Fix/drupal wiki (improve table & url handling) (#4097)

* feat: add support for custom table formatting in htmlToText conversion

* fix tables

* feat: improve plain text table formatting for AI readability

* fix options

* improve drupal wiki connector

* final fix

* adjust leading slash to match code

* linting

---------

Co-authored-by: timothycarambat <rambat1010@gmail.com>

collector/utils/extensions/DrupalWiki/DrupalWiki/index.js

@@ -219,7 +219,9 @@ class DrupalWiki {
       pageId: pageId,
       accessToken: this.accessToken,
     };
-    return `drupalwiki://${this.baseUrl}?payload=${encryptionWorker.encrypt(
+    return `drupalwiki://${
+      this.baseUrl
+    }/node/${pageId}?payload=${encryptionWorker.encrypt(
       JSON.stringify(payload)
     )}`;
   }
@@ -259,17 +261,27 @@ class DrupalWiki {
    * @private
    */
   #processPageBody({ body, url, title, lastModified }) {
-    // use the title as content if there is none
     const textContent = body.trim() !== "" ? body : title;
 
     const plainTextContent = htmlToText(textContent, {
       wordwrap: false,
       preserveNewlines: true,
+      selectors: [
+        {
+          selector: "table",
+          format: "dataTable",
+          options: {
+            colSpacing: 3,
+            rowSpacing: 1,
+            uppercaseHeaderCells: true,
+            maxColumnWidth: Infinity,
+          },
+        },
+      ],
     });
-    // preserve structure
+
     const plainBody = plainTextContent.replace(/\n{3,}/g, "\n\n");
-    // add the link to the document
-    return `Link/URL: ${url}\n\n${plainBody}`;
+    return plainBody;
   }
 
   async #downloadAndProcessAttachments(pageId) {