feat(sdk, plugins): jsx support

packages/plugins/src/engines/habr.tsx

@@ -0,0 +1,13 @@
+import { Engine } from '@txtdot/sdk';
+
+import { JSX } from '@txtdot/sdk';
+
+const Habr = new Engine('Habr', 'Habr parser', ['*']);
+
+Habr.route('*path', async (input, ro) => {
+  return {
+    content: <div>Test</div>,
+  };
+});
+
+export default Habr;

packages/plugins/src/engines/index.ts

@@ -1,5 +1,6 @@
 import StackOverflow from './stackoverflow';
 import Readability from './readability';
 import SearX from './searx';
+import Habr from './habr';
 
-export { StackOverflow, Readability, SearX };
+export { StackOverflow, Readability, SearX, Habr };

packages/plugins/src/engines/readability.ts

@@ -18,7 +18,7 @@ Readability.route('*path', async (input, ro) => {
   }
 
   return {
-    document: parseHTML(parsed.content).document,
+    content: parsed.content,
     textContent: parsed.textContent,
     title: parsed.title,
     lang: parsed.lang,

packages/plugins/src/engines/searx.ts

@@ -46,7 +46,7 @@ async function search(
   const textContent = articles_parsed.map((a) => a.text).join('');
 
   return {
-    document: parseHTML(content).document,
+    content: content,
     textContent,
     title: `${search} - Searx - Page ${page}`,
     lang: document.documentElement.lang,

packages/plugins/src/engines/stackoverflow/questions.ts

@@ -16,9 +16,7 @@ async function questions(
   const answers = allAnswers.map((a) => postParser(a));
 
   return {
-    document: parseHTML(
+    content: `${question}<hr>${answers.length} answers <hr>${answers.join('<hr>')}`,
-      `${question}<hr>${answers.length} answers <hr>${answers.join('<hr>')}`
-    ).document,
     textContent: `${ro.q.id}/${ro.q.slug}\nText output not supported`, // TODO
     title,
     lang: document.documentElement.lang,

packages/plugins/src/engines/stackoverflow/users.ts

@@ -27,8 +27,7 @@ async function users(
     .join('<br/>');
 
   return {
-    document: parseHTML(`${userInfo}<hr><h3>Top Posts</h3>${topPosts}`)
+    content: `${userInfo}<hr><h3>Top Posts</h3>${topPosts}`,
-      .document,
     textContent: `${ro.q.id}/${ro.q.slug}\n`, // TODO
     title: document.querySelector('title')?.textContent || '',
     lang: document.documentElement.lang,

packages/plugins/tsconfig.json

@@ -13,13 +13,13 @@
     /* Language and Environment */
     "target": "ES2020" /* Set the JavaScript language version for emitted JavaScript and include compatible library declarations. */,
     // "lib": [],                                        /* Specify a set of bundled library declaration files that describe the target runtime environment. */
-    // "jsx": "preserve",                                /* Specify what JSX code is generated. */
+    "jsx": "react" /* Specify what JSX code is generated. */,
     // "experimentalDecorators": true,                   /* Enable experimental support for legacy experimental decorators. */
     // "emitDecoratorMetadata": true,                    /* Emit design-type metadata for decorated declarations in source files. */
     // "jsxFactory": "",                                 /* Specify the JSX factory function used when targeting React JSX emit, e.g. 'React.createElement' or 'h'. */
     // "jsxFragmentFactory": "",                         /* Specify the JSX Fragment reference used for fragments when targeting React JSX emit e.g. 'React.Fragment' or 'Fragment'. */
     // "jsxImportSource": "",                            /* Specify module specifier used to import the JSX factory functions when using 'jsx: react-jsx*'. */
-    // "reactNamespace": "",                             /* Specify the object invoked for 'createElement'. This only applies when targeting 'react' JSX emit. */
+    "reactNamespace": "JSX" /* Specify the object invoked for 'createElement'. This only applies when targeting 'react' JSX emit. */,
     // "noLib": true,                                    /* Disable including any library files, including the default lib.d.ts. */
     // "useDefineForClassFields": true,                  /* Emit ECMAScript-standard-compliant class fields. */
     // "moduleDetection": "auto",                        /* Control what method is used to detect module-format JS files. */

packages/sdk/src/jsx.ts

@@ -0,0 +1,24 @@
+// eslint-disable-next-line @typescript-eslint/no-namespace
+export namespace JSX {
+  export type Element = string;
+  export interface IntrinsicElements {
+    // eslint-disable-next-line @typescript-eslint/no-explicit-any
+    [elemName: string]: any;
+  }
+}
+
+export function createElement(
+  name: string,
+  props: { [id: string]: string },
+  ...content: string[]
+) {
+  props = props || {};
+  const propsstr = Object.keys(props)
+    .map((key) => {
+      const value = props[key];
+      if (key === 'className') return `class=${value}`;
+      else return `${key}=${value}`;
+    })
+    .join(' ');
+  return `<${name} ${propsstr}>${content.join('')}</${name}>`;
+}

packages/sdk/src/lib.ts

@@ -18,6 +18,8 @@ import {
   handlerSchema,
 } from './types/handler';
 
+import * as JSX from './jsx';
+
 export {
   Engine,
   EngineParseError,
@@ -32,4 +34,5 @@ export {
   HandlerOutput,
   Route,
   handlerSchema,
+  JSX,
 };

packages/sdk/src/types/handler.ts

@@ -33,7 +33,7 @@ export interface HandlerOutput {
 }
 
 export interface EngineOutput {
-  document: Document;
+  content: string;
   textContent?: string;
   title?: string;
   lang?: string;

packages/server/src/distributor.ts

@@ -9,6 +9,7 @@ import replaceHref from './utils/replace-href';
 import { Engine } from '@txtdot/sdk';
 import { HandlerInput, HandlerOutput } from '@txtdot/sdk';
 import config from './config';
+import { parseHTML } from 'linkedom';
 
 interface IEngineId {
   [key: string]: number;
@@ -51,6 +52,7 @@ export class Distributor {
     }
 
     const engine = this.getFallbackEngine(urlObj.hostname, engineName);
+
     const output = await engine.handle(
       new HandlerInput(
         await decodeStream(data, parseEncodingName(mime)),
@@ -58,22 +60,25 @@ export class Distributor {
       )
     );
 
+    const dom = parseHTML(output.content);
+
     // post-process
     // TODO: generate dom in handler and not parse here twice
     replaceHref(
-      output.document,
+      dom.document,
       requestUrl,
       new URL(remoteUrl),
       engineName,
       redirectPath
     );
 
-    const purify = DOMPurify();
+    const purify = DOMPurify(dom);
-    const content = purify.sanitize(output.document.toString());
+    const content = purify.sanitize(output.content);
 
     return {
       content,
-      textContent: output.textContent || output.document.textContent || '',
+      textContent:
+        output.textContent || dom.document.documentElement.textContent || '',
       title: output.title,
       lang: output.lang,
     };