83 lines
2.9 KiB
TypeScript
83 lines
2.9 KiB
TypeScript
|
// src/lib/metadata-extractor.ts
|
||
|
|
// 1. Update your import to use the new named export
|
||
|
|
//import * as PDFLib from 'pdf-parse';
|
||
|
|
//import { PDFParse } from 'pdf-parse';
|
||
|
|
//import {pdf} from 'pdf-parse';
|
||
|
|
import * as PdfParse from 'pdf-parse-new';
|
||
|
|
import sharp from 'sharp';
|
||
|
|
import exifReader from 'exif-reader';
|
||
|
|
|
||
|
|
/**
|
||
|
|
* Converts EXIF DMS (Degrees, Minutes, Seconds) array to Decimal Degrees.
|
||
|
|
*/
|
||
|
|
function convertDMSToDD(dms: any, ref: string): string {
|
||
|
|
if (!Array.isArray(dms) || dms.length < 3) return String(dms);
|
||
|
|
|
||
|
|
const [degrees, minutes, seconds] = dms;
|
||
|
|
let dd = degrees + (minutes / 60) + (seconds / 3600);
|
||
|
|
|
||
|
|
if (ref === 'S' || ref === 'W') {
|
||
|
|
dd = dd * -1;
|
||
|
|
}
|
||
|
|
return dd.toFixed(6);
|
||
|
|
}
|
||
|
|
|
||
|
|
/**
|
||
|
|
* RECURSIVE SANITIZER:
|
||
|
|
* Converts Buffers to strings, standardizes keys, and handles GPS conversion.
|
||
|
|
*/
|
||
|
|
function sanitizeMetadata(obj: any): any {
|
||
|
|
if (obj === null || typeof obj !== 'object') return obj;
|
||
|
|
if (obj instanceof Date) return obj.toISOString();
|
||
|
|
if (Buffer.isBuffer(obj)) return `[Binary Data: ${obj.length} bytes]`;
|
||
|
|
if (Array.isArray(obj)) return obj.map(sanitizeMetadata);
|
||
|
|
|
||
|
|
const sanitized: any = {};
|
||
|
|
for (const [key, value] of Object.entries(obj)) {
|
||
|
|
const cleanKey = key.charAt(0).toLowerCase() + key.slice(1);
|
||
|
|
|
||
|
|
// Specifically handle GPS Latitude/Longitude Arrays
|
||
|
|
if (cleanKey === 'gPSLatitude' && obj['gPSLatitudeRef']) {
|
||
|
|
sanitized[cleanKey] = convertDMSToDD(value, obj['gPSLatitudeRef']);
|
||
|
|
continue;
|
||
|
|
}
|
||
|
|
if (cleanKey === 'gPSLongitude' && obj['gPSLongitudeRef']) {
|
||
|
|
sanitized[cleanKey] = convertDMSToDD(value, obj['gPSLongitudeRef']);
|
||
|
|
continue;
|
||
|
|
}
|
||
|
|
|
||
|
|
sanitized[cleanKey] = sanitizeMetadata(value);
|
||
|
|
}
|
||
|
|
return sanitized;
|
||
|
|
}
|
||
|
|
|
||
|
|
export async function extractMetadata(buffer: Buffer, filename: string): Promise<any> {
|
||
|
|
const extension = filename.split('.').pop()?.toLowerCase();
|
||
|
|
|
||
|
|
if (extension === 'pdf') {
|
||
|
|
// Create parser instance
|
||
|
|
const parser = new PdfParse.SmartPDFParser({
|
||
|
|
oversaturationFactor: 2.0,
|
||
|
|
enableFastPath: true
|
||
|
|
});
|
||
|
|
console.log(`--- PDF Debug Start: ${filename} ---`);
|
||
|
|
const result = await parser.parse(buffer);
|
||
|
|
console.log(`Parsed ${result.numpages} pages using ${result._meta.method}`);
|
||
|
|
console.log(`Parsed ${result.info} info using ${result._meta.method}`);
|
||
|
|
console.log(`Parsed ${result.info} info using ${result._meta.method}`);
|
||
|
|
console.log(JSON.stringify(result.info, null, 2));
|
||
|
|
//const { text, numpages, info } = await pdf(buffer);
|
||
|
|
//const parser = new PDFParse(buffer);
|
||
|
|
// `text` → full document text
|
||
|
|
// `numpages` → page count
|
||
|
|
// `info` → metadata (author, creation date, etc.)
|
||
|
|
|
||
|
|
// console.log(`Pages: ${numpages}`);
|
||
|
|
// console.log(`Author: ${info.Author}`);
|
||
|
|
// console.log(text.slice(0, 200)); // preview first 200 chars
|
||
|
|
return {
|
||
|
|
type: 'PDF',
|
||
|
|
title: filename
|
||
|
|
};
|
||
|
|
}
|
||
|
|
}
|