bea-edit/src/lib/zip-processing.ts

import { PDFDocument } from 'pdf-lib';

import { parseXJustizMetadata, type XJustizMetadata } from './services/xml-reading.service';
import { unzipArchive } from './services/zip-inflating.service';

export type ZipAttachmentKind = 'pdf' | 'image';

export type ZipAttachment = {
	name: string;
	path: string;
	kind: ZipAttachmentKind;
	data: Uint8Array;
};

export type ProcessedZipArchive = {
	name: string;
	attachments: ZipAttachment[];
};

const IMAGE_EXTENSIONS = new Set([
	'.avif',
	'.bmp',
	'.gif',
	'.heic',
	'.jpeg',
	'.jpg',
	'.jp2',
	'.png',
	'.tif',
	'.tiff',
	'.webp'
]);

const ZIP_META_FILE_NAME = 'xjustiz_nachricht.xml';

const getBaseName = (path: string) => path.split('/').pop() ?? path;

const getCurrentDateString = (date: Date) =>
	`${date.getFullYear()}-${String(date.getMonth() + 1).padStart(2, '0')}-${String(date.getDate()).padStart(2, '0')}`;

const createArchiveName = (date: Date, metadata: XJustizMetadata | null) => {
	const datePrefix = getCurrentDateString(date);

	if (!metadata) {
		return `${datePrefix}_unbekannt_unbekannt.pdf`;
	}

	return `${datePrefix}_${metadata.sender}_${metadata.receiver}.pdf`;
};

const getExtension = (path: string) => {
	const baseName = getBaseName(path).toLowerCase();
	const lastDotIndex = baseName.lastIndexOf('.');

	return lastDotIndex === -1 ? '' : baseName.slice(lastDotIndex);
};

const isPdf = (path: string) => getExtension(path) === '.pdf';

const isImage = (path: string) => IMAGE_EXTENSIONS.has(getExtension(path));

const isZipArchive = (path: string) => getExtension(path) === '.zip';

const normalizeKey = (value: string) => value.replaceAll('\\', '/').toLowerCase().trim();

const isMacMetadataEntry = (path: string) => {
	const normalizedPath = normalizeKey(path);
	const baseName = getBaseName(normalizedPath);

	return (
		normalizedPath.startsWith('__macosx/') ||
		normalizedPath.includes('/__macosx/') ||
		baseName === '.ds_store' ||
		baseName.startsWith('._')
	);
};
const orderAttachmentsByDocumentNames = (attachments: ZipAttachment[], documentNames: string[]) => {
	const buckets = new Map<string, ZipAttachment[]>();

	for (const attachment of attachments) {
		const keys = new Set([normalizeKey(attachment.path), normalizeKey(attachment.name)]);

		for (const key of keys) {
			const bucket = buckets.get(key) ?? [];
			bucket.push(attachment);
			buckets.set(key, bucket);
		}
	}

	const usedAttachments = new Set<ZipAttachment>();
	const orderedAttachments: ZipAttachment[] = [];

	const takeAttachment = (key: string) => {
		const bucket = buckets.get(key);

		if (!bucket) {
			return null;
		}

		const nextAttachment = bucket.find((attachment) => !usedAttachments.has(attachment)) ?? null;

		if (nextAttachment) {
			usedAttachments.add(nextAttachment);
		}

		return nextAttachment;
	};

	for (const documentName of documentNames) {
		const normalizedName = normalizeKey(documentName);
		const normalizedBaseName = normalizeKey(getBaseName(documentName));
		const matchedAttachment = takeAttachment(normalizedName) ?? takeAttachment(normalizedBaseName);

		if (matchedAttachment) {
			orderedAttachments.push(matchedAttachment);
		}
	}

	for (const attachment of attachments) {
		if (!usedAttachments.has(attachment)) {
			orderedAttachments.push(attachment);
		}
	}

	return orderedAttachments;
};

const shouldSkipZipEntry = (path: string) => isMacMetadataEntry(path);

const isMetadataFile = (path: string) => getBaseName(path).toLowerCase() === ZIP_META_FILE_NAME;

const createAttachment = (path: string, data: Uint8Array): ZipAttachment => ({
	name: getBaseName(path),
	path,
	kind: isPdf(path) ? 'pdf' : 'image',
	data
});

const extractZipEntries = async (archiveBytes: Uint8Array): Promise<ProcessedZipArchive[]> => {
	const files = await unzipArchive(archiveBytes);
	const attachments: ZipAttachment[] = [];
	const nestedArchives: ProcessedZipArchive[] = [];
	let xjustizNachrichtXml: Uint8Array | null = null;

	for (const [path, data] of Object.entries(files)) {
		if (shouldSkipZipEntry(path)) {
			continue;
		}

		if (isMetadataFile(path)) {
			xjustizNachrichtXml = data;
			continue;
		}

		if (isZipArchive(path)) {
			const childArchives = await extractZipEntries(data);
			nestedArchives.push(...childArchives);
			continue;
		}

		if (isPdf(path) || isImage(path)) {
			attachments.push(createAttachment(path, data));
		}
	}

	const metadata = xjustizNachrichtXml ? parseXJustizMetadata(xjustizNachrichtXml) : null;
	const documentNames = metadata?.documentNames ?? [];
	const orderedAttachments = orderAttachmentsByDocumentNames(attachments, documentNames);
	const archiveName = createArchiveName(new Date(), metadata);
	const archives: ProcessedZipArchive[] = [];

	if (orderedAttachments.length > 0) {
		archives.push({
			name: archiveName,
			attachments: orderedAttachments
		});
	}

	return [...archives, ...nestedArchives];
};

export const extractZipArchives = async (file: File): Promise<ProcessedZipArchive[]> => {
	const archiveBytes = new Uint8Array(await file.arrayBuffer());

	return extractZipEntries(archiveBytes);
};

const loadImageElement = (src: string) =>
	new Promise<HTMLImageElement>((resolve, reject) => {
		const image = new Image();
		image.decoding = 'async';
		image.onload = () => resolve(image);
		image.onerror = () => reject(new Error(`Unable to decode image asset: ${src}`));
		image.src = src;
	});

const toArrayBuffer = (bytes: Uint8Array) =>
	(() => {
		const copy = new Uint8Array(bytes.byteLength);
		copy.set(bytes);
		return copy.buffer;
	})();

const convertImageBytesToPng = async (imageBytes: Uint8Array) => {
	if (typeof document === 'undefined') {
		throw new Error('Image conversion requires a browser environment.');
	}

	const objectUrl = URL.createObjectURL(new Blob([toArrayBuffer(imageBytes)]));

	try {
		const image = await loadImageElement(objectUrl);
		const canvas = document.createElement('canvas');
		canvas.width = image.naturalWidth || image.width;
		canvas.height = image.naturalHeight || image.height;

		const context = canvas.getContext('2d');

		if (!context) {
			throw new Error('Could not create a canvas context for image conversion.');
		}

		context.drawImage(image, 0, 0);

		const pngBlob = await new Promise<Blob>((resolve, reject) => {
			canvas.toBlob((blob) => {
				if (blob) {
					resolve(blob);
					return;
				}

				reject(new Error('Could not encode image asset as PNG.'));
			}, 'image/png');
		});

		return new Uint8Array(await pngBlob.arrayBuffer());
	} finally {
		URL.revokeObjectURL(objectUrl);
	}
};

const embedAttachmentImage = async (pdfDocument: PDFDocument, attachment: ZipAttachment) => {
	const extension = getExtension(attachment.path);
	const imageData = attachment.data;

	if (extension === '.jpg' || extension === '.jpeg') {
		return pdfDocument.embedJpg(imageData);
	}

	if (extension === '.png') {
		return pdfDocument.embedPng(imageData);
	}

	const pngBytes = await convertImageBytesToPng(imageData);

	return pdfDocument.embedPng(pngBytes);
};

const mergeAttachmentIntoPdf = async (mergedPdf: PDFDocument, attachment: ZipAttachment) => {
	if (attachment.kind === 'pdf') {
		const sourcePdf = await PDFDocument.load(attachment.data, { ignoreEncryption: true });
		const copiedPages = await mergedPdf.copyPages(sourcePdf, sourcePdf.getPageIndices());

		for (const page of copiedPages) {
			mergedPdf.addPage(page);
		}

		return;
	}

	try {
		const embeddedImage = await embedAttachmentImage(mergedPdf, attachment);
		const page = mergedPdf.addPage([embeddedImage.width, embeddedImage.height]);

		page.drawImage(embeddedImage, {
			x: 0,
			y: 0,
			width: page.getWidth(),
			height: page.getHeight()
		});
	} catch (error) {
		throw error;
	}
};

export const mergeProcessedZipArchive = async (archive: ProcessedZipArchive) => {
	if (archive.attachments.length === 0) {
		throw new Error('Cannot create a PDF from an empty archive.');
	}

	const mergedPdf = await PDFDocument.create();

	for (const attachment of archive.attachments) {
		await mergeAttachmentIntoPdf(mergedPdf, attachment);
	}

	return mergedPdf.save();
};