academic-refchecker 1.2.53__py3-none-any.whl → 1.2.55__py3-none-any.whl

{utils → refchecker/utils}/text_utils.py

@@ -173,6 +173,11 @@ def parse_authors_with_initials(authors_text):
     if stripped_text in ['others', 'and others', 'et al', 'et al.']:
         return []
     
+    # Clean LaTeX commands early to prevent parsing issues
+    # This fixes cases like "Hochreiter, Sepp and Schmidhuber, J{\"u}rgen" 
+    # which should parse as 2 authors, not get split incorrectly due to LaTeX braces
+    authors_text = strip_latex_commands(authors_text)
+    
     # Fix spacing around periods in initials (e.g., "Y . Li" -> "Y. Li") before parsing
     authors_text = re.sub(r'(\w)\s+\.', r'\1.', authors_text)
     
@@ -300,9 +305,9 @@ def parse_authors_with_initials(authors_text):
                     comma_parts = [p.strip() for p in part.split(',')]
                     if len(comma_parts) == 2:
                         lastname, firstname = comma_parts
-                        # Both parts should contain only letters, spaces, hyphens, apostrophes, and periods
-                        if (re.match(r'^[A-Za-z\s\-\'.]+$', lastname) and 
-                            re.match(r'^[A-Za-z\s\-\'.]+$', firstname) and
+                        # Both parts should contain only letters (including Unicode), spaces, hyphens, apostrophes, and periods
+                        if (re.match(r'^[\w\s\-\'.]+$', lastname, re.UNICODE) and 
+                            re.match(r'^[\w\s\-\'.]+$', firstname, re.UNICODE) and
                             lastname and firstname):
                             valid_author_parts.append(part)
             
@@ -314,6 +319,50 @@ def parse_authors_with_initials(authors_text):
     # Split on commas first for other formats
     parts = [part.strip() for part in authors_text.split(',') if part.strip()]
     
+    # Handle single author with "Lastname, Firstname" format (exactly 2 parts)
+    if len(parts) == 2:
+        lastname, firstname = parts
+        # Pattern for surnames: capitalized word(s), possibly hyphenated or compound
+        # But exclude common patterns that suggest multiple authors like "Other Author"
+        surname_pattern = r'^[A-Z][a-zA-Z\-\']+$'  # Single surname word (no spaces to avoid "Other Author")
+        # Pattern for first names or initials: either full names or initials with periods
+        # Accept both full names like "David R" and initials like "A. C"
+        firstname_pattern = r'^[A-Z]([a-zA-Z\s\-\'.]*|\.(\s+[A-Z]\.?)*\s*)$'  # Full names or initials
+        
+        # Additional check: if the "firstname" part looks like "Other Author" or similar, 
+        # it's likely multiple authors, not a single "Lastname, Firstname" pattern
+        # We need to distinguish between:
+        # - "David R" (first name + middle initial - single author) 
+        # - "Other Author" (two separate names - multiple authors)
+        if ' ' in firstname:
+            firstname_parts = firstname.split()
+            if len(firstname_parts) == 2:
+                first_part, second_part = firstname_parts
+                # Pattern 1: "David R" - first name + single letter (middle initial)
+                is_name_plus_initial = (
+                    len(first_part) >= 2 and first_part[0].isupper() and first_part[1:].islower() and
+                    len(second_part) <= 2 and second_part.replace('.', '').isalpha()  # Initial like "R" or "R."
+                )
+                # Pattern 2: "Other Author" - two full capitalized words suggesting separate authors
+                looks_like_separate_authors = (
+                    len(first_part) >= 3 and first_part[0].isupper() and first_part[1:].islower() and
+                    len(second_part) >= 3 and second_part[0].isupper() and second_part[1:].islower()
+                )
+                looks_like_multiple_authors = looks_like_separate_authors and not is_name_plus_initial
+            else:
+                # More than 2 parts with spaces likely indicates multiple authors
+                looks_like_multiple_authors = len(firstname_parts) > 2
+        else:
+            looks_like_multiple_authors = False
+        
+        # Check if this looks like a single author in "Lastname, Firstname" format
+        if (re.match(surname_pattern, lastname) and 
+            re.match(firstname_pattern, firstname) and
+            len(lastname) >= 2 and len(firstname) >= 1 and
+            not looks_like_multiple_authors):
+            # This is a single author, return as "Lastname, Firstname"
+            return [f"{lastname}, {firstname}"]
+    
     # Check if this is BibTeX comma-separated format: "Surname, Given, Surname, Given"
     # Enhanced heuristic: even number of parts >= 6, alternating proper surname/given pattern
     # Distinguish between initials (should remain as "Surname, Initial") and full names
@@ -640,7 +689,7 @@ def extract_arxiv_id_from_url(url):
     Returns:
         ArXiv ID or None if not found
     """
-    from utils.url_utils import extract_arxiv_id_from_url as common_extract
+    from refchecker.utils.url_utils import extract_arxiv_id_from_url as common_extract
     return common_extract(url)
 
 def extract_year_from_text(text):
@@ -2092,7 +2141,7 @@ def compare_authors(cited_authors: list, correct_authors: list, normalize_func=N
     # and not penalize for the authoritative source having more authors
     if has_et_al:
         # Import here to avoid circular imports
-        from utils.error_utils import format_author_mismatch
+        from refchecker.utils.error_utils import format_author_mismatch
         # For et al cases, check if each cited author matches ANY author in the correct list
         # rather than comparing positionally, since author order can vary
         for i, cited_author in enumerate(cleaned_cited):
@@ -2126,21 +2175,21 @@ def compare_authors(cited_authors: list, correct_authors: list, normalize_func=N
         
         # Check if cited authors look like parsing fragments
         if looks_like_fragments(cleaned_cited):
-            from utils.error_utils import format_author_count_mismatch
+            from refchecker.utils.error_utils import format_author_count_mismatch
             display_cited = [format_author_for_display(author) for author in cleaned_cited]
             error_msg = format_author_count_mismatch(len(cleaned_cited), len(correct_names), display_cited, correct_names)
             return False, error_msg
         
         # For all count mismatches, show the count mismatch error
         if len(cleaned_cited) < len(correct_names):
-            from utils.error_utils import format_author_count_mismatch
+            from refchecker.utils.error_utils import format_author_count_mismatch
             display_cited = [format_author_for_display(author) for author in cleaned_cited]
             error_msg = format_author_count_mismatch(len(cleaned_cited), len(correct_names), display_cited, correct_names)
             return False, error_msg
         
         # For cases where cited > correct, also show count mismatch
         elif len(cleaned_cited) > len(correct_names):
-            from utils.error_utils import format_author_count_mismatch
+            from refchecker.utils.error_utils import format_author_count_mismatch
             display_cited = [format_author_for_display(author) for author in cleaned_cited]
             error_msg = format_author_count_mismatch(len(cleaned_cited), len(correct_names), display_cited, correct_names)
             return False, error_msg
@@ -2149,7 +2198,7 @@ def compare_authors(cited_authors: list, correct_authors: list, normalize_func=N
         comparison_correct = correct_names
     
     # Use shared three-line formatter (imported lazily to avoid circular imports)
-    from utils.error_utils import format_first_author_mismatch, format_author_mismatch
+    from refchecker.utils.error_utils import format_first_author_mismatch, format_author_mismatch
 
     # Compare first author (most important) using the enhanced name matching
     if comparison_cited and comparison_correct:
@@ -2757,7 +2806,7 @@ def filter_bibtex_by_cited_keys(bib_content, cited_keys):
         return bib_content
     
     # Parse entries and filter
-    from utils.bibtex_parser import parse_bibtex_entries
+    from refchecker.utils.bibtex_parser import parse_bibtex_entries
     entries = parse_bibtex_entries(bib_content)
     filtered_entries = []
     
@@ -3069,7 +3118,7 @@ def extract_latex_references(text, file_path=None):  # pylint: disable=unused-ar
     
     if format_info['format_type'] == 'bibtex':
         # Use the dedicated BibTeX parser for consistent results
-        from utils.bibtex_parser import parse_bibtex_references
+        from refchecker.utils.bibtex_parser import parse_bibtex_references
         return parse_bibtex_references(text)
     
     elif format_info['format_type'] == 'thebibliography':
@@ -3273,7 +3322,7 @@ def extract_latex_references(text, file_path=None):  # pylint: disable=unused-ar
                 # Extract URL if present
                 url_match = re.search(r'\\url\{([^}]+)\}', content)
                 if url_match:
-                    from utils.url_utils import clean_url_punctuation
+                    from refchecker.utils.url_utils import clean_url_punctuation
                     ref['url'] = clean_url_punctuation(url_match.group(1))
             
             # Extract title from \showarticletitle{} or \bibinfo{title}{}
@@ -3335,7 +3384,7 @@ def extract_latex_references(text, file_path=None):  # pylint: disable=unused-ar
             if not ref['url']:
                 url_match = re.search(r'\\url\{([^}]+)\}', content)
                 if url_match:
-                    from utils.url_utils import clean_url_punctuation
+                    from refchecker.utils.url_utils import clean_url_punctuation
                     ref['url'] = clean_url_punctuation(url_match.group(1))
             
             # Extract DOI from \href{https://doi.org/...}

__version__.py

@@ -1,3 +0,0 @@
-"""Version information for RefChecker."""
-
-__version__ = "1.2.53"

academic_refchecker-1.2.53.dist-info/RECORD

@@ -1,47 +0,0 @@
-__version__.py,sha256=iH7i3qnj4nR1gSXECRVUGvJH5oBPWtb7Lb8H9ODFTVc,65
-academic_refchecker-1.2.53.dist-info/licenses/LICENSE,sha256=Kwrx3fePVCeEFDCZvCW4OuoTNBiSoYbpGBI6qzGhWF0,1067
-checkers/__init__.py,sha256=T0PAHTFt6UiGvn-WGoJU8CdhXNmf6zaHmcGVoWHhmJQ,533
-checkers/crossref.py,sha256=cLYmSzE8ehJ5sNko_R3fEiGBGiPH5_HxLhFM-pCfDRM,20378
-checkers/enhanced_hybrid_checker.py,sha256=rbXkzpNkd0bn4e2OooX-CcdGTwwYpgmVaFvX_xCAFsA,27777
-checkers/github_checker.py,sha256=BXJaBC3AloKze04j8EcQz0a79EhtVoi9_871ilV7t60,14233
-checkers/local_semantic_scholar.py,sha256=D8py8-yMCgN1lvhXCiMUOEA4wBkH7AQvrkM4-3LCDsU,21015
-checkers/openalex.py,sha256=Fbc7iscZzmXjAZxH32PDX2r2Nwo9b5Ku-Sh1Ut9KpLA,19550
-checkers/openreview_checker.py,sha256=3ckn6U7TN5nQBjqPacr8W8mm2uMo6aWWB6gsxTDNCPk,40452
-checkers/pdf_paper_checker.py,sha256=L5HRHd3xpo0xDltZGTAA-Wk_arIS9bQV8ITeuxW0bNc,19893
-checkers/semantic_scholar.py,sha256=wk6e8DkYJM_O2nWsi-6EfJT53PzfL8KCmX1rS562KKc,34962
-checkers/webpage_checker.py,sha256=REOotx7Qka86_xbOIMeYj5YVb9D1RVMb4Ye311-28cA,43620
-config/__init__.py,sha256=r7sONsX2-ITviUJRU1KEz76uAuTRqZlzU-TVkvFRGYY,15
-config/logging.conf,sha256=r1tP0ApLHtlz7rV-oKS1MVO7oXJOgahbZFTtYmKnf9U,687
-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=cq_WfzXrF2EI6IKOtJd6_QcwvM1xT3J6a13teg-wSbM,17638
-core/refchecker.py,sha256=-QIT5eUQaPCuQy7S80sXCvtrmcjdH5lf5wdZvsPQO9w,286416
-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
-llm/base.py,sha256=uMF-KOqZ9ZQ7rccOQLpKJiW9sEMMxr7ePXBSF0yYDJY,16782
-llm/providers.py,sha256=A0usJpprCO5D-VX0hqaQzBfi4DG3rdjA39vu02XJsGw,40092
-scripts/__init__.py,sha256=xJwo6afG8s7S888BK2Bxw2d7FX8aLkbl0l_ZoJOFibE,37
-scripts/start_vllm_server.py,sha256=ZepWp2y2cKFW0Kgsoima2RbmF02fTU29UFcLLpsBhFU,4213
-services/__init__.py,sha256=jGi9S74Msak3YR-C4Qb68VU7HB4oLaX9o1rlVAFpOFI,187
-services/pdf_processor.py,sha256=vu_JnhFGZY6jFVbDbPvG-mlQojvB-3Dzc8_946KVV2E,9427
-utils/__init__.py,sha256=1RrGoIIn1_gVzxd56b6a7HeAS-wu7uDP-nxLbR3fJ-8,1199
-utils/arxiv_utils.py,sha256=EzH1PhEAW0df5mmSP-kKHmuwqd4u2CSotRNwQ5IMJx8,19766
-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=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=ezUiRnYRpoO0U_Rqgxv1FxqmeTwPh6X8gLgSDbqg5sY,4874
-utils/error_utils.py,sha256=UJOH7Bp-rPV2JDY_XN38I2pSkqqPdnQoviKa4s4nK_A,12501
-utils/mock_objects.py,sha256=QxU-UXyHSY27IZYN8Sb8ei0JtNkpGSdMXoErrRLHXvE,6437
-utils/text_utils.py,sha256=T3PiiG9-BMPTbdCftG2zypyIeZJl6snuMCKQ0nEOQv0,217834
-utils/unicode_utils.py,sha256=-WBKarXO756p7fd7gCeNsMag4ztDNURwFX5IVniOtwY,10366
-utils/url_utils.py,sha256=HdxIO8QvciP6Jp8Wd4sTSrS8JQrOMwgM7pxdUC8RJb4,9176
-academic_refchecker-1.2.53.dist-info/METADATA,sha256=6j1G-R74oa1900hERaRnJFkV5u4zTuVyLC6YamhXxq4,23256
-academic_refchecker-1.2.53.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-academic_refchecker-1.2.53.dist-info/entry_points.txt,sha256=WdI89tYkIfz-M628PiboOfOLzTBWZAqvlF29qCVCkek,61
-academic_refchecker-1.2.53.dist-info/top_level.txt,sha256=6RlcQEA0kHb7-ndbKMFMZnYnJQVohgsU6BBkbEvJvEs,69
-academic_refchecker-1.2.53.dist-info/RECORD,,

academic_refchecker-1.2.53.dist-info/entry_points.txt

@@ -1,2 +0,0 @@
-[console_scripts]
-academic-refchecker = core.refchecker:main

academic_refchecker-1.2.53.dist-info/top_level.txt

@@ -1,9 +0,0 @@
-__version__
-checkers
-config
-core
-database
-llm
-scripts
-services
-utils