From 29c1bd6202ca517504b34fb741b4b4eaaa435d62 Mon Sep 17 00:00:00 2001
From: Ali Akbar <52413434+Ali-D-Akbar@users.noreply.github.com>
Date: Tue, 23 Apr 2024 19:58:06 +0500
Subject: [PATCH] fix: avoid span tags overlapping (#4330)

---
 course_discovery/apps/course_metadata/tests/test_utils.py | 2 ++
 course_discovery/apps/course_metadata/utils.py            | 4 ++--
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/course_discovery/apps/course_metadata/tests/test_utils.py b/course_discovery/apps/course_metadata/tests/test_utils.py
index a4f35d8808..67cf955213 100644
--- a/course_discovery/apps/course_metadata/tests/test_utils.py
+++ b/course_discovery/apps/course_metadata/tests/test_utils.py
@@ -807,6 +807,8 @@ class UtilsTests(TestCase):
         ('<p><span lang="en">with lang</span></p>', '<p><span lang="en">with lang</span></p>'),
         ('<p><span class="body" lang="en">lang and class</span></p>', '<p><span lang="en">lang and class</span></p>'),
         ('<p><span class="body">class only</span></p>', '<p>class only</p>'),
+        ('<p dir="rtl"><span lang="en">1 span tag inside RTL p</span></p>', '<p dir="rtl"><span lang="en">1 span tag inside RTL p</span></p>'),
+        ('<p dir="rtl"><span lang="en"><span lang="en"><span lang="en">multiple span tags inside RTL p</span></span></span></p>', '<p dir="rtl"><span lang="en">multiple span tags inside RTL p</span></p>'),
 
         # A sample text from real life when pasting from Microsoft Word into a rich text editor
         # pylint: disable=line-too-long
diff --git a/course_discovery/apps/course_metadata/utils.py b/course_discovery/apps/course_metadata/utils.py
index 99bb240ec0..3c945db836 100644
--- a/course_discovery/apps/course_metadata/utils.py
+++ b/course_discovery/apps/course_metadata/utils.py
@@ -732,7 +732,7 @@ def handle_tag(self, tag, attrs, start):
         if not self.is_p_tag_with_dir:
             super().handle_tag(tag, attrs, start)
 
-        elif tag not in HTML_TAGS_ATTRIBUTE_WHITELIST:
+        elif tag not in HTML_TAGS_ATTRIBUTE_WHITELIST and tag != 'span':
             if start:
                 self.outtextf(f'<{tag}')
                 if attrs:
@@ -743,7 +743,7 @@ def handle_tag(self, tag, attrs, start):
                 self.outtextf(f'</{tag}>')
 
         if tag == 'span':
-            if attrs and start and 'lang' in dict(attrs):
+            if attrs and start and not self.in_lang_span and 'lang' in dict(attrs):
                 self.outtextf(f'<span lang="{dict(attrs)["lang"]}">')
                 self.in_lang_span = True
             if not start and self.in_lang_span: