Add document root option

arguments.go

@@ -23,20 +23,22 @@ var defaultConcurrency = func() int {
 const usage = `Link checker for Markdown and HTML
 
 Usage:
-	liche [-c <num-requests>] [-r] [-t <timeout>] [-v] <filenames>...
+	liche [-c <num-requests>] [-d <directory>] [-r] [-t <timeout>] [-v] <filenames>...
 
 Options:
 	-c, --concurrency <num-requests>  Set max number of concurrent HTTP requests. [default: %v]
+	-d, --document-root <directory>  Set document root directory for absolute paths.
 	-r, --recursive  Search Markdown and HTML files recursively
 	-t, --timeout <timeout>  Set timeout for HTTP requests in seconds. Disabled by default.
 	-v, --verbose  Be verbose.`
 
 type arguments struct {
-	filenames   []string
-	concurrency int
-	timeout     time.Duration
-	recursive   bool
-	verbose     bool
+	filenames    []string
+	documentRoot string
+	concurrency  int
+	timeout      time.Duration
+	recursive    bool
+	verbose      bool
 }
 
 func getArgs() (arguments, error) {
@@ -54,6 +56,10 @@ func getArgs() (arguments, error) {
 
 	t := 0.0
 
+	if args["--document-root"] == nil {
+		args["--document-root"] = ""
+	}
+
 	if args["--timeout"] != nil {
 		t, err = strconv.ParseFloat(args["--timeout"].(string), 64)
 
@@ -64,6 +70,7 @@ func getArgs() (arguments, error) {
 
 	return arguments{
 		args["<filenames>"].([]string),
+		args["--document-root"].(string),
 		int(c),
 		time.Duration(t) * time.Second,
 		args["--recursive"].(bool),

file_checker.go

@@ -2,8 +2,10 @@ package main
 
 import (
 	"bytes"
+	"errors"
 	"io/ioutil"
 	"net/url"
+	"path"
 	"strings"
 	"sync"
 	"time"
@@ -13,11 +15,12 @@ import (
 )
 
 type fileChecker struct {
-	urlChecker urlChecker
+	urlChecker   urlChecker
+	documentRoot string
 }
 
-func newFileChecker(timeout time.Duration, s semaphore) fileChecker {
-	return fileChecker{newURLChecker(timeout, s)}
+func newFileChecker(timeout time.Duration, r string, s semaphore) fileChecker {
+	return fileChecker{newURLChecker(timeout, s), r}
 }
 
 func (c fileChecker) Check(f string) ([]urlResult, error) {
@@ -27,7 +30,12 @@ func (c fileChecker) Check(f string) ([]urlResult, error) {
 		return nil, err
 	}
 
-	us := extractURLs(n)
+	us, err := c.extractURLs(n)
+
+	if err != nil {
+		return nil, err
+	}
+
 	rc := make(chan urlResult, len(us))
 	rs := make([]urlResult, 0, len(us))
 
@@ -81,11 +89,23 @@ func parseFile(f string) (*html.Node, error) {
 	return n, nil
 }
 
-func extractURLs(n *html.Node) []string {
+func (c fileChecker) extractURLs(n *html.Node) ([]string, error) {
 	us := make(map[string]bool)
 	ns := make([]*html.Node, 0, 1024)
 	ns = append(ns, n)
 
+	addURL := func(u string) error {
+		u, err := c.resolveURL(u)
+
+		if err != nil {
+			return err
+		}
+
+		us[u] = true
+
+		return nil
+	}
+
 	for len(ns) > 0 {
 		i := len(ns) - 1
 		n := ns[i]
@@ -96,14 +116,20 @@ func extractURLs(n *html.Node) []string {
 			case "a":
 				for _, a := range n.Attr {
 					if a.Key == "href" && isURL(a.Val) {
-						us[a.Val] = true
+						if err := addURL(a.Val); err != nil {
+							return nil, err
+						}
+
 						break
 					}
 				}
 			case "img":
 				for _, a := range n.Attr {
 					if a.Key == "src" && isURL(a.Val) {
-						us[a.Val] = true
+						if err := addURL(a.Val); err != nil {
+							return nil, err
+						}
+
 						break
 					}
 				}
@@ -115,7 +141,19 @@ func extractURLs(n *html.Node) []string {
 		}
 	}
 
-	return stringSetToSlice(us)
+	return stringSetToSlice(us), nil
+}
+
+func (c fileChecker) resolveURL(u string) (string, error) {
+	abs := strings.HasPrefix(u, "/")
+
+	if abs && c.documentRoot != "" {
+		return path.Join(c.documentRoot, u), nil
+	} else if abs {
+		return "", errors.New("document root directory is not specified")
+	}
+
+	return u, nil
 }
 
 func isURL(s string) bool {

file_checker_test.go

@@ -9,35 +9,41 @@ import (
 	"golang.org/x/net/html"
 )
 
-func TestExtractURLs(t *testing.T) {
-	for _, c := range []struct {
+func TestFileCheckerExtractURLs(t *testing.T) {
+	c := newFileChecker(0, "", newSemaphore(42))
+
+	for _, x := range []struct {
 		html    string
 		numURLs int
 	}{
 		{`<a href="https://google.com">Google</a>`, 1},
 		{
 			`
-				<div>
-					<a href="https://google.com">Google</a>
-					<a href="https://google.com">Google</a>
-				</div>
+			<div>
+			<a href="https://google.com">Google</a>
+			<a href="https://google.com">Google</a>
+			</div>
 			`,
 			1,
 		},
 		{
 			`
-				<div>
-					<a href="https://google.com">Google</a>
-					<a href="https://yahoo.com">Yahoo!</a>
-				</div>
+			<div>
+			<a href="https://google.com">Google</a>
+			<a href="https://yahoo.com">Yahoo!</a>
+			</div>
 			`,
 			2,
 		},
 	} {
-		n, err := html.Parse(strings.NewReader(c.html))
+		n, err := html.Parse(strings.NewReader(x.html))
+
+		assert.Equal(t, nil, err)
+
+		us, err := c.extractURLs(n)
 
 		assert.Equal(t, nil, err)
-		assert.Equal(t, c.numURLs, len(extractURLs(n)))
+		assert.Equal(t, x.numURLs, len(us))
 	}
 }
 

main.go

@@ -16,7 +16,7 @@ func main() {
 
 	rc := make(chan fileResult, 1024)
 	s := newSemaphore(args.concurrency)
-	c := newFileChecker(args.timeout, s)
+	c := newFileChecker(args.timeout, args.documentRoot, s)
 
 	go c.CheckMany(fc, rc)