Update app.py

app.py

@@ -84,7 +84,7 @@ DOCUMENT_FORMATTING_RULES = {
         "types": ["Advisory Circular"],
         "italics": True, 
         "quotes": False,
-        "description": "For Advisory Circulars, referenced document titles should be italicized but not quoted",
+        "description": "For Advisory Circulars, referenced document titles should be italicized but not quoted.",
         "example": "See AC 20-135, *Powerplant Installation and Propulsion System Component Fire Protection Test Methods, Standards, and Criteria* for information on X."
     },
     "quotes_only": {
@@ -95,14 +95,14 @@ DOCUMENT_FORMATTING_RULES = {
         ],
         "italics": False, 
         "quotes": True,
-        "description": "For this document type, referenced document titles should be in quotes without italics",
+        "description": "For this document type, referenced document titles should be in quotes without italics.",
         "example": 'See AC 20-135, "Powerplant Installation and Propulsion System Component Fire Protection Test Methods, Standards, and Criteria" for information on X.'
     },
     "no_formatting": {
         "types": ["Policy Statement", "Other"],
         "italics": False, 
         "quotes": False,
-        "description": "For this document type, referenced document titles should not use italics or quotes",
+        "description": "For this document type, referenced document titles should not use italics or quotes.",
         "example": "See AC 20-135, Powerplant Installation and Propulsion System Component Fire Protection Test Methods, Standards, and Criteria for information on X."
     }
 }
@@ -510,6 +510,54 @@ class DocumentCheckerConfig:
                     description="'flight crew' should be 'flightcrew'",
                     is_error=True,
                     replacement="flightcrew"
+                ),
+                PatternConfig(
+                    pattern=r'\bchairman\b',
+                    description="'chairman' should be 'chair'",
+                    is_error=True,
+                    replacement="chair"
+                ),
+                PatternConfig(
+                    pattern=r'\bflagman\b',
+                    description="'flagman' should be 'flagger' or 'flagperson'",
+                    is_error=True,
+                    replacement="flagperson"
+                ),
+                PatternConfig(
+                    pattern=r'\bman\b',
+                    description="'man' should be 'individual' or 'person'",
+                    is_error=True,
+                    replacement="person"
+                ),
+                PatternConfig(
+                    pattern=r'\bmanmade\b',
+                    description="'manmade' should be 'personmade'",
+                    is_error=True,
+                    replacement="personmade"
+                ),
+                PatternConfig(
+                    pattern=r'\bmanpower\b',
+                    description="'manpower' should be 'labor force'",
+                    is_error=True,
+                    replacement="labor force"
+                ),
+                PatternConfig(
+                    pattern=r'\bnotice to airman\b',
+                    description="'notice to airman' should be 'notice to air missions'",
+                    is_error=True,
+                    replacement="notice to air missions"
+                ),
+                PatternConfig(
+                    pattern=r'\bnotice to airmen\b',
+                    description="'notice to airmen' should be 'notice to air missions'",
+                    is_error=True,
+                    replacement="notice to air missions"
+                ),
+                PatternConfig(
+                    pattern=r'\bcockpit\b',
+                    description="'cockpit' should be 'flight deck'",
+                    is_error=True,
+                    replacement="flight deck"
                 )
             ],
             'section_symbol': [
@@ -945,7 +993,13 @@ class FAADocumentChecker(DocumentChecker):
     @profile_performance
     def check_terminology(self, doc: List[str]) -> DocumentCheckResult:
         """
-        Check document terminology and output only the sentences needing correction.
+        Check document terminology and output only unique sentences needing correction.
+        
+        Args:
+            doc (List[str]): List of document paragraphs
+            
+        Returns:
+            DocumentCheckResult: Result containing unique terminology issues with context
         """
         if not self.validate_input(doc):
             return DocumentCheckResult(success=False, issues=[{'error': 'Invalid document input'}])
@@ -954,28 +1008,67 @@ class FAADocumentChecker(DocumentChecker):
         terminology_patterns = self.config_manager.pattern_registry.get('terminology', [])
         prohibited_patterns = self.config_manager.pattern_registry.get('reference_terms', [])
 
-        # Collect sentences with issues
-        issue_sentences = []
+        # Use a dictionary to track unique issues by sentence
+        # Key: sentence, Value: list of issues in that sentence
+        sentence_issues = {}
 
-        # Check each paragraph for terminology and prohibited patterns
+        # Check each paragraph for terminology issues
         for paragraph in doc:
             sentences = re.split(r'(?<=[.!?])\s+', paragraph)
             for sentence in sentences:
+                sentence = sentence.strip()
+                
+                # Skip empty sentences
+                if not sentence:
+                    continue
+
+                current_sentence_issues = []
+
                 # Check for incorrect terms that need replacement
                 for pattern_config in terminology_patterns:
-                    if re.search(pattern_config.pattern, sentence):
-                        issue_sentences.append(sentence.strip())
+                    matches = list(re.finditer(pattern_config.pattern, sentence))
+                    for match in matches:
+                        current_sentence_issues.append({
+                            'incorrect_term': match.group(),
+                            'correct_term': pattern_config.replacement,
+                            'description': pattern_config.description,
+                            'sentence': sentence
+                        })
 
                 # Check for prohibited phrases and constructions
                 for pattern_config in prohibited_patterns:
                     if re.search(pattern_config.pattern, sentence, re.IGNORECASE):
-                        issue_sentences.append(sentence.strip())
+                        current_sentence_issues.append({
+                            'description': pattern_config.description,
+                            'sentence': sentence
+                        })
+
+                # Only add if we found issues in this sentence
+                if current_sentence_issues:
+                    # Use sentence as key to prevent duplicates
+                    if sentence not in sentence_issues:
+                        sentence_issues[sentence] = current_sentence_issues
+                    else:
+                        sentence_issues[sentence].extend(current_sentence_issues)
+
+        # Build the issues per sentence
+        unique_issues = []
+        for sentence, sentence_issue_list in sentence_issues.items():
+            incorrect_terms = set()
+            descriptions = set()
+            for issue in sentence_issue_list:
+                if 'incorrect_term' in issue:
+                    incorrect_terms.add((issue['incorrect_term'], issue.get('correct_term')))
+                if 'description' in issue:
+                    descriptions.add(issue['description'])
+            unique_issues.append({
+                'sentence': sentence,
+                'incorrect_terms': list(incorrect_terms),
+                'descriptions': list(descriptions),
+            })
 
-        # Remove duplicates for cleaner output
-        issue_sentences = list(set(issue_sentences))
+        return DocumentCheckResult(success=not unique_issues, issues=unique_issues)
 
-        # Return only the minimal list of issue sentences
-        return DocumentCheckResult(success=not issue_sentences, issues=issue_sentences)
 
     @profile_performance
     def check_section_symbol_usage(self, doc: List[str]) -> DocumentCheckResult:
@@ -1830,6 +1923,24 @@ class DocumentCheckResultsFormatter:
     def _format_standard_issue(self, issue: Dict[str, Any]) -> str:
         """Format a standard issue consistently."""
         if isinstance(issue, dict):
+            # Handle grouped issues per sentence
+            if 'incorrect_terms' in issue and 'sentence' in issue:
+                # Build the replacements text
+                replacements = '; '.join(
+                    f"'{inc}' with '{corr}'" if corr else f"Remove '{inc}'"
+                    for inc, corr in sorted(issue['incorrect_terms'])
+                )
+                # Start building the output lines
+                lines = []
+                lines.append(f"    • In: {issue['sentence']}")
+                lines.append(f"      Replace {replacements}")
+                # Format each line individually
+                formatted_lines = [
+                    textwrap.fill(line, width=76, subsequent_indent='      ')
+                    for line in lines
+                ]
+                return '\n'.join(formatted_lines)
+            
             # Handle issues with occurrences list
             if 'occurrences' in issue:
                 # Format the first 3 occurrences
@@ -1871,9 +1982,10 @@ class DocumentCheckResultsFormatter:
                 )
                 
             # Handle terminology issues
-            elif 'incorrect_term' in issue and 'correct_term' in issue:
+            if all(k in issue for k in ['incorrect_term', 'correct_term', 'sentence']):
                 return textwrap.fill(
-                    f"    • '{issue['incorrect_term']}' should be '{issue['correct_term']}' in: {issue.get('sentence', '')}",
+                    f"    • Replace '{issue['incorrect_term']}' with '{issue['correct_term']}' in: "
+                    f"{issue['sentence']}",
                     width=76,
                     subsequent_indent='      '
                 )
@@ -1966,7 +2078,7 @@ class DocumentCheckResultsFormatter:
                 "types": ["Advisory Circular"],
                 "italics": True, 
                 "quotes": False,
-                "description": "For Advisory Circulars, referenced document titles should be italicized but not quoted.",
+                "description": "For Advisory Circulars, referenced document titles should be italicized but not quoted",
                 "example": "See AC 25.1309-1B, *System Design and Analysis*, for information on X."
             },
             "quotes_only": {
@@ -1977,14 +2089,14 @@ class DocumentCheckResultsFormatter:
                 ],
                 "italics": False, 
                 "quotes": True,
-                "description": "For this document type, referenced document titles should be in quotes without italics.",
+                "description": "For this document type, referenced document titles should be in quotes without italics",
                 "example": 'See AC 25.1309-1B, "System Design and Analysis," for information on X.'
             },
             "no_formatting": {
                 "types": ["Policy Statement", "Other"],
                 "italics": False, 
                 "quotes": False,
-                "description": "For this document type, referenced document titles should not use italics or quotes.",
+                "description": "For this document type, referenced document titles should not use italics or quotes",
                 "example": "See AC 25.1309-1B, System Design and Analysis, for information on X."
             }
         }