAAG,CAAC,CAAC,mBAAmB;IACrC,SAAS,GAAG,IAAI,CAAC;IAElB,OAAO,CAAS;IAChB,KAAK,CAAS;IAEtB,YAAY,OAAO,GAAG,wBAAwB,EAAE,KAAK,GAAG,kBAAkB;QACxE,IAAI,CAAC,OAAO,GAAG,OAAO,CAAC;QACvB,IAAI,CAAC,KAAK,GAAG,KAAK,CAAC;IACrB,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,UAAU;QACd,MAAM,SAAS,GAAG,MAAM,IAAI,CAAC,WAAW,EAAE,CAAC;QAC3C,IAAI,CAAC,SAAS,EAAE,CAAC;YACf,MAAM,IAAI,KAAK,CAAC,sCAAsC,CAAC,CAAC;QAC1D,CAAC;IACH,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,KAAK,CAAC,IAAY;QACtB,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,GAAG,IAAI,CAAC,OAAO,iBAAiB,EAAE;YAC7D,MAAM,EAAE,MAAM;YACd,OAAO,EAAE,EAAE,cAAc,EAAE,kBAAkB,EAAE;YAC/C,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC;gBACnB,KAAK,EAAE,IAAI,CAAC,KAAK;gBACjB,MAAM,EAAE,IAAI;aACb,CAAC;SACH,CAAC,CAAC;QAEH,IAAI,CAAC,QAAQ,CAAC,EAAE,EAAE,CAAC;YACjB,MAAM,KAAK,GAAG,MAAM,QAAQ,CAAC,IAAI,EAAE,CAAC;YACpC,MAAM,IAAI,KAAK,CAAC,iBAAiB,QAAQ,CAAC,MAAM,IAAI,KAAK,EAAE,CAAC,CAAC;QAC/D,CAAC;QAED,MAAM,IAAI,GAAG,MAAM,QAAQ,CAAC,IAAI,EAA6B,CAAC;QAC9D,OAAO,IAAI,CAAC,SAAS,CAAC;IACxB,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,UAAU,CAAC,KAAe;QAC9B,OAAO,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;IACpD,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,WAAW;QACf,IAAI,CAAC;YACH,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,GAAG,IAAI,CAAC,OAAO,WAAW,CAAC,CAAC;YACzD,OAAO,QAAQ,CAAC,EAAE,CAAC;QACrB,CAAC;QAAC,MAAM,CAAC;YACP,OAAO,KAAK,CAAC;QACf,CAAC;IACH,CAAC;CACF"}

package/dist/embeddings/openai.d.ts

@@ -0,0 +1,36 @@
+/**
+ * OpenAI Embedding Provider
+ *
+ * High-quality embeddings using OpenAI's API.
+ * Uses text-embedding-3-small model (1536 dimensions).
+ */
+import type { IEmbeddingProvider } from './interface.js';
+/**
+ * OpenAI embedding provider
+ */
+export declare class OpenAIEmbeddingProvider implements IEmbeddingProvider {
+    readonly name = "openai";
+    readonly dimensions = 1536;
+    readonly maxTokens = 8191;
+    private apiKey;
+    private model;
+    private baseUrl;
+    constructor(apiKey: string);
+    /**
+     * Initialize the provider
+     */
+    initialize(): Promise<void>;
+    /**
+     * Generate embedding for a single text
+     */
+    embed(text: string): Promise<number[]>;
+    /**
+     * Generate embeddings for multiple texts
+     */
+    embedBatch(texts: string[]): Promise<number[][]>;
+    /**
+     * Check if the provider is available
+     */
+    isAvailable(): Promise<boolean>;
+}
+//# sourceMappingURL=openai.d.ts.map

package/dist/embeddings/openai.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"openai.d.ts","sourceRoot":"","sources":["../../src/embeddings/openai.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,KAAK,EAAE,kBAAkB,EAAE,MAAM,gBAAgB,CAAC;AAEzD;;GAEG;AACH,qBAAa,uBAAwB,YAAW,kBAAkB;IAChE,QAAQ,CAAC,IAAI,YAAY;IACzB,QAAQ,CAAC,UAAU,QAAQ;IAC3B,QAAQ,CAAC,SAAS,QAAQ;IAE1B,OAAO,CAAC,MAAM,CAAS;IACvB,OAAO,CAAC,KAAK,CAA4B;IACzC,OAAO,CAAC,OAAO,CAA+B;gBAElC,MAAM,EAAE,MAAM;IAI1B;;OAEG;IACG,UAAU,IAAI,OAAO,CAAC,IAAI,CAAC;IAQjC;;OAEG;IACG,KAAK,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC;IAyB5C;;OAEG;IACG,UAAU,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC;IAyBtD;;OAEG;IACG,WAAW,IAAI,OAAO,CAAC,OAAO,CAAC;CAatC"}

package/dist/embeddings/openai.js

@@ -0,0 +1,92 @@
+/**
+ * OpenAI Embedding Provider
+ *
+ * High-quality embeddings using OpenAI's API.
+ * Uses text-embedding-3-small model (1536 dimensions).
+ */
+/**
+ * OpenAI embedding provider
+ */
+export class OpenAIEmbeddingProvider {
+    name = 'openai';
+    dimensions = 1536; // text-embedding-3-small
+    maxTokens = 8191;
+    apiKey;
+    model = 'text-embedding-3-small';
+    baseUrl = 'https://api.openai.com/v1';
+    constructor(apiKey) {
+        this.apiKey = apiKey;
+    }
+    /**
+     * Initialize the provider
+     */
+    async initialize() {
+        // Validate API key by making a test request
+        const available = await this.isAvailable();
+        if (!available) {
+            throw new Error('OpenAI API key is invalid or API is unavailable');
+        }
+    }
+    /**
+     * Generate embedding for a single text
+     */
+    async embed(text) {
+        const response = await fetch(`${this.baseUrl}/embeddings`, {
+            method: 'POST',
+            headers: {
+                'Authorization': `Bearer ${this.apiKey}`,
+                'Content-Type': 'application/json',
+            },
+            body: JSON.stringify({
+                model: this.model,
+                input: text,
+            }),
+        });
+        if (!response.ok) {
+            const error = await response.text();
+            throw new Error(`OpenAI API error: ${response.status} ${error}`);
+        }
+        const data = await response.json();
+        return data.data[0]?.embedding ?? [];
+    }
+    /**
+     * Generate embeddings for multiple texts
+     */
+    async embedBatch(texts) {
+        const response = await fetch(`${this.baseUrl}/embeddings`, {
+            method: 'POST',
+            headers: {
+                'Authorization': `Bearer ${this.apiKey}`,
+                'Content-Type': 'application/json',
+            },
+            body: JSON.stringify({
+                model: this.model,
+                input: texts,
+            }),
+        });
+        if (!response.ok) {
+            const error = await response.text();
+            throw new Error(`OpenAI API error: ${response.status} ${error}`);
+        }
+        const data = await response.json();
+        return data.data.map(d => d.embedding);
+    }
+    /**
+     * Check if the provider is available
+     */
+    async isAvailable() {
+        try {
+            // Make a minimal test request
+            const response = await fetch(`${this.baseUrl}/models`, {
+                headers: {
+                    'Authorization': `Bearer ${this.apiKey}`,
+                },
+            });
+            return response.ok;
+        }
+        catch {
+            return false;
+        }
+    }
+}
+//# sourceMappingURL=openai.js.map

package/dist/embeddings/openai.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"openai.js","sourceRoot":"","sources":["../../src/embeddings/openai.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAIH;;GAEG;AACH,MAAM,OAAO,uBAAuB;IACzB,IAAI,GAAG,QAAQ,CAAC;IAChB,UAAU,GAAG,IAAI,CAAC,CAAC,yBAAyB;IAC5C,SAAS,GAAG,IAAI,CAAC;IAElB,MAAM,CAAS;IACf,KAAK,GAAG,wBAAwB,CAAC;IACjC,OAAO,GAAG,2BAA2B,CAAC;IAE9C,YAAY,MAAc;QACxB,IAAI,CAAC,MAAM,GAAG,MAAM,CAAC;IACvB,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,UAAU;QACd,4CAA4C;QAC5C,MAAM,SAAS,GAAG,MAAM,IAAI,CAAC,WAAW,EAAE,CAAC;QAC3C,IAAI,CAAC,SAAS,EAAE,CAAC;YACf,MAAM,IAAI,KAAK,CAAC,iDAAiD,CAAC,CAAC;QACrE,CAAC;IACH,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,KAAK,CAAC,IAAY;QACtB,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,GAAG,IAAI,CAAC,OAAO,aAAa,EAAE;YACzD,MAAM,EAAE,MAAM;YACd,OAAO,EAAE;gBACP,eAAe,EAAE,UAAU,IAAI,CAAC,MAAM,EAAE;gBACxC,cAAc,EAAE,kBAAkB;aACnC;YACD,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC;gBACnB,KAAK,EAAE,IAAI,CAAC,KAAK;gBACjB,KAAK,EAAE,IAAI;aACZ,CAAC;SACH,CAAC,CAAC;QAEH,IAAI,CAAC,QAAQ,CAAC,EAAE,EAAE,CAAC;YACjB,MAAM,KAAK,GAAG,MAAM,QAAQ,CAAC,IAAI,EAAE,CAAC;YACpC,MAAM,IAAI,KAAK,CAAC,qBAAqB,QAAQ,CAAC,MAAM,IAAI,KAAK,EAAE,CAAC,CAAC;QACnE,CAAC;QAED,MAAM,IAAI,GAAG,MAAM,QAAQ,CAAC,IAAI,EAE/B,CAAC;QAEF,OAAO,IAAI,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,SAAS,IAAI,EAAE,CAAC;IACvC,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,UAAU,CAAC,KAAe;QAC9B,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,GAAG,IAAI,CAAC,OAAO,aAAa,EAAE;YACzD,MAAM,EAAE,MAAM;YACd,OAAO,EAAE;gBACP,eAAe,EAAE,UAAU,IAAI,CAAC,MAAM,EAAE;gBACxC,cAAc,EAAE,kBAAkB;aACnC;YACD,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC;gBACnB,KAAK,EAAE,IAAI,CAAC,KAAK;gBACjB,KAAK,EAAE,KAAK;aACb,CAAC;SACH,CAAC,CAAC;QAEH,IAAI,CAAC,QAAQ,CAAC,EAAE,EAAE,CAAC;YACjB,MAAM,KAAK,GAAG,MAAM,QAAQ,CAAC,IAAI,EAAE,CAAC;YACpC,MAAM,IAAI,KAAK,CAAC,qBAAqB,QAAQ,CAAC,MAAM,IAAI,KAAK,EAAE,CAAC,CAAC;QACnE,CAAC;QAED,MAAM,IAAI,GAAG,MAAM,QAAQ,CAAC,IAAI,EAE/B,CAAC;QAEF,OAAO,IAAI,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC;IACzC,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,WAAW;QACf,IAAI,CAAC;YACH,8BAA8B;YAC9B,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,GAAG,IAAI,CAAC,OAAO,SAAS,EAAE;gBACrD,OAAO,EAAE;oBACP,eAAe,EAAE,UAAU,IAAI,CAAC,MAAM,EAAE;iBACzC;aACF,CAAC,CAAC;YACH,OAAO,QAAQ,CAAC,EAAE,CAAC;QACrB,CAAC;QAAC,MAAM,CAAC;YACP,OAAO,KAAK,CAAC;QACf,CAAC;IACH,CAAC;CACF"}