ebk 0.1.0__py3-none-any.whl → 0.3.2__py3-none-any.whl

ebk/ai/question_generator.py

@@ -0,0 +1,328 @@
+"""
+Question generator for active recall and comprehension testing.
+"""
+
+import random
+import re
+from typing import List, Dict, Any, Optional
+from dataclasses import dataclass
+
+
+@dataclass
+class Question:
+    """Represents a question for active recall."""
+    question_text: str
+    answer: str
+    question_type: str  # 'factual', 'conceptual', 'application', 'synthesis'
+    difficulty: str  # 'easy', 'medium', 'hard'
+    context: Optional[str] = None
+    hints: List[str] = None
+
+
+class QuestionGenerator:
+    """
+    Generate questions from text for active recall and comprehension testing.
+    """
+
+    def __init__(self):
+        self.question_templates = {
+            'factual': [
+                "What is {concept}?",
+                "Define {term}.",
+                "Who {action}?",
+                "When did {event} occur?",
+                "List the main characteristics of {topic}."
+            ],
+            'conceptual': [
+                "Explain the relationship between {concept1} and {concept2}.",
+                "Why is {concept} important?",
+                "What is the main idea of {topic}?",
+                "How does {concept} work?",
+                "Compare and contrast {item1} and {item2}."
+            ],
+            'application': [
+                "How would you apply {concept} to {scenario}?",
+                "Give an example of {concept} in practice.",
+                "What would happen if {condition}?",
+                "How could {concept} be used to solve {problem}?"
+            ],
+            'synthesis': [
+                "What conclusions can you draw from {evidence}?",
+                "How would you combine {concept1} and {concept2}?",
+                "What pattern emerges from {data}?",
+                "Predict the outcome if {scenario}."
+            ]
+        }
+
+    def generate_from_text(self, text: str, num_questions: int = 5) -> List[Question]:
+        """Generate questions from a text passage."""
+        questions = []
+
+        # Extract key information
+        sentences = self._split_sentences(text)
+        key_terms = self._extract_key_terms(text)
+        facts = self._extract_facts(sentences)
+
+        # Generate factual questions
+        for fact in facts[:num_questions // 2]:
+            question = self._create_factual_question(fact)
+            if question:
+                questions.append(question)
+
+        # Generate conceptual questions
+        for term in key_terms[:num_questions // 2]:
+            question = self._create_conceptual_question(term, text)
+            if question:
+                questions.append(question)
+
+        return questions[:num_questions]
+
+    def generate_from_highlights(self, highlights: List[str]) -> List[Question]:
+        """Generate questions from user highlights."""
+        questions = []
+
+        for highlight in highlights:
+            # Determine what type of content this is
+            if self._is_definition(highlight):
+                question = self._create_definition_question(highlight)
+            elif self._is_list(highlight):
+                question = self._create_list_question(highlight)
+            else:
+                question = self._create_explanation_question(highlight)
+
+            if question:
+                questions.append(question)
+
+        return questions
+
+    def _create_factual_question(self, fact: str) -> Optional[Question]:
+        """Create a factual question from a fact."""
+        # Simple pattern matching for fact extraction
+        patterns = [
+            (r"(\w+) is (\w+)", "What is {0}?", "{1}"),
+            (r"(\w+) was (\w+)", "What was {0}?", "{1}"),
+            (r"In (\d+), (\w+)", "When did {1} occur?", "{0}"),
+        ]
+
+        for pattern, question_template, answer_template in patterns:
+            match = re.search(pattern, fact, re.IGNORECASE)
+            if match:
+                groups = match.groups()
+                return Question(
+                    question_text=question_template.format(*groups),
+                    answer=answer_template.format(*groups),
+                    question_type='factual',
+                    difficulty='easy',
+                    context=fact
+                )
+
+        return None
+
+    def _create_conceptual_question(self, term: str, context: str) -> Question:
+        """Create a conceptual question about a term."""
+        question_text = f"Explain the concept of {term} based on the text."
+
+        # Extract sentences containing the term for the answer
+        sentences = [s for s in context.split('.') if term.lower() in s.lower()]
+        answer = ' '.join(sentences[:2]) if sentences else f"The text discusses {term}."
+
+        return Question(
+            question_text=question_text,
+            answer=answer,
+            question_type='conceptual',
+            difficulty='medium',
+            context=context[:200]
+        )
+
+    def _create_definition_question(self, highlight: str) -> Optional[Question]:
+        """Create a question from a definition."""
+        # Pattern: "X is defined as Y" or "X: Y"
+        patterns = [
+            r"(\w+) is defined as (.+)",
+            r"(\w+) means (.+)",
+            r"(\w+): (.+)"
+        ]
+
+        for pattern in patterns:
+            match = re.search(pattern, highlight, re.IGNORECASE)
+            if match:
+                term, definition = match.groups()
+                return Question(
+                    question_text=f"Define {term}.",
+                    answer=definition,
+                    question_type='factual',
+                    difficulty='easy',
+                    context=highlight
+                )
+
+        return None
+
+    def _create_list_question(self, highlight: str) -> Question:
+        """Create a question from a list."""
+        # Detect if highlight contains a list
+        list_items = re.findall(r'[•\-\*]\s*(.+)', highlight)
+
+        if list_items:
+            return Question(
+                question_text="List the main points mentioned.",
+                answer='\n'.join(list_items),
+                question_type='factual',
+                difficulty='easy',
+                context=highlight
+            )
+
+        return None
+
+    def _create_explanation_question(self, highlight: str) -> Question:
+        """Create an explanation question from a highlight."""
+        # Extract the main subject
+        first_sentence = highlight.split('.')[0]
+
+        return Question(
+            question_text=f"Explain the following concept: {first_sentence[:50]}...",
+            answer=highlight,
+            question_type='conceptual',
+            difficulty='medium',
+            context=highlight
+        )
+
+    def _is_definition(self, text: str) -> bool:
+        """Check if text is a definition."""
+        definition_patterns = [
+            r'\bis defined as\b',
+            r'\bmeans\b',
+            r'\brefers to\b',
+            r':\s*[A-Z]'  # Colon followed by capital letter
+        ]
+        return any(re.search(pattern, text, re.IGNORECASE) for pattern in definition_patterns)
+
+    def _is_list(self, text: str) -> bool:
+        """Check if text contains a list."""
+        return bool(re.search(r'[•\-\*]\s*\w+', text))
+
+    def _split_sentences(self, text: str) -> List[str]:
+        """Split text into sentences."""
+        sentences = re.split(r'[.!?]\s+', text)
+        return [s.strip() for s in sentences if s.strip()]
+
+    def _extract_key_terms(self, text: str) -> List[str]:
+        """Extract key terms from text."""
+        # Simple noun phrase extraction
+        # In production, use NLP libraries like spaCy
+        words = re.findall(r'\b[A-Z][a-z]+\b', text)
+        return list(set(words))[:10]
+
+    def _extract_facts(self, sentences: List[str]) -> List[str]:
+        """Extract factual statements from sentences."""
+        facts = []
+        fact_patterns = [
+            r'\bis\b',
+            r'\bwas\b',
+            r'\bare\b',
+            r'\bwere\b',
+            r'In \d+',
+            r'\bdefined as\b'
+        ]
+
+        for sentence in sentences:
+            if any(re.search(pattern, sentence) for pattern in fact_patterns):
+                facts.append(sentence)
+
+        return facts
+
+
+class QuizBuilder:
+    """
+    Build and manage quizzes from questions.
+    """
+
+    def __init__(self):
+        self.question_generator = QuestionGenerator()
+
+    def create_quiz(self, questions: List[Question],
+                   quiz_type: str = 'mixed',
+                   num_questions: int = 10) -> Dict[str, Any]:
+        """Create a quiz from questions."""
+        if quiz_type == 'factual':
+            filtered = [q for q in questions if q.question_type == 'factual']
+        elif quiz_type == 'conceptual':
+            filtered = [q for q in questions if q.question_type in ['conceptual', 'synthesis']]
+        else:
+            filtered = questions
+
+        # Randomly select questions
+        selected = random.sample(filtered, min(num_questions, len(filtered)))
+
+        return {
+            'quiz_id': self._generate_quiz_id(),
+            'questions': [
+                {
+                    'id': i,
+                    'question': q.question_text,
+                    'type': q.question_type,
+                    'difficulty': q.difficulty,
+                    'hints': q.hints or []
+                }
+                for i, q in enumerate(selected)
+            ],
+            'answers': {
+                i: q.answer for i, q in enumerate(selected)
+            },
+            'total_questions': len(selected)
+        }
+
+    def grade_quiz(self, quiz: Dict[str, Any],
+                  responses: Dict[int, str]) -> Dict[str, Any]:
+        """Grade a quiz based on responses."""
+        correct = 0
+        results = []
+
+        for q_id, response in responses.items():
+            correct_answer = quiz['answers'].get(q_id, '')
+            is_correct = self._check_answer(response, correct_answer)
+
+            if is_correct:
+                correct += 1
+
+            results.append({
+                'question_id': q_id,
+                'response': response,
+                'correct_answer': correct_answer,
+                'is_correct': is_correct
+            })
+
+        score = (correct / len(responses)) * 100 if responses else 0
+
+        return {
+            'score': score,
+            'correct': correct,
+            'total': len(responses),
+            'results': results
+        }
+
+    def _check_answer(self, response: str, correct: str) -> bool:
+        """Check if response matches correct answer (fuzzy matching)."""
+        # Simple check - can be improved with NLP
+        response_lower = response.lower().strip()
+        correct_lower = correct.lower().strip()
+
+        # Exact match
+        if response_lower == correct_lower:
+            return True
+
+        # Check if key terms are present
+        key_terms = re.findall(r'\b\w+\b', correct_lower)
+        important_terms = [t for t in key_terms if len(t) > 4]
+
+        if important_terms:
+            matches = sum(1 for term in important_terms if term in response_lower)
+            return matches >= len(important_terms) * 0.6
+
+        return False
+
+    def _generate_quiz_id(self) -> str:
+        """Generate unique quiz ID."""
+        import hashlib
+        from datetime import datetime
+        timestamp = datetime.now().isoformat()
+        return hashlib.md5(timestamp.encode()).hexdigest()[:8]

ebk/ai/reading_companion.py

@@ -0,0 +1,224 @@
+"""
+Reading Companion - Track reading sessions and provide intelligent assistance.
+"""
+
+import json
+from pathlib import Path
+from datetime import datetime, timedelta
+from typing import Dict, List, Any, Optional
+from dataclasses import dataclass, field
+import hashlib
+
+
+@dataclass
+class ReadingSession:
+    """Represents a reading session with tracking and insights."""
+    session_id: str
+    book_id: str
+    chapter: Optional[str] = None
+    start_time: datetime = field(default_factory=datetime.now)
+    end_time: Optional[datetime] = None
+    pages_read: int = 0
+    highlights: List[str] = field(default_factory=list)
+    notes: List[str] = field(default_factory=list)
+    comprehension_score: Optional[float] = None
+    quiz_results: List[Dict] = field(default_factory=list)
+
+    @property
+    def duration(self) -> timedelta:
+        """Calculate session duration."""
+        if self.end_time:
+            return self.end_time - self.start_time
+        return datetime.now() - self.start_time
+
+    @property
+    def reading_speed(self) -> float:
+        """Calculate reading speed in pages per hour."""
+        duration_hours = self.duration.total_seconds() / 3600
+        if duration_hours > 0:
+            return self.pages_read / duration_hours
+        return 0
+
+
+class ReadingCompanion:
+    """
+    AI-powered reading companion that tracks sessions and provides assistance.
+    """
+
+    def __init__(self, library_path: Path):
+        self.library_path = Path(library_path)
+        self.sessions_path = self.library_path / '.reading_sessions'
+        self.sessions_path.mkdir(exist_ok=True)
+
+        self.active_sessions: Dict[str, ReadingSession] = {}
+        self.completed_sessions: List[ReadingSession] = []
+
+        self.load_sessions()
+
+    def start_session(self, book_id: str, chapter: str = None) -> ReadingSession:
+        """Start a new reading session."""
+        session_id = self._generate_session_id(book_id)
+
+        session = ReadingSession(
+            session_id=session_id,
+            book_id=book_id,
+            chapter=chapter
+        )
+
+        self.active_sessions[session_id] = session
+        return session
+
+    def end_session(self, session_id: str) -> ReadingSession:
+        """End a reading session and save it."""
+        if session_id not in self.active_sessions:
+            raise ValueError(f"No active session with ID {session_id}")
+
+        session = self.active_sessions[session_id]
+        session.end_time = datetime.now()
+
+        # Move to completed
+        self.completed_sessions.append(session)
+        del self.active_sessions[session_id]
+
+        self.save_sessions()
+        return session
+
+    def add_highlight(self, session_id: str, text: str):
+        """Add a highlight to the current session."""
+        if session_id in self.active_sessions:
+            self.active_sessions[session_id].highlights.append(text)
+
+    def add_note(self, session_id: str, note: str):
+        """Add a note to the current session."""
+        if session_id in self.active_sessions:
+            self.active_sessions[session_id].notes.append(note)
+
+    def get_reading_stats(self, book_id: str = None) -> Dict[str, Any]:
+        """Get reading statistics for a book or all books."""
+        sessions = self.completed_sessions
+
+        if book_id:
+            sessions = [s for s in sessions if s.book_id == book_id]
+
+        if not sessions:
+            return {}
+
+        total_time = sum((s.duration for s in sessions), timedelta())
+        total_pages = sum(s.pages_read for s in sessions)
+        avg_speed = total_pages / (total_time.total_seconds() / 3600) if total_time.total_seconds() > 0 else 0
+
+        return {
+            'total_sessions': len(sessions),
+            'total_time': str(total_time),
+            'total_pages': total_pages,
+            'average_speed': avg_speed,
+            'total_highlights': sum(len(s.highlights) for s in sessions),
+            'total_notes': sum(len(s.notes) for s in sessions)
+        }
+
+    def get_reading_streak(self) -> int:
+        """Calculate current reading streak in days."""
+        if not self.completed_sessions:
+            return 0
+
+        # Sort sessions by date
+        sessions_by_date = {}
+        for session in self.completed_sessions:
+            date = session.start_time.date()
+            sessions_by_date[date] = True
+
+        # Check streak
+        streak = 0
+        current_date = datetime.now().date()
+
+        while current_date in sessions_by_date or current_date == datetime.now().date():
+            if current_date in sessions_by_date:
+                streak += 1
+            current_date -= timedelta(days=1)
+
+            if current_date not in sessions_by_date:
+                break
+
+        return streak
+
+    def save_sessions(self):
+        """Save sessions to disk."""
+        sessions_file = self.sessions_path / 'sessions.json'
+
+        data = {
+            'active': {
+                sid: {
+                    'session_id': s.session_id,
+                    'book_id': s.book_id,
+                    'chapter': s.chapter,
+                    'start_time': s.start_time.isoformat(),
+                    'pages_read': s.pages_read,
+                    'highlights': s.highlights,
+                    'notes': s.notes
+                }
+                for sid, s in self.active_sessions.items()
+            },
+            'completed': [
+                {
+                    'session_id': s.session_id,
+                    'book_id': s.book_id,
+                    'chapter': s.chapter,
+                    'start_time': s.start_time.isoformat(),
+                    'end_time': s.end_time.isoformat() if s.end_time else None,
+                    'pages_read': s.pages_read,
+                    'highlights': s.highlights,
+                    'notes': s.notes,
+                    'comprehension_score': s.comprehension_score,
+                    'quiz_results': s.quiz_results
+                }
+                for s in self.completed_sessions
+            ]
+        }
+
+        with open(sessions_file, 'w') as f:
+            json.dump(data, f, indent=2)
+
+    def load_sessions(self):
+        """Load sessions from disk."""
+        sessions_file = self.sessions_path / 'sessions.json'
+
+        if not sessions_file.exists():
+            return
+
+        with open(sessions_file, 'r') as f:
+            data = json.load(f)
+
+        # Load active sessions
+        for sid, sdata in data.get('active', {}).items():
+            session = ReadingSession(
+                session_id=sdata['session_id'],
+                book_id=sdata['book_id'],
+                chapter=sdata.get('chapter'),
+                start_time=datetime.fromisoformat(sdata['start_time']),
+                pages_read=sdata.get('pages_read', 0),
+                highlights=sdata.get('highlights', []),
+                notes=sdata.get('notes', [])
+            )
+            self.active_sessions[sid] = session
+
+        # Load completed sessions
+        for sdata in data.get('completed', []):
+            session = ReadingSession(
+                session_id=sdata['session_id'],
+                book_id=sdata['book_id'],
+                chapter=sdata.get('chapter'),
+                start_time=datetime.fromisoformat(sdata['start_time']),
+                end_time=datetime.fromisoformat(sdata['end_time']) if sdata.get('end_time') else None,
+                pages_read=sdata.get('pages_read', 0),
+                highlights=sdata.get('highlights', []),
+                notes=sdata.get('notes', []),
+                comprehension_score=sdata.get('comprehension_score'),
+                quiz_results=sdata.get('quiz_results', [])
+            )
+            self.completed_sessions.append(session)
+
+    def _generate_session_id(self, book_id: str) -> str:
+        """Generate unique session ID."""
+        timestamp = datetime.now().isoformat()
+        content = f"{book_id}:{timestamp}"
+        return hashlib.md5(content.encode()).hexdigest()[:12]