feat: migrate to node-html-parser

2025-12-23 23:19:40 +01:00 · 2021-11-25 09:46:45 +05:00
parent 1f6088a864
commit 92a8239899
11 changed files with 299 additions and 5414 deletions
--- a/packages/core/mocks/fs.mock.js
+++ b/packages/core/mocks/fs.mock.js
@@ -17,6 +17,8 @@ async function writeEncrypted(filename, { data, type, key }) {
  if (hasItem(filename)) return { hash, hashType };
  fs[filename] = data;
  return {
+    chunkSize: 512,
+    alg: "xcha-stream",
    hash,
    hashType,
    iv: "some iv",
--- a/packages/core/mocks/storage.mock.js
+++ b/packages/core/mocks/storage.mock.js
@@ -62,6 +62,10 @@ async function hash(password, userId) {
  return password;
 }

+async function generateCryptoKey(password) {
+  return password;
+}
+
 module.exports = {
  read,
  readMulti,
@@ -71,6 +75,7 @@ module.exports = {
  encrypt,
  decrypt,
  deriveCryptoKey,
+  generateCryptoKey,
  getCryptoKey,
  getAllKeys,
  hash,
--- a/packages/core/tests/notes.test.js
+++ b/packages/core/tests/notes.test.js
@@ -63,7 +63,7 @@ test("get all notes", () =>
 test("note without a title should get title from content", () =>
  noteTest().then(async ({ db, id }) => {
    let note = db.notes.note(id);
-    expect(note.title).toBe("Hello This is colorful");
+    expect(note.title).toBe("HelloThis is colorful");
  }));

 test("note should get headline from content", () =>
--- a/packages/core/tests/utils/index.js
+++ b/packages/core/tests/utils/index.js
@@ -34,7 +34,8 @@ var TEST_NOTE = {
  },
 };

-const IMG_CONTENT = `<p>This is a note for me.j</p>\n<p><img src="data:image/png;base64,iVBORw0K" class="attachment" alt="Screenshot_20210915_102333.png" data-mime="image/png" data-hash="d3eab72e94e3cd35" data-filename="Screenshot_20210915_102333.png" data-size="68609" style="float: left;" /> &nbsp;</p>\n<p>&nbsp;</p>`;
+const IMG_CONTENT = `<p>This is a note for me.j</p>\n<p><img src="data:image/png;base64,iVBORw0K" data-hash="d3eab72e94e3cd35" class="attachment" alt="Screenshot_20210915_102333.png" data-mime="image/png" data-filename="Screenshot_20210915_102333.png" data-size="68609" style="float: left;" /> &nbsp;</p>\n<p>&nbsp;</p>`;
+const IMG_CONTENT_WITHOUT_HASH = `<p>This is a note for me.j</p>\n<p><img src="data:image/png;base64,iVBORw0K" class="attachment" alt="Screenshot_20210915_102333.png" data-mime="image/png" data-filename="Screenshot_20210915_102333.png" data-size="68609" style="float: left;" /> &nbsp;</p>\n<p>&nbsp;</p>`;

 const LONG_TEXT =
  "Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book. It has survived not only five centuries, but also the leap into electronic typesetting, remaining essentially unchanged. It was popularised in the 1960s with the release of Letraset sheets containing Lorem Ipsum passages, and more recently with desktop publishing software like Aldus PageMaker including versions of Lorem Ipsum.";
@@ -73,6 +74,7 @@ export {
  noteTest,
  groupedTest,
  IMG_CONTENT,
+  IMG_CONTENT_WITHOUT_HASH,
  StorageInterface,
  TEST_NOTEBOOK,
  TEST_NOTEBOOK2,
--- a/packages/core/collections/content.js
+++ b/packages/core/collections/content.js
@@ -109,7 +109,9 @@ export default class Content extends Collection {

    const allAttachments = this._db.attachments.all;
    const content = getContentFromData(contentItem.type, contentItem.data);
-    const { data, attachments } = content.extractAttachments();
+    const { data, attachments } = await content.extractAttachments(
+      (data, type) => this._db.attachments.save(data, type)
+    );

    const noteAttachments = allAttachments.filter((attachment) =>
      hasItem(attachment.noteIds, contentItem.noteId)
@@ -131,18 +133,6 @@ export default class Content extends Collection {
    }

    for (let attachment of toAdd) {
-      if (attachment.data && attachment.type) {
-        const { key, metadata } = await this._db.attachments.save(
-          attachment.data,
-          attachment.type
-        );
-        attachment = {
-          type: attachment.type,
-          filename: metadata.hash,
-          ...metadata,
-          key,
-        };
-      }
      await this._db.attachments.add(attachment, contentItem.noteId);
    }

--- a/packages/core/content-types/tests/tiny.test.js
+++ b/packages/core/content-types/tests/tiny.test.js
@@ -1,12 +1,15 @@
-import { IMG_CONTENT } from "../../__tests__/utils";
+import { IMG_CONTENT, IMG_CONTENT_WITHOUT_HASH } from "../../__tests__/utils";
 import Tiny from "../tiny";

 test("img src is empty after extract attachments", async () => {
-  const tiny = new Tiny(IMG_CONTENT);
-  const result = tiny.extractAttachments();
+  const tiny = new Tiny(IMG_CONTENT_WITHOUT_HASH);
+  const result = await tiny.extractAttachments(async () => {
+    return { key: "hello", metadata: { hash: "helloworld" } };
+  });
  expect(result.attachments.length).toBe(1);
  expect(result.data).not.toContain(`src="data:image/png;`);
  expect(result.data).not.toContain(`src=`);
+  expect(result.data).toContain(`data-hash=\"helloworld\"`);
 });

 test("img src is present after insert attachments", async () => {
--- a/packages/core/content-types/tiny.js
+++ b/packages/core/content-types/tiny.js
@@ -1,14 +1,11 @@
 import showdown from "showdown";
 import { decode, DecodingMode, EntityLevel } from "entities";
+import dataurl from "../utils/dataurl";
+import { parseHTML } from "../utils/html-parser";

 var converter = new showdown.Converter();
 converter.setFlavor("original");

-const HASH_REGEX = /data-hash="(\S+)"/g;
-const SRC_HASH_REGEX =
-  /src="data:(image\/.+);base64,(\S+)"|data-hash="(\S+)"/gm;
-const TAG_REGEX = /<(span|img)[^>]+/gm;
-
 const splitter = /\W+/gm;

 class Tiny {
@@ -22,10 +19,8 @@ class Tiny {
  }

  toTXT() {
-    return decode(
-      this.data.replace(/<br[^>]*>/gi, "\n").replace(/<[^>]+>/g, ""),
-      { level: EntityLevel.HTML, mode: DecodingMode.Strict }
-    ).trim();
+    const document = parseHTML(this.data);
+    return document.textContent || document.body.textContent;
  }

  toMD() {
@@ -60,90 +55,80 @@ class Tiny {
  }

  async insertMedia(getData) {
-    let document = this.data;
+    let document = parseHTML(this.data);
+    const attachmentElements = document.querySelectorAll("img");

-    const matches = Array.from(this.data.matchAll(TAG_REGEX));
-    for (let i = 0; i < matches.length; ++i) {
-      const match = matches[i];
-      const nodeName = match[1];
-
-      if (nodeName === "img") {
-        const node = match[0];
-
-        const hashMatch = node.match(HASH_REGEX);
-        if (!hashMatch) continue;
-
-        const [attribute, hash] = hashMatch;
+    for (var i = 0; i < attachmentElements.length; ++i) {
+      const attachment = attachmentElements[i];
+      switch (attachment.tagName) {
+        case "IMG": {
+          const hash = getDatasetAttribute(attachment, "hash");
+          if (!hash) continue;

          const src = await getData(hash, {
-          total: matches.length,
+            total: attachmentElements.length,
            current: i,
          });
          if (!src) continue;
-
-        const srcAttribute = createAttribute("src", src);
-        const replacedNode = replaceAt(
-          node,
-          hashMatch.index,
-          attribute,
-          srcAttribute
-        );
-        document = replaceAt(this.data, match.index, node, replacedNode);
+          attachment.setAttribute("src", src);
+          break;
        }
      }
-    return document;
+    }
+    return document.outerHTML || document.body.innerHTML;
  }

-  extractAttachments() {
+  async extractAttachments(store) {
    const attachments = [];
-    let document = this.data;
+    let document = parseHTML(this.data);

-    for (let match of this.data.matchAll(TAG_REGEX)) {
-      const nodeName = match[1];
-      const node = match[0];
-      const attachment = { hash: undefined, data: undefined, type: undefined };
+    const attachmentElements = document.querySelectorAll("img,span");

-      switch (nodeName) {
-        case "img": {
-          const replacedNode = node.replace(
-            SRC_HASH_REGEX,
-            (match, mime, data, hash) => {
-              if (mime) attachment.type = mime;
-              if (data) attachment.data = data;
-              if (hash) attachment.hash = hash;
-              return match.startsWith("src") ? "" : match;
+    for (var i = 0; i < attachmentElements.length; ++i) {
+      const attachment = attachmentElements[i];
+      switch (attachment.tagName) {
+        case "IMG": {
+          if (!getDatasetAttribute(attachment, "hash")) {
+            const src = attachment.getAttribute("src");
+            if (!src) continue;
+
+            const { data, mime } = dataurl.toObject(src);
+            if (!data) continue;
+
+            const type =
+              getDatasetAttribute(attachment, "mime") || mime || "image/jpeg";
+            const { key, metadata } = await store(data, "base64");
+            setDatasetAttribute(attachment, "hash", metadata.hash);
+
+            attachments.push({
+              type,
+              filename:
+                getDatasetAttribute(attachment, "filename") || metadata.hash,
+              ...metadata,
+              key,
+            });
+          } else {
+            attachments.push({
+              hash: getDatasetAttribute(attachment, "hash"),
+            });
          }
-          );
-          document = replaceAt(this.data, match.index, node, replacedNode);
+          attachment.removeAttribute("src");
          break;
        }
-        case "span": {
-          const matches = HASH_REGEX.exec(node);
-          if (!matches) continue;
-          const [_match, hash] = matches;
-          attachments.push({ hash });
+        default: {
+          if (!getDatasetAttribute(attachment, "hash")) continue;
+          attachments.push({
+            hash: getDatasetAttribute(attachment, "hash"),
+          });
          break;
        }
      }
-      if (attachment.hash || attachment.data) attachments.push(attachment);
    }
-
-    return { data: document, attachments };
+    return { data: document.outerHTML || document.body.innerHTML, attachments };
  }
 }
 export default Tiny;

-function replaceAt(str, index, match, replacement) {
-  let start = str.slice(0, index);
-  start += replacement;
-  start += str.slice(index + match.length);
-  return start;
-}
-
-function createAttribute(key, value) {
-  return `${key}="${value}"`;
-}
-
 function getHeadlineFromText(text) {
  for (var i = 0; i < text.length; ++i) {
    const char = text[i];
@@ -176,3 +161,11 @@ function getTitleFromText(text) {
  }
  return title;
 }
+
+function getDatasetAttribute(element, attribute) {
+  return element.getAttribute(`data-${attribute}`);
+}
+
+function setDatasetAttribute(element, attribute, value) {
+  return element.setAttribute(`data-${attribute}`, value);
+}
--- a/packages/core/package-lock.json
+++ b/packages/core/package-lock.json
@@ -1,12 +1,12 @@
 {
  "name": "notes-core",
-  "version": "6.10.0",
+  "version": "6.11.0",
  "lockfileVersion": 2,
  "requires": true,
  "packages": {
    "": {
      "name": "notes-core",
-      "version": "6.10.0",
+      "version": "6.11.0",
      "dependencies": {
        "@stablelib/blake2s": "^1.0.1",
        "async-mutex": "^0.3.2",
@@ -14,6 +14,7 @@
        "entities": "^3.0.1",
        "fast-sort": "^2.0.1",
        "no-internet": "^1.5.2",
+        "node-html-parser": "^5.1.0",
        "qclone": "^1.0.4",
        "showdown": "https://github.com/thecodrr/showdown",
        "spark-md5": "^3.0.1"
@@ -3014,6 +3015,11 @@
      "integrity": "sha1-8kiPMleC9m0XSEL0gZkuL6ulbzg=",
      "dev": true
    },
+    "node_modules/boolbase": {
+      "version": "1.0.0",
+      "resolved": "https://registry.npmjs.org/boolbase/-/boolbase-1.0.0.tgz",
+      "integrity": "sha1-aN/1++YMUes3cl6p4+0xDcwed24="
+    },
    "node_modules/brace-expansion": {
      "version": "1.1.11",
      "resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-1.1.11.tgz",
@@ -3365,6 +3371,32 @@
        "node": ">=4.8"
      }
    },
+    "node_modules/css-select": {
+      "version": "4.1.3",
+      "resolved": "https://registry.npmjs.org/css-select/-/css-select-4.1.3.tgz",
+      "integrity": "sha512-gT3wBNd9Nj49rAbmtFHj1cljIAOLYSX1nZ8CB7TBO3INYckygm5B7LISU/szY//YmdiSLbJvDLOx9VnMVpMBxA==",
+      "dependencies": {
+        "boolbase": "^1.0.0",
+        "css-what": "^5.0.0",
+        "domhandler": "^4.2.0",
+        "domutils": "^2.6.0",
+        "nth-check": "^2.0.0"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/fb55"
+      }
+    },
+    "node_modules/css-what": {
+      "version": "5.1.0",
+      "resolved": "https://registry.npmjs.org/css-what/-/css-what-5.1.0.tgz",
+      "integrity": "sha512-arSMRWIIFY0hV8pIxZMEfmMI47Wj3R/aWpZDDxWYCPEiOMv6tfOrnpDtgxBYPEQD4V0Y/958+1TdC3iWTFcUPw==",
+      "engines": {
+        "node": ">= 6"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/fb55"
+      }
+    },
    "node_modules/cssom": {
      "version": "0.3.8",
      "resolved": "https://registry.npmjs.org/cssom/-/cssom-0.3.8.tgz",
@@ -3592,6 +3624,38 @@
        "node": ">= 6"
      }
    },
+    "node_modules/dom-serializer": {
+      "version": "1.3.2",
+      "resolved": "https://registry.npmjs.org/dom-serializer/-/dom-serializer-1.3.2.tgz",
+      "integrity": "sha512-5c54Bk5Dw4qAxNOI1pFEizPSjVsx5+bpJKmL2kPn8JhBUq2q09tTCa3mjijun2NfK78NMouDYNMBkOrPZiS+ig==",
+      "dependencies": {
+        "domelementtype": "^2.0.1",
+        "domhandler": "^4.2.0",
+        "entities": "^2.0.0"
+      },
+      "funding": {
+        "url": "https://github.com/cheeriojs/dom-serializer?sponsor=1"
+      }
+    },
+    "node_modules/dom-serializer/node_modules/entities": {
+      "version": "2.2.0",
+      "resolved": "https://registry.npmjs.org/entities/-/entities-2.2.0.tgz",
+      "integrity": "sha512-p92if5Nz619I0w+akJrLZH0MX0Pb5DX39XOwQTtXSdQQOaYH03S1uIQp4mhOZtAXrxq4ViO67YTiLBo2638o9A==",
+      "funding": {
+        "url": "https://github.com/fb55/entities?sponsor=1"
+      }
+    },
+    "node_modules/domelementtype": {
+      "version": "2.2.0",
+      "resolved": "https://registry.npmjs.org/domelementtype/-/domelementtype-2.2.0.tgz",
+      "integrity": "sha512-DtBMo82pv1dFtUmHyr48beiuq792Sxohr+8Hm9zoxklYPfa6n0Z3Byjj2IV7bmr2IyqClnqEQhfgHJJ5QF0R5A==",
+      "funding": [
+        {
+          "type": "github",
+          "url": "https://github.com/sponsors/fb55"
+        }
+      ]
+    },
    "node_modules/domexception": {
      "version": "2.0.1",
      "resolved": "https://registry.npmjs.org/domexception/-/domexception-2.0.1.tgz",
@@ -3613,6 +3677,33 @@
        "node": ">=8"
      }
    },
+    "node_modules/domhandler": {
+      "version": "4.2.2",
+      "resolved": "https://registry.npmjs.org/domhandler/-/domhandler-4.2.2.tgz",
+      "integrity": "sha512-PzE9aBMsdZO8TK4BnuJwH0QT41wgMbRzuZrHUcpYncEjmQazq8QEaBWgLG7ZyC/DAZKEgglpIA6j4Qn/HmxS3w==",
+      "dependencies": {
+        "domelementtype": "^2.2.0"
+      },
+      "engines": {
+        "node": ">= 4"
+      },
+      "funding": {
+        "url": "https://github.com/fb55/domhandler?sponsor=1"
+      }
+    },
+    "node_modules/domutils": {
+      "version": "2.8.0",
+      "resolved": "https://registry.npmjs.org/domutils/-/domutils-2.8.0.tgz",
+      "integrity": "sha512-w96Cjofp72M5IIhpjgobBimYEfoPjx1Vx0BSX9P30WBdZW2WIKU0T1Bd0kz2eNZ9ikjKgHbEyKx8BB6H1L3h3A==",
+      "dependencies": {
+        "dom-serializer": "^1.0.1",
+        "domelementtype": "^2.2.0",
+        "domhandler": "^4.2.0"
+      },
+      "funding": {
+        "url": "https://github.com/fb55/domutils?sponsor=1"
+      }
+    },
    "node_modules/dotenv": {
      "version": "10.0.0",
      "resolved": "https://registry.npmjs.org/dotenv/-/dotenv-10.0.0.tgz",
@@ -4211,6 +4302,14 @@
        "node": ">=0.10.0"
      }
    },
+    "node_modules/he": {
+      "version": "1.2.0",
+      "resolved": "https://registry.npmjs.org/he/-/he-1.2.0.tgz",
+      "integrity": "sha512-F/1DnUGPopORZi0ni+CvrCgHQ5FyEAHRLSApuYWMmrbSwoN2Mn/7k+Gl38gJnR7yyDZk6WLXwiGod1JOWNDKGw==",
+      "bin": {
+        "he": "bin/he"
+      }
+    },
    "node_modules/home-or-tmp": {
      "version": "2.0.0",
      "resolved": "https://registry.npmjs.org/home-or-tmp/-/home-or-tmp-2.0.0.tgz",
@@ -6020,6 +6119,15 @@
        "node": "4.x || >=6.0.0"
      }
    },
+    "node_modules/node-html-parser": {
+      "version": "5.1.0",
+      "resolved": "https://registry.npmjs.org/node-html-parser/-/node-html-parser-5.1.0.tgz",
+      "integrity": "sha512-l6C1Gf1o7YuxeMGa17PypEez/rj+ii3q4/NZG37nRmWSLDjHyB0WNrlE4h2UW92D0JSfUSfu+lOvxThttVe7Jw==",
+      "dependencies": {
+        "css-select": "^4.1.3",
+        "he": "1.2.0"
+      }
+    },
    "node_modules/node-int64": {
      "version": "0.4.0",
      "resolved": "https://registry.npmjs.org/node-int64/-/node-int64-0.4.0.tgz",
@@ -6090,6 +6198,17 @@
        "node": ">=4"
      }
    },
+    "node_modules/nth-check": {
+      "version": "2.0.1",
+      "resolved": "https://registry.npmjs.org/nth-check/-/nth-check-2.0.1.tgz",
+      "integrity": "sha512-it1vE95zF6dTT9lBsYbxvqh0Soy4SPowchj0UBGj/V6cTPnXXtQOPUbhZ6CmGzAD/rW22LQK6E96pcdJXk4A4w==",
+      "dependencies": {
+        "boolbase": "^1.0.0"
+      },
+      "funding": {
+        "url": "https://github.com/fb55/nth-check?sponsor=1"
+      }
+    },
    "node_modules/nwsapi": {
      "version": "2.2.0",
      "resolved": "https://registry.npmjs.org/nwsapi/-/nwsapi-2.2.0.tgz",
@@ -10737,6 +10856,11 @@
      "integrity": "sha1-8kiPMleC9m0XSEL0gZkuL6ulbzg=",
      "dev": true
    },
+    "boolbase": {
+      "version": "1.0.0",
+      "resolved": "https://registry.npmjs.org/boolbase/-/boolbase-1.0.0.tgz",
+      "integrity": "sha1-aN/1++YMUes3cl6p4+0xDcwed24="
+    },
    "brace-expansion": {
      "version": "1.1.11",
      "resolved": "https://registry.npmjs.org/brace-expansion/-/brace-expansion-1.1.11.tgz",
@@ -11040,6 +11164,23 @@
        "which": "^1.2.9"
      }
    },
+    "css-select": {
+      "version": "4.1.3",
+      "resolved": "https://registry.npmjs.org/css-select/-/css-select-4.1.3.tgz",
+      "integrity": "sha512-gT3wBNd9Nj49rAbmtFHj1cljIAOLYSX1nZ8CB7TBO3INYckygm5B7LISU/szY//YmdiSLbJvDLOx9VnMVpMBxA==",
+      "requires": {
+        "boolbase": "^1.0.0",
+        "css-what": "^5.0.0",
+        "domhandler": "^4.2.0",
+        "domutils": "^2.6.0",
+        "nth-check": "^2.0.0"
+      }
+    },
+    "css-what": {
+      "version": "5.1.0",
+      "resolved": "https://registry.npmjs.org/css-what/-/css-what-5.1.0.tgz",
+      "integrity": "sha512-arSMRWIIFY0hV8pIxZMEfmMI47Wj3R/aWpZDDxWYCPEiOMv6tfOrnpDtgxBYPEQD4V0Y/958+1TdC3iWTFcUPw=="
+    },
    "cssom": {
      "version": "0.3.8",
      "resolved": "https://registry.npmjs.org/cssom/-/cssom-0.3.8.tgz",
@@ -11224,6 +11365,28 @@
      "integrity": "sha512-Dj6Wk3tWyTE+Fo1rW8v0Xhwk80um6yFYKbuAxc9c3EZxIHFDYwbi34Uk42u1CdnIiVorvt4RmlSDjIPyzGC2ew==",
      "dev": true
    },
+    "dom-serializer": {
+      "version": "1.3.2",
+      "resolved": "https://registry.npmjs.org/dom-serializer/-/dom-serializer-1.3.2.tgz",
+      "integrity": "sha512-5c54Bk5Dw4qAxNOI1pFEizPSjVsx5+bpJKmL2kPn8JhBUq2q09tTCa3mjijun2NfK78NMouDYNMBkOrPZiS+ig==",
+      "requires": {
+        "domelementtype": "^2.0.1",
+        "domhandler": "^4.2.0",
+        "entities": "^2.0.0"
+      },
+      "dependencies": {
+        "entities": {
+          "version": "2.2.0",
+          "resolved": "https://registry.npmjs.org/entities/-/entities-2.2.0.tgz",
+          "integrity": "sha512-p92if5Nz619I0w+akJrLZH0MX0Pb5DX39XOwQTtXSdQQOaYH03S1uIQp4mhOZtAXrxq4ViO67YTiLBo2638o9A=="
+        }
+      }
+    },
+    "domelementtype": {
+      "version": "2.2.0",
+      "resolved": "https://registry.npmjs.org/domelementtype/-/domelementtype-2.2.0.tgz",
+      "integrity": "sha512-DtBMo82pv1dFtUmHyr48beiuq792Sxohr+8Hm9zoxklYPfa6n0Z3Byjj2IV7bmr2IyqClnqEQhfgHJJ5QF0R5A=="
+    },
    "domexception": {
      "version": "2.0.1",
      "resolved": "https://registry.npmjs.org/domexception/-/domexception-2.0.1.tgz",
@@ -11241,6 +11404,24 @@
        }
      }
    },
+    "domhandler": {
+      "version": "4.2.2",
+      "resolved": "https://registry.npmjs.org/domhandler/-/domhandler-4.2.2.tgz",
+      "integrity": "sha512-PzE9aBMsdZO8TK4BnuJwH0QT41wgMbRzuZrHUcpYncEjmQazq8QEaBWgLG7ZyC/DAZKEgglpIA6j4Qn/HmxS3w==",
+      "requires": {
+        "domelementtype": "^2.2.0"
+      }
+    },
+    "domutils": {
+      "version": "2.8.0",
+      "resolved": "https://registry.npmjs.org/domutils/-/domutils-2.8.0.tgz",
+      "integrity": "sha512-w96Cjofp72M5IIhpjgobBimYEfoPjx1Vx0BSX9P30WBdZW2WIKU0T1Bd0kz2eNZ9ikjKgHbEyKx8BB6H1L3h3A==",
+      "requires": {
+        "dom-serializer": "^1.0.1",
+        "domelementtype": "^2.2.0",
+        "domhandler": "^4.2.0"
+      }
+    },
    "dotenv": {
      "version": "10.0.0",
      "resolved": "https://registry.npmjs.org/dotenv/-/dotenv-10.0.0.tgz",
@@ -11703,6 +11884,11 @@
        }
      }
    },
+    "he": {
+      "version": "1.2.0",
+      "resolved": "https://registry.npmjs.org/he/-/he-1.2.0.tgz",
+      "integrity": "sha512-F/1DnUGPopORZi0ni+CvrCgHQ5FyEAHRLSApuYWMmrbSwoN2Mn/7k+Gl38gJnR7yyDZk6WLXwiGod1JOWNDKGw=="
+    },
    "home-or-tmp": {
      "version": "2.0.0",
      "resolved": "https://registry.npmjs.org/home-or-tmp/-/home-or-tmp-2.0.0.tgz",
@@ -13161,6 +13347,15 @@
      "integrity": "sha512-8dG4H5ujfvFiqDmVu9fQ5bOHUC15JMjMY/Zumv26oOvvVJjM67KF8koCWIabKQ1GJIa9r2mMZscBq/TbdOcmNA==",
      "dev": true
    },
+    "node-html-parser": {
+      "version": "5.1.0",
+      "resolved": "https://registry.npmjs.org/node-html-parser/-/node-html-parser-5.1.0.tgz",
+      "integrity": "sha512-l6C1Gf1o7YuxeMGa17PypEez/rj+ii3q4/NZG37nRmWSLDjHyB0WNrlE4h2UW92D0JSfUSfu+lOvxThttVe7Jw==",
+      "requires": {
+        "css-select": "^4.1.3",
+        "he": "1.2.0"
+      }
+    },
    "node-int64": {
      "version": "0.4.0",
      "resolved": "https://registry.npmjs.org/node-int64/-/node-int64-0.4.0.tgz",
@@ -13222,6 +13417,14 @@
        "path-key": "^2.0.0"
      }
    },
+    "nth-check": {
+      "version": "2.0.1",
+      "resolved": "https://registry.npmjs.org/nth-check/-/nth-check-2.0.1.tgz",
+      "integrity": "sha512-it1vE95zF6dTT9lBsYbxvqh0Soy4SPowchj0UBGj/V6cTPnXXtQOPUbhZ6CmGzAD/rW22LQK6E96pcdJXk4A4w==",
+      "requires": {
+        "boolbase": "^1.0.0"
+      }
+    },
    "nwsapi": {
      "version": "2.2.0",
      "resolved": "https://registry.npmjs.org/nwsapi/-/nwsapi-2.2.0.tgz",
--- a/packages/core/package.json
+++ b/packages/core/package.json
@@ -29,6 +29,7 @@
    "entities": "^3.0.1",
    "fast-sort": "^2.0.1",
    "no-internet": "^1.5.2",
+    "node-html-parser": "^5.1.0",
    "qclone": "^1.0.4",
    "showdown": "https://github.com/thecodrr/showdown",
    "spark-md5": "^3.0.1"
--- a/packages/core/utils/html-parser.js
+++ b/packages/core/utils/html-parser.js
@@ -0,0 +1,6 @@
+import { parse } from "node-html-parser";
+
+export const parseHTML =
+  typeof DOMParser === "undefined"
+    ? (input) => parse(input)
+    : (input) => new DOMParser().parseFromString(input, "text/html");
--- a/packages/core/yarn.lock
+++ b/packages/core/yarn.lock