Create regex_checker.py

regex_checker.py

@@ -0,0 +1,86 @@
+# regex_checker.py
+import re
+import traceback
+from typing import List, Dict, Any
+
+def perform_regex_checks(plain_text_from_original_pdf: str) -> Dict[str, Any]:
+    """
+    Performs custom regex checks on plain text from the original PDF.
+    Filters issues to only include those between "abstract" and "references/bibliography"
+    found within this specific text.
+    Assumes plain_text_from_original_pdf is already space-normalized.
+    """
+    text_for_regex_analysis = plain_text_from_original_pdf
+
+    if not text_for_regex_analysis or not text_for_regex_analysis.strip():
+        print("RegexChecker: Input plain text is empty.")
+        return {"total_issues": 0, "issues_list": [], "text_used_for_analysis": ""}
+
+    text_for_regex_analysis_lower = text_for_regex_analysis.lower()
+    
+    abstract_match = re.search(r'\babstract\b', text_for_regex_analysis_lower)
+    content_start_index = abstract_match.start() if abstract_match else 0
+    if abstract_match:
+        print(f"RegexChecker: Found 'abstract' at index {content_start_index} in its text.")
+    else:
+        print(f"RegexChecker: Did not find 'abstract', Regex analysis from index 0 of its text.")
+
+    references_match = re.search(r'\breferences\b', text_for_regex_analysis_lower)
+    bibliography_match = re.search(r'\bbibliography\b', text_for_regex_analysis_lower)
+    content_end_index = len(text_for_regex_analysis)
+
+    if references_match and bibliography_match:
+        content_end_index = min(references_match.start(), bibliography_match.start())
+        print(f"RegexChecker: Found 'references' at {references_match.start()} and 'bibliography' at {bibliography_match.start()}. Using {content_end_index} as end boundary.")
+    elif references_match:
+        content_end_index = references_match.start()
+        print(f"RegexChecker: Found 'references' at {content_end_index}. Using it as end boundary.")
+    elif bibliography_match:
+        content_end_index = bibliography_match.start()
+        print(f"RegexChecker: Found 'bibliography' at {content_end_index}. Using it as end boundary.")
+    else:
+        print(f"RegexChecker: Did not find 'references' or 'bibliography'. Regex analysis up to end of its text (index {content_end_index}).")
+
+
+    if content_start_index >= content_end_index:
+        print(f"RegexChecker: Warning: Content start index ({content_start_index}) is not before end index ({content_end_index}) in its text. No Regex issues will be reported from this range.")
+
+    processed_regex_issues: List[Dict[str, Any]] = []
+    try:
+        # Regex: Missing space before bracketed citation, e.g., "word[1]"
+        regex_pattern = r'\b(\w+)\[(\d+)\]' 
+        regex_matches = list(re.finditer(regex_pattern, text_for_regex_analysis))
+        
+        regex_issues_in_range = 0
+        for reg_idx, match in enumerate(regex_matches):
+            if not (content_start_index <= match.start() < content_end_index):
+                continue
+            regex_issues_in_range += 1
+            
+            word = match.group(1)
+            number = match.group(2)
+            context_str = text_for_regex_analysis[match.start():match.end()]
+            processed_regex_issues.append({
+                '_internal_id': f"regex_{reg_idx}",
+                'ruleId': "SPACE_BEFORE_BRACKET",
+                'message': f"Missing space before '[' in '{context_str}'. Should be '{word} [{number}]'.",
+                'context_text': context_str,
+                'offset_in_text': match.start(),
+                'error_length': match.end() - match.start(),
+                'replacements_suggestion': [f"{word} [{number}]"],
+                'category_name': "Formatting",
+                'source_check_type': 'CustomRegex',
+                'is_mapped_to_pdf': False,
+                'pdf_coordinates_list': [],
+                'mapped_page_number': -1
+            })
+        print(f"RegexChecker: Regex check found {len(regex_matches)} raw matches, {regex_issues_in_range} issues within defined content range of its text.")
+        
+        return {
+            "total_issues": len(processed_regex_issues),
+            "issues_list": processed_regex_issues,
+            "text_used_for_analysis": text_for_regex_analysis
+        }
+    except Exception as e:
+        print(f"Error in perform_regex_checks: {e}\n{traceback.format_exc()}")
+        return {"error": str(e), "total_issues": 0, "issues_list": [], "text_used_for_analysis": text_for_regex_analysis}