pdfdataextract/dist/tesseractjsocr.js

Extract data from a pdf with pure javascript

"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
exports.TesseractJsOcr = void 0;
//@ts-ignore: ignore import errors because its dynamicly loaded from pdfdataextractor
const tesseract_js_1 = require("tesseract.js");
/**
 * implementation for tesseractjs
 */
class TesseractJsOcr {
    /**
     * recognize characters of buffers
     *
     * @param {Buffer[]} buffers - the image buffers
     * @param {OCRLang[]} langs - the language traineddata used for recognition
     * @returns {Promise<string[]>} an array with text from each side
     */
    async ocrBuffers(buffers, langs) {
        if (buffers.length == 0)
            return [];
        if (buffers.length == 1) {
            const lang = langs.join('+');
            const worker = await (0, tesseract_js_1.createWorker)(lang);
            const data = await worker.recognize(buffers[0]);
            await worker.terminate();
            return [data.data.text];
        }
        const lang = langs.join('+');
        const scheduler = (0, tesseract_js_1.createScheduler)();
        for (let i = 0; i < buffers.length; i++) {
            const worker = await (0, tesseract_js_1.createWorker)(lang);
            scheduler.addWorker(worker);
        }
        const result = await Promise.all(buffers.map(async (buffer) => scheduler.addJob('recognize', buffer)));
        await scheduler.terminate();
        return result.map((r) => r.data.text);
    }
}
exports.TesseractJsOcr = TesseractJsOcr;
//# sourceMappingURL=tesseractjsocr.js.map