feat(api,web): support CBR — extraction RAR, scan et lecture d'albums
- api: cbr utilitaire (extraction RAR), scanner/métadonnées (+ tests), books, schéma - web: ReaderPage/locators (+ tests), mockData, types partagés - deps: pnpm-lock Refs: #17
This commit is contained in:
@ -28,12 +28,12 @@ export class BooksController {
|
||||
|
||||
@Get(":id/pages")
|
||||
pages(@Param("id") id: string) {
|
||||
return this.books.listCbzPages(Number(id));
|
||||
return this.books.listComicPages(Number(id));
|
||||
}
|
||||
|
||||
@Get(":id/pages/:page")
|
||||
page(@Param("id") id: string, @Param("page") page: string, @Res() reply: FastifyReply) {
|
||||
const result = this.books.readCbzPage(Number(id), Number(page));
|
||||
async page(@Param("id") id: string, @Param("page") page: string, @Res() reply: FastifyReply) {
|
||||
const result = await this.books.readComicPage(Number(id), Number(page));
|
||||
reply.header("Content-Type", result.contentType);
|
||||
reply.header("Cache-Control", "private, max-age=3600");
|
||||
return reply.send(result.data);
|
||||
|
||||
@ -3,6 +3,7 @@ import { createReadStream, existsSync } from "node:fs";
|
||||
import { extname } from "node:path";
|
||||
import { and, eq, sql } from "drizzle-orm";
|
||||
import { BookQueryDto } from "@readabook/shared";
|
||||
import { listCbrImageEntries, readCbrPage } from "../common/cbr.js";
|
||||
import { listCbzImageEntries, readCbzPage } from "../common/cbz.js";
|
||||
import { DatabaseService } from "../database/database.service.js";
|
||||
import { books } from "../database/schema.js";
|
||||
@ -68,10 +69,10 @@ export class BooksService {
|
||||
return { book, stream: createReadStream(book.coverPath), coverPath: book.coverPath };
|
||||
}
|
||||
|
||||
listCbzPages(id: number) {
|
||||
async listComicPages(id: number) {
|
||||
const book = this.get(id);
|
||||
this.assertCbzBook(book);
|
||||
const pages = listCbzImageEntries(book.filePath);
|
||||
this.assertComicArchiveBook(book);
|
||||
const pages = book.format === "cbr" ? await listCbrImageEntries(book.filePath) : listCbzImageEntries(book.filePath);
|
||||
return {
|
||||
bookId: book.id,
|
||||
pageCount: pages.length,
|
||||
@ -79,14 +80,17 @@ export class BooksService {
|
||||
};
|
||||
}
|
||||
|
||||
readCbzPage(id: number, page: number) {
|
||||
async readComicPage(id: number, page: number) {
|
||||
const book = this.get(id);
|
||||
this.assertCbzBook(book);
|
||||
this.assertComicArchiveBook(book);
|
||||
try {
|
||||
const result = readCbzPage(book.filePath, page);
|
||||
const result =
|
||||
book.format === "cbr"
|
||||
? await readCbrPage(book.filePath, page, this.database.config.storageDir)
|
||||
: readCbzPage(book.filePath, page);
|
||||
return { book, page, contentType: lookupMime(result.entryName), data: result.data };
|
||||
} catch (error) {
|
||||
throw new NotFoundException(error instanceof Error ? error.message : "CBZ page not found");
|
||||
throw new NotFoundException(error instanceof Error ? error.message : "Comic page not found");
|
||||
}
|
||||
}
|
||||
|
||||
@ -94,9 +98,9 @@ export class BooksService {
|
||||
return this.database.db.select({ count: sql<number>`count(*)` }).from(books).get()?.count ?? 0;
|
||||
}
|
||||
|
||||
private assertCbzBook(book: typeof books.$inferSelect): void {
|
||||
if (book.format !== "cbz") {
|
||||
throw new BadRequestException("Book is not a CBZ archive");
|
||||
private assertComicArchiveBook(book: typeof books.$inferSelect): void {
|
||||
if (book.format !== "cbz" && book.format !== "cbr") {
|
||||
throw new BadRequestException("Book is not a comic archive");
|
||||
}
|
||||
if (!existsSync(book.filePath)) {
|
||||
throw new NotFoundException("Book file not found on disk");
|
||||
|
||||
63
apps/api/src/common/cbr.ts
Normal file
63
apps/api/src/common/cbr.ts
Normal file
@ -0,0 +1,63 @@
|
||||
import { mkdtempSync, readFileSync, rmSync } from "node:fs";
|
||||
import { extname, join } from "node:path";
|
||||
import { createExtractorFromFile } from "node-unrar-js";
|
||||
import { COMIC_IMAGE_EXTENSIONS, MAX_COMIC_ARCHIVE_ENTRIES } from "./cbz.js";
|
||||
import type { CbzPageEntry } from "./cbz.js";
|
||||
|
||||
export async function listCbrImageEntries(filePath: string): Promise<CbzPageEntry[]> {
|
||||
const extractor = await createExtractorFromFile({ filepath: filePath });
|
||||
const list = extractor.getFileList();
|
||||
if (list.arcHeader.flags.volume) {
|
||||
throw new Error("Multi-volume CBR archives are not supported");
|
||||
}
|
||||
if (list.arcHeader.flags.headerEncrypted) {
|
||||
throw new Error("Encrypted CBR archives are not supported");
|
||||
}
|
||||
|
||||
const headers = [...list.fileHeaders];
|
||||
if (headers.length > MAX_COMIC_ARCHIVE_ENTRIES) {
|
||||
throw new Error("CBR archive has too many entries");
|
||||
}
|
||||
|
||||
const images = headers
|
||||
.filter((header) => !header.flags.directory && !header.flags.encrypted)
|
||||
.filter((header) => COMIC_IMAGE_EXTENSIONS.has(extname(header.name).toLowerCase()))
|
||||
.map((header) => ({ entryName: header.name, name: header.name.split(/[\\/]/).pop() ?? header.name }))
|
||||
.sort((a, b) => a.entryName.localeCompare(b.entryName, undefined, { numeric: true, sensitivity: "base" }));
|
||||
|
||||
if (!images.length) {
|
||||
throw new Error("CBR archive does not contain readable image pages");
|
||||
}
|
||||
|
||||
return images;
|
||||
}
|
||||
|
||||
export async function readCbrPage(
|
||||
filePath: string,
|
||||
pageNumber: number,
|
||||
storageDir: string
|
||||
): Promise<{ entryName: string; data: Buffer }> {
|
||||
if (!Number.isInteger(pageNumber) || pageNumber < 1) {
|
||||
throw new Error("CBR page number must be a positive integer");
|
||||
}
|
||||
const pages = await listCbrImageEntries(filePath);
|
||||
const page = pages[pageNumber - 1];
|
||||
if (!page) {
|
||||
throw new Error("CBR page not found");
|
||||
}
|
||||
|
||||
const tempDir = mkdtempSync(join(storageDir, "cbr-page-"));
|
||||
const safeName = `page${extname(page.entryName).toLowerCase() || ".jpg"}`;
|
||||
try {
|
||||
const extractor = await createExtractorFromFile({
|
||||
filepath: filePath,
|
||||
targetPath: tempDir,
|
||||
filenameTransform: () => safeName
|
||||
});
|
||||
const extracted = extractor.extract({ files: [page.entryName] });
|
||||
[...extracted.files];
|
||||
return { entryName: page.entryName, data: readFileSync(join(tempDir, safeName)) };
|
||||
} finally {
|
||||
rmSync(tempDir, { recursive: true, force: true });
|
||||
}
|
||||
}
|
||||
@ -6,18 +6,18 @@ export type CbzPageEntry = {
|
||||
name: string;
|
||||
};
|
||||
|
||||
const IMAGE_EXTENSIONS = new Set([".jpg", ".jpeg", ".png", ".webp", ".gif", ".avif"]);
|
||||
const MAX_CBZ_ENTRIES = 20000;
|
||||
export const COMIC_IMAGE_EXTENSIONS = new Set([".jpg", ".jpeg", ".png", ".webp", ".gif", ".avif"]);
|
||||
export const MAX_COMIC_ARCHIVE_ENTRIES = 20000;
|
||||
|
||||
export function listCbzImageEntries(filePath: string): CbzPageEntry[] {
|
||||
const zip = new AdmZip(filePath);
|
||||
const entries = zip.getEntries();
|
||||
if (entries.length > MAX_CBZ_ENTRIES) {
|
||||
if (entries.length > MAX_COMIC_ARCHIVE_ENTRIES) {
|
||||
throw new Error("CBZ archive has too many entries");
|
||||
}
|
||||
|
||||
const images = entries
|
||||
.filter((entry) => !entry.isDirectory && IMAGE_EXTENSIONS.has(extname(entry.entryName).toLowerCase()))
|
||||
.filter((entry) => !entry.isDirectory && COMIC_IMAGE_EXTENSIONS.has(extname(entry.entryName).toLowerCase()))
|
||||
.map((entry) => ({ entryName: entry.entryName, name: entry.name }))
|
||||
.sort((a, b) => a.entryName.localeCompare(b.entryName, undefined, { numeric: true, sensitivity: "base" }));
|
||||
|
||||
|
||||
@ -59,7 +59,7 @@ export class DatabaseService implements OnModuleDestroy {
|
||||
language TEXT,
|
||||
publisher TEXT,
|
||||
published_date TEXT,
|
||||
format TEXT NOT NULL CHECK (format IN ('epub','pdf','cbz')),
|
||||
format TEXT NOT NULL CHECK (format IN ('epub','pdf','cbz','cbr')),
|
||||
file_path TEXT NOT NULL UNIQUE,
|
||||
cover_path TEXT,
|
||||
file_size INTEGER NOT NULL,
|
||||
@ -119,15 +119,15 @@ export class DatabaseService implements OnModuleDestroy {
|
||||
VALUES (new.id, new.title, new.author, new.description, new.isbn);
|
||||
END;
|
||||
`);
|
||||
this.ensureBooksSupportsCbz();
|
||||
this.ensureBooksSupportsComicArchives();
|
||||
this.sqlite.exec("INSERT INTO book_fts(book_fts) VALUES('rebuild')");
|
||||
}
|
||||
|
||||
private ensureBooksSupportsCbz(): void {
|
||||
private ensureBooksSupportsComicArchives(): void {
|
||||
const table = this.sqlite
|
||||
.prepare("SELECT sql FROM sqlite_master WHERE type = 'table' AND name = 'books'")
|
||||
.get() as { sql?: string } | undefined;
|
||||
if (!table?.sql || table.sql.includes("'cbz'")) return;
|
||||
if (!table?.sql || (table.sql.includes("'cbz'") && table.sql.includes("'cbr'"))) return;
|
||||
|
||||
this.sqlite.exec(`
|
||||
PRAGMA foreign_keys = OFF;
|
||||
@ -149,7 +149,7 @@ export class DatabaseService implements OnModuleDestroy {
|
||||
language TEXT,
|
||||
publisher TEXT,
|
||||
published_date TEXT,
|
||||
format TEXT NOT NULL CHECK (format IN ('epub','pdf','cbz')),
|
||||
format TEXT NOT NULL CHECK (format IN ('epub','pdf','cbz','cbr')),
|
||||
file_path TEXT NOT NULL UNIQUE,
|
||||
cover_path TEXT,
|
||||
file_size INTEGER NOT NULL,
|
||||
|
||||
@ -37,7 +37,7 @@ export const books = sqliteTable(
|
||||
language: text("language"),
|
||||
publisher: text("publisher"),
|
||||
publishedDate: text("published_date"),
|
||||
format: text("format", { enum: ["epub", "pdf", "cbz"] }).notNull(),
|
||||
format: text("format", { enum: ["epub", "pdf", "cbz", "cbr"] }).notNull(),
|
||||
filePath: text("file_path").notNull(),
|
||||
coverPath: text("cover_path"),
|
||||
fileSize: integer("file_size").notNull(),
|
||||
|
||||
@ -7,12 +7,12 @@ import { listCbzImageEntries } from "../common/cbz.js";
|
||||
import { extractMetadata } from "./metadata.js";
|
||||
|
||||
describe("pdf metadata extraction", () => {
|
||||
it("falls back to file name and reads simple PDF info fields", () => {
|
||||
it("falls back to file name and reads simple PDF info fields", async () => {
|
||||
const dir = mkdtempSync(join(tmpdir(), "readabook-"));
|
||||
const file = join(dir, "Example.pdf");
|
||||
writeFileSync(file, "%PDF-1.4\n1 0 obj << /Title (My Book) /Author (Ada) >> endobj");
|
||||
|
||||
const metadata = extractMetadata(file, dir);
|
||||
const metadata = await extractMetadata(file, dir);
|
||||
|
||||
expect(metadata.title).toBe("My Book");
|
||||
expect(metadata.author).toBe("Ada");
|
||||
@ -20,7 +20,7 @@ describe("pdf metadata extraction", () => {
|
||||
});
|
||||
|
||||
describe("cbz metadata extraction", () => {
|
||||
it("uses the file name as title and first image as cover", () => {
|
||||
it("uses the file name as title and first image as cover", async () => {
|
||||
const dir = mkdtempSync(join(tmpdir(), "readabook-"));
|
||||
const file = join(dir, "Comic One.cbz");
|
||||
const zip = new AdmZip();
|
||||
@ -28,7 +28,7 @@ describe("cbz metadata extraction", () => {
|
||||
zip.addFile("001.jpg", Buffer.from([0xff, 0xd8, 0xff, 0xd9]));
|
||||
zip.writeZip(file);
|
||||
|
||||
const metadata = extractMetadata(file, dir);
|
||||
const metadata = await extractMetadata(file, dir);
|
||||
const pages = listCbzImageEntries(file);
|
||||
|
||||
expect(metadata.title).toBe("Comic One");
|
||||
|
||||
@ -3,6 +3,7 @@ import { mkdirSync, readFileSync, writeFileSync } from "node:fs";
|
||||
import { basename, dirname, extname, join } from "node:path";
|
||||
import AdmZip from "adm-zip";
|
||||
import { XMLParser } from "fast-xml-parser";
|
||||
import { listCbrImageEntries, readCbrPage } from "../common/cbr.js";
|
||||
import { listCbzImageEntries } from "../common/cbz.js";
|
||||
|
||||
export type BookMetadata = {
|
||||
@ -22,7 +23,7 @@ const xmlParser = new XMLParser({
|
||||
textNodeName: "#text"
|
||||
});
|
||||
|
||||
export function extractMetadata(filePath: string, storageDir: string): BookMetadata {
|
||||
export async function extractMetadata(filePath: string, storageDir: string): Promise<BookMetadata> {
|
||||
const extension = extname(filePath).toLowerCase();
|
||||
if (extension === ".epub") {
|
||||
return extractEpubMetadata(filePath, storageDir);
|
||||
@ -30,6 +31,9 @@ export function extractMetadata(filePath: string, storageDir: string): BookMetad
|
||||
if (extension === ".cbz") {
|
||||
return extractCbzMetadata(filePath, storageDir);
|
||||
}
|
||||
if (extension === ".cbr") {
|
||||
return extractCbrMetadata(filePath, storageDir);
|
||||
}
|
||||
return extractPdfMetadata(filePath);
|
||||
}
|
||||
|
||||
@ -92,6 +96,20 @@ function extractCbzMetadata(filePath: string, storageDir: string): BookMetadata
|
||||
};
|
||||
}
|
||||
|
||||
async function extractCbrMetadata(filePath: string, storageDir: string): Promise<BookMetadata> {
|
||||
const firstPage = (await listCbrImageEntries(filePath))[0];
|
||||
const page = await readCbrPage(filePath, 1, storageDir);
|
||||
const extension = extname(firstPage.entryName) || ".jpg";
|
||||
const hash = createHash("sha256").update(filePath).digest("hex").slice(0, 24);
|
||||
const target = join(storageDir, "covers", `${hash}${extension}`);
|
||||
mkdirSync(dirname(target), { recursive: true });
|
||||
writeFileSync(target, page.data);
|
||||
return {
|
||||
...fallbackMetadata(filePath),
|
||||
coverPath: target
|
||||
};
|
||||
}
|
||||
|
||||
function fallbackMetadata(filePath: string): BookMetadata {
|
||||
return {
|
||||
title: basename(filePath, extname(filePath)),
|
||||
|
||||
@ -40,7 +40,7 @@ export class ScannerService {
|
||||
|
||||
private async ingestFile(libraryId: number, filePath: string): Promise<void> {
|
||||
const stats = statSync(filePath);
|
||||
let metadata = extractMetadata(filePath, this.database.config.storageDir);
|
||||
let metadata = await extractMetadata(filePath, this.database.config.storageDir);
|
||||
if (this.database.config.openLibraryEnabled) {
|
||||
try {
|
||||
metadata = { ...metadata, ...(await this.openLibrary.enrich(metadata)) };
|
||||
@ -84,15 +84,16 @@ function* walkBooks(root: string): Generator<string> {
|
||||
}
|
||||
if (!entry.isFile()) continue;
|
||||
const extension = extname(entry.name).toLowerCase();
|
||||
if (extension === ".epub" || extension === ".pdf" || extension === ".cbz") {
|
||||
if (extension === ".epub" || extension === ".pdf" || extension === ".cbz" || extension === ".cbr") {
|
||||
yield path;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
function bookFormatFromPath(filePath: string): "epub" | "pdf" | "cbz" {
|
||||
function bookFormatFromPath(filePath: string): "epub" | "pdf" | "cbz" | "cbr" {
|
||||
const extension = extname(filePath).toLowerCase();
|
||||
if (extension === ".epub") return "epub";
|
||||
if (extension === ".cbz") return "cbz";
|
||||
if (extension === ".cbr") return "cbr";
|
||||
return "pdf";
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user