legalmind-ai 1.1.0__py3-none-any.whl

legalmind/providers/fallback/versalaw2_core/data_loader.py

@@ -0,0 +1,308 @@
+#!/usr/bin/env python3
+"""
+Enhanced MayaLaw Data Loader
+Supports multiple markdown formats
+"""
+
+import re
+from pathlib import Path
+from typing import List, Dict, Optional
+import logging
+
+logger = logging.getLogger(__name__)
+
+class MayaLawDataLoader:
+    """Enhanced data loader with multiple parsers"""
+    
+    def __init__(self, mayalaw_path="/root/dragon/global/mayalaw"):
+        self.mayalaw_path = Path(mayalaw_path)
+        self.cases = []
+        self.stats = {
+            'files_loaded': 0,
+            'cases_loaded': 0,
+            'errors': []
+        }
+        self.load_all_cases()
+    
+    def load_all_cases(self):
+        """Load all study cases from MayaLaw"""
+        files_to_load = [
+            ("JAWABAN_LENGKAP_20_PERTANYAAN_HUKUM.md", "pertanyaan_format"),
+            ("LAW_LIBRARY_BATCH_IV_ANSWERS.md", "qa_format"),
+            ("LAW_LIBRARY_BATCH_V_FINAL.md", "qa_format"),
+            ("LAW_LIBRARY_BATCH_IV_PART2.md", "qa_format"),
+            ("ADVANCED_LEGAL_QUESTIONS_20_PART2.md", "pertanyaan_simple"),
+        ]
+        
+        for filename, parser_type in files_to_load:
+            filepath = self.mayalaw_path / filename
+            if filepath.exists():
+                try:
+                    cases = self.parse_file(filepath, parser_type)
+                    self.cases.extend(cases)
+                    self.stats['files_loaded'] += 1
+                    self.stats['cases_loaded'] += len(cases)
+                    logger.info(f"✅ Loaded {len(cases)} cases from {filename}")
+                    print(f"✅ Loaded {len(cases)} cases from {filename}")
+                except Exception as e:
+                    error_msg = f"Error loading {filename}: {e}"
+                    self.stats['errors'].append(error_msg)
+                    logger.error(error_msg)
+                    print(f"⚠️  {error_msg}")
+            else:
+                logger.warning(f"File not found: {filename}")
+        
+        print(f"\n📚 Total: {len(self.cases)} study cases loaded from {self.stats['files_loaded']} files\n")
+        logger.info(f"Total cases loaded: {len(self.cases)}")
+    
+    def parse_file(self, filepath: Path, parser_type: str) -> List[Dict]:
+        """Parse file based on type"""
+        with open(filepath, 'r', encoding='utf-8') as f:
+            content = f.read()
+        
+        if parser_type == "pertanyaan_format":
+            return self.parse_pertanyaan_format(content, filepath.name)
+        elif parser_type == "qa_format":
+            return self.parse_qa_format(content, filepath.name)
+        elif parser_type == "pertanyaan_simple":
+            return self.parse_pertanyaan_simple(content, filepath.name)
+        else:
+            logger.warning(f"Unknown parser type: {parser_type}")
+            return []
+    
+    def parse_pertanyaan_format(self, content: str, filename: str) -> List[Dict]:
+        """Parse '## 📋 PERTANYAAN #N' format"""
+        cases = []
+        pattern = r'##\s+📋\s+PERTANYAAN\s+#(\d+)'
+        parts = re.split(pattern, content)
+        
+        for i in range(1, len(parts), 2):
+            if i+1 < len(parts):
+                case_num = parts[i]
+                case_content = parts[i+1]
+                case = self.parse_case_content(case_num, case_content, filename)
+                if case:
+                    cases.append(case)
+        
+        return cases
+    
+    def parse_qa_format(self, content: str, filename: str) -> List[Dict]:
+        """Parse Q&A format like '**Q1:**'"""
+        cases = []
+        
+        # Try multiple Q&A patterns
+        patterns = [
+            r'\*\*Q(\d+):\s*(.*?)\*\*',
+            r'##\s+\*\*Q(\d+):',
+            r'Q(\d+):\s+(.+?)(?=Q\d+:|$)',
+        ]
+        
+        for pattern in patterns:
+            matches = re.finditer(pattern, content, re.DOTALL)
+            for match in matches:
+                case_num = match.group(1)
+                # Get content after this Q until next Q or end
+                start_pos = match.end()
+                next_q = re.search(r'\*\*Q\d+:', content[start_pos:])
+                end_pos = start_pos + next_q.start() if next_q else len(content)
+                case_content = content[start_pos:end_pos]
+                
+                case = self.parse_qa_content(case_num, case_content, filename)
+                if case:
+                    cases.append(case)
+            
+            if cases:  # If we found cases with this pattern, stop trying others
+                break
+        
+        return cases
+    
+    def parse_pertanyaan_simple(self, content: str, filename: str) -> List[Dict]:
+        """Parse simple '## PERTANYAAN N' format"""
+        cases = []
+        pattern = r'##\s+PERTANYAAN\s+(\d+)'
+        parts = re.split(pattern, content)
+        
+        for i in range(1, len(parts), 2):
+            if i+1 < len(parts):
+                case_num = parts[i]
+                case_content = parts[i+1]
+                case = self.parse_case_content(case_num, case_content, filename)
+                if case:
+                    cases.append(case)
+        
+        return cases
+    
+    def parse_case_content(self, num: str, content: str, filename: str) -> Optional[Dict]:
+        """Parse detailed case content with sections"""
+        
+        # Extract KASUS
+        kasus_match = re.search(r'\*\*KASUS:\*\*\s*(.*?)(?=\*\*PERTANYAAN:|\*\*JAWABAN:|###|\Z)', content, re.DOTALL)
+        kasus = kasus_match.group(1).strip() if kasus_match else ""
+        
+        # Extract PERTANYAAN
+        pertanyaan_match = re.search(r'\*\*PERTANYAAN:\*\*\s*(.*?)(?=---|###|\*\*JAWABAN:|\Z)', content, re.DOTALL)
+        pertanyaan = pertanyaan_match.group(1).strip() if pertanyaan_match else ""
+        
+        # Extract JAWABAN
+        jawaban_match = re.search(r'###\s+⚖️\s+JAWABAN\s*(.*?)(?=###|\Z)', content, re.DOTALL)
+        if not jawaban_match:
+            jawaban_match = re.search(r'\*\*JAWABAN:\*\*\s*(.*?)(?=###|\Z)', content, re.DOTALL)
+        jawaban = jawaban_match.group(1).strip() if jawaban_match else ""
+        
+        # Extract DASAR HUKUM
+        dasar_hukum_match = re.search(r'###\s+📖\s+DASAR HUKUM\s*(.*?)(?=###|\Z)', content, re.DOTALL)
+        dasar_hukum = dasar_hukum_match.group(1).strip() if dasar_hukum_match else ""
+        
+        # Extract ANALISIS
+        analisis_match = re.search(r'###\s+🔍\s+ANALISIS\s*(.*?)(?=##|\Z)', content, re.DOTALL)
+        analisis = analisis_match.group(1).strip() if analisis_match else ""
+        
+        # Extract references
+        pasal = self.extract_pasal(content)
+        uu = self.extract_uu(content)
+        
+        # Build case
+        case = {
+            'id': f'case_{num}',
+            'number': num,
+            'file': filename,
+            'kasus': kasus[:500] if kasus else "",
+            'pertanyaan': pertanyaan[:500] if pertanyaan else "",
+            'jawaban': jawaban[:800] if jawaban else "",
+            'dasar_hukum': dasar_hukum[:500] if dasar_hukum else "",
+            'analisis': analisis[:1000] if analisis else "",
+            'pasal': pasal,
+            'uu': uu,
+            'full_content': content[:2000]
+        }
+        
+        # Only return if we have meaningful content
+        if kasus or pertanyaan or jawaban:
+            return case
+        return None
+    
+    def parse_qa_content(self, num: str, content: str, filename: str) -> Optional[Dict]:
+        """Parse Q&A style content"""
+        
+        # For Q&A format, the question is usually at the start
+        lines = content.strip().split('\n')
+        question = lines[0].strip() if lines else ""
+        
+        # Answer is usually after "JAWABAN:" or "A:"
+        answer_match = re.search(r'(?:JAWABAN:|A:)\s*(.*?)(?=\*\*Q\d+:|\Z)', content, re.DOTALL)
+        answer = answer_match.group(1).strip() if answer_match else content[:500]
+        
+        # Extract references
+        pasal = self.extract_pasal(content)
+        uu = self.extract_uu(content)
+        
+        case = {
+            'id': f'case_{num}',
+            'number': num,
+            'file': filename,
+            'kasus': "",
+            'pertanyaan': question[:500],
+            'jawaban': answer[:800],
+            'dasar_hukum': "",
+            'analisis': "",
+            'pasal': pasal,
+            'uu': uu,
+            'full_content': content[:2000]
+        }
+        
+        if question or answer:
+            return case
+        return None
+    
+    def extract_pasal(self, content: str) -> List[str]:
+        """Extract Pasal references"""
+        pasal_patterns = [
+            r'Pasal\s+(\d+(?:\s+ayat\s+\(\d+\))?)\s+([A-Z]+)',
+            r'Pasal\s+(\d+)',
+        ]
+        
+        pasal_list = []
+        for pattern in pasal_patterns:
+            matches = re.findall(pattern, content)
+            for match in matches:
+                if isinstance(match, tuple):
+                    if len(match) == 2:
+                        pasal_list.append(f"Pasal {match[0]} {match[1]}")
+                    else:
+                        pasal_list.append(f"Pasal {match[0]}")
+                else:
+                    pasal_list.append(f"Pasal {match}")
+        
+        # Remove duplicates
+        seen = set()
+        unique = []
+        for p in pasal_list:
+            if p not in seen:
+                seen.add(p)
+                unique.append(p)
+        
+        return unique[:10]
+    
+    def extract_uu(self, content: str) -> List[str]:
+        """Extract UU references"""
+        uu_patterns = [
+            r'UU\s+No\.?\s*(\d+)\s+Tahun\s+(\d+)',
+            r'UU\s+No\.?\s*(\d+)/(\d+)',
+        ]
+        
+        uu_list = []
+        for pattern in uu_patterns:
+            matches = re.findall(pattern, content)
+            for match in matches:
+                uu_list.append(f"UU No. {match[0]} Tahun {match[1]}")
+        
+        return list(set(uu_list))[:5]
+    
+    def search(self, query: str, top_k: int = 3) -> List[Dict]:
+        """Search for relevant cases"""
+        query_lower = query.lower()
+        
+        scored_cases = []
+        for case in self.cases:
+            score = self.calculate_relevance(query_lower, case)
+            if score > 0:
+                scored_cases.append((score, case))
+        
+        scored_cases.sort(reverse=True, key=lambda x: x[0])
+        return [case for score, case in scored_cases[:top_k]]
+    
+    def calculate_relevance(self, query: str, case: Dict) -> float:
+        """Calculate relevance score"""
+        score = 0.0
+        
+        # High weight for pertanyaan
+        if query in case.get('pertanyaan', '').lower():
+            score += 20
+        
+        # Medium weight for kasus
+        if query in case.get('kasus', '').lower():
+            score += 10
+        
+        # Lower weight for jawaban/analisis
+        if query in case.get('jawaban', '').lower():
+            score += 5
+        if query in case.get('analisis', '').lower():
+            score += 3
+        
+        # Keyword matching
+        keywords = [k for k in query.split() if len(k) > 3]
+        for keyword in keywords:
+            full_content = case.get('full_content', '').lower()
+            count = full_content.count(keyword)
+            score += count * 2
+        
+        return score
+    
+    def get_stats(self) -> Dict:
+        """Get loading statistics"""
+        return {
+            'total_cases': len(self.cases),
+            'files_loaded': self.stats['files_loaded'],
+            'cases_by_file': {},
+            'errors': self.stats['errors']
+        }

legalmind/providers/fallback/versalaw2_core/enhanced_system.py

@@ -0,0 +1,151 @@
+#!/usr/bin/env python3
+"""
+Enhanced VersaLaw2 System with Maya Wisdom Integration
+Combines MayaLaw cases + Maya Wisdom + AI
+"""
+
+import sys
+from pathlib import Path
+import logging
+
+# Add maya-legal-system to path
+maya_system_path = Path("/root/dragon/global/mayalaw/maya-legal-system")
+if maya_system_path.exists():
+    sys.path.insert(0, str(maya_system_path))
+
+from .system import VersaLaw2System
+from .config import Config
+
+logger = logging.getLogger(__name__)
+
+class EnhancedVersaLaw2System(VersaLaw2System):
+    """
+    Enhanced system with Maya Wisdom integration
+    
+    Features:
+    - MayaLaw cases (126 cases)
+    - Maya Wisdom knowledge base
+    - AI processing
+    - Combined context
+    """
+    
+    def __init__(self, config=None):
+        """Initialize enhanced system"""
+        super().__init__(config)
+        
+        # Try to load Maya Wisdom
+        self.wisdom = None
+        self.wisdom_available = False
+        
+        try:
+            from core.maya_wisdom_processor import MayaWisdomProcessor
+            self.wisdom = MayaWisdomProcessor()
+            self.wisdom_available = True
+            print("✅ Maya Wisdom Processor loaded")
+            logger.info("Maya Wisdom Processor loaded successfully")
+        except ImportError as e:
+            print("⚠️  Maya Wisdom not available (optional)")
+            logger.warning(f"Maya Wisdom not loaded: {e}")
+        except Exception as e:
+            print(f"⚠️  Error loading Maya Wisdom: {e}")
+            logger.error(f"Maya Wisdom error: {e}")
+    
+    def ask(self, question, use_cache=True, include_wisdom=True):
+        """
+        Enhanced ask with Maya Wisdom
+        
+        Args:
+            question: Legal question
+            use_cache: Use caching
+            include_wisdom: Include Maya Wisdom in context
+        
+        Returns:
+            Enhanced result with wisdom
+        """
+        
+        # Get wisdom if available and requested
+        wisdom_response = None
+        if include_wisdom and self.wisdom_available:
+            try:
+                wisdom_response = self.wisdom.process_legal_question(question)
+                logger.info("Maya Wisdom response generated")
+            except Exception as e:
+                logger.warning(f"Maya Wisdom processing error: {e}")
+        
+        # Get standard result
+        result = super().ask(question, use_cache)
+        
+        # Enhance result with wisdom
+        if wisdom_response:
+            result['wisdom'] = wisdom_response
+            result['enhanced'] = True
+            
+            # Add wisdom to metadata
+            result['metadata']['wisdom_type'] = wisdom_response.get('type', 'unknown')
+            result['metadata']['wisdom_confidence'] = wisdom_response.get('confidence', 0.0)
+        else:
+            result['enhanced'] = False
+        
+        return result
+    
+    def print_answer(self, result):
+        """Enhanced print with wisdom"""
+        
+        # Print wisdom if available
+        if result.get('enhanced') and result.get('wisdom'):
+            wisdom = result['wisdom']
+            
+            print(f"{'='*60}")
+            print("🧠 MAYA WISDOM")
+            print(f"{'='*60}\n")
+            
+            print(f"Type: {wisdom.get('type', 'N/A')}")
+            print(f"Confidence: {wisdom.get('confidence', 0):.0%}\n")
+            
+            if 'answer' in wisdom:
+                print(f"Basic Knowledge:")
+                print(f"{wisdom['answer']}\n")
+            
+            if 'details' in wisdom:
+                print(f"Details:")
+                for key, value in wisdom['details'].items():
+                    print(f"  • {key}: {value}")
+                print()
+        
+        # Print standard answer
+        super().print_answer(result)
+    
+    def get_stats(self):
+        """Enhanced stats with wisdom info"""
+        stats = super().get_stats()
+        
+        stats['wisdom'] = {
+            'available': self.wisdom_available,
+            'loaded': self.wisdom is not None
+        }
+        
+        return stats
+
+def create_enhanced_system(ai_provider='mock', api_key=None):
+    """
+    Helper function to create enhanced system
+    
+    Args:
+        ai_provider: 'openai', 'deepseek', 'qodo', or 'mock'
+        api_key: API key for the provider
+    
+    Returns:
+        EnhancedVersaLaw2System instance
+    """
+    config = Config()
+    config['ai_provider'] = ai_provider
+    
+    if api_key:
+        if ai_provider == 'openai':
+            config['openai_api_key'] = api_key
+        elif ai_provider == 'deepseek':
+            config['deepseek_api_key'] = api_key
+        elif ai_provider == 'qodo':
+            config['qodo_api_key'] = api_key
+    
+    return EnhancedVersaLaw2System(config)