ableton-mcp-ai/mcp_server/test_arrangement.py

"""
Arrangement View Verification and Testing System for AbletonMCP_AI

Provides comprehensive verification, automated validation, and test scenarios
for Arrangement View functionality including clip creation, positioning,
integrity checks, and recording validation.

Author: AbletonMCP_AI
"""
import json
import logging
import os
import sqlite3
import socket
import time
import traceback
from dataclasses import dataclass, field
from datetime import datetime
from pathlib import Path
from typing import Any, Dict, List, Optional, Tuple, Callable, Union

logger = logging.getLogger("ArrangementVerifier")

# =============================================================================
# CONSTANTS AND CONFIGURATION
# =============================================================================

ABLETON_HOST = "127.0.0.1"
ABLETON_PORT = 9877
DEFAULT_TIMEOUT = 30.0
MAX_VERIFICATION_WAIT = 60.0

DB_PATH = Path(__file__).parent / "arrangement_tests.db"


# =============================================================================
# DATA CLASSES
# =============================================================================

@dataclass
class VerificationResult:
    """Result of a single verification check."""
    success: bool
    check_name: str
    message: str
    details: Dict[str, Any] = field(default_factory=dict)
    timestamp: float = field(default_factory=time.time)
    duration_ms: float = 0.0
    
    def to_dict(self) -> Dict[str, Any]:
        return {
            "success": self.success,
            "check_name": self.check_name,
            "message": self.message,
            "details": self.details,
            "timestamp": datetime.fromtimestamp(self.timestamp).isoformat(),
            "duration_ms": round(self.duration_ms, 2),
        }


@dataclass
class ClipInfo:
    """Information about a clip in Arrangement View."""
    name: str
    track_index: int
    track_name: str
    start_time: float
    end_time: float
    length: float
    is_midi: bool
    color: int = 0
    muted: bool = False
    looping: bool = False
    
    @classmethod
    def from_dict(cls, data: Dict[str, Any]) -> "ClipInfo":
        return cls(
            name=data.get("name", ""),
            track_index=data.get("track_index", 0),
            track_name=data.get("track_name", ""),
            start_time=data.get("start_time", 0.0),
            end_time=data.get("end_time", 0.0),
            length=data.get("length", 0.0),
            is_midi=data.get("is_midi", False),
            color=data.get("color", 0),
            muted=data.get("muted", False),
            looping=data.get("looping", False),
        )
    
    def to_dict(self) -> Dict[str, Any]:
        return {
            "name": self.name,
            "track_index": self.track_index,
            "track_name": self.track_name,
            "start_time": self.start_time,
            "end_time": self.end_time,
            "length": self.length,
            "is_midi": self.is_midi,
            "color": self.color,
            "muted": self.muted,
            "looping": self.looping,
        }


@dataclass
class TestScenario:
    """A test scenario with pre and post conditions."""
    name: str
    description: str
    pre_conditions: List[Callable[[], VerificationResult]]
    test_action: Callable[[], Dict[str, Any]]
    post_conditions: List[Callable[[], VerificationResult]]
    timeout_seconds: float = 30.0
    

@dataclass
class TestReport:
    """Complete test report with all results."""
    test_name: str
    started_at: str
    completed_at: str
    duration_seconds: float
    results: List[VerificationResult]
    summary: Dict[str, Any]
    
    def to_dict(self) -> Dict[str, Any]:
        return {
            "test_name": self.test_name,
            "started_at": self.started_at,
            "completed_at": self.completed_at,
            "duration_seconds": round(self.duration_seconds, 3),
            "results": [r.to_dict() for r in self.results],
            "summary": self.summary,
        }
    
    def to_json(self, indent: int = 2) -> str:
        return json.dumps(self.to_dict(), indent=indent)


# =============================================================================
# ARRANGEMENT VERIFIER CLASS
# =============================================================================

class ArrangementVerifier:
    """
    Main verification class for Arrangement View testing.
    
    Provides comprehensive verification methods for:
    - Clip creation and counting
    - Clip positioning and timing
    - Content validation
    - Integrity checks
    """
    
    def __init__(self, ableton_host: str = ABLETON_HOST, ableton_port: int = ABLETON_PORT):
        """
        Initialize the ArrangementVerifier.
        
        Args:
            ableton_host: Host where Ableton Live is running
            ableton_port: TCP port for Ableton connection
        """
        self.host = ableton_host
        self.port = ableton_port
        self._verification_results: List[VerificationResult] = []
        self._last_clips_snapshot: List[ClipInfo] = []
        self._db_connection: Optional[sqlite3.Connection] = None
        
    def _send_command(self, cmd_type: str, params: Dict[str, Any] = None, 
                      timeout: float = DEFAULT_TIMEOUT) -> Dict[str, Any]:
        """Send a command to Ableton and return the response."""
        sock = None
        try:
            sock = socket.create_connection((self.host, self.port), timeout=timeout)
            sock.settimeout(timeout)
            
            msg = json.dumps({"type": cmd_type, "params": params or {}}) + "\n"
            sock.sendall(msg.encode("utf-8"))
            
            buf = b""
            while True:
                chunk = sock.recv(65536)
                if not chunk:
                    break
                buf += chunk
                if b"\n" in buf:
                    raw, _, _ = buf.partition(b"\n")
                    response = json.loads(raw.decode("utf-8"))
                    return response
            
            return {"status": "error", "message": "No response received"}
        except socket.timeout:
            return {"status": "error", "message": f"Timeout after {timeout}s"}
        except ConnectionRefusedError:
            return {"status": "error", "message": f"Connection refused to {self.host}:{self.port}"}
        except Exception as e:
            return {"status": "error", "message": str(e)}
        finally:
            if sock:
                try:
                    sock.close()
                except Exception:
                    pass
    
    def _get_arrangement_clips(self, track_index: int = None) -> List[ClipInfo]:
        """Get all clips from Arrangement View."""
        params = {}
        if track_index is not None:
            params["track_index"] = track_index
            
        resp = self._send_command("get_arrangement_clips", params, timeout=15.0)
        
        if resp.get("status") != "success":
            return []
        
        result = resp.get("result", {})
        clips_data = result.get("clips", [])
        
        clips = []
        for clip_data in clips_data:
            if "start_time" in clip_data:
                clips.append(ClipInfo.from_dict(clip_data))
        
        return clips
    
    def verify_clips_created(self, expected_count: int, 
                            track_index: int = None) -> bool:
        """
        Verify that the expected number of clips exists in Arrangement View.
        
        Args:
            expected_count: Number of clips expected
            track_index: Optional track index to check (None = all tracks)
            
        Returns:
            True if clip count matches expected, False otherwise
        """
        start_time = time.time()
        clips = self._get_arrangement_clips(track_index)
        actual_count = len(clips)
        
        success = actual_count == expected_count
        duration_ms = (time.time() - start_time) * 1000
        
        result = VerificationResult(
            success=success,
            check_name="verify_clips_created",
            message=(f"Expected {expected_count} clips, found {actual_count}"
                    if not success else f"Found exactly {expected_count} clips"),
            details={
                "expected_count": expected_count,
                "actual_count": actual_count,
                "track_index": track_index,
                "clips": [c.name for c in clips],
            },
            duration_ms=duration_ms,
        )
        
        self._verification_results.append(result)
        
        if not success:
            logger.error(f"Clip count mismatch: expected {expected_count}, got {actual_count}")
        
        return success
    
    def verify_clip_positions(self, expected_positions: List[Dict[str, Any]], 
                             tolerance_beats: float = 0.01) -> bool:
        """
        Verify that clips are at expected positions.
        
        Args:
            expected_positions: List of dicts with keys:
                - track_index: int
                - start_time: float (in beats)
                - name: str (optional)
            tolerance_beats: Tolerance for position matching in beats
            
        Returns:
            True if all clips at expected positions, False otherwise
        """
        start_time = time.time()
        clips = self._get_arrangement_clips()
        
        errors = []
        matched = []
        
        for expected in expected_positions:
            exp_track = expected.get("track_index")
            exp_start = expected.get("start_time")
            exp_name = expected.get("name", "")
            
            # Find matching clip
            found = False
            for clip in clips:
                if exp_track is not None and clip.track_index != exp_track:
                    continue
                if exp_start is not None:
                    if abs(clip.start_time - exp_start) <= tolerance_beats:
                        if not exp_name or exp_name in clip.name:
                            found = True
                            matched.append({
                                "expected": expected,
                                "found": clip.to_dict(),
                            })
                            break
            
            if not found:
                errors.append({
                    "expected": expected,
                    "error": "No matching clip found",
                    "available_clips": [c.to_dict() for c in clips if exp_track is None or c.track_index == exp_track],
                })
        
        success = len(errors) == 0
        duration_ms = (time.time() - start_time) * 1000
        
        result = VerificationResult(
            success=success,
            check_name="verify_clip_positions",
            message=(f"All {len(expected_positions)} clips at expected positions"
                    if success else f"Failed to find {len(errors)} clips at expected positions"),
            details={
                "expected_count": len(expected_positions),
                "matched_count": len(matched),
                "error_count": len(errors),
                "matched": matched,
                "errors": errors,
                "tolerance_beats": tolerance_beats,
            },
            duration_ms=duration_ms,
        )
        
        self._verification_results.append(result)
        
        if not success:
            for err in errors:
                logger.error(f"Position mismatch: expected {err['expected']}, not found in arrangement")
        
        return success
    
    def verify_arrangement_has_content(self, min_clips: int = 1,
                                       min_length_beats: float = 0.0) -> bool:
        """
        Verify that Arrangement View has content (clips exist and have length).
        
        Args:
            min_clips: Minimum number of clips required
            min_length_beats: Minimum total length in beats
            
        Returns:
            True if arrangement has content, False otherwise
        """
        start_time = time.time()
        clips = self._get_arrangement_clips()
        
        clip_count = len(clips)
        total_length = max((c.end_time for c in clips), default=0.0)
        
        has_clips = clip_count >= min_clips
        has_length = total_length >= min_length_beats
        success = has_clips and has_length
        
        duration_ms = (time.time() - start_time) * 1000
        
        result = VerificationResult(
            success=success,
            check_name="verify_arrangement_has_content",
            message=(f"Arrangement has {clip_count} clips, total length {total_length:.1f} beats"
                    if success else f"Insufficient content: {clip_count} clips, {total_length:.1f} beats"),
            details={
                "clip_count": clip_count,
                "total_length_beats": total_length,
                "min_clips_required": min_clips,
                "min_length_required": min_length_beats,
                "has_clips": has_clips,
                "has_length": has_length,
            },
            duration_ms=duration_ms,
        )
        
        self._verification_results.append(result)
        
        if not success:
            logger.error(f"Arrangement lacks content: {clip_count} clips, {total_length:.1f} beats")
        
        return success
    
    def verify_clip_integrity(self, clip_info: Dict[str, Any]) -> bool:
        """
        Verify integrity of a specific clip.
        
        Checks:
        - Clip exists at specified location
        - Start time < End time
        - Length is positive
        - Track index is valid
        
        Args:
            clip_info: Dict with clip information to verify
            
        Returns:
            True if clip integrity verified, False otherwise
        """
        start_time = time.time()
        errors = []
        
        # Required fields
        required = ["track_index", "start_time", "end_time", "length"]
        for field in required:
            if field not in clip_info:
                errors.append(f"Missing required field: {field}")
        
        if errors:
            success = False
        else:
            # Validate values
            track_idx = clip_info.get("track_index")
            start = clip_info.get("start_time")
            end = clip_info.get("end_time")
            length = clip_info.get("length")
            
            if start >= end:
                errors.append(f"Invalid timing: start_time ({start}) >= end_time ({end})")
            
            if length <= 0:
                errors.append(f"Invalid length: {length} (must be positive)")
            
            expected_length = end - start
            if abs(length - expected_length) > 0.01:
                errors.append(f"Length mismatch: declared {length}, calculated {expected_length}")
            
            # Check track exists
            tracks_resp = self._send_command("get_tracks", timeout=10.0)
            if tracks_resp.get("status") == "success":
                track_count = len(tracks_resp.get("result", {}).get("tracks", []))
                if track_idx < 0 or track_idx >= track_count:
                    errors.append(f"Invalid track_index: {track_idx} (0-{track_count-1} available)")
            
            success = len(errors) == 0
        
        duration_ms = (time.time() - start_time) * 1000
        
        result = VerificationResult(
            success=success,
            check_name="verify_clip_integrity",
            message=("Clip integrity verified"
                    if success else f"Integrity check failed: {'; '.join(errors)}"),
            details={
                "clip_info": clip_info,
                "errors": errors,
            },
            duration_ms=duration_ms,
        )
        
        self._verification_results.append(result)
        
        if not success:
            logger.error(f"Clip integrity failed: {errors}")
        
        return success
    
    def get_verification_report(self) -> Dict[str, Any]:
        """
        Get comprehensive verification report.
        
        Returns:
            Dict with all verification results and summary statistics
        """
        total = len(self._verification_results)
        passed = sum(1 for r in self._verification_results if r.success)
        failed = total - passed
        
        total_duration_ms = sum(r.duration_ms for r in self._verification_results)
        
        # Group by check type
        by_type: Dict[str, List[VerificationResult]] = {}
        for r in self._verification_results:
            by_type.setdefault(r.check_name, []).append(r)
        
        summary = {
            "total_checks": total,
            "passed": passed,
            "failed": failed,
            "success_rate": round(passed / total * 100, 1) if total > 0 else 0.0,
            "total_duration_ms": round(total_duration_ms, 2),
            "by_check_type": {
                name: {
                    "total": len(results),
                    "passed": sum(1 for r in results if r.success),
                    "failed": sum(1 for r in results if not r.success),
                }
                for name, results in by_type.items()
            },
        }
        
        return {
            "timestamp": datetime.now().isoformat(),
            "results": [r.to_dict() for r in self._verification_results],
            "summary": summary,
        }
    
    def clear_results(self):
        """Clear all stored verification results."""
        self._verification_results = []
    
    def save_results_to_db(self, test_name: str) -> bool:
        """
        Save verification results to SQLite database.
        
        Args:
            test_name: Name identifier for this test run
            
        Returns:
            True if saved successfully, False otherwise
        """
        try:
            conn = sqlite3.connect(DB_PATH)
            cursor = conn.cursor()
            
            # Create table if not exists
            cursor.execute("""
                CREATE TABLE IF NOT EXISTS verification_results (
                    id INTEGER PRIMARY KEY AUTOINCREMENT,
                    test_name TEXT,
                    check_name TEXT,
                    success BOOLEAN,
                    message TEXT,
                    details TEXT,
                    timestamp TEXT,
                    duration_ms REAL
                )
            """)
            
            # Insert results
            for result in self._verification_results:
                cursor.execute("""
                    INSERT INTO verification_results 
                    (test_name, check_name, success, message, details, timestamp, duration_ms)
                    VALUES (?, ?, ?, ?, ?, ?, ?)
                """, (
                    test_name,
                    result.check_name,
                    result.success,
                    result.message,
                    json.dumps(result.details),
                    datetime.fromtimestamp(result.timestamp).isoformat(),
                    result.duration_ms,
                ))
            
            conn.commit()
            conn.close()
            return True
        except Exception as e:
            logger.error(f"Failed to save results to DB: {e}")
            return False


# =============================================================================
# HELPER FUNCTIONS
# =============================================================================

def wait_for_arrangement_content(verifier: ArrangementVerifier, 
                                  timeout: float = 30.0,
                                  poll_interval: float = 0.5,
                                  min_clips: int = 1) -> Tuple[bool, List[ClipInfo]]:
    """
    Wait for Arrangement View to have content.
    
    Polls Ableton until clips appear or timeout is reached.
    
    Args:
        verifier: ArrangementVerifier instance
        timeout: Maximum wait time in seconds
        poll_interval: Time between polls in seconds
        min_clips: Minimum number of clips to consider successful
        
    Returns:
        Tuple of (success, list of clips found)
    """
    start_time = time.time()
    
    while (time.time() - start_time) < timeout:
        clips = verifier._get_arrangement_clips()
        if len(clips) >= min_clips:
            logger.info(f"Found {len(clips)} clips after {time.time() - start_time:.1f}s")
            return True, clips
        time.sleep(poll_interval)
    
    logger.warning(f"Timeout waiting for content after {timeout}s")
    return False, []


def compare_arrangement_before_after(verifier: ArrangementVerifier,
                                     action: Callable[[], Any],
                                     expected_changes: Dict[str, Any] = None) -> Dict[str, Any]:
    """
    Compare Arrangement View before and after an action.
    
    Args:
        verifier: ArrangementVerifier instance
        action: Callable that performs the action
        expected_changes: Dict with expected changes:
            - min_new_clips: int
            - expected_positions: list of clip positions
            
    Returns:
        Comparison report with before/after state
    """
    # Capture before state
    before_clips = verifier._get_arrangement_clips()
    before_count = len(before_clips)
    before_end_time = max((c.end_time for c in before_clips), default=0.0)
    
    # Execute action
    action_start = time.time()
    try:
        action_result = action()
        action_success = True
    except Exception as e:
        action_result = str(e)
        action_success = False
    action_duration = time.time() - action_start
    
    # Wait briefly for arrangement to update
    time.sleep(0.5)
    
    # Capture after state
    after_clips = verifier._get_arrangement_clips()
    after_count = len(after_clips)
    after_end_time = max((c.end_time for c in after_clips), default=0.0)
    
    # Calculate differences
    new_clips = after_count - before_count
    length_added = after_end_time - before_end_time
    
    # Find new clip details
    before_positions = {(c.track_index, round(c.start_time, 2)): c for c in before_clips}
    new_clip_details = []
    for clip in after_clips:
        key = (clip.track_index, round(clip.start_time, 2))
        if key not in before_positions:
            new_clip_details.append(clip.to_dict())
    
    report = {
        "action_success": action_success,
        "action_result": action_result,
        "action_duration_seconds": round(action_duration, 3),
        "before": {
            "clip_count": before_count,
            "end_time_beats": before_end_time,
        },
        "after": {
            "clip_count": after_count,
            "end_time_beats": after_end_time,
        },
        "changes": {
            "new_clips": new_clips,
            "length_added_beats": length_added,
            "new_clip_details": new_clip_details[:10],  # Limit to first 10
        },
    }
    
    # Validate against expected changes
    if expected_changes:
        min_clips = expected_changes.get("min_new_clips", 0)
        report["validation"] = {
            "expected_min_new_clips": min_clips,
            "actual_new_clips": new_clips,
            "meets_expectations": new_clips >= min_clips,
        }
    
    return report


def assert_clip_properties(clip: Union[ClipInfo, Dict[str, Any]],
                          expected: Dict[str, Any],
                          tolerance: float = 0.01) -> VerificationResult:
    """
    Assert that a clip has expected properties.
    
    Args:
        clip: ClipInfo or dict with clip data
        expected: Dict of expected property values
        tolerance: Tolerance for floating point comparisons
        
    Returns:
        VerificationResult with success/failure details
    """
    start_time = time.time()
    
    if isinstance(clip, dict):
        clip_data = clip
    else:
        clip_data = clip.to_dict()
    
    mismatches = []
    
    for key, expected_value in expected.items():
        actual_value = clip_data.get(key)
        
        if actual_value is None:
            mismatches.append(f"Missing property: {key}")
            continue
        
        # Compare with tolerance for floats
        if isinstance(expected_value, float):
            if abs(actual_value - expected_value) > tolerance:
                mismatches.append(f"{key}: expected {expected_value}, got {actual_value}")
        elif actual_value != expected_value:
            mismatches.append(f"{key}: expected {expected_value}, got {actual_value}")
    
    success = len(mismatches) == 0
    duration_ms = (time.time() - start_time) * 1000
    
    return VerificationResult(
        success=success,
        check_name="assert_clip_properties",
        message=("All properties match" if success else f"Property mismatches: {mismatches}"),
        details={
            "clip": clip_data,
            "expected": expected,
            "mismatches": mismatches,
            "tolerance": tolerance,
        },
        duration_ms=duration_ms,
    )


# =============================================================================
# AUTOMATED VALIDATION
# =============================================================================

class ArrangementValidator:
    """
    Automated validation system for Arrangement View operations.
    
    Provides:
    - Pre-condition checks
    - Post-condition checks
    - Error collection and reporting
    """
    
    def __init__(self, verifier: ArrangementVerifier):
        self.verifier = verifier
        self.pre_check_results: List[VerificationResult] = []
        self.post_check_results: List[VerificationResult] = []
        self.errors: List[str] = []
    
    def pre_condition_checks(self) -> bool:
        """
        Run all pre-condition checks before performing arrangement operations.
        
        Checks:
        - Ableton is running and reachable
        - arrangement_overdub is available (via health check)
        - No corruption in current arrangement
        
        Returns:
            True if all pre-conditions met, False otherwise
        """
        self.pre_check_results = []
        
        # Check 1: Ableton is running
        resp = self.verifier._send_command("health_check", timeout=10.0)
        ableton_ok = resp.get("status") == "success"
        
        result = VerificationResult(
            success=ableton_ok,
            check_name="pre_ableton_running",
            message="Ableton is running and responding" if ableton_ok else "Ableton is not reachable",
            details={"health_response": resp.get("result", {}) if ableton_ok else resp.get("message")},
        )
        self.pre_check_results.append(result)
        
        if not ableton_ok:
            self.errors.append("Pre-condition failed: Ableton not running")
            return False
        
        # Check 2: Session info available
        resp = self.verifier._send_command("get_session_info", timeout=5.0)
        session_ok = resp.get("status") == "success"
        
        result = VerificationResult(
            success=session_ok,
            check_name="pre_session_info",
            message="Session info accessible" if session_ok else "Cannot read session info",
            details={"session": resp.get("result", {}) if session_ok else resp.get("message")},
        )
        self.pre_check_results.append(result)
        
        if not session_ok:
            self.errors.append("Pre-condition failed: Cannot read session info")
        
        # Check 3: Tracks accessible
        resp = self.verifier._send_command("get_tracks", timeout=5.0)
        tracks_ok = resp.get("status") == "success"
        track_count = len(resp.get("result", {}).get("tracks", [])) if tracks_ok else 0
        
        result = VerificationResult(
            success=tracks_ok and track_count > 0,
            check_name="pre_tracks_accessible",
            message=f"{track_count} tracks accessible" if tracks_ok else "Cannot read tracks",
            details={"track_count": track_count},
        )
        self.pre_check_results.append(result)
        
        if not tracks_ok or track_count == 0:
            self.errors.append(f"Pre-condition failed: No tracks available ({track_count} found)")
        
        # Check 4: arrangement_overdub availability (via session capabilities)
        session_result = resp.get("result", {}) if session_ok else {}
        # arrangement_overdub is typically available in Live 12
        overdub_available = session_ok  # Simplified check
        
        result = VerificationResult(
            success=overdub_available,
            check_name="pre_arrangement_overdub",
            message="Arrangement overdub available" if overdub_available else "Arrangement overdub not confirmed",
            details={},
        )
        self.pre_check_results.append(result)
        
        return all(r.success for r in self.pre_check_results)
    
    def post_condition_checks(self, expected_clips: int = None,
                              expected_duration: float = None) -> bool:
        """
        Run all post-condition checks after performing arrangement operations.
        
        Args:
            expected_clips: Expected number of clips (None = any)
            expected_duration: Expected total duration in beats (None = any)
            
        Returns:
            True if all post-conditions met, False otherwise
        """
        self.post_check_results = []
        
        # Check 1: Clips exist
        clips = self.verifier._get_arrangement_clips()
        clips_exist = len(clips) > 0
        
        result = VerificationResult(
            success=clips_exist,
            check_name="post_clips_exist",
            message=f"{len(clips)} clips in arrangement" if clips_exist else "No clips found in arrangement",
            details={"clip_count": len(clips), "clips": [c.name for c in clips[:5]]},
        )
        self.post_check_results.append(result)
        
        if expected_clips is not None and len(clips) != expected_clips:
            self.errors.append(f"Post-condition failed: Expected {expected_clips} clips, got {len(clips)}")
        
        # Check 2: Clip positions are valid (no negative start times)
        invalid_positions = [c for c in clips if c.start_time < 0]
        positions_valid = len(invalid_positions) == 0
        
        result = VerificationResult(
            success=positions_valid,
            check_name="post_positions_valid",
            message="All clip positions valid" if positions_valid else f"{len(invalid_positions)} clips with invalid positions",
            details={"invalid_count": len(invalid_positions), "invalid_clips": [c.to_dict() for c in invalid_positions[:3]]},
        )
        self.post_check_results.append(result)
        
        if not positions_valid:
            self.errors.append(f"Post-condition failed: {len(invalid_positions)} clips have negative start times")
        
        # Check 3: No corruption (overlapping clips on same track - may be valid but flagged)
        # This is informational as overlapping clips can be intentional
        overlaps = []
        clips_by_track: Dict[int, List[ClipInfo]] = {}
        for c in clips:
            clips_by_track.setdefault(c.track_index, []).append(c)
        
        for track_idx, track_clips in clips_by_track.items():
            sorted_clips = sorted(track_clips, key=lambda x: x.start_time)
            for i in range(len(sorted_clips) - 1):
                if sorted_clips[i].end_time > sorted_clips[i + 1].start_time:
                    overlaps.append({
                        "track": track_idx,
                        "clip1": sorted_clips[i].name,
                        "clip2": sorted_clips[i + 1].name,
                        "overlap_beats": sorted_clips[i].end_time - sorted_clips[i + 1].start_time,
                    })
        
        result = VerificationResult(
            success=True,  # Overlaps are not necessarily errors
            check_name="post_no_corruption",
            message=f"{len(overlaps)} overlapping clips detected (informational)" if overlaps else "No clip overlaps detected",
            details={"overlaps": overlaps[:5]},
        )
        self.post_check_results.append(result)
        
        # Check 4: Total duration
        if clips:
            total_duration = max(c.end_time for c in clips)
        else:
            total_duration = 0.0
        
        duration_ok = expected_duration is None or abs(total_duration - expected_duration) < 1.0
        
        result = VerificationResult(
            success=duration_ok,
            check_name="post_duration_check",
            message=f"Total duration: {total_duration:.1f} beats" if duration_ok else f"Duration mismatch: expected ~{expected_duration}, got {total_duration}",
            details={"total_duration_beats": total_duration, "expected": expected_duration},
        )
        self.post_check_results.append(result)
        
        if not duration_ok:
            self.errors.append(f"Post-condition failed: Duration {total_duration} != expected {expected_duration}")
        
        return all(r.success for r in self.post_check_results)
    
    def get_validation_report(self) -> Dict[str, Any]:
        """Get complete validation report with all checks and errors."""
        return {
            "pre_checks": [r.to_dict() for r in self.pre_check_results],
            "post_checks": [r.to_dict() for r in self.post_check_results],
            "errors": self.errors,
            "all_pre_conditions_met": all(r.success for r in self.pre_check_results),
            "all_post_conditions_met": all(r.success for r in self.post_check_results),
        }


# =============================================================================
# TEST SCENARIOS
# =============================================================================

class ArrangementTestScenarios:
    """
    Collection of test scenarios for Arrangement View.
    
    Each scenario includes:
    - Pre-condition checks
    - Test action execution
    - Post-condition verification
    """
    
    def __init__(self, verifier: ArrangementVerifier):
        self.verifier = verifier
        self.validator = ArrangementValidator(verifier)
    
    def test_simple_arrangement_recording(self, duration_bars: int = 4) -> TestReport:
        """
        T023: Test simple arrangement recording.
        
        Records from Session to Arrangement for specified bars and verifies:
        - Recording completes successfully
        - Clips appear in Arrangement View
        - Clip positions are correct
        
        Args:
            duration_bars: Number of bars to record
            
        Returns:
            TestReport with full results
        """
        started_at = datetime.now().isoformat()
        start_time = time.time()
        
        self.verifier.clear_results()
        results = []
        
        # Step 1: Pre-conditions
        logger.info(f"[test_simple_arrangement_recording] Checking pre-conditions...")
        if not self.validator.pre_condition_checks():
            for result in self.validator.pre_check_results:
                results.append(result)
            
            return TestReport(
                test_name="test_simple_arrangement_recording",
                started_at=started_at,
                completed_at=datetime.now().isoformat(),
                duration_seconds=time.time() - start_time,
                results=results,
                summary={
                    "status": "FAILED",
                    "reason": "Pre-conditions not met",
                    "total_checks": len(results),
                    "passed": sum(1 for r in results if r.success),
                    "failed": sum(1 for r in results if not r.success),
                },
            )
        
        for result in self.validator.pre_check_results:
            results.append(result)
        
        # Step 2: Record to arrangement
        logger.info(f"[test_simple_arrangement_recording] Recording {duration_bars} bars...")
        
        def record_action():
            # This simulates the MCP command - in real test, this would call the actual MCP tool
            resp = self.verifier._send_command(
                "record_to_arrangement",
                {"duration_bars": duration_bars},
                timeout=60.0
            )
            return resp
        
        # Use compare_before_after pattern
        comparison = compare_arrangement_before_after(
            self.verifier,
            record_action,
            expected_changes={"min_new_clips": 1}
        )
        
        # Verify clips were created
        success = self.verifier.verify_arrangement_has_content(min_clips=1)
        
        # Step 3: Post-conditions
        logger.info(f"[test_simple_arrangement_recording] Checking post-conditions...")
        self.validator.post_condition_checks()
        for result in self.validator.post_check_results:
            results.append(result)
        
        completed_at = datetime.now().isoformat()
        duration = time.time() - start_time
        
        # Add verifier results
        results.extend(self.verifier._verification_results)
        
        summary = {
            "status": "PASSED" if all(r.success for r in results) else "FAILED",
            "total_checks": len(results),
            "passed": sum(1 for r in results if r.success),
            "failed": sum(1 for r in results if not r.success),
            "recording_comparison": comparison,
        }
        
        report = TestReport(
            test_name="test_simple_arrangement_recording",
            started_at=started_at,
            completed_at=completed_at,
            duration_seconds=duration,
            results=results,
            summary=summary,
        )
        
        logger.info(f"[test_simple_arrangement_recording] Completed: {summary['status']}")
        return report
    
    def test_build_arrangement_timeline(self) -> TestReport:
        """
        T021: Test building arrangement timeline structure.
        
        Creates a full arrangement structure (Intro→Build→Drop→Break→Outro)
        and verifies timeline positions.
        
        Returns:
            TestReport with full results
        """
        started_at = datetime.now().isoformat()
        start_time = time.time()
        
        self.verifier.clear_results()
        results = []
        
        # Pre-conditions
        if not self.validator.pre_condition_checks():
            for result in self.validator.pre_check_results:
                results.append(result)
            return TestReport(
                test_name="test_build_arrangement_timeline",
                started_at=started_at,
                completed_at=datetime.now().isoformat(),
                duration_seconds=time.time() - start_time,
                results=results,
                summary={"status": "FAILED", "reason": "Pre-conditions not met"},
            )
        
        for result in self.validator.pre_check_results:
            results.append(result)
        
        # Build arrangement
        song_config = {
            "bpm": 95,
            "structure": "intro_build_drop_break_outro",
            "tracks": [
                {
                    "name": "Kick",
                    "clips": [
                        {"name": "Kick Pattern", "start_time": 0, "duration": 64, "notes": []}
                    ]
                },
                {
                    "name": "Snare",
                    "clips": [
                        {"name": "Snare Pattern", "start_time": 16, "duration": 48, "notes": []}
                    ]
                }
            ]
        }
        
        def build_action():
            from engines.arrangement_engine import ArrangementBuilder
            builder = ArrangementBuilder()
            arrangement = builder.fill_arrangement_with_song(song_config)
            return arrangement.to_dict()
        
        try:
            arrangement_data = build_action()
            
            # Verify structure
            expected_positions = [
                {"track_index": 0, "start_time": 0.0, "name": "Kick"},
                {"track_index": 1, "start_time": 64.0, "name": "Snare"},  # Bar 16 * 4 beats
            ]
            
            success = self.verifier.verify_clip_positions(expected_positions, tolerance_beats=4.0)
            
        except Exception as e:
            logger.error(f"Build arrangement failed: {e}")
            results.append(VerificationResult(
                success=False,
                check_name="build_arrangement",
                message=f"Failed to build arrangement: {str(e)}",
                details={"traceback": traceback.format_exc()},
            ))
        
        # Post-conditions
        self.validator.post_condition_checks()
        for result in self.validator.post_check_results:
            results.append(result)
        
        results.extend(self.verifier._verification_results)
        
        summary = {
            "status": "PASSED" if all(r.success for r in results) else "FAILED",
            "total_checks": len(results),
            "passed": sum(1 for r in results if r.success),
            "failed": sum(1 for r in results if not r.success),
        }
        
        return TestReport(
            test_name="test_build_arrangement_timeline",
            started_at=started_at,
            completed_at=datetime.now().isoformat(),
            duration_seconds=time.time() - start_time,
            results=results,
            summary=summary,
        )
    
    def test_section_at_bar(self, section_bar: int = 8, section_name: str = "drop") -> TestReport:
        """
        Test creating a specific section at a bar position.
        
        Creates a section and verifies it's at the correct location.
        
        Args:
            section_bar: Bar where section should start
            section_name: Name of the section
            
        Returns:
            TestReport with full results
        """
        started_at = datetime.now().isoformat()
        start_time = time.time()
        
        self.verifier.clear_results()
        results = []
        
        # Pre-conditions
        if not self.validator.pre_condition_checks():
            for result in self.validator.pre_check_results:
                results.append(result)
            return TestReport(
                test_name="test_section_at_bar",
                started_at=started_at,
                completed_at=datetime.now().isoformat(),
                duration_seconds=time.time() - start_time,
                results=results,
                summary={"status": "FAILED", "reason": "Pre-conditions not met"},
            )
        
        for result in self.validator.pre_check_results:
            results.append(result)
        
        # Create section
        def create_section():
            from engines.arrangement_engine import ArrangementBuilder
            builder = ArrangementBuilder()
            marker = builder.create_section_marker(section_name, section_bar)
            return marker.to_dict()
        
        try:
            marker_data = create_section()
            
            # Verify section position
            actual_start = marker_data.get("start_bar")
            actual_end = marker_data.get("end_bar")
            
            position_correct = actual_start == section_bar
            duration_positive = actual_end > actual_start
            
            results.append(VerificationResult(
                success=position_correct and duration_positive,
                check_name="section_position",
                message=f"Section '{section_name}' at bar {actual_start}, ends at {actual_end}",
                details={
                    "expected_bar": section_bar,
                    "actual_start": actual_start,
                    "actual_end": actual_end,
                    "position_correct": position_correct,
                    "duration_positive": duration_positive,
                },
            ))
            
        except Exception as e:
            results.append(VerificationResult(
                success=False,
                check_name="create_section",
                message=f"Failed to create section: {str(e)}",
                details={"traceback": traceback.format_exc()},
            ))
        
        # Post-conditions
        self.validator.post_condition_checks()
        for result in self.validator.post_check_results:
            results.append(result)
        
        summary = {
            "status": "PASSED" if all(r.success for r in results) else "FAILED",
            "total_checks": len(results),
            "passed": sum(1 for r in results if r.success),
            "failed": sum(1 for r in results if not r.success),
        }
        
        return TestReport(
            test_name="test_section_at_bar",
            started_at=started_at,
            completed_at=datetime.now().isoformat(),
            duration_seconds=time.time() - start_time,
            results=results,
            summary=summary,
        )
    
    def test_without_numpy(self) -> TestReport:
        """
        Test that all functionality works without numpy dependency.
        
        Runs core verification methods using only SQLite and standard library.
        
        Returns:
            TestReport with full results
        """
        started_at = datetime.now().isoformat()
        start_time = time.time()
        
        self.verifier.clear_results()
        results = []
        
        # Verify no numpy is imported
        import sys
        numpy_loaded = "numpy" in sys.modules
        
        results.append(VerificationResult(
            success=not numpy_loaded,
            check_name="no_numpy_dependency",
            message="numpy not loaded" if not numpy_loaded else "numpy is loaded (may cause issues)",
            details={"numpy_in_sys_modules": numpy_loaded},
        ))
        
        # Run basic verifications that don't need numpy
        try:
            # Test database operations
            db_success = self.verifier.save_results_to_db("test_without_numpy")
            
            results.append(VerificationResult(
                success=db_success,
                check_name="sqlite_operations",
                message="SQLite operations successful" if db_success else "SQLite operations failed",
                details={},
            ))
            
            # Test clip counting
            clips = self.verifier._get_arrangement_clips()
            results.append(VerificationResult(
                success=True,  # Even 0 clips is valid
                check_name="clip_counting",
                message=f"Retrieved {len(clips)} clips without numpy",
                details={"clip_count": len(clips)},
            ))
            
        except Exception as e:
            results.append(VerificationResult(
                success=False,
                check_name="without_numpy_execution",
                message=f"Error running without numpy: {str(e)}",
                details={"traceback": traceback.format_exc()},
            ))
        
        summary = {
            "status": "PASSED" if all(r.success for r in results) else "FAILED",
            "total_checks": len(results),
            "passed": sum(1 for r in results if r.success),
            "failed": sum(1 for r in results if not r.success),
        }
        
        return TestReport(
            test_name="test_without_numpy",
            started_at=started_at,
            completed_at=datetime.now().isoformat(),
            duration_seconds=time.time() - start_time,
            results=results,
            summary=summary,
        )


# =============================================================================
# MCP INTEGRATION
# =============================================================================

def create_mcp_test_tools() -> List[Dict[str, Any]]:
    """
    Create test tool definitions for MCP integration.
    
    Returns:
        List of tool definitions that can be registered with MCP server
    """
    return [
        {
            "name": "run_arrangement_test",
            "description": "Run a specific Arrangement View test scenario",
            "parameters": {
                "type": "object",
                "properties": {
                    "test_name": {
                        "type": "string",
                        "enum": ["simple_recording", "build_timeline", "section_at_bar", "without_numpy"],
                        "description": "Name of test to run",
                    },
                    "duration_bars": {
                        "type": "number",
                        "default": 4,
                        "description": "Duration for recording tests",
                    },
                    "section_bar": {
                        "type": "number",
                        "default": 8,
                        "description": "Bar position for section tests",
                    },
                },
                "required": ["test_name"],
            },
        },
        {
            "name": "verify_arrangement_state",
            "description": "Verify current state of Arrangement View",
            "parameters": {
                "type": "object",
                "properties": {
                    "expected_clips": {
                        "type": "number",
                        "description": "Expected number of clips",
                    },
                    "expected_duration": {
                        "type": "number",
                        "description": "Expected total duration in beats",
                    },
                },
            },
        },
        {
            "name": "get_arrangement_report",
            "description": "Get comprehensive arrangement verification report",
            "parameters": {
                "type": "object",
                "properties": {},
            },
        },
    ]


def run_mcp_test(test_name: str, **kwargs) -> str:
    """
    Execute a test via MCP and return JSON result.
    
    This function is designed to be called from the MCP server as a tool handler.
    
    Args:
        test_name: Name of test to run
        **kwargs: Additional test parameters
        
    Returns:
        JSON string with test results
    """
    verifier = ArrangementVerifier()
    scenarios = ArrangementTestScenarios(verifier)
    
    test_map = {
        "simple_recording": lambda: scenarios.test_simple_arrangement_recording(
            duration_bars=kwargs.get("duration_bars", 4)
        ),
        "build_timeline": lambda: scenarios.test_build_arrangement_timeline(),
        "section_at_bar": lambda: scenarios.test_section_at_bar(
            section_bar=kwargs.get("section_bar", 8)
        ),
        "without_numpy": lambda: scenarios.test_without_numpy(),
    }
    
    if test_name not in test_map:
        return json.dumps({
            "status": "error",
            "message": f"Unknown test: {test_name}. Available: {list(test_map.keys())}",
        }, indent=2)
    
    try:
        report = test_map[test_name]()
        return report.to_json()
    except Exception as e:
        return json.dumps({
            "status": "error",
            "message": str(e),
            "traceback": traceback.format_exc(),
        }, indent=2)


def generate_test_report_json(verifier: ArrangementVerifier, 
                              test_name: str = "arrangement_verification") -> str:
    """
    Generate a comprehensive JSON report for MCP consumption.
    
    Args:
        verifier: ArrangementVerifier with results
        test_name: Name of the test run
        
    Returns:
        JSON string with complete report
    """
    report_data = verifier.get_verification_report()
    report_data["test_name"] = test_name
    report_data["generated_at"] = datetime.now().isoformat()
    
    return json.dumps(report_data, indent=2)


# =============================================================================
# MAIN / TEST ENTRY POINT
# =============================================================================

def run_all_tests() -> Dict[str, TestReport]:
    """
    Run all test scenarios and return reports.
    
    Returns:
        Dict mapping test names to TestReport objects
    """
    verifier = ArrangementVerifier()
    scenarios = ArrangementTestScenarios(verifier)
    
    reports = {}
    
    logger.info("=" * 70)
    logger.info("RUNNING ALL ARRANGEMENT VIEW TESTS")
    logger.info("=" * 70)
    
    # Test 1: Simple recording
    logger.info("\n[1/4] Running test_simple_arrangement_recording...")
    reports["simple_recording"] = scenarios.test_simple_arrangement_recording(duration_bars=4)
    
    # Test 2: Build timeline
    logger.info("\n[2/4] Running test_build_arrangement_timeline...")
    reports["build_timeline"] = scenarios.test_build_arrangement_timeline()
    
    # Test 3: Section at bar
    logger.info("\n[3/4] Running test_section_at_bar...")
    reports["section_at_bar"] = scenarios.test_section_at_bar(section_bar=8)
    
    # Test 4: Without numpy
    logger.info("\n[4/4] Running test_without_numpy...")
    reports["without_numpy"] = scenarios.test_without_numpy()
    
    # Summary
    logger.info("\n" + "=" * 70)
    logger.info("TEST SUMMARY")
    logger.info("=" * 70)
    
    for name, report in reports.items():
        status = report.summary.get("status", "UNKNOWN")
        passed = report.summary.get("passed", 0)
        total = report.summary.get("total_checks", 0)
        logger.info(f"  {name}: {status} ({passed}/{total} checks passed)")
    
    return reports


def main():
    """Main entry point for running tests from command line."""
    logging.basicConfig(
        level=logging.INFO,
        format="%(asctime)s [%(name)s] %(levelname)s: %(message)s"
    )
    
    print("=" * 70)
    print("ARRANGEMENT VIEW VERIFICATION AND TESTING SYSTEM")
    print("=" * 70)
    print()
    
    # Run all tests
    reports = run_all_tests()
    
    # Save results
    print("\n" + "=" * 70)
    print("SAVING RESULTS")
    print("=" * 70)
    
    for name, report in reports.items():
        json_path = Path(f"test_report_{name}.json")
        with open(json_path, "w") as f:
            f.write(report.to_json())
        print(f"  Saved: {json_path}")
    
    print("\nDone!")
    
    return reports


if __name__ == "__main__":
    main()