llm-extract/dist/processors/pdf.js

Modular SDK for structured text extraction from documents using LLMs

"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
exports.PDFProcessor = void 0;
const pdf = require('pdf-parse');
class PDFProcessor {
    constructor() {
        this.supportedTypes = ['application/pdf'];
    }
    async parseDocument(buffer, options = {}) {
        const startTime = Date.now();
        try {
            const data = await pdf(buffer, {
                max: options.maxPages || 0,
            });
            const extractedText = data.text.trim();
            const finalText = extractedText;
            return {
                extractedText: finalText,
                metadata: {
                    pageCount: data.numpages,
                    hasImages: false,
                    processingInfo: {
                        textLength: finalText.length,
                    }
                },
                processingTimeMs: Date.now() - startTime
            };
        }
        catch (error) {
            throw new Error(`PDF parsing failed: ${error.message}`);
        }
    }
}
exports.PDFProcessor = PDFProcessor;
//# sourceMappingURL=pdf.js.map