core: use upstream htmlparser2 instead of in-house

This is mostly because I am not 100% certain about how stable our own
optimized version is. While it does perform a lot better, there is
significant risk that things might break in production. To avoid
that, I am replacing it with the upstream version which is much more
heavily tested.

packages/core/__tests__/notes.test.js

@@ -348,7 +348,7 @@ test("note content should not contain image base64 data after save", () =>
     const note = db.notes.note(id);
     const content = await note.content();
     expect(content).not.toContain(`src="data:image/png;`);
-    expect(content).toContain(`src=""`);
+    expect(content).not.toContain(`src=`);
   }));
 
 test("adding a note with an invalid tag should clean the tag array", () =>

packages/core/content-types/__tests__/tiptap.test.js

@@ -27,7 +27,7 @@ test("img src is empty after extract attachments", async () => {
   });
   expect(result.attachments).toHaveLength(1);
   expect(result.data).not.toContain(`src="data:image/png;`);
-  expect(result.data).toContain(`src=""`);
+  expect(result.data).not.toContain(`src=""`);
   expect(result.data).toContain(`data-hash="helloworld"`);
 });
 

packages/core/content-types/tiptap.js

@@ -20,7 +20,7 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
 import showdown from "@streetwriters/showdown";
 import dataurl from "../utils/dataurl";
 import { extractFirstParagraph, getDummyDocument } from "../utils/html-parser";
-import { Attributes, HTMLParser, HTMLRewriter } from "../utils/html-rewriter";
+import { HTMLParser, HTMLRewriter } from "../utils/html-rewriter";
 import { convert } from "html-to-text";
 import { list } from "html-to-text/lib/formatter";
 
@@ -101,7 +101,7 @@ export class Tiptap {
     let hashes = [];
     new HTMLParser({
       ontag: (name, attr) => {
-        const hash = Attributes.get(attr, ATTRIBUTES.hash);
+        const hash = attr[ATTRIBUTES.hash];
         if (name === "img" && hash) hashes.push(hash);
       }
     }).parse(this.data);
@@ -119,12 +119,11 @@ export class Tiptap {
 
     return new HTMLRewriter({
       ontag: (name, attr) => {
-        const hash = Attributes.get(attr, ATTRIBUTES.hash);
+        const hash = attr[ATTRIBUTES.hash];
         if (name === "img" && hash) {
-          const src = images[Attributes.get(attr, ATTRIBUTES.hash)];
+          const src = images[hash];
           if (!src) return;
-
-          return { name, attr: Attributes.set(attr, ATTRIBUTES.src, src) };
+          attr[ATTRIBUTES.src] = src;
         }
       }
     }).transform(this.data);
@@ -137,8 +136,7 @@ export class Tiptap {
   removeAttachments(hashes) {
     return new HTMLRewriter({
       ontag: (_name, attr) => {
-        if (hashes.includes(Attributes.get(attr, ATTRIBUTES.hash)))
-          return false;
+        if (hashes.includes(attr[ATTRIBUTES.hash])) return false;
       }
     }).transform(this.data);
   }
@@ -147,8 +145,8 @@ export class Tiptap {
     let sources = [];
     new HTMLParser({
       ontag: (name, attr, pos) => {
-        const hash = Attributes.get(attr, ATTRIBUTES.hash);
-        const src = Attributes.get(attr, ATTRIBUTES.src);
+        const hash = attr[ATTRIBUTES.hash];
+        const src = attr[ATTRIBUTES.src];
         if (name === "img" && !hash && src) {
           sources.push({
             src,
@@ -178,16 +176,13 @@ export class Tiptap {
       ontag: (name, attr, pos) => {
         switch (name) {
           case "img": {
-            const hash = Attributes.get(attr, ATTRIBUTES.hash);
+            const hash = attr[ATTRIBUTES.hash];
 
             if (hash) {
               attachments.push({
                 hash
               });
-              return {
-                name,
-                attr: Attributes.set(attr, ATTRIBUTES.src, "")
-              };
+              delete attr[ATTRIBUTES.src];
             } else {
               const imageData = images[`${pos.start}${pos.end}`];
               if (!imageData) return imageData;
@@ -195,10 +190,8 @@ export class Tiptap {
               const { key, metadata, mime } = imageData;
               if (!metadata.hash) return;
 
-              const type =
-                Attributes.get(attr, ATTRIBUTES.mime) || mime || "image/jpeg";
-              const filename =
-                Attributes.get(attr, ATTRIBUTES.filename) || metadata.hash;
+              const type = attr[ATTRIBUTES.mime] || mime || "image/jpeg";
+              const filename = attr[ATTRIBUTES.filename] || metadata.hash;
 
               attachments.push({
                 type,
@@ -207,18 +200,13 @@ export class Tiptap {
                 key
               });
 
-              return {
-                name,
-                attr: Attributes.set(
-                  Attributes.set(attr, ATTRIBUTES.hash, metadata.hash),
-                  ATTRIBUTES.src,
-                  ""
-                )
-              };
+              attr[ATTRIBUTES.hash] = metadata.hash;
+              delete attr[ATTRIBUTES.src];
             }
+            break;
           }
           case "span": {
-            const hash = Attributes.get(attr, ATTRIBUTES.hash);
+            const hash = attr[ATTRIBUTES.hash];
             if (!hash) return;
             attachments.push({
               hash

packages/core/package.json

@@ -40,13 +40,16 @@
     "async-mutex": "^0.3.2",
     "dayjs": "^1.11.3",
     "entities": "^4.3.1",
-    "htmlparser2": "github:thecodrr/htmlparser2",
     "html-to-text": "github:thecodrr/node-html-to-text",
+    "htmlparser2": "^8.0.1",
     "linkedom": "^0.14.17",
     "liqe": "^1.13.0",
     "qclone": "^1.2.0",
     "spark-md5": "^3.0.2"
   },
+  "overrides": {
+    "htmlparser2": "^8.0.1"
+  },
   "np": {
     "releaseDraft": false,
     "message": "chore: bump version to %s"

packages/core/utils/html-rewriter.js

@@ -17,7 +17,7 @@ You should have received a copy of the GNU General Public License
 along with this program.  If not, see <http://www.gnu.org/licenses/>.
 */
 
-import { Parser, Attributes } from "htmlparser2";
+import { Parser } from "htmlparser2";
 
 export class HTMLRewriter {
   /**
@@ -73,7 +73,12 @@ export class HTMLRewriter {
           }
 
           this.write(`<${name}`);
-          if (attr) this.write(` ${attr}`);
+          if (attr) {
+            for (const key in attr) {
+              if (!key) continue;
+              this.write(` ${key}="${attr[key]}"`);
+            }
+          }
           this.currentTag = name;
         },
         onclosetag: (name, isImplied) => {
@@ -111,8 +116,7 @@ export class HTMLRewriter {
         decodeEntities: false,
         lowerCaseAttributeNames: false,
         lowerCaseTags: false,
-        recognizeCDATA: false,
-        parseAttributes: false
+        recognizeCDATA: false
       }
     );
   }
@@ -171,8 +175,7 @@ export class HTMLParser {
         decodeEntities: false,
         lowerCaseAttributeNames: false,
         lowerCaseTags: false,
-        recognizeCDATA: false,
-        parseAttributes: false
+        recognizeCDATA: false
       }
     );
   }
@@ -182,5 +185,3 @@ export class HTMLParser {
     this.parser.reset();
   }
 }
-
-export { Attributes };