empathy-framework 4.6.6__py3-none-any.whl → 4.7.1__py3-none-any.whl

empathy_software_plugin/wizards/performance/profiler_parsers.py

@@ -1,278 +0,0 @@
-"""Profiler Output Parsers
-
-Parses output from various profilers (cProfile, perf, Chrome DevTools, etc.)
-
-Copyright 2025 Smart AI Memory, LLC
-Licensed under Fair Source 0.9
-"""
-
-import json
-import re
-from dataclasses import dataclass
-from enum import Enum
-from typing import Any
-
-
-class ProfilerType(Enum):
-    """Types of profilers"""
-
-    CPROFILE = "cprofile"
-    PYINSTRUMENT = "pyinstrument"
-    CHROME_DEVTOOLS = "chrome_devtools"
-    NODE_PROFILER = "node_profiler"
-    GOLANG_PPROF = "golang_pprof"
-
-
-@dataclass
-class FunctionProfile:
-    """Standardized function profile data.
-
-    Universal format across all profilers.
-    """
-
-    function_name: str
-    file_path: str
-    line_number: int
-    total_time: float  # seconds
-    self_time: float  # seconds (excluding called functions)
-    call_count: int
-    cumulative_time: float  # seconds
-    percent_total: float
-    profiler: str
-    children: list["FunctionProfile"] | None = None
-    metadata: dict[str, Any] | None = None
-
-    def to_dict(self) -> dict[str, Any]:
-        """Convert to dictionary"""
-        return {
-            "function_name": self.function_name,
-            "file_path": self.file_path,
-            "line_number": self.line_number,
-            "total_time": self.total_time,
-            "self_time": self.self_time,
-            "call_count": self.call_count,
-            "cumulative_time": self.cumulative_time,
-            "percent_total": self.percent_total,
-            "profiler": self.profiler,
-            "children": [c.to_dict() for c in (self.children or [])],
-            "metadata": self.metadata or {},
-        }
-
-
-class BaseProfilerParser:
-    """Base class for profiler parsers"""
-
-    def __init__(self, profiler_name: str):
-        self.profiler_name = profiler_name
-
-    def parse(self, data: str) -> list[FunctionProfile]:
-        """Parse profiler output"""
-        raise NotImplementedError(
-            f"{self.__class__.__name__}.parse() must be implemented. "
-            "Create a subclass of BaseProfilerParser and implement the parse() method. "
-            f"See CProfileParser, ChromeDevToolsParser, or SimpleJSONProfilerParser for examples."
-        )
-
-
-class CProfileParser(BaseProfilerParser):
-    """Parse Python cProfile output.
-
-    Handles both text and pstats format.
-    """
-
-    def __init__(self):
-        super().__init__("cprofile")
-
-    def parse(self, data: str) -> list[FunctionProfile]:
-        """Parse cProfile text output"""
-        profiles = []
-
-        # Pattern for cProfile text output:
-        # ncalls  tottime  percall  cumtime  percall filename:lineno(function)
-        pattern = r"(\d+)\s+([\d.]+)\s+([\d.]+)\s+([\d.]+)\s+([\d.]+)\s+(.+?):(\d+)\((.+?)\)"
-
-        for line in data.split("\n"):
-            match = re.match(pattern, line.strip())
-            if match:
-                (
-                    ncalls,
-                    tottime,
-                    percall_tot,
-                    cumtime,
-                    percall_cum,
-                    filepath,
-                    lineno,
-                    funcname,
-                ) = match.groups()
-
-                profiles.append(
-                    FunctionProfile(
-                        function_name=funcname,
-                        file_path=filepath,
-                        line_number=int(lineno),
-                        total_time=float(tottime),
-                        self_time=float(tottime),  # tottime is self time in cProfile
-                        call_count=int(ncalls),
-                        cumulative_time=float(cumtime),
-                        percent_total=0.0,  # Calculate later
-                        profiler=self.profiler_name,
-                    ),
-                )
-
-        # Calculate percentages
-        if profiles:
-            total_time = sum(p.total_time for p in profiles)
-            for profile in profiles:
-                profile.percent_total = (
-                    (profile.total_time / total_time * 100) if total_time > 0 else 0
-                )
-
-        # Sort by total time descending
-        profiles.sort(key=lambda p: p.total_time, reverse=True)
-
-        return profiles
-
-
-class ChromeDevToolsParser(BaseProfilerParser):
-    """Parse Chrome DevTools Performance profile.
-
-    JSON format from Chrome DevTools Performance tab.
-    """
-
-    def __init__(self):
-        super().__init__("chrome_devtools")
-
-    def parse(self, data: str) -> list[FunctionProfile]:
-        """Parse Chrome DevTools JSON"""
-        profiles = []
-
-        try:
-            profile_data = json.loads(data)
-
-            # Chrome DevTools format is complex - simplified parsing
-            # Look for function calls in trace events
-            events = profile_data.get("traceEvents", [])
-
-            function_times = {}
-
-            for event in events:
-                if event.get("ph") == "X":  # Complete events
-                    name = event.get("name", "unknown")
-                    dur = event.get("dur", 0) / 1000000  # Convert microseconds to seconds
-
-                    if name not in function_times:
-                        function_times[name] = {"total_time": 0, "call_count": 0}
-
-                    function_times[name]["total_time"] += dur
-                    function_times[name]["call_count"] += 1
-
-            # Convert to FunctionProfile
-            total_time = sum(stats["total_time"] for stats in function_times.values())
-
-            for func_name, stats in function_times.items():
-                profiles.append(
-                    FunctionProfile(
-                        function_name=func_name,
-                        file_path="",  # Chrome doesn't always provide
-                        line_number=0,
-                        total_time=stats["total_time"],
-                        self_time=stats["total_time"],
-                        call_count=stats["call_count"],
-                        cumulative_time=stats["total_time"],
-                        percent_total=(
-                            (stats["total_time"] / total_time * 100) if total_time > 0 else 0
-                        ),
-                        profiler=self.profiler_name,
-                    ),
-                )
-
-            profiles.sort(key=lambda p: p.total_time, reverse=True)
-
-        except json.JSONDecodeError:
-            pass
-
-        return profiles
-
-
-class SimpleJSONProfilerParser(BaseProfilerParser):
-    """Parse simple JSON profiler format.
-
-    For custom or simplified profiling data.
-    """
-
-    def __init__(self):
-        super().__init__("simple_json")
-
-    def parse(self, data: str) -> list[FunctionProfile]:
-        """Parse simple JSON format"""
-        profiles = []
-
-        try:
-            parsed = json.loads(data)
-
-            functions = parsed.get("functions", [])
-
-            for func in functions:
-                profiles.append(
-                    FunctionProfile(
-                        function_name=func.get("name", "unknown"),
-                        file_path=func.get("file", ""),
-                        line_number=func.get("line", 0),
-                        total_time=func.get("total_time", 0.0),
-                        self_time=func.get("self_time", 0.0),
-                        call_count=func.get("calls", 0),
-                        cumulative_time=func.get("cumulative_time", 0.0),
-                        percent_total=func.get("percent", 0.0),
-                        profiler=self.profiler_name,
-                    ),
-                )
-
-            profiles.sort(key=lambda p: p.total_time, reverse=True)
-
-        except json.JSONDecodeError:
-            pass
-
-        return profiles
-
-
-class ProfilerParserFactory:
-    """Factory for creating profiler parsers"""
-
-    _parsers = {
-        "cprofile": CProfileParser,
-        "chrome_devtools": ChromeDevToolsParser,
-        "simple_json": SimpleJSONProfilerParser,
-    }
-
-    @classmethod
-    def create(cls, profiler_type: str) -> BaseProfilerParser:
-        """Create parser for profiler type"""
-        parser_class = cls._parsers.get(profiler_type.lower())
-
-        if not parser_class:
-            raise ValueError(
-                f"Unsupported profiler: {profiler_type}. "
-                f"Supported: {', '.join(cls._parsers.keys())}",
-            )
-
-        return parser_class()
-
-
-def parse_profiler_output(profiler_type: str, data: str) -> list[FunctionProfile]:
-    """Convenience function to parse profiler output.
-
-    Args:
-        profiler_type: Type of profiler ("cprofile", "chrome_devtools", etc.)
-        data: Raw profiler output
-
-    Returns:
-        List of FunctionProfile objects
-
-    Example:
-        >>> profiles = parse_profiler_output("cprofile", profile_data)
-        >>> for profile in profiles[:5]:  # Top 5 slowest
-        ...     print(f"{profile.function_name}: {profile.total_time:.3f}s")
-
-    """
-    parser = ProfilerParserFactory.create(profiler_type)
-    return parser.parse(data)

empathy_software_plugin/wizards/performance/trajectory_analyzer.py

@@ -1,429 +0,0 @@
-"""Performance Trajectory Analyzer (Level 4)
-
-Analyzes performance trends to predict future bottlenecks.
-
-This is Level 4 Anticipatory Empathy - predicting performance degradation
-BEFORE it becomes critical.
-
-Copyright 2025 Smart AI Memory, LLC
-Licensed under Fair Source 0.9
-"""
-
-from dataclasses import dataclass
-from typing import Any
-
-
-@dataclass
-class PerformanceTrend:
-    """Trend analysis for a metric"""
-
-    metric_name: str
-    current_value: float
-    previous_value: float
-    change: float
-    change_percent: float
-    direction: str  # "improving", "degrading", "stable"
-    rate_of_change: float  # per day/hour
-    concerning: bool
-    reasoning: str
-
-
-@dataclass
-class TrajectoryPrediction:
-    """Performance trajectory prediction.
-
-    This is Level 4 - predicting BEFORE hitting limits.
-    """
-
-    trajectory_state: str  # "optimal", "degrading", "critical"
-    estimated_time_to_critical: str | None
-    trends: list[PerformanceTrend]
-    overall_assessment: str
-    confidence: float
-    recommendations: list[str]
-
-    def to_dict(self) -> dict[str, Any]:
-        """Convert to dictionary format for compatibility"""
-        return {
-            "trajectory_state": self.trajectory_state,
-            "estimated_time_to_critical": self.estimated_time_to_critical,
-            "trends": [
-                {
-                    "metric_name": t.metric_name,
-                    "current_value": t.current_value,
-                    "previous_value": t.previous_value,
-                    "change": t.change,
-                    "change_percent": t.change_percent,
-                    "direction": t.direction,
-                    "rate_of_change": t.rate_of_change,
-                    "concerning": t.concerning,
-                    "reasoning": t.reasoning,
-                    "severity": "HIGH" if t.concerning else "LOW",
-                }
-                for t in self.trends
-            ],
-            "overall_assessment": self.overall_assessment,
-            "confidence": self.confidence,
-            "recommendations": self.recommendations,
-        }
-
-    def __getitem__(self, key):
-        """Allow dict-style access for backward compatibility"""
-        d = self.to_dict()
-        return d[key]
-
-
-class PerformanceTrajectoryAnalyzer:
-    """Analyzes performance trajectory to predict degradation.
-
-    Level 4 Anticipatory Empathy implementation.
-    """
-
-    def __init__(self):
-        # Define acceptable ranges
-        self.acceptable_ranges = {
-            "response_time": (0, 1.0),  # seconds
-            "throughput": (100, float("inf")),  # requests/sec
-            "error_rate": (0, 0.01),  # 1%
-            "cpu_usage": (0, 0.80),  # 80%
-            "memory_usage": (0, 0.85),  # 85%
-        }
-
-        # Define concerning growth rates
-        self.concerning_rates = {
-            "response_time": 0.1,  # 100ms increase per day
-            "error_rate": 0.005,  # 0.5% increase
-            "memory_usage": 0.05,  # 5% increase
-        }
-
-    def analyze_trajectory(
-        self,
-        current_metrics,
-        historical_metrics=None,  # Can be dict or list  # Can be dict or list
-    ) -> TrajectoryPrediction:
-        """Analyze performance trajectory.
-
-        Accepts parameters in either order for backward compatibility:
-        - analyze_trajectory(current_metrics: dict, historical_metrics: list)
-        - analyze_trajectory(historical_metrics: list, current_metrics: dict)
-
-        Args:
-            current_metrics: Current performance metrics (dict) OR historical data (list)
-            historical_metrics: Historical data (list) OR current metrics (dict)
-
-        Returns:
-            TrajectoryPrediction with assessment
-
-        Example:
-            >>> history = [
-            ...     {"time": "day1", "response_time": 0.2, "error_rate": 0.001},
-            ...     {"time": "day2", "response_time": 0.45, "error_rate": 0.003},
-            ...     {"time": "day3", "response_time": 0.8, "error_rate": 0.007}
-            ... ]
-            >>> prediction = analyzer.analyze_trajectory(current_metrics, history)
-            >>> if prediction.trajectory_state == "degrading":
-            ...     print(f"ALERT: {prediction.overall_assessment}")
-
-        """
-        # Auto-detect parameter order (backward compatibility)
-        if isinstance(current_metrics, list) and isinstance(historical_metrics, dict):
-            # Parameters were passed in reverse order
-            current_metrics, historical_metrics = historical_metrics, current_metrics
-        elif isinstance(current_metrics, list) and historical_metrics is None:
-            # Only historical metrics provided
-            historical_metrics = current_metrics
-            current_metrics = historical_metrics[-1] if historical_metrics else {}
-
-        if not historical_metrics:
-            return TrajectoryPrediction(
-                trajectory_state="optimal",
-                estimated_time_to_critical=None,
-                trends=[],
-                overall_assessment="Insufficient historical data for trajectory analysis",
-                confidence=0.3,
-                recommendations=["Collect performance metrics over time"],
-            )
-
-        # Analyze trends for each metric
-        trends = []
-
-        for metric_name, current_value in current_metrics.items():
-            trend = self._analyze_metric_trend(metric_name, current_value, historical_metrics)
-
-            if trend:
-                trends.append(trend)
-
-        # Determine overall trajectory state
-        trajectory_state = self._determine_trajectory_state(trends)
-
-        # Estimate time to critical (if degrading)
-        time_to_critical = None
-        if trajectory_state in ["degrading", "critical"]:
-            time_to_critical = self._estimate_time_to_critical(trends, current_metrics)
-
-        # Generate assessment
-        assessment = self._generate_assessment(trajectory_state, trends, time_to_critical)
-
-        # Generate recommendations
-        recommendations = self._generate_recommendations(trajectory_state, trends)
-
-        # Calculate confidence
-        confidence = self._calculate_confidence(historical_metrics, trends)
-
-        return TrajectoryPrediction(
-            trajectory_state=trajectory_state,
-            estimated_time_to_critical=time_to_critical,
-            trends=trends,
-            overall_assessment=assessment,
-            confidence=confidence,
-            recommendations=recommendations,
-        )
-
-    def _analyze_metric_trend(
-        self,
-        metric_name: str,
-        current_value: float,
-        historical_metrics: list[dict[str, Any]],
-    ) -> PerformanceTrend | None:
-        """Analyze trend for single metric"""
-        # Validate current_value is numeric
-        try:
-            current_value = float(current_value)
-        except (ValueError, TypeError):
-            # Skip non-numeric fields (timestamps, etc.)
-            return None
-
-        # Extract historical values
-        historical_values = []
-        for entry in historical_metrics:
-            if metric_name in entry and entry[metric_name] is not None:
-                try:
-                    historical_values.append(float(entry[metric_name]))
-                except (ValueError, TypeError):
-                    # Skip non-numeric fields (timestamps, etc.)
-                    return None
-
-        if not historical_values:
-            return None
-
-        # Calculate overall trend from first to current
-        first_value = historical_values[0]
-        previous_value = historical_values[-1]
-
-        # Overall change from start to current
-        total_change = current_value - first_value
-        change_percent = (total_change / first_value * 100) if first_value != 0 else 0
-
-        # Determine direction based on overall trend
-        if abs(change_percent) < 5:
-            direction = "stable"
-        elif total_change > 0:
-            # For metrics like response_time, error_rate - increase is bad
-            if metric_name in ["response_time", "error_rate", "cpu_usage", "memory_usage"]:
-                direction = "degrading"
-            else:
-                direction = "improving"
-        elif metric_name in ["response_time", "error_rate", "cpu_usage", "memory_usage"]:
-            direction = "improving"
-        else:
-            direction = "degrading"
-
-        # Calculate rate of change (per time period)
-        time_periods = len(historical_values)
-        rate_of_change = abs(total_change) / time_periods if time_periods > 0 else 0
-
-        # Determine if concerning
-        concerning, reasoning = self._is_trend_concerning(
-            metric_name,
-            current_value,
-            total_change,
-            rate_of_change,
-            direction,
-        )
-
-        return PerformanceTrend(
-            metric_name=metric_name,
-            current_value=current_value,
-            previous_value=previous_value,
-            change=total_change,
-            change_percent=change_percent,
-            direction=direction,
-            rate_of_change=rate_of_change,
-            concerning=concerning,
-            reasoning=reasoning,
-        )
-
-    def _is_trend_concerning(
-        self,
-        metric_name: str,
-        current_value: float,
-        change: float,
-        rate_of_change: float,
-        direction: str,
-    ) -> tuple[bool, str]:
-        """Determine if trend is concerning"""
-        # Check if currently out of acceptable range
-        if metric_name in self.acceptable_ranges:
-            min_val, max_val = self.acceptable_ranges[metric_name]
-
-            if current_value < min_val:
-                return True, f"{metric_name} below acceptable range"
-            if current_value > max_val:
-                return True, f"{metric_name} above acceptable range ({max_val})"
-
-        # Check rate of change
-        if metric_name in self.concerning_rates:
-            threshold = self.concerning_rates[metric_name]
-
-            if direction == "degrading" and rate_of_change > threshold:
-                return True, f"{metric_name} degrading rapidly ({change:+.3f} per period)"
-
-        return False, "Within acceptable trajectory"
-
-    def _determine_trajectory_state(self, trends: list[PerformanceTrend]) -> str:
-        """Determine overall trajectory state"""
-        concerning_trends = [t for t in trends if t.concerning]
-
-        if not concerning_trends:
-            return "optimal"
-
-        # Count critical metrics
-        critical_metrics = ["response_time", "error_rate"]
-        critical_concerning = sum(1 for t in concerning_trends if t.metric_name in critical_metrics)
-
-        if critical_concerning >= 1:
-            return "critical"
-
-        if len(concerning_trends) >= 2:
-            return "degrading"
-
-        return "degrading"
-
-    def _estimate_time_to_critical(
-        self,
-        trends: list[PerformanceTrend],
-        current_metrics: dict[str, float],
-    ) -> str | None:
-        """Estimate time until metrics hit critical thresholds.
-
-        Core Level 4 - predicting the future.
-        """
-        for trend in trends:
-            if not trend.concerning:
-                continue
-
-            # Response time prediction
-            if trend.metric_name == "response_time" and trend.direction == "degrading":
-                critical_threshold = 1.0  # 1 second
-                current = trend.current_value
-                rate = trend.rate_of_change
-
-                if rate > 0 and current < critical_threshold:
-                    periods_to_critical = (critical_threshold - current) / rate
-                    if 0 < periods_to_critical < 30:  # Within 30 periods
-                        return f"~{int(periods_to_critical)} time periods"
-
-            # Memory usage prediction
-            if trend.metric_name == "memory_usage" and trend.direction == "degrading":
-                critical_threshold = 0.95  # 95%
-                current = trend.current_value
-                rate = trend.rate_of_change
-
-                if rate > 0 and current < critical_threshold:
-                    periods_to_critical = (critical_threshold - current) / rate
-                    if 0 < periods_to_critical < 30:
-                        return f"~{int(periods_to_critical)} time periods"
-
-        return None
-
-    def _generate_assessment(
-        self,
-        trajectory_state: str,
-        trends: list[PerformanceTrend],
-        time_to_critical: str | None,
-    ) -> str:
-        """Generate overall assessment"""
-        if trajectory_state == "optimal":
-            return "Performance metrics stable. System operating within acceptable ranges."
-
-        concerning = [t for t in trends if t.concerning]
-
-        if trajectory_state == "critical":
-            trends_desc = ", ".join(f"{t.metric_name} {t.direction}" for t in concerning[:3])
-            return (
-                f"CRITICAL performance trajectory: {trends_desc}. Immediate investigation required."
-            )
-
-        if trajectory_state == "degrading":
-            trends_desc = ", ".join(f"{t.metric_name} {t.direction}" for t in concerning[:3])
-
-            if time_to_critical:
-                return (
-                    f"Performance degrading: {trends_desc}. "
-                    f"In our experience, this pattern leads to service degradation. "
-                    f"Estimated time to critical: {time_to_critical}. "
-                    "Early optimization recommended."
-                )
-
-            return (
-                f"Performance degrading: {trends_desc}. "
-                "In our experience, this warrants investigation."
-            )
-
-        return "Performance trajectory under assessment."
-
-    def _generate_recommendations(
-        self,
-        trajectory_state: str,
-        trends: list[PerformanceTrend],
-    ) -> list[str]:
-        """Generate actionable recommendations"""
-        if trajectory_state == "optimal":
-            return ["Continue monitoring performance metrics"]
-
-        recommendations = []
-
-        if trajectory_state in ["degrading", "critical"]:
-            recommendations.append("Investigate performance degradation immediately")
-            recommendations.append("Review recent code changes")
-
-        concerning = [t for t in trends if t.concerning]
-
-        for trend in concerning:
-            if trend.metric_name == "response_time":
-                recommendations.append("Profile slow endpoints to identify bottlenecks")
-                recommendations.append("Consider adding caching or database optimization")
-            elif trend.metric_name == "memory_usage":
-                recommendations.append("Check for memory leaks")
-                recommendations.append("Review object lifecycle and garbage collection")
-            elif trend.metric_name == "error_rate":
-                recommendations.append("Review error logs for patterns")
-                recommendations.append("Add error monitoring and alerting")
-
-        if trajectory_state == "critical":
-            recommendations.append("Consider scaling resources immediately")
-
-        return list(dict.fromkeys(recommendations))  # Deduplicate (preserves order)
-
-    def _calculate_confidence(
-        self,
-        historical_metrics: list[dict[str, Any]],
-        trends: list[PerformanceTrend],
-    ) -> float:
-        """Calculate confidence in prediction"""
-        # More data = higher confidence
-        data_points = len(historical_metrics)
-        data_confidence = min(data_points / 10, 1.0)
-
-        # More consistent trends = higher confidence
-        if trends:
-            concerning_count = sum(1 for t in trends if t.concerning)
-            trend_confidence = concerning_count / len(trends) if trends else 0.5
-        else:
-            trend_confidence = 0.5
-
-        return (data_confidence + trend_confidence) / 2
-
-
-# Alias for backward compatibility
-TrajectoryAnalyzer = PerformanceTrajectoryAnalyzer