* style: add eslint-unicorn * style: fix unicorn/no-useless-spread * style: fix unicorn/no-useless-promise-resolve-reject * style: fix unicorn/no-for-loop * fix: codeql bad HTML filtering regexp * fix: codeql incomplete replace * fix: unicorn/no-abusive-eslint-disable * style: fix unicorn/no-new-array * style: fix unicorn/no-typeof-undefined * style: fix unicorn/no-zero-fractions * style: fix unicorn/no-empty-file * style: fix unicorn/prefer-date-now * revert: auto fix unicorn/prefer-switch on lib/v2/kuaidi100/utils.js * style: fix unicorn/prefer-array-find * style: fix unicorn/prefer-array-flat * style: fix unicorn/prefer-array-flat-map * style: fix unicorn/prefer-at * style: fix unicorn/prefer-string-starts-ends-with * style: fix unicorn/prefer-includes * fix: codeql URL substring sanitization * style: fix unicorn/prefer-optional-catch-binding * style: fix unicorn/catch-error-name * style: fix unicorn/escape-case * style: fix unicorn/prefer-native-coercion-functions * style: fix unicorn/prefer-regexp-test * style: fix unicorn/require-array-join-separator * style: fix unicorn/prefer-math-trunc * style: fix unicorn/prefer-negative-index * style: fix unicorn/prefer-dom-node-dataset * style: fix unicorn/prefer-dom-node-text-content * style: fix unicorn/prefer-query-selector * style: fix unicorn/no-array-for-each * style: fix unicorn/no-negated-condition * style: fix unicorn/prefer-add-event-listener * style: fix unicorn/import-style * style: fix prefer-regex-literals * style: disable unicorn/no-useless-switch-case * style: disable unicorn/text-encoding-identifier-case * style: fix unicorn/prefer-set-has * style: fix unicorn/prefer-spread * revert: auto fix on lib/routes/universities/ynnu/edu/base64.js * style: fix unicorn/no-useless-undefined * style: fix unicorn/no-array-push-push * style: fix unicorn/no-useless-undefined again * style: fix unicorn/no-lonely-if * style: fix unicorn/prefer-reflect-apply * style: fix unicorn/switch-case-braces * style: fix unicorn/prefer-switch * style: fix unicorn/prefer-array-some * fix: deepscan UNUSED_VAR_ASSIGN * style: fix unicorn/prefer-ternary * fix: follow-up of unicorn/prefer-ternary * revert: auto fix of unicorn/prefer-string-slice for substring() * style: disable unicorn/prefer-string-slice fix: auto fix slice over deprecated substr * style: fix unicorn/throw-new-error * style: fix unicorn/filename-case * test: fix dateParser renaming * style: fix unicorn/better-regex * style: fix unicorn/prefer-string-replace-all * fix(deps): add sanitize-html * style: fix no-prototype-builtins * style: fix unicorn/consistent-destructuring * style: fix unicorn/consistent-function-scoping * style: fix unicorn/prefer-regexp-test * style: fix unicorn/prefer-logical-operator-over-ternary * style: fix unicorn/no-array-callback-reference * style: add prefer-object-has-own * style: warn unicorn/no-empty-file * style: fix unicorn/prefer-number-properties * style: fix no-useless-undefined again * style: fix unicorn/numeric-separators-style * style: disable unicorn/no-array-callback-reference false postive with cheerio
142 lines
5.8 KiB
JavaScript
142 lines
5.8 KiB
JavaScript
const got = require('@/utils/got');
|
|
const cheerio = require('cheerio');
|
|
const { parseDate } = require('@/utils/parse-date');
|
|
const { art } = require('@/utils/render');
|
|
const path = require('path');
|
|
|
|
module.exports = async (ctx) => {
|
|
const MUST_FETCH_BY_TOPICS = new Set(['authors']);
|
|
const CAN_USE_SOPHI = ['world'];
|
|
|
|
const category = ctx.params.category;
|
|
const topic = ctx.params.topic ?? (category === 'authors' ? 'reuters' : '');
|
|
const limit = ctx.query.limit ? Number.parseInt(ctx.query.limit) : 20;
|
|
const useSophi = ctx.query.sophi === 'true' && 'topic' !== '' && CAN_USE_SOPHI.includes(category);
|
|
|
|
const section_id = `/${category}/${topic ? `${topic}/` : ''}`;
|
|
const { title, description, rootUrl, response } = await (async () => {
|
|
if (MUST_FETCH_BY_TOPICS.has(category)) {
|
|
const rootUrl = 'https://www.reuters.com/pf/api/v3/content/fetch/articles-by-topic-v1';
|
|
const response = await got(rootUrl, {
|
|
searchParams: {
|
|
query: JSON.stringify({
|
|
offset: 0,
|
|
size: limit,
|
|
topic_url: section_id,
|
|
website: 'reuters',
|
|
}),
|
|
},
|
|
}).json();
|
|
|
|
return {
|
|
title: `${response.result.topics[0].name} | Reuters`,
|
|
description: response.result.topics[0].entity_id,
|
|
rootUrl,
|
|
response,
|
|
};
|
|
} else {
|
|
const rootUrl = 'https://www.reuters.com/pf/api/v3/content/fetch/articles-by-section-alias-or-id-v1';
|
|
const response = await got(rootUrl, {
|
|
searchParams: {
|
|
query: JSON.stringify({
|
|
offset: 0,
|
|
size: limit,
|
|
section_id,
|
|
website: 'reuters',
|
|
...(useSophi
|
|
? {
|
|
fetch_type: 'sophi',
|
|
sophi_page: '*',
|
|
sophi_widget: 'topic',
|
|
}
|
|
: {}),
|
|
}),
|
|
},
|
|
}).json();
|
|
return {
|
|
title: response.result.section.title,
|
|
description: response.result.section.section_about,
|
|
rootUrl,
|
|
response,
|
|
};
|
|
}
|
|
})();
|
|
|
|
let items = response.result.articles.map((e) => ({
|
|
title: e.title,
|
|
link: new URL(e.canonical_url, rootUrl).href,
|
|
guid: e.id,
|
|
pubDate: parseDate(e.published_time),
|
|
updated: parseDate(e.updated_time),
|
|
author: e.authors.map((e) => e.name).join(', '),
|
|
category: e.kicker.names,
|
|
description: e.description,
|
|
}));
|
|
|
|
items = items.filter((e, i) => items.findIndex((f) => e.guid === f.guid) === i);
|
|
|
|
const results = await Promise.allSettled(
|
|
items.map((item) =>
|
|
ctx.query.mode === 'fulltext'
|
|
? ctx.cache.tryGet(item.link, async () => {
|
|
const detailResponse = await got(item.link);
|
|
const content = cheerio.load(detailResponse.data);
|
|
|
|
if (detailResponse.url.startsWith('https://www.reuters.com/investigates/')) {
|
|
const ldJson = JSON.parse(content('script[type="application/ld+json"]').text());
|
|
content('.special-report-article-container .container, #slide-dek, #slide-end, .share-in-article-container').remove();
|
|
|
|
item.title = ldJson.headline;
|
|
item.pubDate = parseDate(ldJson.dateCreated);
|
|
item.author = ldJson.creator;
|
|
item.category = ldJson.keywords;
|
|
item.description = content('.special-report-article-container').html();
|
|
|
|
return item;
|
|
}
|
|
|
|
const matches = content('script#fusion-metadata')
|
|
.text()
|
|
.match(/Fusion.globalContent=({[\S\s]*?});/);
|
|
|
|
if (matches) {
|
|
const data = JSON.parse(matches[1]);
|
|
|
|
item.title = data.result.title || item.title;
|
|
item.description = art(path.join(__dirname, 'templates/description.art'), {
|
|
result: data.result,
|
|
});
|
|
item.pubDate = parseDate(data.result.display_time);
|
|
item.author = data.result.authors.map((author) => author.name).join(', ');
|
|
item.category = data.result.taxonomy.keywords;
|
|
|
|
return item;
|
|
}
|
|
|
|
content('.title').remove();
|
|
content('.article-metadata').remove();
|
|
|
|
item.title = content('meta[property="og:title"]').attr('content');
|
|
item.pubDate = parseDate(detailResponse.data.match(/"datePublished":"(.*?)","dateModified/)[1]);
|
|
item.author = detailResponse.data
|
|
.match(/{"@type":"Person","name":"(.*?)"}/g)
|
|
.map((p) => p.match(/"name":"(.*?)"/)[1])
|
|
.join(', ');
|
|
item.description = content('article').html();
|
|
|
|
return item;
|
|
})
|
|
: item
|
|
)
|
|
);
|
|
items = results.filter((r) => r.status === 'fulfilled').map((r) => r.value);
|
|
|
|
ctx.state.data = {
|
|
title,
|
|
description,
|
|
image: 'https://www.reuters.com/pf/resources/images/reuters/logo-vertical-default-512x512.png?d=116',
|
|
link: `https://www.reuters.com${section_id}`,
|
|
item: items,
|
|
};
|
|
};
|