[FIX] website, website_blog: remove stars from blog description

Current behavior:
When adding a "Heading" in the first 200 characters of a blog with
/Heading, the short description of the blog preview would have
unwanted stars (*) around the heading

Steps to reproduce:
- Create a blog article
- In the first 200 characters of the blog use a heading (e.g. /Heading1)
- Go back to the blog list, the description of the new article contains
  unwanted stars (*)

opw-2798595

closes odoo/odoo#88352

X-original-commit: 6cb57d2f1ffb0adceaa3199cc0bf5140589d9328
Signed-off-by: Romain Derie (rde) <rde@odoo.com>

addons/website/models/mixins.py

@@ -366,7 +366,6 @@ class WebsiteSearchableMixin(models.AbstractModel):
             for result, data in zip(self, results_data):
                 for html_field in html_fields:
                     if data[html_field]:
-                        text = text_from_html(data[html_field])
-                        text = re.sub('\\s+', ' ', text).strip()
+                        text = text_from_html(data[html_field], True)
                         data[html_field] = text
         return results_data

addons/website/tools.py

@@ -1,5 +1,6 @@
 # Part of Odoo. See LICENSE file for full copyright and licensing details.
 import contextlib
+import re
 from lxml import etree
 from unittest.mock import Mock, MagicMock, patch
 
@@ -144,7 +145,7 @@ def similarity_score(s1, s2):
     score -= len(set1.symmetric_difference(s2)) / (len(s1) + len(s2))
     return score
 
-def text_from_html(html_fragment):
+def text_from_html(html_fragment, collapse_whitespace=False):
     """
     Returns the plain non-tag text from an html
 
@@ -154,4 +155,7 @@ def text_from_html(html_fragment):
     """
     # lxml requires one single root element
     tree = etree.fromstring('<p>%s</p>' % html_fragment, etree.XMLParser(recover=True))
-    return ' '.join(tree.itertext())
+    content = ' '.join(tree.itertext())
+    if collapse_whitespace:
+        content = re.sub('\\s+', ' ', content).strip()
+    return content

addons/website_blog/models/website_blog.py

@@ -6,9 +6,9 @@ import random
 
 from odoo import api, models, fields, _
 from odoo.addons.http_routing.models.ir_http import slug, unslug
+from odoo.addons.website.tools import text_from_html
 from odoo.tools.json import scriptsafe as json_scriptsafe
 from odoo.tools.translate import html_translate
-from odoo.tools import html2plaintext
 
 
 class Blog(models.Model):
@@ -197,7 +197,7 @@ class BlogPost(models.Model):
             if blog_post.teaser_manual:
                 blog_post.teaser = blog_post.teaser_manual
             else:
-                content = html2plaintext(blog_post.content).replace('\n', ' ')
+                content = text_from_html(blog_post.content, True)
                 blog_post.teaser = content[:200] + '...'
 
     def _set_teaser(self):

addons/website_blog/tests/test_website_blog_flow.py

@@ -108,3 +108,11 @@ class TestWebsiteBlogFlow(TestWebsiteBlogCommon):
 
         self.assertFalse(self.env['mail.message'].sudo().search(
             [('model', '=', 'blog.post'), ('attachment_ids', 'in', second_attachment.ids)]))
+
+    def test_website_blog_teaser_content(self):
+        """ Make sure that the content of the post is correctly rendered in
+            proper plain text. """
+
+        self.test_blog_post.content = "<h2>Test Content</h2>"
+
+        self.assertEqual(self.test_blog_post.teaser, "Test Content...")