feat(library): core types + rss/atom chapter feed parser

This commit is contained in:
afiqzudinhadi 2026-08-03 15:07:06 +08:00
parent 1e7e6d1032
commit c398669ae6
3 changed files with 260 additions and 0 deletions

View file

@ -0,0 +1,97 @@
export interface FeedItem {
title: string;
date: string; // ISO (YYYY-MM-DD) or ''
id: string;
}
const BLOCK_RE = /<(item|entry)\b[^>]*>([\s\S]*?)<\/\1>/gi;
function extractTag(block: string, tag: string): string | null {
const re = new RegExp(`<${tag}\\b[^>]*>([\\s\\S]*?)<\\/${tag}>`, 'i');
const m = re.exec(block);
return m ? m[1] : null;
}
function unwrapCdata(raw: string): string {
const m = /^\s*<!\[CDATA\[([\s\S]*?)\]\]>\s*$/.exec(raw);
return m ? m[1] : raw;
}
function decodeEntities(raw: string): string {
return raw
.replace(/&lt;/g, '<')
.replace(/&gt;/g, '>')
.replace(/&quot;/g, '"')
.replace(/&#39;/g, "'")
.replace(/&amp;/g, '&');
}
function cleanText(raw: string | null): string {
if (raw === null) return '';
return decodeEntities(unwrapCdata(raw)).trim();
}
function toIsoDate(raw: string | null): string {
if (!raw) return '';
const d = new Date(cleanText(raw));
if (isNaN(d.getTime())) return '';
return d.toISOString().slice(0, 10);
}
/**
* RSS 2.0 <item> + Atom <entry>, regex-based (no DOM libs). Entities decoded
* (&amp; &lt; &gt; &quot; &#39;), CDATA unwrapped. Invalid/missing dates ''.
*/
export function parseFeed(xml: string): FeedItem[] {
const items: FeedItem[] = [];
if (!xml) return items;
let match: RegExpExecArray | null;
BLOCK_RE.lastIndex = 0;
while ((match = BLOCK_RE.exec(xml)) !== null) {
const block = match[2];
const title = cleanText(extractTag(block, 'title'));
const rawDate = extractTag(block, 'pubDate') ?? extractTag(block, 'updated');
const id = cleanText(extractTag(block, 'guid') ?? extractTag(block, 'id'));
items.push({ title, date: toIsoDate(rawDate), id });
}
return items;
}
const CHAPTER_RE = /(chapter|ch\.?|#)\s*(\d+(\.\d+)?)/i;
/** /(chapter|ch\.?|#)\s*(\d+(\.\d+)?)/i → parseFloat; null when no match */
export function extractChapterNumber(title: string): number | null {
const m = CHAPTER_RE.exec(title);
if (!m) return null;
return parseFloat(m[2]);
}
/**
* Items assumed newest-first. Picks the item with the highest parseable
* chapter number; falls back to date ordering (max ISO date) when no item
* title yields a parseable chapter number.
*/
export function latestChapter(items: FeedItem[]): { chapter: number | null; date: string; title: string } | null {
if (items.length === 0) return null;
let best: FeedItem | null = null;
let bestChapter: number | null = null;
for (const item of items) {
const chapter = extractChapterNumber(item.title);
if (chapter !== null && (bestChapter === null || chapter > bestChapter)) {
bestChapter = chapter;
best = item;
}
}
if (best) return { chapter: bestChapter, date: best.date, title: best.title };
let latest = items[0];
for (const item of items) {
if (item.date && (!latest.date || item.date > latest.date)) latest = item;
}
return { chapter: null, date: latest.date, title: latest.title };
}

View file

@ -0,0 +1,28 @@
export type HttpJsonFn = (url: string, headers: Record<string, string>) => Promise<any>;
export type HttpTextFn = (url: string, headers: Record<string, string>) => Promise<string>;
export interface LibraryNoteCtx {
frontmatter: Record<string, string>;
body: string;
filename: string;
}
export interface MediaTypeSpec {
typeName: 'manga' | 'book' | 'game' | 'comic';
itemType: string; // 'manga_item' etc.
folderSettingKey: string;
enabledSettingKey: string;
throttleMs: number;
hasId(fm: Record<string, string>): boolean;
isActive(fm: Record<string, string>): boolean;
resolve(ctx: LibraryNoteCtx, deps: SpecDeps): Promise<Record<string, string> | null>; // returns fm patches (id fields)
sync(ctx: LibraryNoteCtx, deps: SpecDeps): Promise<{ content: string; flipped: boolean } | null>; // full new note content; null = skip
}
export interface SpecDeps {
http: HttpJsonFn;
httpText: HttpTextFn;
getKey(name: 'rawg' | 'comicvine'): string; // '' when unset
log(msg: string): void;
notify(msg: string): void;
}

135
tests/library-rss.test.ts Normal file
View file

@ -0,0 +1,135 @@
import { describe, expect, test } from 'bun:test';
import { extractChapterNumber, latestChapter, parseFeed } from 'packages/obsidian/src/library/rss';
const rssSample = `<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
<channel>
<title>Chainsaw Man Updates</title>
<item>
<title>Chainsaw Man Chapter 214</title>
<pubDate>Thu, 30 Jul 2026 12:00:00 GMT</pubDate>
<guid>https://example.com/csm-214</guid>
</item>
<item>
<title>Chainsaw Man Chapter 213.5</title>
<pubDate>Thu, 23 Jul 2026 12:00:00 GMT</pubDate>
<guid>https://example.com/csm-213-5</guid>
</item>
<item>
<title>Chainsaw Man Chapter 213</title>
<pubDate>Thu, 16 Jul 2026 12:00:00 GMT</pubDate>
<guid>https://example.com/csm-213</guid>
</item>
</channel>
</rss>`;
const atomSample = `<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
<title>Some Manga Feed</title>
<entry>
<title>Ch. 12 name</title>
<updated>2026-08-01T09:30:00Z</updated>
<id>https://example.com/entry/12</id>
</entry>
</feed>`;
const cdataSample = `<rss version="2.0"><channel><item>
<title><![CDATA[One Piece Chapter 1120]]></title>
<pubDate>Mon, 01 Jun 2026 00:00:00 GMT</pubDate>
<guid>id-cdata</guid>
</item></channel></rss>`;
const entitySample = `<rss version="2.0"><channel><item>
<title>Attack &amp; Titan &#39;Special&#39; Chapter 5</title>
<pubDate>Mon, 01 Jun 2026 00:00:00 GMT</pubDate>
<guid>id-entity</guid>
</item></channel></rss>`;
const numberlessSample = `<rss version="2.0"><channel>
<item>
<title>Chainsaw Man - Extra Announcement</title>
<pubDate>Thu, 30 Jul 2026 12:00:00 GMT</pubDate>
<guid>a</guid>
</item>
<item>
<title>Chainsaw Man - Fan Art Contest</title>
<pubDate>Thu, 16 Jul 2026 12:00:00 GMT</pubDate>
<guid>b</guid>
</item>
</channel></rss>`;
describe('parseFeed', () => {
test('RSS 2.0 — 3 items, title/date/id mapped, newest first as-authored', () => {
const items = parseFeed(rssSample);
expect(items).toEqual([
{ title: 'Chainsaw Man Chapter 214', date: '2026-07-30', id: 'https://example.com/csm-214' },
{ title: 'Chainsaw Man Chapter 213.5', date: '2026-07-23', id: 'https://example.com/csm-213-5' },
{ title: 'Chainsaw Man Chapter 213', date: '2026-07-16', id: 'https://example.com/csm-213' },
]);
});
test('Atom — entry title/updated/id mapped', () => {
const items = parseFeed(atomSample);
expect(items).toEqual([{ title: 'Ch. 12 — name', date: '2026-08-01', id: 'https://example.com/entry/12' }]);
});
test('CDATA title unwrapped', () => {
const items = parseFeed(cdataSample);
expect(items[0].title).toBe('One Piece Chapter 1120');
});
test('entity-encoded title decoded', () => {
const items = parseFeed(entitySample);
expect(items[0].title).toBe(`Attack & Titan 'Special' Chapter 5`);
});
test('empty xml → []', () => {
expect(parseFeed('')).toEqual([]);
});
test('garbage xml → []', () => {
expect(parseFeed('not xml at all, just some random text')).toEqual([]);
});
});
describe('extractChapterNumber', () => {
test('"Chapter 214" → 214', () => {
expect(extractChapterNumber('Chapter 214')).toBe(214);
});
test('"ch.213.5" → 213.5', () => {
expect(extractChapterNumber('ch.213.5')).toBe(213.5);
});
test('"#77" → 77', () => {
expect(extractChapterNumber('#77')).toBe(77);
});
test('"Episode 5" → null', () => {
expect(extractChapterNumber('Episode 5')).toBeNull();
});
test('"Vol. 3 Chapter 21" → 21', () => {
expect(extractChapterNumber('Vol. 3 Chapter 21')).toBe(21);
});
test('numberless title → null', () => {
expect(extractChapterNumber('Chainsaw Man - Extra Announcement')).toBeNull();
});
});
describe('latestChapter', () => {
test('empty items → null', () => {
expect(latestChapter([])).toBeNull();
});
test('RSS sample → highest chapter (214) wins', () => {
const result = latestChapter(parseFeed(rssSample));
expect(result).toEqual({ chapter: 214, date: '2026-07-30', title: 'Chainsaw Man Chapter 214' });
});
test('Atom sample → single entry chapter 12', () => {
const result = latestChapter(parseFeed(atomSample));
expect(result).toEqual({ chapter: 12, date: '2026-08-01', title: 'Ch. 12 — name' });
});
test('numberless titles → date-based fallback, chapter null', () => {
const result = latestChapter(parseFeed(numberlessSample));
expect(result).toEqual({ chapter: null, date: '2026-07-30', title: 'Chainsaw Man - Extra Announcement' });
});
});