benchkit 0.21.0

//! Parser-specific analysis utilities
//!
//! This module provides specialized analysis capabilities for parser benchmarking,
//! including command throughput metrics, parsing pipeline analysis, and 
//! parser-specific quality metrics.

use crate ::measurement ::BenchmarkResult;
use std ::time ::Duration;
use std ::collections ::HashMap;

/// Parser performance analyzer with command-specific metrics
#[ derive(Debug, Clone) ]
pub struct ParserAnalyzer
{
  /// Name of the parser being analyzed
  pub parser_name: String,
  /// Number of commands processed
  pub command_count: u64,
  /// Total characters in input
  pub character_count: u64,
  /// Average command complexity (arguments, nesting, etc.)
  pub average_complexity: f64,
}

impl ParserAnalyzer
{
  /// Create a new parser analyzer
  pub fn new(parser_name: impl Into< String >, command_count: u64, character_count: u64) -> Self
  {
  Self
  {
   parser_name: parser_name.into(),
   command_count,
   character_count,
   average_complexity: 1.0,
 }
 }
  
  /// Set the average command complexity
  pub fn with_complexity(mut self, complexity: f64) -> Self
  {
  self.average_complexity = complexity;
  self
 }
  
  /// Analyze parser performance metrics
  pub fn analyze(&self, result: &BenchmarkResult) -> ParserMetrics
  {
  let mean_duration = result.mean_time();
  let mean_seconds = mean_duration.as_secs_f64();
  
  let commands_per_second = if mean_seconds > 0.0 
  {
   self.command_count as f64 / mean_seconds
 }
  else
  {
   0.0
 };
  
  let characters_per_second = if mean_seconds > 0.0
  {
   self.character_count as f64 / mean_seconds  
 }
  else
  {
   0.0
 };
  
  let tokens_per_second = commands_per_second * self.average_complexity;
  
  ParserMetrics
  {
   parser_name: self.parser_name.clone(),
   processing_time: mean_duration,
   commands_per_second,
   characters_per_second,
   tokens_per_second,
   throughput_mb_per_second: characters_per_second / 1_048_576.0,
   command_count: self.command_count,
   character_count: self.character_count,
   average_complexity: self.average_complexity,
 }
 }
  
  /// Compare multiple parser implementations
  pub fn compare_parsers(&self, results: &HashMap< String, BenchmarkResult >) -> ParserComparison
  {
  let mut metrics = HashMap ::new();
  
  for (name, result) in results
  {
   let parser_metrics = self.analyze(result);
   metrics.insert(name.clone(), parser_metrics);
 }
  
  ParserComparison
  {
   parser_name: self.parser_name.clone(),
   metrics,
 }
 }
}

/// Parser performance metrics
#[ derive(Debug, Clone) ]
pub struct ParserMetrics
{
  /// Parser name
  pub parser_name: String,
  /// Processing time
  pub processing_time: Duration,
  /// Commands processed per second
  pub commands_per_second: f64,
  /// Characters processed per second  
  pub characters_per_second: f64,
  /// Tokens processed per second (estimated)
  pub tokens_per_second: f64,
  /// Throughput in MB/s
  pub throughput_mb_per_second: f64,
  /// Total commands processed
  pub command_count: u64,
  /// Total characters processed
  pub character_count: u64,
  /// Average command complexity
  pub average_complexity: f64,
}

impl ParserMetrics
{
  /// Get human-readable commands per second description
  pub fn commands_description( &self ) -> String
  {
  if self.commands_per_second >= 1_000_000.0
  {
   format!("{:.1}M cmd/s", self.commands_per_second / 1_000_000.0)
 }
  else if self.commands_per_second >= 1_000.0
  {
   format!("{:.1}K cmd/s", self.commands_per_second / 1_000.0)
 }
  else
  {
   format!("{:.0} cmd/s", self.commands_per_second)
 }
 }
  
  /// Get human-readable tokens per second description
  pub fn tokens_description( &self ) -> String
  {
  if self.tokens_per_second >= 1_000_000.0
  {
   format!("{:.1}M tokens/s", self.tokens_per_second / 1_000_000.0)
 }
  else if self.tokens_per_second >= 1_000.0
  {
   format!("{:.1}K tokens/s", self.tokens_per_second / 1_000.0)
 }
  else
  {
   format!("{:.0} tokens/s", self.tokens_per_second)
 }
 }
  
  /// Get human-readable throughput description
  pub fn throughput_description( &self ) -> String
  {
  if self.throughput_mb_per_second >= 1.0
  {
   format!("{:.1} MB/s", self.throughput_mb_per_second)
 }
  else
  {
   format!("{:.0} KB/s", self.characters_per_second / 1024.0)
 }
 }
  
  /// Generate markdown report for parser metrics
  pub fn to_markdown( &self ) -> String
  {
  let mut report = String ::new();
  
  report.push_str(&format!("### {} Parser Analysis\n\n", self.parser_name));
  
  report.push_str(&format!("- **Commands processed** : {} ({:.1} avg complexity)\n", 
  self.command_count, self.average_complexity));
  report.push_str(&format!("- **Characters processed** : {} ({:.1} chars/cmd)\n", 
  self.character_count, 
  self.character_count as f64 / self.command_count as f64));
  report.push_str(&format!("- **Processing time** : {:.3?}\n", self.processing_time));
  
  report.push_str(&format!("- **Performance** : \n"));
  report.push_str(&format!("  - Commands: {}\n", self.commands_description()));
  report.push_str(&format!("  - Tokens: {}\n", self.tokens_description()));
  report.push_str(&format!("  - Throughput: {}\n", self.throughput_description()));
  
  report.push('\n');
  report
 }
}

/// Comparison of parser performance across implementations
#[ derive(Debug, Clone) ]
pub struct ParserComparison
{
  /// Parser name being compared
  pub parser_name: String,
  /// Parser metrics for each implementation
  pub metrics: HashMap< String, ParserMetrics >,
}

impl ParserComparison
{
  /// Get the fastest parser by commands per second
  pub fn fastest_parser( &self ) -> Option< (&String, &ParserMetrics) >
  {
  self.metrics
   .iter()
   .max_by(|a, b| a.1.commands_per_second.partial_cmp(&b.1.commands_per_second).unwrap())
 }
  
  /// Get the highest throughput parser
  pub fn highest_throughput( &self ) -> Option< (&String, &ParserMetrics) >
  {
  self.metrics
   .iter()
   .max_by(|a, b| a.1.throughput_mb_per_second.partial_cmp(&b.1.throughput_mb_per_second).unwrap())
 }
  
  /// Calculate performance speedups relative to baseline
  pub fn calculate_speedups(&self, baseline: &str) -> Option< HashMap<String, f64 >>
  {
  let baseline_rate = self.metrics.get(baseline)?.commands_per_second;
  
  if baseline_rate <= 0.0
  {
   return None;
 }
  
  let mut speedups = HashMap ::new();
  
  for (name, metrics) in &self.metrics
  {
   let speedup = metrics.commands_per_second / baseline_rate;
   speedups.insert(name.clone(), speedup);
 }
  
  Some(speedups)
 }
  
  /// Generate comprehensive parser comparison report
  pub fn to_markdown( &self ) -> String
  {
  let mut report = String ::new();
  
  report.push_str(&format!("## {} Parser Comparison\n\n", self.parser_name));
  
  // Executive summary
  if let Some((fastest_name, fastest_metrics)) = self.fastest_parser()
  {
   report.push_str(&format!("**Best performing parser** : {} ({})\n\n", 
   fastest_name,
   fastest_metrics.commands_description()));
 }
  
  // Detailed results table
  report.push_str("| Implementation | Commands/sec | Tokens/sec | Throughput | Avg Latency | Complexity |\n");
  report.push_str("|----------------|--------------|------------|------------|-------------|------------|\n");
  
  // Sort by commands per second (fastest first)
  let mut sorted_metrics: Vec< _ > = self.metrics.iter().collect();
  sorted_metrics.sort_by(|a, b| b.1.commands_per_second.partial_cmp(&a.1.commands_per_second).unwrap());
  
  for (name, metrics) in &sorted_metrics
  {
   let avg_latency = if metrics.commands_per_second > 0.0 
   {
  format!("{:.1} μs", 1_000_000.0 / metrics.commands_per_second)
 }
   else
   {
  "N/A".to_string()
 };
   
   report.push_str(&format!(
  "| {} | {} | {} | {} | {} | {:.1} |\n",
  name,
  metrics.commands_description(),
  metrics.tokens_description(),
  metrics.throughput_description(),
  avg_latency,
  metrics.average_complexity
 ));
 }
  
  report.push('\n');
  
  // Speedup analysis
  if self.metrics.len() > 1
  {
   let slowest_name = sorted_metrics.last().unwrap().0;
   if let Some(speedups) = self.calculate_speedups(slowest_name)
   {
  report.push_str("### Performance Speedups\n\n");
  report.push_str(&format!("*Relative to {} (baseline)*\n\n", slowest_name));
  
  for (name, _metrics) in &sorted_metrics
  {
   if let Some(speedup) = speedups.get(*name)
   {
  if *name != slowest_name
  {
   report.push_str(&format!("- **{}** : {:.1}x faster\n", name, speedup));
 }
 }
 }
  report.push('\n');
 }
 }
  
  report
 }
}

/// Parser pipeline stage analysis
#[ derive(Debug, Clone) ]
pub struct ParserPipelineAnalyzer
{
  /// Pipeline stages and their names
  stage_names: Vec< String >,
  /// Results for each pipeline stage
  stage_results: HashMap< String, BenchmarkResult >,
}

impl ParserPipelineAnalyzer
{
  /// Create a new pipeline analyzer
  pub fn new() -> Self
  {
  Self
  {
   stage_names: Vec ::new(),
   stage_results: HashMap ::new(),
 }
 }
  
  /// Add a pipeline stage result
  pub fn add_stage(&mut self, name: impl Into< String >, result: BenchmarkResult) -> &mut Self
  {
  let stage_name = name.into();
  self.stage_names.push(stage_name.clone());
  self.stage_results.insert(stage_name, result);
  self
 }
  
  /// Analyze pipeline bottlenecks
  pub fn analyze_bottlenecks( &self ) -> PipelineAnalysis
  {
  let mut stage_times = HashMap ::new();
  let mut total_time = Duration ::ZERO;
  
  for (name, result) in &self.stage_results
  {
   let mean_time = result.mean_time();
   stage_times.insert(name.clone(), mean_time);
   total_time += mean_time;
 }
  
  // Find bottleneck (slowest stage)
  let bottleneck = stage_times
   .iter()
   .max_by(|a, b| a.1.cmp(b.1))
   .map(|(name, time)| (name.clone(), *time));
  
  // Calculate stage percentages
  let mut stage_percentages = HashMap ::new();
  if total_time > Duration ::ZERO
  {
   for (name, time) in &stage_times
   {
  let percentage = time.as_secs_f64() / total_time.as_secs_f64() * 100.0;
  stage_percentages.insert(name.clone(), percentage);
 }
 }
  
  PipelineAnalysis
  {
   stage_times,
   stage_percentages,
   bottleneck,
   total_time,
   stage_count: self.stage_results.len(),
 }
 }
  
  /// Generate pipeline analysis report
  pub fn to_markdown( &self ) -> String
  {
  let analysis = self.analyze_bottlenecks();
  let mut report = String ::new();
  
  report.push_str("## Parser Pipeline Analysis\n\n");
  
  if let Some((bottleneck_name, bottleneck_time)) = &analysis.bottleneck
  {
   report.push_str(&format!("**Primary bottleneck** : {} ({:.2?})\n", 
   bottleneck_name, bottleneck_time));
   
   if let Some(percentage) = analysis.stage_percentages.get(bottleneck_name)
   {
  report.push_str(&format!("**Bottleneck impact** : {:.1}% of total processing time\n\n", percentage));
 }
 }
  
  // Pipeline breakdown table
  report.push_str("| Pipeline Stage | Time | Percentage | Reliability |\n");
  report.push_str("|----------------|------|------------|-------------|\n");
  
  // Sort stages by order they were added (pipeline order)
  for stage_name in &self.stage_names
  {
   if let (Some(time), Some(percentage), Some(result)) = (
  analysis.stage_times.get(stage_name),
  analysis.stage_percentages.get(stage_name),
  self.stage_results.get(stage_name)
 ) {
  let reliability = if result.is_reliable() { "✅ Reliable" } else { "⚠️ Variable" };
  
  report.push_str(&format!(
   "| {} | {:.2?} | {:.1}% | {} |\n",
   stage_name, time, percentage, reliability
 ));
 }
 }
  
  report.push('\n');
  
  // Performance recommendations
  report.push_str("### Optimization Recommendations\n\n");
  
  if let Some((bottleneck_name, _)) = &analysis.bottleneck
  {
   if let Some(percentage) = analysis.stage_percentages.get(bottleneck_name)
   {
  if *percentage > 50.0
  {
   report.push_str(&format!("🎯 **High Priority** : Optimize {} stage ({:.1}% of total time)\n", 
   bottleneck_name, percentage));
 }
  else if *percentage > 25.0
  {
   report.push_str(&format!("⚡ **Medium Priority** : Consider optimizing {} stage ({:.1}% of total time)\n", 
   bottleneck_name, percentage));
 }
 }
 }
  
  // Check for unreliable stages
  for (stage_name, result) in &self.stage_results
  {
   if !result.is_reliable()
   {
  let cv = result.coefficient_of_variation() * 100.0;
  report.push_str(&format!("📊 **Reliability Issue** : {} stage has high variability (CV: {:.1}%)\n", 
  stage_name, cv));
 }
 }
  
  report.push('\n');
  report
 }
}

/// Analysis results for parser pipeline
#[ derive(Debug, Clone) ]
pub struct PipelineAnalysis
{
  /// Time taken by each stage
  pub stage_times: HashMap< String, Duration >,
  /// Percentage of total time for each stage
  pub stage_percentages: HashMap< String, f64 >,
  /// Primary bottleneck (stage name and time)
  pub bottleneck: Option< (String, Duration) >,
  /// Total pipeline time
  pub total_time: Duration,
  /// Number of stages analyzed
  pub stage_count: usize,
}

impl Default for ParserPipelineAnalyzer
{
  fn default() -> Self
  {
  Self ::new()
 }
}