glin-profanity 3.1.5 → 3.2.2

package/dist/multimodal/ocr.d.cts

@@ -0,0 +1,157 @@
+import { F as Filter } from '../Filter-D34Wsmrj.cjs';
+import { L as Language, F as FilterConfig, C as CheckProfanityResult } from '../types-B9c_ik4k.cjs';
+
+/**
+ * OCR (Optical Character Recognition) Integration for glin-profanity
+ *
+ * Extracts text from images and checks for profanity.
+ * Uses Tesseract.js as an optional peer dependency.
+ *
+ * @example
+ * ```typescript
+ * import { createOCRChecker, checkImageForProfanity } from 'glin-profanity/ocr';
+ *
+ * // Quick check
+ * const result = await checkImageForProfanity(imageBuffer);
+ * console.log(result.containsProfanity);
+ *
+ * // With custom config
+ * const checker = createOCRChecker({
+ *   languages: ['english', 'spanish'],
+ *   tesseractLangs: ['eng', 'spa'],
+ *   detectLeetspeak: true,
+ * });
+ * const result = await checker.checkImage(imageBuffer);
+ * ```
+ *
+ * @packageDocumentation
+ * @module glin-profanity/ocr
+ */
+
+type ImageLike = string | Buffer | Uint8Array | Blob | File | HTMLImageElement | HTMLCanvasElement;
+/**
+ * OCR checker configuration
+ */
+interface OCRCheckerConfig {
+    /** Languages for profanity detection */
+    languages?: Language[];
+    /** Tesseract language codes (e.g., 'eng', 'spa', 'fra') */
+    tesseractLangs?: string[];
+    /** Enable leetspeak detection */
+    detectLeetspeak?: boolean;
+    /** Enable Unicode normalization */
+    normalizeUnicode?: boolean;
+    /** Minimum OCR confidence to process text (0-100) */
+    minConfidence?: number;
+    /** Custom filter configuration */
+    filterConfig?: Partial<FilterConfig>;
+}
+/**
+ * OCR check result
+ */
+interface OCRCheckResult {
+    /** Whether profanity was found in the extracted text */
+    containsProfanity: boolean;
+    /** Extracted text from the image */
+    extractedText: string;
+    /** OCR confidence score (0-100) */
+    ocrConfidence: number;
+    /** Profane words found */
+    profaneWords: string[];
+    /** Full profanity check result */
+    profanityResult: CheckProfanityResult;
+    /** Processing time in milliseconds */
+    processingTimeMs: number;
+}
+/**
+ * Creates an OCR profanity checker
+ *
+ * @example
+ * ```typescript
+ * const checker = createOCRChecker({
+ *   languages: ['english'],
+ *   tesseractLangs: ['eng'],
+ *   detectLeetspeak: true,
+ * });
+ *
+ * // Check an image
+ * const result = await checker.checkImage('./screenshot.png');
+ *
+ * // Check multiple images
+ * const results = await checker.checkImages([image1, image2, image3]);
+ *
+ * // Clean up when done
+ * await checker.terminate();
+ * ```
+ */
+declare function createOCRChecker(config?: OCRCheckerConfig): {
+    /**
+     * Check a single image for profanity
+     */
+    checkImage(image: ImageLike): Promise<OCRCheckResult>;
+    /**
+     * Check multiple images for profanity
+     */
+    checkImages(images: ImageLike[]): Promise<OCRCheckResult[]>;
+    /**
+     * Extract text from image without profanity check
+     */
+    extractText(image: ImageLike): Promise<{
+        text: string;
+        confidence: number;
+    }>;
+    /**
+     * Check extracted text (if you already have text from another OCR)
+     */
+    checkText(text: string): CheckProfanityResult;
+    /**
+     * Terminate the Tesseract worker (clean up resources)
+     */
+    terminate(): Promise<void>;
+    /**
+     * Get the underlying filter instance
+     */
+    getFilter(): Filter;
+};
+/**
+ * Quick function to check an image for profanity
+ * Creates a temporary worker, checks the image, and terminates
+ *
+ * @example
+ * ```typescript
+ * import { checkImageForProfanity } from 'glin-profanity/ocr';
+ *
+ * const result = await checkImageForProfanity('./meme.png');
+ * if (result.containsProfanity) {
+ *   console.log('Found profanity:', result.profaneWords);
+ * }
+ * ```
+ */
+declare function checkImageForProfanity(image: ImageLike, config?: OCRCheckerConfig): Promise<OCRCheckResult>;
+/**
+ * Batch check multiple images for profanity
+ * More efficient than calling checkImageForProfanity multiple times
+ *
+ * @example
+ * ```typescript
+ * import { batchCheckImages } from 'glin-profanity/ocr';
+ *
+ * const images = ['./img1.png', './img2.png', './img3.png'];
+ * const results = await batchCheckImages(images);
+ *
+ * const flagged = results.filter(r => r.containsProfanity);
+ * console.log(`${flagged.length} images contain profanity`);
+ * ```
+ */
+declare function batchCheckImages(images: ImageLike[], config?: OCRCheckerConfig): Promise<OCRCheckResult[]>;
+/**
+ * Supported Tesseract language codes
+ * Map from glin-profanity language to Tesseract code
+ */
+declare const LANGUAGE_TO_TESSERACT: Record<Language, string>;
+/**
+ * Helper to convert glin-profanity languages to Tesseract codes
+ */
+declare function languagesToTesseract(languages: Language[]): string[];
+
+export { CheckProfanityResult, FilterConfig, LANGUAGE_TO_TESSERACT, Language, type OCRCheckResult, type OCRCheckerConfig, batchCheckImages, checkImageForProfanity, createOCRChecker, languagesToTesseract };

package/dist/multimodal/ocr.d.ts

@@ -0,0 +1,157 @@
+import { F as Filter } from '../Filter-BGcyIAvO.js';
+import { L as Language, F as FilterConfig, C as CheckProfanityResult } from '../types-B9c_ik4k.js';
+
+/**
+ * OCR (Optical Character Recognition) Integration for glin-profanity
+ *
+ * Extracts text from images and checks for profanity.
+ * Uses Tesseract.js as an optional peer dependency.
+ *
+ * @example
+ * ```typescript
+ * import { createOCRChecker, checkImageForProfanity } from 'glin-profanity/ocr';
+ *
+ * // Quick check
+ * const result = await checkImageForProfanity(imageBuffer);
+ * console.log(result.containsProfanity);
+ *
+ * // With custom config
+ * const checker = createOCRChecker({
+ *   languages: ['english', 'spanish'],
+ *   tesseractLangs: ['eng', 'spa'],
+ *   detectLeetspeak: true,
+ * });
+ * const result = await checker.checkImage(imageBuffer);
+ * ```
+ *
+ * @packageDocumentation
+ * @module glin-profanity/ocr
+ */
+
+type ImageLike = string | Buffer | Uint8Array | Blob | File | HTMLImageElement | HTMLCanvasElement;
+/**
+ * OCR checker configuration
+ */
+interface OCRCheckerConfig {
+    /** Languages for profanity detection */
+    languages?: Language[];
+    /** Tesseract language codes (e.g., 'eng', 'spa', 'fra') */
+    tesseractLangs?: string[];
+    /** Enable leetspeak detection */
+    detectLeetspeak?: boolean;
+    /** Enable Unicode normalization */
+    normalizeUnicode?: boolean;
+    /** Minimum OCR confidence to process text (0-100) */
+    minConfidence?: number;
+    /** Custom filter configuration */
+    filterConfig?: Partial<FilterConfig>;
+}
+/**
+ * OCR check result
+ */
+interface OCRCheckResult {
+    /** Whether profanity was found in the extracted text */
+    containsProfanity: boolean;
+    /** Extracted text from the image */
+    extractedText: string;
+    /** OCR confidence score (0-100) */
+    ocrConfidence: number;
+    /** Profane words found */
+    profaneWords: string[];
+    /** Full profanity check result */
+    profanityResult: CheckProfanityResult;
+    /** Processing time in milliseconds */
+    processingTimeMs: number;
+}
+/**
+ * Creates an OCR profanity checker
+ *
+ * @example
+ * ```typescript
+ * const checker = createOCRChecker({
+ *   languages: ['english'],
+ *   tesseractLangs: ['eng'],
+ *   detectLeetspeak: true,
+ * });
+ *
+ * // Check an image
+ * const result = await checker.checkImage('./screenshot.png');
+ *
+ * // Check multiple images
+ * const results = await checker.checkImages([image1, image2, image3]);
+ *
+ * // Clean up when done
+ * await checker.terminate();
+ * ```
+ */
+declare function createOCRChecker(config?: OCRCheckerConfig): {
+    /**
+     * Check a single image for profanity
+     */
+    checkImage(image: ImageLike): Promise<OCRCheckResult>;
+    /**
+     * Check multiple images for profanity
+     */
+    checkImages(images: ImageLike[]): Promise<OCRCheckResult[]>;
+    /**
+     * Extract text from image without profanity check
+     */
+    extractText(image: ImageLike): Promise<{
+        text: string;
+        confidence: number;
+    }>;
+    /**
+     * Check extracted text (if you already have text from another OCR)
+     */
+    checkText(text: string): CheckProfanityResult;
+    /**
+     * Terminate the Tesseract worker (clean up resources)
+     */
+    terminate(): Promise<void>;
+    /**
+     * Get the underlying filter instance
+     */
+    getFilter(): Filter;
+};
+/**
+ * Quick function to check an image for profanity
+ * Creates a temporary worker, checks the image, and terminates
+ *
+ * @example
+ * ```typescript
+ * import { checkImageForProfanity } from 'glin-profanity/ocr';
+ *
+ * const result = await checkImageForProfanity('./meme.png');
+ * if (result.containsProfanity) {
+ *   console.log('Found profanity:', result.profaneWords);
+ * }
+ * ```
+ */
+declare function checkImageForProfanity(image: ImageLike, config?: OCRCheckerConfig): Promise<OCRCheckResult>;
+/**
+ * Batch check multiple images for profanity
+ * More efficient than calling checkImageForProfanity multiple times
+ *
+ * @example
+ * ```typescript
+ * import { batchCheckImages } from 'glin-profanity/ocr';
+ *
+ * const images = ['./img1.png', './img2.png', './img3.png'];
+ * const results = await batchCheckImages(images);
+ *
+ * const flagged = results.filter(r => r.containsProfanity);
+ * console.log(`${flagged.length} images contain profanity`);
+ * ```
+ */
+declare function batchCheckImages(images: ImageLike[], config?: OCRCheckerConfig): Promise<OCRCheckResult[]>;
+/**
+ * Supported Tesseract language codes
+ * Map from glin-profanity language to Tesseract code
+ */
+declare const LANGUAGE_TO_TESSERACT: Record<Language, string>;
+/**
+ * Helper to convert glin-profanity languages to Tesseract codes
+ */
+declare function languagesToTesseract(languages: Language[]): string[];
+
+export { CheckProfanityResult, FilterConfig, LANGUAGE_TO_TESSERACT, Language, type OCRCheckResult, type OCRCheckerConfig, batchCheckImages, checkImageForProfanity, createOCRChecker, languagesToTesseract };