soxoj · soxoj · Apr 4, 2026 · Apr 4, 2026
diff --git a/socid_extractor/main.py b/socid_extractor/main.py
@@ -1,3 +1,4 @@
+import html as html_module
 import logging
 from http.cookies import SimpleCookie
 
@@ -120,7 +121,15 @@ def extract(page):
             else:
                 groupdict = regexp_group.groupdict()
                 if groupdict:
-                    values = groupdict
+                    values = {}
+                    for k, v in groupdict.items():
+                        if k.endswith('_raw'):
+                            base = k[:-4]
+                            if not values.get(base):
+                                values[base] = v
+                        else:
+                            if not values.get(k):
+                                values[k] = v
                 else:
                     extracted = regexp_group.group(1)
                     logging.debug('Extracted: %s', extracted)
@@ -145,7 +154,8 @@ def extract(page):
             except PROCESS_ERRORS as e:
                 logging.debug('Postprocess error: ', e)
 
-        return {a: b for a, b in values.items() if b or type(b) == bool}
+        return {a: html_module.unescape(b) if isinstance(b, str) else b
+                for a, b in values.items() if b or type(b) == bool}
 
     # all schemes have been checked
     return {}
diff --git a/socid_extractor/schemes.py b/socid_extractor/schemes.py
@@ -2490,7 +2490,7 @@
     'Threads': {
         'url_hints': ('threads.net', 'threads.com'),
         'flags': ['Threads, Say more', 'barcelona'],
-        'regex': r'og:title" content="(?P<fullname>[^"]*?) \((?:@|&#064;)(?P<username>[^)]+)\)[^"]*Threads[\s\S]*?og:description" content="(?P<follower_count>[\d,]+) Followers[^"]*?(?P<posts_count>[\d,]+) Threads[^"]*?(?:&quot;|")(?P<bio>[^"&]*)',
+        'regex': r'og:title" content="(?P<fullname>[^"]*?) \((?:@|&#064;)(?P<username>[^)]+)\)[^"]*Threads[\s\S]*?og:description" content="(?P<follower_count>[\d,]+) Followers[^"]*?(?P<posts_count>[\d,]+) Threads(?:[^"]*?(?:&#x2022;|\xb7) (?:&quot;(?P<bio>[^&]+)&quot;|(?P<bio_raw>[^".]*)\.))?',
     },
     'Smule': {
         'url_hints': ('smule.com',),

diff --git a/tests/test_socid_improvements.py b/tests/test_socid_improvements.py
@@ -931,8 +931,8 @@ def test_threads_profile_extraction():
     """Threads: extract fullname, username, follower/post counts, bio from OG tags."""
     html = (
         '<!DOCTYPE html><html><head>'
-        '<meta property="og:title" content="Mark Zuckerberg (@zuck) · Threads, Say more">'
-        '<meta property="og:description" content="12,500 Followers · 340 Threads · &quot;CEO of Meta&quot;">'
+        '<meta property="og:title" content="Mark Zuckerberg (&#064;zuck) &#x2022; Threads, Say more">'
+        '<meta property="og:description" content="12,500 Followers &#x2022; 340 Threads &#x2022; &quot;CEO of Meta&quot;. See the latest.">'
         '</head><body>'
         '<div class="barcelona">content</div>'
         '</body></html>'
@@ -1190,12 +1190,66 @@ def test_threads_profile():
     )
     info = extract(html)
     assert info.get('username') == 'fusteee'
-    assert info.get('fullname') == 'Marc Fuste&#xe9;'
+    assert info.get('fullname') == 'Marc Fusteé'
     assert info.get('follower_count') == '33'
     assert info.get('posts_count') == '0'
     assert info.get('bio') == 'Regalame tus mejores noches'
 
 
+def test_threads_profile_no_bio():
+    """Threads: profile without bio should not capture HTML garbage."""
+    html = (
+        '<!DOCTYPE html><html><head>'
+        '<meta property="og:title" content="Alice (&#064;alice_test) &#x2022; Threads, Say more">'
+        '<meta property="og:description" content="5 Followers &#x2022; 0 Threads. See the latest conversations with &#064;alice_test.">'
+        '</head><body>'
+        '<div class="barcelona">content</div>'
+        '</body></html>'
+    )
+    info = extract(html)
+    assert info.get('username') == 'alice_test'
+    assert info.get('fullname') == 'Alice'
+    assert info.get('follower_count') == '5'
+    assert info.get('posts_count') == '0'
+    assert not info.get('bio')
+
+
+def test_threads_profile_unicode_name():
+    """Threads: fullname with unicode HTML entities should be decoded."""
+    html = (
+        '<!DOCTYPE html><html><head>'
+        '<meta property="og:title" content="&#x1d4d0;&#x1d4fb;&#x1d4fd; (&#064;bob_test) &#x2022; Threads, Say more">'
+        '<meta property="og:description" content="10 Followers &#x2022; 3 Threads &#x2022; &quot;hello&quot;. See the latest conversations.">'
+        '</head><body>'
+        '<div class="barcelona">content</div>'
+        '</body></html>'
+    )
+    info = extract(html)
+    assert info.get('username') == 'bob_test'
+    assert info.get('fullname') == '\U0001d4d0\U0001d4fb\U0001d4fd'
+    assert info.get('follower_count') == '10'
+    assert info.get('posts_count') == '3'
+    assert info.get('bio') == 'hello'
+
+
+def test_threads_profile_emoji_bio():
+    """Threads: emoji bio without quotes should be extracted, not lost."""
+    html = (
+        '<!DOCTYPE html><html><head>'
+        '<meta property="og:title" content="Eve (&#064;eve_test) &#x2022; Threads, Say more">'
+        '<meta property="og:description" content="21 Followers &#x2022; 4 Threads &#x2022; &#x1f92b;. See the latest conversations.">'
+        '</head><body>'
+        '<div class="barcelona">content</div>'
+        '</body></html>'
+    )
+    info = extract(html)
+    assert info.get('username') == 'eve_test'
+    assert info.get('follower_count') == '21'
+    assert info.get('posts_count') == '4'
+    assert info.get('bio') == '🤫'
+    assert not info.get('bio_raw')
+
+
 def test_chess_com_html_profile():
     """Chess.com HTML: extract fullname, username and image from og:meta."""
     html = (