[CORL-735] Scraper support for JSON Linked Data (#2671)

* feat: add support for application/json+ld parsing

- upgraded scraper packages to use new jsonld parser
- waiting on https://github.com/microlinkhq/metascraper/pull/225
  for merging
- fixes #2670

* chore: metascraper upgrades

src/core/server/services/stories/scraper/rules/helpers.ts

@@ -0,0 +1,4 @@
+import { Rule, Ruler } from "metascraper";
+
+export const wrap = (rule: Rule): Ruler => ({ htmlDom, url }) =>
+  rule(htmlDom, url);

src/core/server/services/stories/scraper/rules/modified.ts

@@ -1,8 +1,12 @@
+import { $jsonld } from "@metascraper/helpers";
 import { Rules } from "metascraper";
 
+import { wrap } from "./helpers";
+
 export const modifiedScraper = (): Rules => ({
   modified: [
     // From: http://ogp.me/#type_article
-    ({ htmlDom: $ }) => $('meta[property="article:modified"]').attr("content"),
+    wrap($jsonld("dateModified")),
+    wrap($ => $('meta[property="article:modified"]').attr("content")),
   ],
 });

src/core/server/services/stories/scraper/rules/section.ts

@@ -1,8 +1,12 @@
+import { $jsonld } from "@metascraper/helpers";
 import { Rules } from "metascraper";
 
+import { wrap } from "./helpers";
+
 export const sectionScraper = (): Rules => ({
   section: [
     // From: http://ogp.me/#type_article
-    ({ htmlDom: $ }) => $('meta[property="article:section"]').attr("content"),
+    wrap($jsonld("articleSection")),
+    wrap($ => $('meta[property="article:section"]').attr("content")),
   ],
 });

src/types/metascraper.d.ts

@@ -1,17 +1,21 @@
 declare module "metascraper" {
   export interface Scraper {
-    (
-      options: {
-        url: string;
-        html: string;
-      }
-    ): Promise<Record<string, string | undefined>>;
+    (options: { url: string; html: string }): Promise<
+      Record<string, string | undefined>
+    >;
   }
 
-  export type Rules = Record<
-    string,
-    Array<(options: { htmlDom: CheerioSelector }) => string | undefined>
-  >;
+  export type Ruler = (options: {
+    htmlDom: CheerioSelector;
+    url: string;
+  }) => string | undefined;
+
+  export type Rule = (
+    htmlDom: CheerioSelector,
+    url: string
+  ) => string | undefined;
+
+  export type Rules = Record<string, Array<Ruler>>;
 
   export function load(rules: Rules[]): Scraper;
 }
@@ -40,3 +44,7 @@ declare module "metascraper-title" {
   import { Rules } from "metascraper";
   export default function def(): Rules;
 }
+
+declare module "@metascraper/helpers" {
+  export const $jsonld: any;
+}