sparkforensics-mcp 0.3.0 → 0.4.0

package/vendor-core/threshold-summary.js

@@ -1,35 +1,13 @@
-const THRESHOLD_SUMMARIES                         = {
-  spill: 'single-task disk spill above 1 GiB',
-  shuffle: 'shuffle read above the configured minimum byte threshold',
-  skew: 'task duration skew above the configured ratio',
-  gc: 'JVM GC time share above the configured ratio',
-  slowHost: 'a host running 2x+ slower than its peers by mean task duration (per-executor byte/time dimensions use a separate, narrower ratio ladder starting at 1.33x; only those can reach critical on ratio alone)',
-  stageSlowness: 'a stage running far longer than its peers, not attributable to a single slow host',
-  straggler: 'one or more tasks finishing far after the rest of their stage',
-  speculationWaste: 'speculative task attempts that completed after the original',
-  tinyTask: 'median task duration below the configured floor',
-  partitionSizing: 'partition byte size outside the configured target range',
-  stageShape: 'low parallelism, data explosion, or task-count skew relative to core count',
-  stageFailed: 'a stage that failed outright',
-  failures: 'task failures above the configured rate',
-  retryWaste: 'retried task attempts consuming executor time',
-  coldStart: 'executor startup time above the configured floor',
-  incompleteRun: 'an event log missing its terminal ApplicationEnd/job-completion event',
-  utilization: 'core occupancy below the configured floor across the run',
-  memoryUtilization: 'executor heap usage outside the configured band',
-  cacheUtilization: 'cached partitions evicted or spilled to disk',
-  coreLocality: 'task placement missing data-local core assignment',
-  cachingOpportunity: 'a dataset re-read from source multiple times with no cache/persist',
-  jobFailureRate: 'job failure rate above the configured threshold',
-  autoscalingChurn: 'executor add/remove churn above the configured rate',
-  configAudit: 'a Spark conf value outside the recommended range',
-  duplicatePlanSubtree: 'the same physical plan subtree executed more than once',
-  smallFiles: 'output files below the configured target size',
-  overBroadcast: 'a broadcast join above the configured size ceiling',
-  underBroadcast: 'a join below the configured size floor that skipped broadcast',
-  broadcastSizing: 'a broadcast join outside the configured size range in either direction',
-};
+import { ENTRY_BY_TYPE,                         } from './detectors.js';
+import { presentationOf } from './finding-presentation.js';
+import { effectiveThresholds } from './threshold-overrides.js';
 
-export function getThresholdSummary(type        )         {
-  return THRESHOLD_SUMMARIES[type] ?? 'criteria not met';
+/** The clean-check criterion for a finding type, built from the thresholds of the first DETECTORS
+ * entry that emits it (configAudit's four entries share one summary), with `overrides` merged in
+ * when the run was tuned. */
+export function getThresholdSummary(type        , overrides                     )         {
+  const presentation = presentationOf(type);
+  const entry = ENTRY_BY_TYPE.get(type);
+  if (!presentation || !entry) return 'criteria not met';
+  return presentation.thresholdSummary(effectiveThresholds(entry, overrides)         );
 }

package/vendor-core/types.js

@@ -1,3 +1,6 @@
+                                                  
+                                                                
+
                              
                                                          
 
@@ -287,48 +290,9 @@
                             
  
 
-                          
-              
-                           
-               
-                                                                            
-                                                                  
-                                                                            
-                                                                              
-                                                                             
-                                                       
-                           
-                         
-                  
-                                                                        
-                                                                          
-                                                                            
-                                                                         
-                                                                              
-                    
-                          
-                          
-                     
-                      
-                              
-                                  
-                    
-                                          
-                    
-                  
-                          
-                          
-                                                     
-                              
-                                                                                    
-                      
-                                                                        
-                                                                             
-                                                                            
-                                                                         
-                         
-                         
- 
+// `Finding` is a union discriminated on `type`, one member per emitted finding type: see
+// finding-types.ts for each detector's shape and which of its fields are public evidence.
+                                                                                                                  
 
                                                                                      
 

package/vendor-core/wall-clock.js

@@ -1,15 +1,4 @@
-export function mergeIntervals(intervals                    )                     {
-  if (intervals.length === 0) return [];
-  const sorted = [...intervals].sort((a, b) => a[0] - b[0]);
-  const out                     = [[sorted[0][0], sorted[0][1]]];
-  for (let i = 1; i < sorted.length; i++) {
-    const last = out[out.length - 1];
-    const cur = sorted[i];
-    if (cur[0] <= last[1]) last[1] = Math.max(last[1], cur[1]);
-    else out.push([cur[0], cur[1]]);
-  }
-  return out;
-}
+import { mergeIntervals } from './intervals.js';
 
 export function computeWallClock(app                                                             , stages                                                           )                                                                                     {
   const start = app?.startTime ?? 0;

package/vendor-core/wasted-core-hours.js

@@ -3,11 +3,11 @@
 // core-time that actually ran tasks. Feeds a report widget only, no detector.
 
 import { computeTotalCores } from './core-count.js';
+import { MS_PER_CORE_HOUR } from './format-utils.js';
 
-export const MS_PER_CORE_HOUR         = 3.6e6;
 const TOP_N = 5;
 
-                                 
+