Fix encoding issues in paper_data function and add tests for text decoding

main.py

@@ -31,14 +31,20 @@ def paper_data(paper_data, wait_time=5):
                     continue
                 try:
                     try:
-                        escaped_title = repr(title.encode('latin1').decode('unicode-escape')).strip()
+                        try:
+                            escaped_title = repr(title.encode('latin1').decode('unicode-escape', errors='replace')).strip()
+                        except:
+                            escaped_title = repr(title).strip().encode('latin1', errors='replace').decode('utf-8', errors='replace')
                     except:
                         escaped_title = repr(title).strip()
                     title = html.escape(str(escaped_title).strip()[1:-1])
                     try:
-                        encoded_bytes = citation.encode('latin1').decode('utf-8', errors='replace')
+                        try:
+                            encoded_bytes = citation.encode('latin1').decode('unicode-escape', errors='replace')
+                        except:
+                            encoded_bytes = repr(citation).strip().encode('latin1').decode('utf-8', errors='replace')
                     except:
-                        encoded_bytes = repr(citation)
+                        encoded_bytes = repr(citation).strip()
                     citation = html.unescape(encoded_bytes)
                     status = post_blog(doi, title, category, summary, mindmap, citation, access_key, wait_time)
                 except Exception as e:

test.py

@@ -0,0 +1,7 @@
+# Original text with incorrect encoding
+text = "Itâ\x80\x99s the AIâ\x80\x99s fault, not mine: Mind perception increases blame attribution to AI"
+
+# Decode as 'latin1' and re-encode as 'utf-8'
+fixed_text = text.encode('latin1').decode('utf-8')
+
+print(fixed_text)