feat(route/pixiv): add R18 novels support and full content toggle for… (#17391)
* feat(route/pixiv): add R18 novels support and full content toggle for user novels * fix: information & image placeholders * refactor: split novels fetching into SFW/NSFW modules and improve type definitions * feat: add info for sfw * feat: add radar * refactor: use jsdom instead of regex * feat: add limit support for nsfw novels * docs: rename radar title * revert: part of #17440 Object.entries(options.searchParams) returns `[]` * fix: clean up * feat: early exit when no SFW novels found * refactor: combine novel parsing logic into utils * docs: restore pixiv doc link * feat: cache novel content * refactor: cleanup * refactor: full content function ---------
This commit is contained in:
Tsuyumi
GitHub
parent
6451863ee9
commit
cf7ce2470b
|
|
@ -0,0 +1,22 @@
|
|||
import got from '../pixiv-got';
|
||||
import { maskHeader } from '../constants';
|
||||
import queryString from 'query-string';
|
||||
|
||||
/**
|
||||
* 获取插画详细信息
|
||||
* @param {string} illust_id 插画作品 id
|
||||
* @param {string} token pixiv oauth token
|
||||
* @returns {Promise<got.AxiosResponse<{illust: IllustDetail}>>}
|
||||
*/
|
||||
export default function getIllustDetail(illust_id: string, token: string) {
|
||||
return got('https://app-api.pixiv.net/v1/illust/detail', {
|
||||
headers: {
|
||||
...maskHeader,
|
||||
Authorization: 'Bearer ' + token,
|
||||
},
|
||||
searchParams: queryString.stringify({
|
||||
illust_id,
|
||||
filter: 'for_ios',
|
||||
}),
|
||||
});
|
||||
}
|
||||
|
|
@ -0,0 +1,247 @@
|
|||
import got from '../pixiv-got';
|
||||
import { maskHeader } from '../constants';
|
||||
import queryString from 'query-string';
|
||||
import { config } from '@/config';
|
||||
import { JSDOM, VirtualConsole } from 'jsdom';
|
||||
|
||||
import pixivUtils from '../utils';
|
||||
import ConfigNotFoundError from '@/errors/types/config-not-found';
|
||||
import cache from '@/utils/cache';
|
||||
import { parseDate } from 'tough-cookie';
|
||||
import { getToken } from '../token';
|
||||
|
||||
interface nsfwNovelWork {
|
||||
id: string;
|
||||
title: string;
|
||||
caption: string;
|
||||
restrict: number;
|
||||
x_restrict: number;
|
||||
is_original: boolean;
|
||||
image_urls: {
|
||||
square_medium: string;
|
||||
medium: string;
|
||||
large: string;
|
||||
};
|
||||
create_date: string;
|
||||
tags: Array<{
|
||||
name: string;
|
||||
translated_name: string | null;
|
||||
added_by_uploaded_user: boolean;
|
||||
}>;
|
||||
page_count: number;
|
||||
text_length: number;
|
||||
user: {
|
||||
id: number;
|
||||
name: string;
|
||||
account: string;
|
||||
profile_image_urls: {
|
||||
medium: string;
|
||||
};
|
||||
is_followed: boolean;
|
||||
is_access_blocking_user: boolean;
|
||||
};
|
||||
series?: {
|
||||
id?: number;
|
||||
title?: string;
|
||||
};
|
||||
total_bookmarks: number;
|
||||
total_view: number;
|
||||
total_comments: number;
|
||||
}
|
||||
|
||||
interface nsfwNovelsResponse {
|
||||
data: {
|
||||
user: {
|
||||
id: number;
|
||||
name: string;
|
||||
account: string;
|
||||
profile_image_urls: {
|
||||
medium: string;
|
||||
};
|
||||
is_followed: boolean;
|
||||
is_access_blocking_user: boolean;
|
||||
};
|
||||
novels: nsfwNovelWork[];
|
||||
};
|
||||
}
|
||||
|
||||
interface nsfwNovelDetail {
|
||||
id: string;
|
||||
title: string;
|
||||
seriesId: string | null;
|
||||
seriesTitle: string | null;
|
||||
seriesIsWatched: boolean | null;
|
||||
userId: string;
|
||||
coverUrl: string;
|
||||
tags: string[];
|
||||
caption: string;
|
||||
cdate: string;
|
||||
rating: {
|
||||
like: number;
|
||||
bookmark: number;
|
||||
view: number;
|
||||
};
|
||||
text: string;
|
||||
marker: null;
|
||||
illusts: string[];
|
||||
images: {
|
||||
[key: string]: {
|
||||
novelImageId: string;
|
||||
sl: string;
|
||||
urls: {
|
||||
'240mw': string;
|
||||
'480mw': string;
|
||||
'1200x1200': string;
|
||||
'128x128': string;
|
||||
original: string;
|
||||
};
|
||||
};
|
||||
};
|
||||
seriesNavigation: {
|
||||
nextNovel: null;
|
||||
prevNovel: {
|
||||
id: number;
|
||||
viewable: boolean;
|
||||
contentOrder: string;
|
||||
title: string;
|
||||
coverUrl: string;
|
||||
viewableMessage: null;
|
||||
} | null;
|
||||
} | null;
|
||||
glossaryItems: string[];
|
||||
replaceableItemIds: string[];
|
||||
aiType: number;
|
||||
isOriginal: boolean;
|
||||
}
|
||||
|
||||
function getNovels(user_id: string, token: string): Promise<nsfwNovelsResponse> {
|
||||
return got('https://app-api.pixiv.net/v1/user/novels', {
|
||||
headers: {
|
||||
...maskHeader,
|
||||
Authorization: 'Bearer ' + token,
|
||||
},
|
||||
searchParams: queryString.stringify({
|
||||
user_id,
|
||||
filter: 'for_ios',
|
||||
}),
|
||||
});
|
||||
}
|
||||
|
||||
async function getNovelFullContent(novel_id: string, token: string): Promise<nsfwNovelDetail> {
|
||||
return (await cache.tryGet(`https://app-api.pixiv.net/webview/v2/novel:${novel_id}`, async () => {
|
||||
// https://github.com/mikf/gallery-dl/blob/main/gallery_dl/extractor/pixiv.py
|
||||
// https://github.com/mikf/gallery-dl/commit/db507e30c7431d4ed7e23c153a044ce1751c2847
|
||||
const response = await got('https://app-api.pixiv.net/webview/v2/novel', {
|
||||
headers: {
|
||||
...maskHeader,
|
||||
Authorization: 'Bearer ' + token,
|
||||
},
|
||||
searchParams: queryString.stringify({
|
||||
id: novel_id,
|
||||
viewer_version: '20221031_ai',
|
||||
}),
|
||||
});
|
||||
|
||||
const virtualConsole = new VirtualConsole().on('error', () => void 0);
|
||||
|
||||
const { window } = new JSDOM(response.data, {
|
||||
runScripts: 'dangerously',
|
||||
virtualConsole,
|
||||
});
|
||||
|
||||
const novelDetail = window.pixiv?.novel as nsfwNovelDetail;
|
||||
|
||||
window.close();
|
||||
|
||||
if (!novelDetail) {
|
||||
throw new Error('No novel data found');
|
||||
}
|
||||
|
||||
return novelDetail;
|
||||
})) as nsfwNovelDetail;
|
||||
}
|
||||
|
||||
function convertPixivProtocolExtended(caption: string): string {
|
||||
const protocolMap = new Map([
|
||||
[/pixiv:\/\/novels\/(\d+)/g, 'https://www.pixiv.net/novel/show.php?id=$1'],
|
||||
[/pixiv:\/\/illusts\/(\d+)/g, 'https://www.pixiv.net/artworks/$1'],
|
||||
[/pixiv:\/\/users\/(\d+)/g, 'https://www.pixiv.net/users/$1'],
|
||||
[/pixiv:\/\/novel\/series\/(\d+)/g, 'https://www.pixiv.net/novel/series/$1'],
|
||||
]);
|
||||
|
||||
let convertedText = caption;
|
||||
|
||||
for (const [pattern, replacement] of protocolMap) {
|
||||
convertedText = convertedText.replace(pattern, replacement);
|
||||
}
|
||||
|
||||
return convertedText;
|
||||
}
|
||||
|
||||
export async function getR18Novels(id: string, fullContent: boolean, limit: number = 100) {
|
||||
if (!config.pixiv || !config.pixiv.refreshToken) {
|
||||
throw new ConfigNotFoundError(
|
||||
'該用戶爲 R18 創作者，需要 PIXIV_REFRESHTOKEN。This user is an R18 creator, PIXIV_REFRESHTOKEN is required - pixiv RSS is disabled due to the lack of <a href="https://docs.rsshub.app/deploy/config#route-specific-configurations">relevant config</a>'
|
||||
);
|
||||
}
|
||||
|
||||
const token = await getToken(cache.tryGet);
|
||||
if (!token) {
|
||||
throw new ConfigNotFoundError('pixiv not login');
|
||||
}
|
||||
|
||||
const response = await getNovels(id, token);
|
||||
const novels = limit ? response.data.novels.slice(0, limit) : response.data.novels;
|
||||
const username = novels[0].user.name;
|
||||
|
||||
const items = await Promise.all(
|
||||
novels.map(async (novel) => {
|
||||
const baseItem = {
|
||||
title: novel.series?.title ? `${novel.series.title} - ${novel.title}` : novel.title,
|
||||
description: `
|
||||
<img src="${pixivUtils.getProxiedImageUrl(novel.image_urls.large)}" />
|
||||
<p>${convertPixivProtocolExtended(novel.caption) || ''}</p>
|
||||
<p>
|
||||
字數：${novel.text_length}<br>
|
||||
閱覽數：${novel.total_view}<br>
|
||||
收藏數：${novel.total_bookmarks}<br>
|
||||
評論數：${novel.total_comments}<br>
|
||||
</p>`,
|
||||
author: novel.user.name,
|
||||
pubDate: parseDate(novel.create_date),
|
||||
link: `https://www.pixiv.net/novel/show.php?id=${novel.id}`,
|
||||
category: novel.tags.map((t) => t.name),
|
||||
};
|
||||
|
||||
if (!fullContent) {
|
||||
return baseItem;
|
||||
}
|
||||
|
||||
try {
|
||||
const novelDetail = await getNovelFullContent(novel.id, token);
|
||||
const images = Object.fromEntries(
|
||||
Object.entries(novelDetail.images)
|
||||
.filter(([, image]) => image?.urls?.original)
|
||||
.map(([id, image]) => [id, image.urls.original.replace('https://i.pximg.net', config.pixiv.imgProxy || '')])
|
||||
);
|
||||
|
||||
const content = await pixivUtils.parseNovelContent(novelDetail.text, images, token);
|
||||
|
||||
return {
|
||||
...baseItem,
|
||||
description: `${baseItem.description}<hr>${content}`,
|
||||
};
|
||||
} catch {
|
||||
return baseItem;
|
||||
}
|
||||
})
|
||||
);
|
||||
|
||||
return {
|
||||
title: `${username}'s novels - pixiv`,
|
||||
description: `${username} 的 pixiv 最新小说`,
|
||||
image: pixivUtils.getProxiedImageUrl(novels[0].user.profile_image_urls.medium),
|
||||
link: `https://www.pixiv.net/users/${id}/novels`,
|
||||
item: items,
|
||||
};
|
||||
}
|
||||
|
|
@ -0,0 +1,201 @@
|
|||
import got from '@/utils/got';
|
||||
import cache from '@/utils/cache';
|
||||
import pixivUtils from '../utils';
|
||||
import { parseDate } from '@/utils/parse-date';
|
||||
|
||||
const baseUrl = 'https://www.pixiv.net';
|
||||
interface sfwNovelWork {
|
||||
id: string;
|
||||
title: string;
|
||||
genre: string;
|
||||
xRestrict: number;
|
||||
restrict: number;
|
||||
url: string;
|
||||
tags: string[];
|
||||
userId: string;
|
||||
userName: string;
|
||||
profileImageUrl: string;
|
||||
textCount: number;
|
||||
wordCount: number;
|
||||
readingTime: number;
|
||||
useWordCount: boolean;
|
||||
description: string;
|
||||
isBookmarkable: boolean;
|
||||
bookmarkData: null;
|
||||
bookmarkCount: number;
|
||||
isOriginal: boolean;
|
||||
marker: null;
|
||||
titleCaptionTranslation: {
|
||||
workTitle: null;
|
||||
workCaption: null;
|
||||
};
|
||||
createDate: string;
|
||||
updateDate: string;
|
||||
isMasked: boolean;
|
||||
aiType: number;
|
||||
seriesId: string;
|
||||
seriesTitle: string;
|
||||
isUnlisted: boolean;
|
||||
}
|
||||
|
||||
interface sfwNovelsResponse {
|
||||
data: {
|
||||
error: boolean;
|
||||
message: string;
|
||||
body: {
|
||||
works: Record<string, sfwNovelWork>;
|
||||
extraData: {
|
||||
meta: {
|
||||
title: string;
|
||||
description: string;
|
||||
canonical: string;
|
||||
ogp: {
|
||||
description: string;
|
||||
image: string;
|
||||
title: string;
|
||||
type: string;
|
||||
};
|
||||
twitter: {
|
||||
description: string;
|
||||
image: string;
|
||||
title: string;
|
||||
card: string;
|
||||
};
|
||||
alternateLanguages: {
|
||||
ja: string;
|
||||
en: string;
|
||||
};
|
||||
descriptionHeader: string;
|
||||
};
|
||||
};
|
||||
};
|
||||
};
|
||||
}
|
||||
|
||||
interface sfwNovelDetail {
|
||||
body: {
|
||||
content: string;
|
||||
textEmbeddedImages: Record<
|
||||
string,
|
||||
{
|
||||
novelImageId: string;
|
||||
sl: string;
|
||||
urls: {
|
||||
original: string;
|
||||
'1200x1200': string;
|
||||
'480mw': string;
|
||||
'240mw': string;
|
||||
'128x128': string;
|
||||
};
|
||||
}
|
||||
>;
|
||||
};
|
||||
}
|
||||
|
||||
async function getNovelFullContent(novel_id: string): Promise<{ content: string; images: Record<string, string> }> {
|
||||
const url = `${baseUrl}/ajax/novel/${novel_id}`;
|
||||
return (await cache.tryGet(url, async () => {
|
||||
const response = await got(url, {
|
||||
headers: {
|
||||
referer: `${baseUrl}/novel/show.php?id=${novel_id}`,
|
||||
},
|
||||
});
|
||||
|
||||
const novelDetail = response.data as sfwNovelDetail;
|
||||
|
||||
if (!novelDetail) {
|
||||
throw new Error('No novel data found');
|
||||
}
|
||||
|
||||
const images: Record<string, string> = {};
|
||||
|
||||
if (novelDetail.body.textEmbeddedImages) {
|
||||
for (const [id, image] of Object.entries(novelDetail.body.textEmbeddedImages)) {
|
||||
images[id] = pixivUtils.getProxiedImageUrl(image.urls.original);
|
||||
}
|
||||
}
|
||||
|
||||
return {
|
||||
content: novelDetail.body.content,
|
||||
images,
|
||||
};
|
||||
})) as { content: string; images: Record<string, string> };
|
||||
}
|
||||
|
||||
export async function getNonR18Novels(id: string, fullContent: boolean, limit: number = 100) {
|
||||
const url = `${baseUrl}/users/${id}/novels`;
|
||||
const { data: allData } = await got(`${baseUrl}/ajax/user/${id}/profile/all`, {
|
||||
headers: {
|
||||
referer: url,
|
||||
},
|
||||
});
|
||||
|
||||
const novels = Object.keys(allData.body.novels)
|
||||
.sort((a, b) => Number(b) - Number(a))
|
||||
.slice(0, Number.parseInt(String(limit), 10));
|
||||
|
||||
if (novels.length === 0) {
|
||||
throw new Error('No novels found, fallback to R18 API');
|
||||
// Throw error early to avoid unnecessary API requests
|
||||
// Since hasPixivAuth() check failed earlier and R18 API requires authentication, this will result in ConfigNotFoundError
|
||||
}
|
||||
|
||||
const searchParams = new URLSearchParams();
|
||||
for (const novel of novels) {
|
||||
searchParams.append('ids[]', novel);
|
||||
}
|
||||
|
||||
const { data } = (await got(`${baseUrl}/ajax/user/${id}/profile/novels`, {
|
||||
headers: {
|
||||
referer: url,
|
||||
},
|
||||
searchParams,
|
||||
})) as sfwNovelsResponse;
|
||||
|
||||
const items = await Promise.all(
|
||||
Object.values(data.body.works).map(async (item) => {
|
||||
const baseItem = {
|
||||
title: item.title,
|
||||
description: `
|
||||
<img src=${pixivUtils.getProxiedImageUrl(item.url)} />
|
||||
<p>${item.description}</p>
|
||||
<p>
|
||||
字數：${item.textCount}<br>
|
||||
閱讀時間：${item.readingTime} 分鐘<br>
|
||||
收藏數：${item.bookmarkCount}<br>
|
||||
</p>
|
||||
`,
|
||||
link: `${baseUrl}/novel/show.php?id=${item.id}`,
|
||||
author: item.userName,
|
||||
pubDate: parseDate(item.createDate),
|
||||
updated: parseDate(item.updateDate),
|
||||
category: item.tags,
|
||||
};
|
||||
|
||||
if (!fullContent) {
|
||||
return baseItem;
|
||||
}
|
||||
|
||||
try {
|
||||
const { content: initialContent, images } = await getNovelFullContent(item.id);
|
||||
|
||||
const content = await pixivUtils.parseNovelContent(initialContent, images);
|
||||
|
||||
return {
|
||||
...baseItem,
|
||||
description: `${baseItem.description}<hr>${content}`,
|
||||
};
|
||||
} catch {
|
||||
return baseItem;
|
||||
}
|
||||
})
|
||||
);
|
||||
|
||||
return {
|
||||
title: data.body.extraData.meta.title,
|
||||
description: data.body.extraData.meta.ogp.description,
|
||||
image: pixivUtils.getProxiedImageUrl(Object.values(data.body.works)[0].profileImageUrl),
|
||||
link: url,
|
||||
item: items,
|
||||
};
|
||||
}
|
||||
|
|
@ -1,15 +1,36 @@
|
|||
import { Route } from '@/types';
|
||||
import got from '@/utils/got';
|
||||
import { parseDate } from '@/utils/parse-date';
|
||||
const baseUrl = 'https://www.pixiv.net';
|
||||
import { Data, Route, ViewType } from '@/types';
|
||||
import { fallback, queryToBoolean } from '@/utils/readable-social';
|
||||
import { getR18Novels } from './api/get-novels-nsfw';
|
||||
import { getNonR18Novels } from './api/get-novels-sfw';
|
||||
import { config } from '@/config';
|
||||
|
||||
export const route: Route = {
|
||||
path: '/user/novels/:id',
|
||||
path: '/user/novels/:id/:full_content?',
|
||||
categories: ['social-media'],
|
||||
view: ViewType.Articles,
|
||||
example: '/pixiv/user/novels/27104704',
|
||||
parameters: { id: "User id, available in user's homepage URL" },
|
||||
parameters: {
|
||||
id: "User id, available in user's homepage URL",
|
||||
full_content: {
|
||||
description: 'Enable or disable the display of full content. ',
|
||||
options: [
|
||||
{ value: 'true', label: 'true' },
|
||||
{ value: 'false', label: 'false' },
|
||||
],
|
||||
default: 'false',
|
||||
},
|
||||
},
|
||||
features: {
|
||||
requireConfig: false,
|
||||
requireConfig: [
|
||||
{
|
||||
name: 'PIXIV_REFRESHTOKEN',
|
||||
optional: true,
|
||||
description: `
|
||||
Pixiv 登錄後的 refresh_token，用於獲取 R18 小說
|
||||
refresh_token after Pixiv login, required for accessing R18 novels
|
||||
[https://docs.rsshub.app/deploy/config#pixiv](https://docs.rsshub.app/deploy/config#pixiv)`,
|
||||
},
|
||||
],
|
||||
requirePuppeteer: false,
|
||||
antiCrawler: false,
|
||||
supportBT: false,
|
||||
|
|
@ -18,54 +39,53 @@ export const route: Route = {
|
|||
},
|
||||
radar: [
|
||||
{
|
||||
title: 'User Novels (簡介 Basic info)',
|
||||
source: ['www.pixiv.net/users/:id/novels'],
|
||||
target: '/user/novels/:id',
|
||||
},
|
||||
{
|
||||
title: 'User Novels (全文 Full text)',
|
||||
source: ['www.pixiv.net/users/:id/novels'],
|
||||
target: '/user/novels/:id/true',
|
||||
},
|
||||
],
|
||||
name: 'User Novels',
|
||||
maintainers: ['TonyRL'],
|
||||
maintainers: ['TonyRL', 'SnowAgar25'],
|
||||
handler,
|
||||
description: `
|
||||
| 小說類型 Novel Type | full_content | PIXIV_REFRESHTOKEN | 返回內容 Content |
|
||||
|-------------------|--------------|-------------------|-----------------|
|
||||
| Non R18 | false | 不需要 Not Required | 簡介 Basic info |
|
||||
| Non R18 | true | 不需要 Not Required | 全文 Full text |
|
||||
| R18 | false | 需要 Required | 簡介 Basic info |
|
||||
| R18 | true | 需要 Required | 全文 Full text |
|
||||
|
||||
Default value for \`full_content\` is \`false\` if not specified.
|
||||
|
||||
Example:
|
||||
- \`/pixiv/user/novels/79603797\` → 簡介 Basic info
|
||||
- \`/pixiv/user/novels/79603797/true\` → 全文 Full text`,
|
||||
};
|
||||
|
||||
async function handler(ctx) {
|
||||
const id = ctx.req.param('id');
|
||||
const { limit = 100 } = ctx.req.query();
|
||||
const url = `${baseUrl}/users/${id}/novels`;
|
||||
const { data: allData } = await got(`${baseUrl}/ajax/user/${id}/profile/all`, {
|
||||
headers: {
|
||||
referer: url,
|
||||
},
|
||||
});
|
||||
const hasPixivAuth = () => Boolean(config.pixiv && config.pixiv.refreshToken);
|
||||
|
||||
const novels = Object.keys(allData.body.novels)
|
||||
.sort((a, b) => b - a)
|
||||
.slice(0, Number.parseInt(limit, 10));
|
||||
const searchParams = new URLSearchParams();
|
||||
for (const novel of novels) {
|
||||
searchParams.append('ids[]', novel);
|
||||
async function handler(ctx): Promise<Data> {
|
||||
const id = ctx.req.param('id');
|
||||
const fullContent = fallback(undefined, queryToBoolean(ctx.req.param('full_content')), false);
|
||||
|
||||
const { limit } = ctx.req.query();
|
||||
|
||||
// Use R18 API first if auth exists
|
||||
if (hasPixivAuth()) {
|
||||
return await getR18Novels(id, fullContent, limit);
|
||||
}
|
||||
|
||||
const { data } = await got(`${baseUrl}/ajax/user/${id}/profile/novels`, {
|
||||
headers: {
|
||||
referer: url,
|
||||
},
|
||||
searchParams,
|
||||
});
|
||||
// Attempt non-R18 API when Pixiv auth is missing
|
||||
const nonR18Result = await getNonR18Novels(id, fullContent, limit).catch(() => null);
|
||||
if (nonR18Result) {
|
||||
return nonR18Result;
|
||||
}
|
||||
|
||||
const items = Object.values(data.body.works).map((item) => ({
|
||||
title: item.seriesTitle || item.title,
|
||||
description: item.description || item.title,
|
||||
link: `${baseUrl}/novel/series/${item.id}`,
|
||||
author: item.userName,
|
||||
pubDate: parseDate(item.createDate),
|
||||
updated: parseDate(item.updateDate),
|
||||
category: item.tags,
|
||||
}));
|
||||
|
||||
return {
|
||||
title: data.body.extraData.meta.title,
|
||||
description: data.body.extraData.meta.ogp.description,
|
||||
image: Object.values(data.body.works)[0].profileImageUrl,
|
||||
link: url,
|
||||
item: items,
|
||||
};
|
||||
// Fallback to R18 API as last resort
|
||||
return await getR18Novels(id, fullContent, limit);
|
||||
}
|
||||
|
|
|
|||
|
|
@ -1,4 +1,6 @@
|
|||
import { config } from '@/config';
|
||||
import { load } from 'cheerio';
|
||||
import getIllustDetail from './api/get-illust-detail';
|
||||
|
||||
export default {
|
||||
getImgs(illust) {
|
||||
|
|
@ -14,4 +16,121 @@ export default {
|
|||
}
|
||||
return images;
|
||||
},
|
||||
getProxiedImageUrl(originalUrl: string): string {
|
||||
return originalUrl.replace('https://i.pximg.net', config.pixiv.imgProxy || '');
|
||||
},
|
||||
// docs: https://www.pixiv.help/hc/ja/articles/235584168-小説作品の本文内に使える特殊タグとは
|
||||
async parseNovelContent(content: string, images: Record<string, string>, token?: string): Promise<string> {
|
||||
try {
|
||||
// 如果有 token，處理 pixiv 圖片引用
|
||||
// If token exists, process pixiv image references
|
||||
if (token) {
|
||||
const imageMatches = [...content.matchAll(/\[pixivimage:(\d+)(?:-(\d+))?\]/g)];
|
||||
const imageIdToUrl = new Map<string, string>();
|
||||
|
||||
// 批量獲取圖片資訊
|
||||
// Batch fetch image information
|
||||
await Promise.all(
|
||||
imageMatches.map(async ([, illustId, pageNum]) => {
|
||||
if (!illustId) {
|
||||
return;
|
||||
}
|
||||
|
||||
try {
|
||||
const illust = (await getIllustDetail(illustId, token)).data.illust;
|
||||
const pixivimages = this.getImgs(illust).map((img) => img.match(/src="([^"]+)"/)?.[1] || '');
|
||||
|
||||
const imageUrl = pixivimages[Number(pageNum) || 0];
|
||||
if (imageUrl) {
|
||||
imageIdToUrl.set(pageNum ? `${illustId}-${pageNum}` : illustId, imageUrl);
|
||||
}
|
||||
} catch (error) {
|
||||
// 記錄錯誤但不中斷處理
|
||||
// Log error but don't interrupt processing
|
||||
logger.warn(`Failed to fetch illust detail for ID ${illustId}: ${error instanceof Error ? error.message : String(error)}`);
|
||||
}
|
||||
})
|
||||
);
|
||||
|
||||
// 替換 pixiv 圖片引用為 img 標籤
|
||||
// Replace pixiv image references with img tags
|
||||
content = content.replaceAll(/\[pixivimage:(\d+)(?:-(\d+))?\]/g, (match, illustId, pageNum) => {
|
||||
const key = pageNum ? `${illustId}-${pageNum}` : illustId;
|
||||
const imageUrl = imageIdToUrl.get(key);
|
||||
return imageUrl ? `<img src="${imageUrl}" alt="pixiv illustration ${illustId}${pageNum ? ` page ${pageNum}` : ''}">` : match;
|
||||
});
|
||||
} else {
|
||||
/*
|
||||
* 處理 get-novels-sfw 的情況
|
||||
* 當沒有 PIXIV_REFRESHTOKEN 時，將 [pixivimage:(\d+)] 格式轉換為 artwork 連結
|
||||
* 因無法獲取 Pixiv 作品詳情，改為提供直接連結到原始作品頁面
|
||||
*
|
||||
* Handle get-novels-sfw case
|
||||
* When PIXIV_REFRESHTOKEN is not available, convert [pixivimage:(\d+)] format to artwork link
|
||||
* Provide direct link to original artwork page since artwork details cannot be retrieved
|
||||
*/
|
||||
content = content.replaceAll(/\[pixivimage:(\d+)(?:-(\d+))?\]/g, (_, illustId) => `<a href="https://www.pixiv.net/artworks/${illustId}" target="_blank" rel="noopener noreferrer">Pixiv Artwork #${illustId}</a>`);
|
||||
}
|
||||
|
||||
// 處理作者上傳的圖片
|
||||
// Process author uploaded images
|
||||
content = content.replaceAll(/\[uploadedimage:(\d+)\]/g, (match, imageId) => {
|
||||
if (images[imageId]) {
|
||||
return `<img src="${images[imageId]}" alt="novel illustration ${imageId}">`;
|
||||
}
|
||||
return match;
|
||||
});
|
||||
|
||||
// 基本格式處理
|
||||
// Basic formatting
|
||||
content = content
|
||||
// 換行轉換為 HTML 換行
|
||||
// Convert newlines to HTML breaks
|
||||
.replaceAll('\n', '<br>')
|
||||
// 連續換行轉換為段落
|
||||
// Convert consecutive breaks to paragraphs
|
||||
.replaceAll(/(<br>){2,}/g, '</p><p>')
|
||||
// ruby 標籤（為日文漢字標註讀音）
|
||||
// ruby tags (for Japanese kanji readings)
|
||||
.replaceAll(/\[\[rb:(.*?)>(.*?)\]\]/g, '<ruby>$1<rt>$2</rt></ruby>')
|
||||
// 外部連結
|
||||
// external links
|
||||
.replaceAll(/\[\[jumpuri:(.*?)>(.*?)\]\]/g, '<a href="$2" target="_blank" rel="noopener noreferrer">$1</a>')
|
||||
// 頁面跳轉，但由於 [newpage] 使用 hr 分隔，沒有頁數，沒必要跳轉，所以只顯示文字
|
||||
// Page jumps, but since [newpage] uses hr separators, without the page numbers, jumping isn't needed, so just display text
|
||||
.replaceAll(/\[jump:(\d+)\]/g, 'Jump to page $1')
|
||||
// 章節標題
|
||||
// chapter titles
|
||||
.replaceAll(/\[chapter:(.*?)\]/g, '<h2>$1</h2>')
|
||||
// 分頁符
|
||||
// page breaks
|
||||
.replaceAll('[newpage]', '<hr>');
|
||||
|
||||
// 使用 cheerio 進行 HTML 清理和優化
|
||||
// Use cheerio for HTML cleanup and optimization
|
||||
const $content = load(`<article><p>${content}</p></article>`);
|
||||
|
||||
// 處理嵌套段落：移除多餘的嵌套
|
||||
// Handle nested paragraphs: remove unnecessary nesting
|
||||
$content('p p').each((_, elem) => {
|
||||
const $elem = $content(elem);
|
||||
$elem.replaceWith($elem.html() || '');
|
||||
});
|
||||
|
||||
// 處理段落中的標題：確保正確的 HTML 結構
|
||||
// Handle headings in paragraphs: ensure correct HTML structure
|
||||
$content('p h2').each((_, elem) => {
|
||||
const $elem = $content(elem);
|
||||
const $parent = $elem.parent('p');
|
||||
const html = $elem.prop('outerHTML');
|
||||
if ($parent.length && html) {
|
||||
$parent.replaceWith(`</p>${html}<p>`);
|
||||
}
|
||||
});
|
||||
|
||||
return $content.html() || '';
|
||||
} catch (error) {
|
||||
throw new Error(`Error parsing novel content: ${error instanceof Error ? error.message : String(error)}`);
|
||||
}
|
||||
},
|
||||
};
|
||||
|
|
|
|||
Loading…
Reference in New Issue