tweak uloc

boyter · Apr 30, 2024 · 594e3a7 · 594e3a7
1 parent 4782495
commit 594e3a7
Show file tree

Hide file tree

Showing 3 changed files with 26 additions and 86 deletions.
diff --git a/processor/formatters.go b/processor/formatters.go
@@ -36,6 +36,8 @@ var tabularShortFormatBodyNoComplexity = "%-22s %11d %11d %10d %11d %9d\n"
 var tabularShortFormatFileNoComplexity = "%s %11d %10d %11d %9d\n"
 var longNameTruncate = 22
 
+var tabularUlocFormatBody = "Total Unique Source Lines of Code (ULOC) %38d\n"
+
 var tabularWideBreak = "─────────────────────────────────────────────────────────────────────────────────────────────────────────────\n"
 var tabularWideBreakCi = "-------------------------------------------------------------------------------------------------------------\n"
 var tabularWideFormatHead = "%-33s %9s %9s %8s %9s %8s %10s %16s\n"
@@ -673,18 +675,6 @@ create table t        (
 	return str.String()
 }
 
-// var tabularWideFormatBody = "%-33s %9d %9d %8d %9d %8d %10d %16.2f\n"
-var tabularUlocFormatBody = "Total Unique Source Lines of Code (ULOC) %38d\n"
-
-func ulocDisplay(input int) string {
-	var str strings.Builder
-	str.WriteString(getTabularShortBreak())
-	str.WriteString(fmt.Sprintf(tabularUlocFormatBody, input))
-	str.WriteString(getTabularShortBreak())
-
-	return str.String()
-}
-
 func fileSummarize(input chan *FileJob) string {
 	if FormatMulti != "" {
 		return fileSummarizeMulti(input)
@@ -1077,6 +1067,11 @@ func fileSummarizeShort(input chan *FileJob) string {
 	}
 	str.WriteString(getTabularShortBreak())
 
+	if UlocMode {
+		str.WriteString(fmt.Sprintf(tabularUlocFormatBody, len(ulocGlobalCount)))
+		str.WriteString(getTabularShortBreak())
+	}
+
 	if !Cocomo {
 		if SLOCCountFormat {
 			calculateCocomoSLOCCount(sumCode, &str)

diff --git a/processor/processor.go b/processor/processor.go
@@ -522,6 +522,11 @@ func printLanguages() {
 	}
 }
 
+// global variables to deal with ULOC calculations
+var ulocMutex = sync.Mutex{}
+var ulocGlobalCount = map[string]struct{}{}
+var ulocLanguageCount = map[string]map[string]struct{}{}
+
 // Process is the main entry point of the command line it sets everything up and starts running
 func Process() {
 	if Languages {

diff --git a/processor/workers.go b/processor/workers.go
@@ -645,8 +645,6 @@ func fileProcessorWorker(input chan *FileJob, output chan *FileJob) {
 	var fileCount int64
 	var gcEnabled int64
 	var wg sync.WaitGroup
-	var ulocMutex = sync.Mutex{}
-	uloc := map[string]struct{}{}
 
 	for i := 0; i < FileProcessJobWorkers; i++ {
 		wg.Add(1)
@@ -682,13 +680,6 @@ func fileProcessorWorker(input chan *FileJob, output chan *FileJob) {
 					if processFile(job) {
 						output <- job
 					}
-					if UlocMode {
-						ulocMutex.Lock()
-						for _, l := range strings.Split(string(content), "\n") {
-							uloc[l] = struct{}{}
-						}
-						ulocMutex.Unlock()
-					}
 				} else {
 					if Verbose {
 						printWarn(fmt.Sprintf("error reading: %s %s", job.Location, err))
@@ -703,78 +694,13 @@ func fileProcessorWorker(input chan *FileJob, output chan *FileJob) {
 	go func() {
 		wg.Wait()
 		close(output)
-		fmt.Println(ulocDisplay(len(uloc)))
 
 		if Debug {
 			printDebug(fmt.Sprintf("milliseconds reading files into memory: %d", makeTimestampMilli()-startTime))
 		}
 	}()
 }
 
-// Processes files using ULOC - Unique lines of code
-func fileProcessorWorkerUloc(input chan *FileJob, output chan *FileJob) {
-	var startTime int64
-	var fileCount int64
-	var gcEnabled int64
-	var wg sync.WaitGroup
-
-	var ulocMutex = sync.Mutex{}
-	uloc := map[string]struct{}{}
-
-	for i := 0; i < FileProcessJobWorkers; i++ {
-		wg.Add(1)
-		go func() {
-			reader := NewFileReader()
-
-			for job := range input {
-				atomic.CompareAndSwapInt64(&startTime, 0, makeTimestampMilli())
-
-				loc := job.Location
-				if job.Symlocation != "" {
-					loc = job.Symlocation
-				}
-
-				fileStartTime := makeTimestampNano()
-				content, err := reader.ReadFile(loc, int(job.Bytes))
-				atomic.AddInt64(&fileCount, 1)
-
-				if atomic.LoadInt64(&gcEnabled) == 0 && atomic.LoadInt64(&fileCount) >= int64(GcFileCount) {
-					debug.SetGCPercent(gcPercent)
-					atomic.AddInt64(&gcEnabled, 1)
-					if Verbose {
-						printWarn("read file limit exceeded GC re-enabled")
-					}
-				}
-
-				if Trace {
-					printTrace(fmt.Sprintf("nanoseconds read into memory: %s: %d", job.Location, makeTimestampNano()-fileStartTime))
-				}
-
-				if err == nil {
-					ulocMutex.Lock()
-					for _, l := range strings.Split(string(content), "\n") {
-						uloc[l] = struct{}{}
-					}
-					ulocMutex.Unlock()
-				} else {
-					if Verbose {
-						printWarn(fmt.Sprintf("error reading: %s %s", job.Location, err))
-					}
-				}
-			}
-
-			wg.Done()
-		}()
-	}
-
-	wg.Wait()
-	if Debug {
-		printDebug(fmt.Sprintf("milliseconds reading files into memory: %d", makeTimestampMilli()-startTime))
-	}
-
-	fmt.Println(ulocDisplay(len(uloc)))
-}
-
 // Process a single file
 // File must have been read to job.Content already
 func processFile(job *FileJob) bool {
@@ -823,6 +749,20 @@ func processFile(job *FileJob) bool {
 
 	CountStats(job)
 
+	if UlocMode {
+		ulocMutex.Lock()
+		for _, l := range strings.Split(string(job.Content), "\n") {
+			ulocGlobalCount[l] = struct{}{}
+
+			_, ok := ulocLanguageCount[job.Language]
+			if !ok {
+				ulocLanguageCount[job.Language] = map[string]struct{}{}
+			}
+			ulocLanguageCount[job.Language][l] = struct{}{}
+		}
+		ulocMutex.Unlock()
+	}
+
 	if Duplicates {
 		duplicates.mux.Lock()
 		jobHash := job.Hash.Sum(nil)