annu-kute-ced/src/lib/server/castopod.ts

import { castopod } from '$lib/config';
import { fetchText } from './http';
import type { Episode } from '$lib/types';

function tag(xml: string, name: string): string {
	const cdata = new RegExp(`<${name}[^>]*><!\\[CDATA\\[([\\s\\S]*?)\\]\\]></${name}>`, 'i');
	const plain = new RegExp(`<${name}[^>]*>([\\s\\S]*?)</${name}>`, 'i');
	const match = xml.match(cdata) ?? xml.match(plain);
	return match?.[1]?.trim() ?? '';
}

function attr(xml: string, tagName: string, attrName: string): string {
	const tagMatch = xml.match(new RegExp(`<${tagName}\\b[^>]*>`, 'i'));
	if (!tagMatch) return '';
	const attrMatch = tagMatch[0].match(new RegExp(`${attrName}="([^"]*)"`, 'i'));
	return attrMatch?.[1] ?? '';
}

function stripTags(html: string): string {
	return html
		.replace(/<[^>]+>/g, ' ')
		.replace(/&nbsp;/g, ' ')
		.replace(/&amp;/g, '&')
		.replace(/&lt;/g, '<')
		.replace(/&gt;/g, '>')
		.replace(/&quot;/g, '"')
		.replace(/&#0?39;|&apos;/g, "'")
		.replace(/\s+/g, ' ')
		.trim();
}

function parseDuration(raw: string): number {
	if (!raw) return 0;
	const parts = raw.split(':').map(Number);
	if (parts.some(Number.isNaN)) return 0;
	if (parts.length === 3) return parts[0] * 3600 + parts[1] * 60 + parts[2];
	if (parts.length === 2) return parts[0] * 60 + parts[1];
	return parts[0] ?? 0;
}

async function getEpisodesForSlug(slug: string): Promise<Episode[]> {
	const feedUrl = `${castopod.url}/@${slug}/feed`;
	const xml = await fetchText(feedUrl);
	if (!xml) return [];

	const channelImage =
		tag(xml.split('<item')[0] ?? '', 'url') || attr(xml.split('<item')[0] ?? '', 'itunes:image', 'href');
	const items = xml.split(/<item[\s>]/i).slice(1);

	return items.map((item, index) => {
		const body = item.split('</item>')[0] ?? item;
		const title = tag(body, 'title');
		const link = tag(body, 'link');
		const guid = tag(body, 'guid') || link || `${slug}-${index}`;
		const description = stripTags(tag(body, 'description') || tag(body, 'itunes:summary'));
		return {
			id: guid,
			title,
			description,
			audioUrl: attr(body, 'enclosure', 'url'),
			audioType: attr(body, 'enclosure', 'type') || 'audio/mpeg',
			duration: parseDuration(tag(body, 'itunes:duration')),
			date: tag(body, 'pubDate'),
			image: attr(body, 'itunes:image', 'href') || channelImage || null,
			link,
			podcast: slug
		};
	});
}

export async function getCastopodEpisodes(count = castopod.episodesCount): Promise<Episode[]> {
	const all: Episode[] = [];
	for (const slug of castopod.podcastSlugs) {
		const episodes = await getEpisodesForSlug(slug);
		all.push(...episodes);
		if (castopod.podcastSlugs.length > 1) {
			await new Promise((resolve) => setTimeout(resolve, 5000));
		}
	}
	return all
		.filter((episode) => episode.audioUrl)
		.sort((a, b) => new Date(b.date).getTime() - new Date(a.date).getTime())
		.slice(0, count);
}