Add cross-compilation Makefile targets and tar-based releases.

Revamp the build system to be more inline with other Prometheus exporters.
Notably add Darwin and Windows build targets, and add support for releases
using tar files.

tools/vendor/github.com/mibk/dupl/main.go

@@ -12,7 +12,7 @@ import (
 	"strings"
 
 	"github.com/mibk/dupl/job"
-	"github.com/mibk/dupl/output"
+	"github.com/mibk/dupl/printer"
 	"github.com/mibk/dupl/syntax"
 )
 
@@ -20,18 +20,18 @@ const defaultThreshold = 15
 
 var (
 	paths     = []string{"."}
-	vendor    = flag.Bool("vendor", false, "check files in vendor directory")
-	verbose   = flag.Bool("verbose", false, "explain what is being done")
-	threshold = flag.Int("threshold", defaultThreshold, "minimum token sequence as a clone")
-	files     = flag.Bool("files", false, "files names from stdin")
+	vendor    = flag.Bool("vendor", false, "")
+	verbose   = flag.Bool("verbose", false, "")
+	threshold = flag.Int("threshold", defaultThreshold, "")
+	files     = flag.Bool("files", false, "")
 
-	html     = flag.Bool("html", false, "html output")
-	plumbing = flag.Bool("plumbing", false, "plumbing output for consumption by scripts or tools")
+	html     = flag.Bool("html", false, "")
+	plumbing = flag.Bool("plumbing", false, "")
 )
 
 const (
 	vendorDirPrefix = "vendor" + string(filepath.Separator)
-	vendorDirInPath = string(filepath.Separator) + "vendor" + string(filepath.Separator)
+	vendorDirInPath = string(filepath.Separator) + vendorDirPrefix
 )
 
 func init() {
@@ -39,43 +39,6 @@ func init() {
 	flag.IntVar(threshold, "t", defaultThreshold, "alias for -threshold")
 }
 
-func usage() {
-	fmt.Fprintln(os.Stderr, `Usage of dupl:
-  dupl [flags] [paths]
-
-Paths:
-  If the given path is a file, dupl will use it regardless of
-  the file extension. If it is a directory it will recursively
-  search for *.go files in that directory.
-
-  If no path is given dupl will recursively search for *.go
-  files in the current directory.
-
-Flags:
-  -files
-    	read file names from stdin one at each line
-  -html
-    	output the results as HTML, including duplicate code fragments
-  -plumbing
-    	plumbing (easy-to-parse) output for consumption by scripts or tools
-  -t, -threshold size
-    	minimum token sequence size as a clone (default 15)
-  -vendor
-    	check files in vendor directory
-  -v, -verbose
-    	explain what is being done
-
-Examples:
-  dupl -t 100
-    	Search clones in the current directory of size at least
-    	100 tokens.
-  dupl $(find app/ -name '*_test.go')
-    	Search for clones in tests in the app directory.
-  find app/ -name '*_test.go' |dupl -files
-    	The same as above.`)
-	os.Exit(2)
-}
-
 func main() {
 	flag.Usage = usage
 	flag.Parse()
@@ -110,7 +73,17 @@ func main() {
 		}
 		close(duplChan)
 	}()
-	printDupls(duplChan)
+
+	newPrinter := printer.NewText
+	if *html {
+		newPrinter = printer.NewHTML
+	} else if *plumbing {
+		newPrinter = printer.NewPlumbing
+	}
+	p := newPrinter(os.Stdout, ioutil.ReadFile)
+	if err := printDupls(p, duplChan); err != nil {
+		log.Fatal(err)
+	}
 }
 
 func filesFeed() chan string {
@@ -120,10 +93,7 @@ func filesFeed() chan string {
 			s := bufio.NewScanner(os.Stdin)
 			for s.Scan() {
 				f := s.Text()
-				if strings.HasPrefix(f, "./") {
-					f = f[2:]
-				}
-				fchan <- f
+				fchan <- strings.TrimPrefix(f, "./")
 			}
 			close(fchan)
 		}()
@@ -160,7 +130,7 @@ func crawlPaths(paths []string) chan string {
 	return fchan
 }
 
-func printDupls(duplChan <-chan syntax.Match) {
+func printDupls(p printer.Printer, duplChan <-chan syntax.Match) error {
 	groups := make(map[string][][]*syntax.Node)
 	for dupl := range duplChan {
 		groups[dupl.Hash] = append(groups[dupl.Hash], dupl.Frags...)
@@ -171,32 +141,18 @@ func printDupls(duplChan <-chan syntax.Match) {
 	}
 	sort.Strings(keys)
 
-	p := getPrinter()
+	if err := p.PrintHeader(); err != nil {
+		return err
+	}
 	for _, k := range keys {
 		uniq := unique(groups[k])
 		if len(uniq) > 1 {
-			if err := p.Print(uniq); err != nil {
-				log.Fatal(err)
+			if err := p.PrintClones(uniq); err != nil {
+				return err
 			}
 		}
 	}
-	p.Finish()
-}
-
-func getPrinter() output.Printer {
-	var fr fileReader
-	if *html {
-		return output.NewHTMLPrinter(os.Stdout, fr)
-	} else if *plumbing {
-		return output.NewPlumbingPrinter(os.Stdout, fr)
-	}
-	return output.NewTextPrinter(os.Stdout, fr)
-}
-
-type fileReader struct{}
-
-func (fileReader) ReadFile(filename string) ([]byte, error) {
-	return ioutil.ReadFile(filename)
+	return p.PrintFooter()
 }
 
 func unique(group [][]*syntax.Node) [][]*syntax.Node {
@@ -217,3 +173,39 @@ func unique(group [][]*syntax.Node) [][]*syntax.Node {
 	}
 	return newGroup
 }
+
+func usage() {
+	fmt.Fprintln(os.Stderr, `Usage: dupl [flags] [paths]
+
+Paths:
+  If the given path is a file, dupl will use it regardless of
+  the file extension. If it is a directory, it will recursively
+  search for *.go files in that directory.
+
+  If no path is given, dupl will recursively search for *.go
+  files in the current directory.
+
+Flags:
+  -files
+    	read file names from stdin one at each line
+  -html
+    	output the results as HTML, including duplicate code fragments
+  -plumbing
+    	plumbing (easy-to-parse) output for consumption by scripts or tools
+  -t, -threshold size
+    	minimum token sequence size as a clone (default 15)
+  -vendor
+    	check files in vendor directory
+  -v, -verbose
+    	explain what is being done
+
+Examples:
+  dupl -t 100
+    	Search clones in the current directory of size at least
+    	100 tokens.
+  dupl $(find app/ -name '*_test.go')
+    	Search for clones in tests in the app directory.
+  find app/ -name '*_test.go' |dupl -files
+    	The same as above.`)
+	os.Exit(2)
+}

tools/vendor/github.com/mibk/dupl/printer/html.go

@@ -0,0 +1,120 @@
+package printer
+
+import (
+	"bytes"
+	"fmt"
+	"io"
+	"regexp"
+	"sort"
+
+	"github.com/mibk/dupl/syntax"
+)
+
+type html struct {
+	iota int
+	w    io.Writer
+	ReadFile
+}
+
+func NewHTML(w io.Writer, fread ReadFile) Printer {
+	return &html{w: w, ReadFile: fread}
+}
+
+func (p *html) PrintHeader() error {
+	_, err := fmt.Fprint(p.w, `<!DOCTYPE html>
+<meta charset="utf-8"/>
+<title>Duplicates</title>
+<style>
+	pre {
+		background-color: #FFD;
+		border: 1px solid #E2E2E2;
+		padding: 1ex;
+	}
+</style>
+`)
+	return err
+}
+
+func (p *html) PrintClones(dups [][]*syntax.Node) error {
+	p.iota++
+	fmt.Fprintf(p.w, "<h1>#%d found %d clones</h1>\n", p.iota, len(dups))
+
+	clones := make([]clone, len(dups))
+	for i, dup := range dups {
+		cnt := len(dup)
+		if cnt == 0 {
+			panic("zero length dup")
+		}
+		nstart := dup[0]
+		nend := dup[cnt-1]
+
+		file, err := p.ReadFile(nstart.Filename)
+		if err != nil {
+			return err
+		}
+
+		lineStart, _ := blockLines(file, nstart.Pos, nend.End)
+		cl := clone{filename: nstart.Filename, lineStart: lineStart}
+		start := findLineBeg(file, nstart.Pos)
+		content := append(toWhitespace(file[start:nstart.Pos]), file[nstart.Pos:nend.End]...)
+		cl.fragment = deindent(content)
+		clones[i] = cl
+	}
+
+	sort.Sort(byNameAndLine(clones))
+	for _, cl := range clones {
+		fmt.Fprintf(p.w, "<h2>%s:%d</h2>\n<pre>%s</pre>\n", cl.filename, cl.lineStart, cl.fragment)
+	}
+	return nil
+}
+
+func (*html) PrintFooter() error { return nil }
+
+func findLineBeg(file []byte, index int) int {
+	for i := index; i >= 0; i-- {
+		if file[i] == '\n' {
+			return i + 1
+		}
+	}
+	return 0
+}
+
+func toWhitespace(str []byte) []byte {
+	var out []byte
+	for _, c := range bytes.Runes(str) {
+		if c == '\t' {
+			out = append(out, '\t')
+		} else {
+			out = append(out, ' ')
+		}
+	}
+	return out
+}
+
+func deindent(block []byte) []byte {
+	const maxVal = 99
+	min := maxVal
+	re := regexp.MustCompile(`(^|\n)(\t*)\S`)
+	for _, line := range re.FindAllSubmatch(block, -1) {
+		indent := line[2]
+		if len(indent) < min {
+			min = len(indent)
+		}
+	}
+	if min == 0 || min == maxVal {
+		return block
+	}
+	block = block[min:]
+Loop:
+	for i := 0; i < len(block); i++ {
+		if block[i] == '\n' && i != len(block)-1 {
+			for j := 0; j < min; j++ {
+				if block[i+j+1] != '\t' {
+					continue Loop
+				}
+			}
+			block = append(block[:i+1], block[i+1+min:]...)
+		}
+	}
+	return block
+}

tools/vendor/github.com/mibk/dupl/printer/plumbing.go

@@ -0,0 +1,36 @@
+package printer
+
+import (
+	"fmt"
+	"io"
+	"sort"
+
+	"github.com/mibk/dupl/syntax"
+)
+
+type plumbing struct {
+	w io.Writer
+	ReadFile
+}
+
+func NewPlumbing(w io.Writer, fread ReadFile) Printer {
+	return &plumbing{w, fread}
+}
+
+func (p *plumbing) PrintHeader() error { return nil }
+
+func (p *plumbing) PrintClones(dups [][]*syntax.Node) error {
+	clones, err := prepareClonesInfo(p.ReadFile, dups)
+	if err != nil {
+		return err
+	}
+	sort.Sort(byNameAndLine(clones))
+	for i, cl := range clones {
+		nextCl := clones[(i+1)%len(clones)]
+		fmt.Fprintf(p.w, "%s:%d-%d: duplicate of %s:%d-%d\n", cl.filename, cl.lineStart, cl.lineEnd,
+			nextCl.filename, nextCl.lineStart, nextCl.lineEnd)
+	}
+	return nil
+}
+
+func (p *plumbing) PrintFooter() error { return nil }

tools/vendor/github.com/mibk/dupl/printer/printer.go

@@ -0,0 +1,11 @@
+package printer
+
+import "github.com/mibk/dupl/syntax"
+
+type ReadFile func(filename string) ([]byte, error)
+
+type Printer interface {
+	PrintHeader() error
+	PrintClones(dups [][]*syntax.Node) error
+	PrintFooter() error
+}

tools/vendor/github.com/mibk/dupl/printer/text.go

@@ -0,0 +1,100 @@
+package printer
+
+import (
+	"fmt"
+	"io"
+	"sort"
+
+	"github.com/mibk/dupl/syntax"
+)
+
+type text struct {
+	cnt int
+	w   io.Writer
+	ReadFile
+}
+
+func NewText(w io.Writer, fread ReadFile) Printer {
+	return &text{w: w, ReadFile: fread}
+}
+
+func (p *text) PrintHeader() error { return nil }
+
+func (p *text) PrintClones(dups [][]*syntax.Node) error {
+	p.cnt++
+	fmt.Fprintf(p.w, "found %d clones:\n", len(dups))
+	clones, err := prepareClonesInfo(p.ReadFile, dups)
+	if err != nil {
+		return err
+	}
+	sort.Sort(byNameAndLine(clones))
+	for _, cl := range clones {
+		fmt.Fprintf(p.w, "  %s:%d,%d\n", cl.filename, cl.lineStart, cl.lineEnd)
+	}
+	return nil
+}
+
+func (p *text) PrintFooter() error {
+	_, err := fmt.Fprintf(p.w, "\nFound total %d clone groups.\n", p.cnt)
+	return err
+}
+
+func prepareClonesInfo(fread ReadFile, dups [][]*syntax.Node) ([]clone, error) {
+	clones := make([]clone, len(dups))
+	for i, dup := range dups {
+		cnt := len(dup)
+		if cnt == 0 {
+			panic("zero length dup")
+		}
+		nstart := dup[0]
+		nend := dup[cnt-1]
+
+		file, err := fread(nstart.Filename)
+		if err != nil {
+			return nil, err
+		}
+
+		cl := clone{filename: nstart.Filename}
+		cl.lineStart, cl.lineEnd = blockLines(file, nstart.Pos, nend.End)
+		clones[i] = cl
+	}
+	return clones, nil
+}
+
+func blockLines(file []byte, from, to int) (int, int) {
+	line := 1
+	lineStart, lineEnd := 0, 0
+	for offset, b := range file {
+		if b == '\n' {
+			line++
+		}
+		if offset == from {
+			lineStart = line
+		}
+		if offset == to-1 {
+			lineEnd = line
+			break
+		}
+	}
+	return lineStart, lineEnd
+}
+
+type clone struct {
+	filename  string
+	lineStart int
+	lineEnd   int
+	fragment  []byte
+}
+
+type byNameAndLine []clone
+
+func (c byNameAndLine) Len() int { return len(c) }
+
+func (c byNameAndLine) Swap(i, j int) { c[i], c[j] = c[j], c[i] }
+
+func (c byNameAndLine) Less(i, j int) bool {
+	if c[i].filename == c[j].filename {
+		return c[i].lineStart < c[j].lineStart
+	}
+	return c[i].filename < c[j].filename
+}