academic-refchecker 1.2.41__tar.gz → 1.2.43__tar.gz

{academic_refchecker-1.2.41/src/academic_refchecker.egg-info → academic_refchecker-1.2.43}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academic-refchecker
-Version: 1.2.41
+Version: 1.2.43
 Summary: A comprehensive tool for validating reference accuracy in academic papers
 Author-email: Mark Russinovich <markrussinovich@hotmail.com>
 License-Expression: MIT

{academic_refchecker-1.2.41 → academic_refchecker-1.2.43}/src/__version__.py

@@ -1,3 +1,3 @@
 """Version information for RefChecker."""
 
-__version__ = "1.2.41"
+__version__ = "1.2.43"

{academic_refchecker-1.2.41 → academic_refchecker-1.2.43/src/academic_refchecker.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academic-refchecker
-Version: 1.2.41
+Version: 1.2.43
 Summary: A comprehensive tool for validating reference accuracy in academic papers
 Author-email: Mark Russinovich <markrussinovich@hotmail.com>
 License-Expression: MIT

{academic_refchecker-1.2.41 → academic_refchecker-1.2.43}/src/checkers/openreview_checker.py

@@ -473,9 +473,10 @@ class OpenReviewReferenceChecker:
         
         if cited_venue and paper_venue:
             if are_venues_substantially_different(cited_venue, paper_venue):
+                from utils.error_utils import clean_venue_for_comparison
                 errors.append({
                     "warning_type": "venue",
-                    "warning_details": f"Venue mismatch: cited as '{cited_venue}' but OpenReview shows '{paper_venue}'"
+                    "warning_details": f"Venue mismatch: cited as '{clean_venue_for_comparison(cited_venue)}' but OpenReview shows '{clean_venue_for_comparison(paper_venue)}'"
                 })
         
         # Create verified data structure

{academic_refchecker-1.2.41 → academic_refchecker-1.2.43}/src/checkers/semantic_scholar.py

@@ -544,11 +544,8 @@ class NonArxivReferenceChecker:
         if cited_venue and paper_venue:
             # Use the utility function to check if venues are substantially different
             if are_venues_substantially_different(cited_venue, paper_venue):
-                errors.append({
-                    'warning_type': 'venue',
-                    'warning_details': f"Venue mismatch: cited as '{cited_venue}' but actually '{paper_venue}'",
-                    'ref_venue_correct': paper_venue
-                })
+                from utils.error_utils import create_venue_warning
+                errors.append(create_venue_warning(cited_venue, paper_venue))
         elif not cited_venue and paper_venue:
             # Check if this is an arXiv paper first
             external_ids = paper_data.get('externalIds', {})

{academic_refchecker-1.2.41 → academic_refchecker-1.2.43}/src/core/parallel_processor.py

@@ -279,8 +279,11 @@ class ParallelReferenceProcessor:
         from utils.text_utils import format_authors_for_display
         authors = format_authors_for_display(reference.get('authors', []))
         year = reference.get('year', '')
-        # Get venue from either 'venue' or 'journal' field
+        # Get venue from either 'venue' or 'journal' field and clean it up
         venue = reference.get('venue', '') or reference.get('journal', '')
+        if venue:
+            from utils.error_utils import clean_venue_for_comparison
+            venue = clean_venue_for_comparison(venue)
         url = reference.get('url', '')
         doi = reference.get('doi', '')
         

{academic_refchecker-1.2.41 → academic_refchecker-1.2.43}/src/core/refchecker.py

@@ -5407,25 +5407,6 @@ class ArxivReferenceChecker:
                 if error_details:
                     subreason = self._categorize_unverified_reason(error_details)
                     print(f"         Subreason: {subreason}")
-            
-            year_str = self._format_year_string(reference.get('year'))
-            
-            # Apply LaTeX cleaning and formatting to authors for display
-            authors = reference.get('authors', [])
-            if authors:
-                from utils.text_utils import strip_latex_commands, format_authors_for_display
-                cleaned_authors = [strip_latex_commands(author) for author in authors]
-                authors_display = format_authors_for_display(cleaned_authors)
-            else:
-                authors_display = 'Unknown authors'
-                
-            # Only show URL if it exists and is different from reference_url
-            ref_url = reference.get('url', '').strip()
-            if ref_url and ref_url != reference_url:
-                # Clean trailing punctuation from URL display
-                from utils.url_utils import clean_url_punctuation
-                clean_ref_url = clean_url_punctuation(ref_url)
-                print(f"          URL: {clean_ref_url}")
 
     def _categorize_unverified_reason(self, error_details):
         """Categorize the unverified error into checker error or not found"""

{academic_refchecker-1.2.41 → academic_refchecker-1.2.43}/src/utils/bibtex_parser.py

@@ -103,37 +103,99 @@ def parse_bibtex_entry_content(entry_type: str, entry_key: str, content: str) ->
     Returns:
         Dictionary with parsed entry data
     """
-    # Extract fields using regex
     fields = {}
     
-    # Pattern to match field = {value} or field = "value"
-    # Handle nested braces properly
-    field_pattern = r'(\w+)\s*=\s*(?:\{([^{}]*(?:\{[^{}]*\}[^{}]*)*)\}|"([^"]*)")'
-    
-    for match in re.finditer(field_pattern, content, re.DOTALL):
-        field_name = match.group(1).lower()
-        field_value = match.group(2) or match.group(3) or ""
-        # Strip outer quotes if present (handles cases like title = {"Some Title"})
-        field_value = field_value.strip()
-        if field_value.startswith('"') and field_value.endswith('"'):
-            field_value = field_value[1:-1]
-        fields[field_name] = field_value
+    # Use a more robust approach with manual parsing
+    i = 0
+    while i < len(content):
+        # Skip whitespace
+        while i < len(content) and content[i].isspace():
+            i += 1
+        
+        if i >= len(content):
+            break
+        
+        # Look for field name
+        field_start = i
+        while i < len(content) and (content[i].isalnum() or content[i] == '_'):
+            i += 1
+        
+        if i == field_start:
+            i += 1  # Skip non-alphanumeric character
+            continue
+        
+        field_name = content[field_start:i].lower()
+        
+        # Skip whitespace
+        while i < len(content) and content[i].isspace():
+            i += 1
+        
+        # Look for equals sign
+        if i >= len(content) or content[i] != '=':
+            continue
+        i += 1  # Skip '='
+        
+        # Skip whitespace
+        while i < len(content) and content[i].isspace():
+            i += 1
+        
+        if i >= len(content):
+            break
+        
+        # Parse field value
+        field_value = ""
+        if content[i] == '"':
+            # Handle quoted strings
+            i += 1  # Skip opening quote
+            value_start = i
+            while i < len(content) and content[i] != '"':
+                i += 1
+            if i < len(content):
+                field_value = content[value_start:i]
+                i += 1  # Skip closing quote
+        elif content[i] == '{':
+            # Handle braced strings with proper nesting
+            brace_count = 0
+            value_start = i + 1  # Skip opening brace
+            i += 1
+            while i < len(content):
+                if content[i] == '{':
+                    brace_count += 1
+                elif content[i] == '}':
+                    if brace_count == 0:
+                        break
+                    brace_count -= 1
+                i += 1
+            
+            if i < len(content):
+                field_value = content[value_start:i]
+                i += 1  # Skip closing brace
+        
+        if field_value:
+            field_value = field_value.strip()
+            # Strip outer quotes if present (handles cases like title = {"Some Title"})
+            if field_value.startswith('"') and field_value.endswith('"'):
+                field_value = field_value[1:-1]
+            fields[field_name] = field_value
+        
+        # Skip to next field (look for comma)
+        while i < len(content) and content[i] not in ',}':
+            i += 1
+        if i < len(content) and content[i] == ',':
+            i += 1
     
-    # If field extraction failed, try a simpler approach
+    # Fallback to regex if manual parsing failed
     if not fields:
-        logger.debug("Field extraction failed, trying line-by-line approach")
-        lines = content.split('\n')
-        for line in lines:
-            line = line.strip()
-            if '=' in line:
-                field_match = re.match(r'(\w+)\s*=\s*[{"]([^{}"]*)[}"]', line)
-                if field_match:
-                    field_name = field_match.group(1).lower()
-                    field_value = field_match.group(2).strip()
-                    # Strip outer quotes if present
-                    if field_value.startswith('"') and field_value.endswith('"'):
-                        field_value = field_value[1:-1]
-                    fields[field_name] = field_value
+        logger.debug("Manual parsing failed, trying regex approach")
+        field_pattern = r'(\w+)\s*=\s*(?:\{([^{}]*(?:\{[^{}]*\}[^{}]*)*)\}|"([^"]*)")'
+        
+        for match in re.finditer(field_pattern, content, re.DOTALL):
+            field_name = match.group(1).lower()
+            field_value = match.group(2) or match.group(3) or ""
+            field_value = field_value.strip()
+            if field_value.startswith('"') and field_value.endswith('"'):
+                field_value = field_value[1:-1]
+            fields[field_name] = field_value
     
     return {
         'type': entry_type,
@@ -216,6 +278,9 @@ def parse_bibtex_references(bibliography_text: str) -> List[Dict[str, Any]]:
         
         # Extract journal/venue
         journal = fields.get('journal', fields.get('booktitle', fields.get('venue', '')))
+        # Remove braces from journal/venue names
+        if journal and journal.startswith('{') and journal.endswith('}'):
+            journal = journal[1:-1]
         
         # Extract DOI and construct URL
         doi = fields.get('doi', '')
@@ -225,6 +290,9 @@ def parse_bibtex_references(bibliography_text: str) -> List[Dict[str, Any]]:
         
         # Extract other URLs
         url = fields.get('url', '')
+        if url:
+            from utils.url_utils import clean_url
+            url = clean_url(url)
         
         # Handle special @misc entries with only howpublished field
         if not title and not authors and entry_type == 'misc':
@@ -249,6 +317,10 @@ def parse_bibtex_references(bibliography_text: str) -> List[Dict[str, Any]]:
                         else:
                             url = howpublished
                         
+                        # Clean the reconstructed URL
+                        from utils.url_utils import clean_url
+                        url = clean_url(url)
+                        
                         # Generate title from domain/path
                         if 'jailbreakchat.com' in domain:
                             title = 'JailbreakChat Website'
@@ -275,6 +347,11 @@ def parse_bibtex_references(bibliography_text: str) -> List[Dict[str, Any]]:
             
         if url.startswith('\\url{') and url.endswith('}'):
             url = url[5:-1]  # Remove \url{...}
+            
+        # Clean any URL we extracted
+        if url:
+            from utils.url_utils import clean_url
+            url = clean_url(url)
         
         # Construct ArXiv URL from eprint field if no URL present
         if not url and not doi_url:

{academic_refchecker-1.2.41 → academic_refchecker-1.2.43}/src/utils/error_utils.py

@@ -89,6 +89,20 @@ def create_title_error(error_details: str, correct_title: str) -> Dict[str, str]
     }
 
 
+def clean_venue_for_comparison(venue: str) -> str:
+    """
+    Clean venue name for display in warnings using the shared normalization logic.
+    
+    Args:
+        venue: Raw venue string
+        
+    Returns:
+        Cleaned venue name suitable for display
+    """
+    from utils.text_utils import normalize_venue_for_display
+    return normalize_venue_for_display(venue)
+
+
 def create_venue_warning(cited_venue: str, correct_venue: str) -> Dict[str, str]:
     """
     Create a standardized venue warning dictionary.
@@ -100,9 +114,13 @@ def create_venue_warning(cited_venue: str, correct_venue: str) -> Dict[str, str]
     Returns:
         Standardized warning dictionary
     """
+    # Clean both venues for display in the warning
+    clean_cited = clean_venue_for_comparison(cited_venue)
+    clean_correct = clean_venue_for_comparison(correct_venue)
+    
     return {
         'warning_type': 'venue',
-        'warning_details': f"Venue mismatch: cited as '{cited_venue}' but actually '{correct_venue}'",
+        'warning_details': f"Venue mismatch: cited as '{clean_cited}' but actually '{clean_correct}'",
         'ref_venue_correct': correct_venue
     }
 

{academic_refchecker-1.2.41 → academic_refchecker-1.2.43}/src/utils/text_utils.py

@@ -11,6 +11,69 @@ from typing import List
 logger = logging.getLogger(__name__)
 
 
+def expand_abbreviations(text: str) -> str:
+    """
+    Generic abbreviation expansion using common academic patterns.
+    
+    This function expands common academic abbreviations to their full forms
+    to improve venue name matching and comparison.
+    
+    Args:
+        text: Text containing potential abbreviations
+        
+    Returns:
+        Text with abbreviations expanded
+    """
+    if not text:
+        return text
+        
+    common_abbrevs = {
+        # IEEE specific abbreviations (only expand with periods, not full words)
+        'robot.': 'robotics', 'autom.': 'automation', 'lett.': 'letters',
+        'trans.': 'transactions', 'syst.': 'systems', 'netw.': 'networks',
+        'learn.': 'learning', 'ind.': 'industrial', 'electron.': 'electronics',
+        'mechatron.': 'mechatronics', 'intell.': 'intelligence',
+        'transp.': 'transportation', 'contr.': 'control', 'mag.': 'magazine',
+        # General academic abbreviations (only expand with periods)
+        'int.': 'international', 'intl.': 'international', 'conf.': 'conference',
+        'j.': 'journal', 'proc.': 'proceedings', 'assoc.': 'association',
+        'comput.': 'computing', 'sci.': 'science', 'eng.': 'engineering',
+        'tech.': 'technology', 'artif.': 'artificial', 'mach.': 'machine',
+        'stat.': 'statistics', 'math.': 'mathematics', 'phys.': 'physics',
+        'chem.': 'chemistry', 'bio.': 'biology', 'med.': 'medicine',
+        'adv.': 'advances', 'ann.': 'annual', 'symp.': 'symposium',
+        'workshop': 'workshop', 'worksh.': 'workshop',
+        'natl.': 'national', 'acad.': 'academy', 'rev.': 'review',
+        # Physics journal abbreviations
+        'phys.': 'physics', 'phys. rev.': 'physical review', 
+        'phys. rev. lett.': 'physical review letters',
+        'phys. rev. a': 'physical review a', 'phys. rev. b': 'physical review b',
+        'phys. rev. c': 'physical review c', 'phys. rev. d': 'physical review d',
+        'phys. rev. e': 'physical review e', 'phys. lett.': 'physics letters',
+        'phys. lett. b': 'physics letters b', 'nucl. phys.': 'nuclear physics',
+        'nucl. phys. a': 'nuclear physics a', 'nucl. phys. b': 'nuclear physics b',
+        'j. phys.': 'journal of physics', 'ann. phys.': 'annals of physics',
+        'mod. phys. lett.': 'modern physics letters', 'eur. phys. j.': 'european physical journal',
+        # Nature journals
+        'nature phys.': 'nature physics', 'sci. adv.': 'science advances',
+        # Handle specific multi-word patterns and well-known acronyms
+        'proc. natl. acad. sci.': 'proceedings of the national academy of sciences',
+        'pnas': 'proceedings of the national academy of sciences',
+        'neurips': 'neural information processing systems',
+    }
+    
+    # Sort by length (longest first) to ensure longer matches take precedence
+    for abbrev, expansion in sorted(common_abbrevs.items(), key=lambda x: len(x[0]), reverse=True):
+        # For abbreviations ending in period, use word boundary at start only
+        if abbrev.endswith('.'):
+            pattern = r'\b' + re.escape(abbrev)
+        else:
+            pattern = r'\b' + re.escape(abbrev) + r'\b'
+        text = re.sub(pattern, expansion, text)
+    
+    return text
+
+
 def normalize_apostrophes(text):
     """
     Normalize all apostrophe variants to standard ASCII apostrophe
@@ -2255,8 +2318,13 @@ def format_author_for_display(author_name):
     if not author_name:
         return author_name
     
+    # Clean up any stray punctuation that might have been attached during parsing
+    author_name = author_name.strip()
+    # Remove trailing semicolons that sometimes get attached during bibliographic parsing
+    author_name = re.sub(r'[;,]\s*$', '', author_name)
+    
     # Normalize apostrophes for consistent display
-    author_name = normalize_apostrophes(author_name.strip())
+    author_name = normalize_apostrophes(author_name)
     
     # Check if it's in "Lastname, Firstname" format
     if ',' in author_name:
@@ -3667,8 +3735,77 @@ def are_venues_substantially_different(venue1: str, venue2: str) -> bool:
         return bool(venue1 != venue2)
     
     # Clean LaTeX commands from both venues first
-    venue1 = strip_latex_commands(venue1)
-    venue2 = strip_latex_commands(venue2)
+    venue1_latex_cleaned = strip_latex_commands(venue1)
+    venue2_latex_cleaned = strip_latex_commands(venue2)
+    
+    # For comparison, we need lowercase normalized versions
+    def normalize_for_comparison(venue_text):
+        # Get the cleaned display version first
+        cleaned = normalize_venue_for_display(venue_text)
+        # Then normalize for comparison: lowercase, expand abbreviations, remove punctuation
+        venue_lower = cleaned.lower()
+        
+        # Handle LaTeX penalty commands before abbreviation expansion
+        venue_lower = re.sub(r'\\penalty\d+\s*', ' ', venue_lower)  # Remove \\penalty0 etc
+        venue_lower = re.sub(r'\s+', ' ', venue_lower).strip()  # Clean up extra spaces
+        
+        # Expand abbreviations for comparison
+        def expand_abbreviations(text):
+            common_abbrevs = {
+                # IEEE specific abbreviations (only expand with periods, not full words)
+                'robot.': 'robotics', 'autom.': 'automation', 'lett.': 'letters',
+                'trans.': 'transactions', 'syst.': 'systems', 'netw.': 'networks',
+                'learn.': 'learning', 'ind.': 'industrial', 'electron.': 'electronics',
+                'mechatron.': 'mechatronics', 'intell.': 'intelligence',
+                'transp.': 'transportation', 'contr.': 'control', 'mag.': 'magazine',
+                # General academic abbreviations (only expand with periods)
+                'int.': 'international', 'intl.': 'international', 'conf.': 'conference',
+                'j.': 'journal', 'proc.': 'proceedings', 'assoc.': 'association',
+                'comput.': 'computing', 'sci.': 'science', 'eng.': 'engineering',
+                'tech.': 'technology', 'artif.': 'artificial', 'mach.': 'machine',
+                'stat.': 'statistics', 'math.': 'mathematics', 'phys.': 'physics',
+                'chem.': 'chemistry', 'bio.': 'biology', 'med.': 'medicine',
+                'adv.': 'advances', 'ann.': 'annual', 'symp.': 'symposium',
+                'workshop': 'workshop', 'worksh.': 'workshop',
+                'natl.': 'national', 'acad.': 'academy', 'rev.': 'review',
+                # Physics journal abbreviations
+                'phys.': 'physics', 'phys. rev.': 'physical review', 
+                'phys. rev. lett.': 'physical review letters',
+                'phys. rev. a': 'physical review a', 'phys. rev. b': 'physical review b',
+                'phys. rev. c': 'physical review c', 'phys. rev. d': 'physical review d',
+                'phys. rev. e': 'physical review e', 'phys. lett.': 'physics letters',
+                'phys. lett. b': 'physics letters b', 'nucl. phys.': 'nuclear physics',
+                'nucl. phys. a': 'nuclear physics a', 'nucl. phys. b': 'nuclear physics b',
+                'j. phys.': 'journal of physics', 'ann. phys.': 'annals of physics',
+                'mod. phys. lett.': 'modern physics letters', 'eur. phys. j.': 'european physical journal',
+                # Nature journals
+                'nature phys.': 'nature physics', 'sci. adv.': 'science advances',
+                # Handle specific multi-word patterns and well-known acronyms
+                'proc. natl. acad. sci.': 'proceedings of the national academy of sciences',
+                'pnas': 'proceedings of the national academy of sciences',
+            }
+            # Sort by length (longest first) to ensure longer matches take precedence
+            for abbrev, expansion in sorted(common_abbrevs.items(), key=lambda x: len(x[0]), reverse=True):
+                # For abbreviations ending in period, use word boundary at start only
+                if abbrev.endswith('.'):
+                    pattern = r'\b' + re.escape(abbrev)
+                else:
+                    pattern = r'\b' + re.escape(abbrev) + r'\b'
+                text = re.sub(pattern, expansion, text)
+            return text
+        
+        venue_lower = expand_abbreviations(venue_lower)
+        
+        # Remove punctuation and normalize spacing for comparison
+        venue_lower = re.sub(r'[.,;:]', '', venue_lower)  # Remove punctuation
+        venue_lower = re.sub(r'\\s+on\\s+', ' ', venue_lower)  # Remove \"on\" preposition
+        venue_lower = re.sub(r'\\s+for\\s+', ' ', venue_lower)  # Remove \"for\" preposition
+        venue_lower = re.sub(r'\\s+', ' ', venue_lower).strip()  # Normalize whitespace
+        
+        return venue_lower
+    
+    normalized_venue1 = normalize_for_comparison(venue1_latex_cleaned)
+    normalized_venue2 = normalize_for_comparison(venue2_latex_cleaned)
     
     def expand_abbreviations(text):
         """Generic abbreviation expansion using common academic patterns"""
@@ -3985,8 +4122,8 @@ def are_venues_substantially_different(venue1: str, venue2: str) -> bool:
             if not acronym or not full_text:
                 return False
             
-            # Normalize the full text
-            normalized_full = normalize_venue(full_text)
+            # Use the internal comparison normalization function
+            normalized_full = normalize_for_comparison(full_text)
             
             # Generate all possible acronyms from the full text
             possible_acronyms = []
@@ -4100,9 +4237,9 @@ def are_venues_substantially_different(venue1: str, venue2: str) -> bool:
         if (arxiv1 == 'arxiv' and arxiv2.startswith('https://arxiv.org')) or (arxiv2 == 'arxiv' and arxiv1.startswith('https://arxiv.org')):
             return False
     
-    # Normalize both venues first
-    norm1 = normalize_venue(venue1)
-    norm2 = normalize_venue(venue2)
+    # Use normalized venues from shared function
+    norm1 = normalized_venue1
+    norm2 = normalized_venue2
     
     # Direct match after normalization (highest priority)
     if norm1 == norm2:
@@ -4356,4 +4493,144 @@ def is_year_substantially_different(cited_year: int, correct_year: int, context:
     
     # Any year difference should be flagged as a warning for manual review
     warning_msg = f"Year mismatch: cited as {cited_year} but actually {correct_year}"
-    return True, warning_msg
+    return True, warning_msg
+
+
+def normalize_venue_for_display(venue: str) -> str:
+    """
+    Normalize venue names for consistent display and comparison.
+    
+    This function is used both for display in warnings and for venue comparison
+    to ensure consistent normalization across the system.
+    
+    Args:
+        venue: Raw venue string
+        
+    Returns:
+        Normalized venue string with prefixes removed and abbreviations expanded
+    """
+    if not venue:
+        return ""
+    
+    def expand_abbreviations(text):
+        """Generic abbreviation expansion using common academic patterns"""
+        # Common academic abbreviations mapping
+        common_abbrevs = {
+            # IEEE specific abbreviations (only expand with periods, not full words)
+            'robot.': 'robotics',
+            'autom.': 'automation',
+            'lett.': 'letters',
+            'trans.': 'transactions',
+            'syst.': 'systems',
+            'netw.': 'networks',
+            'learn.': 'learning',
+            'ind.': 'industrial',
+            'electron.': 'electronics',
+            'mechatron.': 'mechatronics',
+            'intell.': 'intelligence',
+            'transp.': 'transportation',
+            'contr.': 'control',
+            'mag.': 'magazine',
+            
+            # General academic abbreviations (only expand with periods)
+            'int.': 'international',
+            'intl.': 'international', 
+            'conf.': 'conference',
+            'j.': 'journal',
+            'proc.': 'proceedings',
+            'assoc.': 'association',
+            'comput.': 'computing',
+            'sci.': 'science',
+            'eng.': 'engineering',
+            'tech.': 'technology',
+            'artif.': 'artificial',
+            'mach.': 'machine',
+            'stat.': 'statistics',
+            'math.': 'mathematics',
+            'phys.': 'physics',
+            'chem.': 'chemistry',
+            'bio.': 'biology',
+            'med.': 'medicine',
+            'adv.': 'advances',
+            'ann.': 'annual',
+            'symp.': 'symposium',
+            'workshop': 'workshop',
+            'worksh.': 'workshop',
+        }
+        
+        text_lower = text.lower()
+        for abbrev, expansion in common_abbrevs.items():
+            # Only replace if it's a word boundary to avoid partial replacements
+            pattern = r'\b' + re.escape(abbrev) + r'\b'
+            text_lower = re.sub(pattern, expansion, text_lower)
+        
+        return text_lower
+    
+    venue_text = venue.strip()
+    
+    # Extract venue from complex editor strings (e.g. "In Smith, J.; and Doe, K., eds., Conference Name, volume 1")
+    # This handles patterns like "In [authors], eds., [venue], [optional metadata]" (case-insensitive)
+    editor_match = re.search(r'in\s+[^,]+(?:,\s*[^,]*)*,\s*eds?\.,\s*(.+?)(?:,\s*volume\s*\d+|,\s*pp?\.|$)', venue_text, re.IGNORECASE)
+    if editor_match:
+        # Extract the venue part from editor string (preserve original case)
+        venue_text = editor_match.group(1).strip()
+        # Clean up any remaining metadata like "volume X of Proceedings..." (case-insensitive)
+        venue_text = re.sub(r',\s*volume\s+\d+.*$', '', venue_text, flags=re.IGNORECASE)
+        venue_text = re.sub(r'\s+of\s+proceedings.*$', '', venue_text, flags=re.IGNORECASE)
+    
+    # Remove years, volumes, pages, and other citation metadata
+    # But preserve arXiv IDs (don't remove digits after arXiv:)
+    if not re.match(r'arxiv:', venue_text, re.IGNORECASE):
+        venue_text = re.sub(r',?\s*\d{4}[a-z]?\s*$', '', venue_text)  # Years like "2024" or "2024b"
+        venue_text = re.sub(r',?\s*\(\d{4}\)$', '', venue_text)  # Years in parentheses
+        venue_text = re.sub(r"'\d{2}$", '', venue_text)  # Year suffixes like 'CVPR'16'
+    venue_text = re.sub(r',?\s*(vol\.?\s*|volume\s*)\d+.*$', '', venue_text, flags=re.IGNORECASE)  # Volume info
+    venue_text = re.sub(r',?\s*\d+\s*\([^)]*\).*$', '', venue_text)  # Issue info with optional spaces
+    venue_text = re.sub(r',?\s*pp?\.\s*\d+.*$', '', venue_text, flags=re.IGNORECASE)  # Page info
+    venue_text = re.sub(r'\s*\(print\).*$', '', venue_text, flags=re.IGNORECASE)  # Print designation
+    venue_text = re.sub(r'\s*\(\d{4}\.\s*print\).*$', '', venue_text, flags=re.IGNORECASE)  # Year.Print
+    
+    # Remove procedural prefixes (case-insensitive)
+    prefixes_to_remove = [
+        r'^\d{4}\s+\d+(st|nd|rd|th)\s+',  # "2012 IEEE/RSJ"
+        r'^\d{4}\s+',                     # "2024 "
+        r'^proceedings\s+(of\s+)?(the\s+)?(\d+(st|nd|rd|th)\s+)?(ieee\s+)?',  # "Proceedings of the IEEE"
+        r'^proc\.\s+of\s+(the\s+)?(\d+(st|nd|rd|th)\s+)?(ieee\s+)?',        # "Proc. of the IEEE" (require "of")
+        r'^procs\.\s+of\s+(the\s+)?(\d+(st|nd|rd|th)\s+)?(ieee\s+)?',       # "Procs. of the IEEE" (require "of")
+        r'^in\s+',
+        r'^advances\s+in\s+',             # "Advances in Neural Information Processing Systems"
+        r'^adv\.\s+',                     # "Adv. Neural Information Processing Systems"
+        # Handle ordinal prefixes: "The Twelfth", "The Ninth", etc.
+        r'^the\s+(first|second|third|fourth|fifth|sixth|seventh|eighth|ninth|tenth|eleventh|twelfth|thirteenth|fourteenth|fifteenth|sixteenth|seventeenth|eighteenth|nineteenth|twentieth|twenty-first|twenty-second|twenty-third|twenty-fourth|twenty-fifth|twenty-sixth|twenty-seventh|twenty-eighth|twenty-ninth|thirtieth|thirty-first|thirty-second|thirty-third|thirty-fourth|thirty-fifth|thirty-sixth|thirty-seventh|thirty-eighth|thirty-ninth|fortieth|forty-first|forty-second|forty-third|forty-fourth|forty-fifth|forty-sixth|forty-seventh|forty-eighth|forty-ninth|fiftieth)\s+',
+        # Handle numeric ordinals: "The 41st", "The 12th", etc.
+        r'^the\s+\d+(st|nd|rd|th)\s+',
+        # Handle standalone "The" prefix
+        r'^the\s+',
+    ]
+    
+    for prefix_pattern in prefixes_to_remove:
+        venue_text = re.sub(prefix_pattern, '', venue_text, flags=re.IGNORECASE)
+    
+    # Note: For display purposes, we preserve case and don't expand abbreviations
+    # Only do minimal cleaning needed for proper display
+    
+    # Remove organization prefixes/suffixes that don't affect identity (case-insensitive)
+    # But preserve IEEE when it's part of a journal name like \"IEEE Transactions\"
+    if not re.match(r'ieee\s+transactions', venue_text, re.IGNORECASE):
+        venue_text = re.sub(r'^(ieee|acm|aaai|usenix|sigcomm|sigkdd|sigmod|vldb|osdi|sosp|eurosys)\s+', '', venue_text, flags=re.IGNORECASE)  # Remove org prefixes
+    venue_text = re.sub(r'^ieee/\w+\s+', '', venue_text, flags=re.IGNORECASE)  # Remove "IEEE/RSJ " etc
+    venue_text = re.sub(r'\s+(ieee|acm|aaai|usenix)\s*$', '', venue_text, flags=re.IGNORECASE)  # Remove org suffixes
+    venue_text = re.sub(r'/\w+\s+', ' ', venue_text)  # Remove "/ACM " style org separators
+    
+    # IMPORTANT: Don't remove "Conference on" or "International" - they're needed for display
+    # Only remove specific org-prefixed conference patterns where the org is clear
+    venue_text = re.sub(r'^(ieee|acm|aaai|nips)(/\w+)?\s+conference\s+on\s+', '', venue_text, flags=re.IGNORECASE)
+    
+    # Note: Don't remove "Conference on" as it's often part of the actual venue name
+    # Only remove it if it's clearly a procedural prefix (handled in prefixes_to_remove above)
+    
+    # Clean up spacing (preserve punctuation and case for display)
+    venue_text = re.sub(r'\s+', ' ', venue_text)     # Normalize whitespace
+    venue_text = venue_text.strip()
+    
+    return venue_text

{academic_refchecker-1.2.41 → academic_refchecker-1.2.43}/src/utils/url_utils.py

@@ -209,7 +209,13 @@ def validate_url_format(url: str) -> bool:
 
 def clean_url(url: str) -> str:
     """
-    Clean a URL by removing common issues like extra spaces, fragments, etc.
+    Clean a URL by removing common issues like extra spaces, fragments, malformed LaTeX, etc.
+    
+    This function handles:
+    - Whitespace trimming
+    - Malformed LaTeX URL wrappers like \\url{https://...}
+    - Trailing punctuation from academic references
+    - DOI URL query parameter cleanup
     
     Args:
         url: URL to clean
@@ -223,6 +229,18 @@ def clean_url(url: str) -> str:
     # Remove leading/trailing whitespace
     url = url.strip()
     
+    # Handle malformed URLs that contain \url{} wrappers within the URL text
+    # e.g., "https://\url{https://www.example.com/}" -> "https://www.example.com/"
+    import re
+    url_pattern = r'https?://\\url\{(https?://[^}]+)\}'
+    url_match = re.search(url_pattern, url)
+    if url_match:
+        url = url_match.group(1)
+    
+    # Remove trailing punctuation that's commonly part of sentence structure
+    # but preserve legitimate URL characters
+    url = url.rstrip('.,;!?)')
+    
     # Note: Preserving query parameters for all URLs now
     # Previously this function removed query parameters for non-DOI URLs,
     # but this was causing issues with OpenReview and other URLs that need their parameters
@@ -254,6 +272,14 @@ def clean_url_punctuation(url: str) -> str:
     # Remove leading/trailing whitespace
     url = url.strip()
     
+    # Handle malformed URLs that contain \\url{} wrappers within the URL text
+    # e.g., "https://\\url{https://www.example.com/}" -> "https://www.example.com/"
+    import re
+    url_pattern = r'https?://\\url\{(https?://[^}]+)\}'
+    url_match = re.search(url_pattern, url)
+    if url_match:
+        url = url_match.group(1)
+    
     # Remove trailing punctuation that's commonly part of sentence structure
     # but preserve legitimate URL characters
     url = url.rstrip('.,;!?)')