academic-refchecker 1.2.42__py3-none-any.whl → 1.2.43__py3-none-any.whl

__version__.py

@@ -1,3 +1,3 @@
 """Version information for RefChecker."""
 
-__version__ = "1.2.42"
+__version__ = "1.2.43"

{academic_refchecker-1.2.42.dist-info → academic_refchecker-1.2.43.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: academic-refchecker
-Version: 1.2.42
+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-1.2.42.dist-info → academic_refchecker-1.2.43.dist-info}/RECORD

@@ -1,5 +1,5 @@
-__version__.py,sha256=jrP5O1rb9OpfyEnz9IJjKo7ZhdOr-9_yzLGwvjDTLWA,65
-academic_refchecker-1.2.42.dist-info/licenses/LICENSE,sha256=Kwrx3fePVCeEFDCZvCW4OuoTNBiSoYbpGBI6qzGhWF0,1067
+__version__.py,sha256=JbybFux4Juuafz1jN0cgsedPmzBO8U9DJ874tJu2saA,65
+academic_refchecker-1.2.43.dist-info/licenses/LICENSE,sha256=Kwrx3fePVCeEFDCZvCW4OuoTNBiSoYbpGBI6qzGhWF0,1067
 checkers/__init__.py,sha256=T0PAHTFt6UiGvn-WGoJU8CdhXNmf6zaHmcGVoWHhmJQ,533
 checkers/crossref.py,sha256=Hzq4dlf1CSn0aZWU8CMOnLxIvaSivTabLoepIOkgkmY,20585
 checkers/enhanced_hybrid_checker.py,sha256=6yf5tV4jLSVzjX1xR_kQq0NOgQIst-z_WmkiqqMc8hQ,23469
@@ -15,7 +15,7 @@ config/settings.py,sha256=-vODFoXbWbGPUElpmchE5zbCj_n4Vtxr8HU1hQDFp_c,6164
 core/__init__.py,sha256=1T2MSQyDk0u_PupbHvm4CvNNN--dxsw78fqKUrqoYrM,157
 core/db_connection_pool.py,sha256=XRiOdehikkSz3obH4WKgf8woa3694if50Q15rBT-4XQ,4697
 core/parallel_processor.py,sha256=5V2iJDBFwwryMCnCNU_oRt2u5he1wpy-_9qapC_6f00,17043
-core/refchecker.py,sha256=ElXgD1iPI-rDDFZmCPMZpkIP4UeX3nPAJVCfsVPNgcw,274640
+core/refchecker.py,sha256=sVRg3PUzrs2vLFlEBoi4bxUy-TpO5iQHCkokGas-ygQ,273616
 database/__init__.py,sha256=mEuVHlEBuS44t_2ZT_JnvQQrlRCjo1SJq1NmaJ6r8OY,125
 database/download_semantic_scholar_db.py,sha256=waN4I97KC_36YMiPbiBDUUmgfzu1nub5yeKdAsIR2aw,75276
 llm/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -30,17 +30,17 @@ utils/arxiv_utils.py,sha256=Y8sDJgDwHxp1L33BkQoDumIl0Pkp-BuYZb1PwWYsmak,18251
 utils/author_utils.py,sha256=DLTo1xsxef2wxoe4s_MWrh36maj4fgnvFlsDLpDE-qQ,5507
 utils/biblatex_parser.py,sha256=OkHXQcjiBrEDuhBfEk0RtmAYxufu5lAxAjb8__DzMjI,25537
 utils/bibliography_utils.py,sha256=mpmdAklzAs1CT3gqrOcjujGhouL95OuliCx0LE9Pg90,11705
-utils/bibtex_parser.py,sha256=jsQ87lkzmBmJO3VEN3itw22CJ1Hesei4IvM2sfsaFKI,12867
+utils/bibtex_parser.py,sha256=a89NLy_q2kwED4QFJgxWFgPQOJBV73bIUL3RS_Urmro,15231
 utils/config_validator.py,sha256=rxf7K3DYmJ-BNPsmtaCNipY2BTVT-pJZ7wN-M9Y3GC8,11167
 utils/db_utils.py,sha256=_wSupfBlm0ILFvntQTvoj7tLDCbrYPRQrp9NDvphF_E,6281
 utils/doi_utils.py,sha256=8f3iE4RdSNkzeqa9wJfoKcVEiBVse3_uf643biLudmw,4134
 utils/error_utils.py,sha256=JqnRg4z-O9GcJ1eJGeTMzmOQwPWbWo2Lf6Duwj-ymHQ,6258
 utils/mock_objects.py,sha256=QxU-UXyHSY27IZYN8Sb8ei0JtNkpGSdMXoErrRLHXvE,6437
-utils/text_utils.py,sha256=F5o-37KUkkr-ie4sg6ld5om3-uDpAxPUSjDFxY0fsL4,203063
+utils/text_utils.py,sha256=jPgCOBTVboLRJyypoOtL-dg1wBDQrKBux2ImvC6wL58,206296
 utils/unicode_utils.py,sha256=-WBKarXO756p7fd7gCeNsMag4ztDNURwFX5IVniOtwY,10366
-utils/url_utils.py,sha256=n0m5rXKV0-UrE8lI85VEx23KmfGwky57sI6gFPuu78I,7358
-academic_refchecker-1.2.42.dist-info/METADATA,sha256=k7fzk4fhb-kz-CdJE-gaeU2I5xM16D1rNNeEuer_9Hk,22298
-academic_refchecker-1.2.42.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-academic_refchecker-1.2.42.dist-info/entry_points.txt,sha256=WdI89tYkIfz-M628PiboOfOLzTBWZAqvlF29qCVCkek,61
-academic_refchecker-1.2.42.dist-info/top_level.txt,sha256=6RlcQEA0kHb7-ndbKMFMZnYnJQVohgsU6BBkbEvJvEs,69
-academic_refchecker-1.2.42.dist-info/RECORD,,
+utils/url_utils.py,sha256=aq1hSYEA888bOKuBOGWRclgTFIjw32rpFdsBO_Ja8ZM,8402
+academic_refchecker-1.2.43.dist-info/METADATA,sha256=ZsJhIw1n7Yjoug6mpV4zpAPf-eSW5xSMdd3Dl_WTOlI,22298
+academic_refchecker-1.2.43.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+academic_refchecker-1.2.43.dist-info/entry_points.txt,sha256=WdI89tYkIfz-M628PiboOfOLzTBWZAqvlF29qCVCkek,61
+academic_refchecker-1.2.43.dist-info/top_level.txt,sha256=6RlcQEA0kHb7-ndbKMFMZnYnJQVohgsU6BBkbEvJvEs,69
+academic_refchecker-1.2.43.dist-info/RECORD,,

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"""

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:

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

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('.,;!?)')