code grouming, migration to go1.21 (#75)

.github/workflows/ci.yml

@@ -20,7 +20,7 @@ jobs:
       - name: setup golang
         uses: actions/setup-go@v4
         with:
-          go-version: ^1.20
+          go-version: ^1.21
           check-latest: true
           cache: true
       - name: golangci-lint
@@ -40,7 +40,7 @@ jobs:
       - name: setup golang
         uses: actions/setup-go@v4
         with:
-          go-version: ^1.20
+          go-version: ^1.21
           check-latest: true
           cache: true
       - name: test-coverage
@@ -62,7 +62,7 @@ jobs:
       - name: setup golang
         uses: actions/setup-go@v4
         with:
-          go-version: ^1.20
+          go-version: ^1.21
           check-latest: true
           cache: true
       - name: init codeql

.github/workflows/release.yml

@@ -15,7 +15,7 @@ jobs:
       - name: set up golang
         uses: actions/setup-go@v4
         with:
-          go-version: ^1.20
+          go-version: ^1.21
           check-latest: true
           cache: true
       - name: build

.golangci.yml

@@ -1,6 +1,6 @@
 run:
   allow-parallel-runners: true
-  go: '1.20'
+  go: '1.21'
 
 output:
   format: 'colored-line-number'

.goreleaser.yml

@@ -51,8 +51,8 @@ nfpms:
     license: MIT
     vendor: Crawley
     formats:
-    - deb
-    - rpm
+      - deb
+      - rpm
 
 changelog:
   filters:

cmd/crawley/main.go

@@ -21,7 +21,6 @@ import (
 
 const (
 	appName        = "Crawley"
-	appHelp        = "the unix-way web crawler"
 	appSite        = "https://github.com/s0rg/crawley"
 	defaultDelay   = 150 * time.Millisecond
 	defaultTimeout = 5 * time.Second
@@ -63,20 +62,9 @@ func version() string {
 func usage() {
 	var sb strings.Builder
 
-	const twoCR = "\n\n"
-
-	sb.WriteString(appName)
-	sb.WriteString(" - ")
-	sb.WriteString(appHelp)
-	sb.WriteString(", usage:")
-	sb.WriteString(twoCR)
-
-	sb.WriteString(filepath.Base(os.Args[0]))
-	sb.WriteString(" [flags] url")
-	sb.WriteString(twoCR)
-
-	sb.WriteString("possible flags with default values:")
-	sb.WriteString(twoCR)
+	fmt.Fprintf(&sb, "%s - the unix-way web crawler, usage:\n\n", appName)
+	fmt.Fprintf(&sb, "%s [flags] url\n\n", filepath.Base(os.Args[0]))
+	fmt.Fprint(&sb, "possible flags with default values:\n\n")
 
 	_, _ = os.Stderr.WriteString(sb.String())
 

go.mod

@@ -1,6 +1,6 @@
 module github.com/s0rg/crawley
 
-go 1.20
+go 1.21
 
 require (
 	github.com/s0rg/compflag v1.1.0

pkg/client/http.go

@@ -6,11 +6,8 @@ import (
 	"io"
 	"net"
 	"net/http"
-	"time"
 )
 
-const transportTimeout = 10 * time.Second
-
 // HTTP holds pre-configured http.Client.
 type HTTP struct {
 	c       *http.Client
@@ -24,13 +21,13 @@ func New(cfg *Config) (h *HTTP) {
 	transport := &http.Transport{
 		Proxy: http.ProxyFromEnvironment,
 		Dial: (&net.Dialer{
-			Timeout: transportTimeout,
+			Timeout: cfg.Timeout,
 		}).Dial,
 		TLSClientConfig: &tls.Config{
 			InsecureSkipVerify: cfg.SkipSSL,
 		},
-		IdleConnTimeout:     transportTimeout,
-		TLSHandshakeTimeout: transportTimeout,
+		IdleConnTimeout:     cfg.Timeout,
+		TLSHandshakeTimeout: cfg.Timeout,
 		MaxConnsPerHost:     cfg.Workers,
 		MaxIdleConns:        cfg.Workers,
 		MaxIdleConnsPerHost: cfg.Workers,

pkg/crawler/config.go

@@ -9,6 +9,7 @@ import (
 )
 
 const (
+	minDepth   = -1
 	minWorkers = 1
 	maxWorkers = 64
 	minDelay   = time.Duration(0)
@@ -30,44 +31,27 @@ type config struct {
 }
 
 func (c *config) validate() {
-	switch {
-	case c.Client.Workers < minWorkers:
-		c.Client.Workers = minWorkers
-	case c.Client.Workers > maxWorkers:
-		c.Client.Workers = maxWorkers
-	}
-
-	switch {
-	case c.Client.Timeout < minTimeout:
-		c.Client.Timeout = minTimeout
-	case c.Client.Timeout > maxTimeout:
-		c.Client.Timeout = maxTimeout
-	}
-
-	if c.Delay < minDelay {
-		c.Delay = minDelay
-	}
-
-	if c.Depth < 0 {
-		c.Depth = -1
-	}
+	c.Client.Workers = min(maxWorkers, max(minWorkers, c.Client.Workers))
+	c.Client.Timeout = min(maxTimeout, max(minTimeout, c.Client.Timeout))
+	c.Delay = max(minDelay, c.Delay)
+	c.Depth = max(minDepth, c.Depth)
 }
 
 func (c *config) String() (rv string) {
 	var sb strings.Builder
 
-	_, _ = sb.WriteString(fmt.Sprintf("workers: %d depth: %d timeout: %s", c.Client.Workers, c.Depth, c.Client.Timeout))
+	fmt.Fprintf(&sb, "workers: %d depth: %d timeout: %s", c.Client.Workers, c.Depth, c.Client.Timeout)
 
 	if c.Brute {
-		_, _ = sb.WriteString(" brute: on")
+		sb.WriteString(" brute: on")
 	}
 
 	if c.ScanJS {
-		_, _ = sb.WriteString(" js: on")
+		sb.WriteString(" js: on")
 	}
 
 	if c.Delay > 0 {
-		_, _ = sb.WriteString(fmt.Sprintf(" delay: %s", c.Delay))
+		fmt.Fprintf(&sb, " delay: %s", c.Delay)
 	}
 
 	return sb.String()

pkg/crawler/crawler.go

@@ -8,6 +8,7 @@ import (
 	"log"
 	"net/http"
 	"net/url"
+	"slices"
 	"strings"
 	"sync"
 	"time"
@@ -76,7 +77,7 @@ func New(opts ...Option) (c *Crawler) {
 }
 
 // Run starts crawling process for given base uri.
-func (c *Crawler) Run(uri string, fn func(string)) (err error) {
+func (c *Crawler) Run(uri string, urlcb func(string)) (err error) {
 	var base *url.URL
 
 	if base, err = url.Parse(uri); err != nil {
@@ -106,7 +107,7 @@ func (c *Crawler) Run(uri string, fn func(string)) (err error) {
 
 	go func() {
 		for s := range c.handleCh {
-			fn(s)
+			urlcb(s)
 		}
 
 		c.wg.Done()
@@ -195,7 +196,7 @@ func (c *Crawler) close() {
 	close(c.crawlCh)
 	c.wg.Wait() // wait for crawlers
 
-	c.wg.Add(1)
+	c.wg.Add(1) // for handler`s Done()
 	close(c.handleCh)
 	c.wg.Wait() // wait for handler
 
@@ -265,13 +266,9 @@ func (c *Crawler) isIgnored(v string) (yes bool) {
 		return
 	}
 
-	for _, s := range c.cfg.Ignored {
-		if strings.Contains(v, s) {
-			return true
-		}
-	}
-
-	return false
+	return slices.ContainsFunc(c.cfg.Ignored, func(s string) bool {
+		return strings.Contains(v, s)
+	})
 }
 
 func (c *Crawler) linkHandler(a atom.Atom, s string) {

pkg/crawler/crawler_test.go

@@ -105,7 +105,6 @@ func TestCrawlerOK(t *testing.T) {
 	}
 
 	if len(results) != 3 {
-		t.Log(results)
 		t.Error("results: less than expected")
 	}
 
@@ -250,7 +249,6 @@ sitemap: http://other.host/sitemap.xml`
 	}
 
 	if len(resA) != 5 {
-		t.Log(resA)
 		t.Fatal("unexpected len for A")
 	}
 

pkg/links/clean_test.go

@@ -2,7 +2,6 @@ package links
 
 import (
 	"net/url"
-	"reflect"
 	"testing"
 )
 
@@ -47,7 +46,7 @@ func TestClean(t *testing.T) {
 			}
 
 			if gotOk {
-				if !reflect.DeepEqual(gotU, tc.wantU) {
+				if gotU != tc.wantU {
 					t.Errorf("clean() gotU = %v, want %v", gotU, tc.wantU)
 				}
 			}

pkg/links/html_test.go

@@ -3,7 +3,7 @@ package links
 import (
 	"bytes"
 	"net/url"
-	"reflect"
+	"slices"
 	"strings"
 	"testing"
 
@@ -49,7 +49,7 @@ func TestExtractTag(t *testing.T) {
 
 			gotU := extractTag(tc.args.b, tc.args.t, tc.args.k)
 
-			if !reflect.DeepEqual(gotU, tc.wantU) {
+			if gotU != tc.wantU {
 				t.Errorf("extractTag() gotU = %v, want %v", gotU, tc.wantU)
 			}
 		})
@@ -205,7 +205,7 @@ func TestExtractToken(t *testing.T) {
 				t.Errorf("extractToken() key gotU = %v, want %v", key, tc.keyWant)
 			}
 
-			if !reflect.DeepEqual(res, tc.wantURL) {
+			if res != tc.wantURL {
 				t.Errorf("extractToken() link gotU = %v, want %v", res, tc.wantURL)
 			}
 		})
@@ -297,7 +297,7 @@ func TestExtractURLS(t *testing.T) {
 			})
 
 			if tc.hasLink {
-				if !reflect.DeepEqual(res, tc.lnk) {
+				if res != tc.lnk {
 					t.Errorf("extractToken() link gotU = %v, want %v", res, tc.lnk)
 				}
 			}
@@ -326,7 +326,7 @@ loremipsumhTTp://foo fdfdfs HttPs://bar
 		t.Error("unexpected len")
 	}
 
-	if !reflect.DeepEqual(res, want) {
+	if slices.Compare(res, want) != 0 {
 		t.Error("unexpected result")
 	}
 }

pkg/links/js.go

@@ -40,7 +40,7 @@ func cleanResult(s []byte) (rv string, ok bool) {
 	rv = string(bytes.Trim(s, codeCleanChars))
 
 	if strings.HasPrefix(rv, mimeAppPrefix) || strings.HasPrefix(rv, mimeTxtPrefix) {
-		return "", false
+		return
 	}
 
 	return rv, true

pkg/links/sitemap_test.go

@@ -117,7 +117,6 @@ func TestExtractSitemapURLError(t *testing.T) {
 	})
 
 	if len(l) != 0 {
-		t.Log(l)
 		t.Error("unexpected results count")
 	}
 }

pkg/robots/parser.go

@@ -64,12 +64,11 @@ func extractToken(b []byte) (k tokenKind, v string) {
 		return
 	}
 
-	var val []byte
-	if val = bytes.TrimSpace(b[pos+1:]); len(val) == 0 {
-		return
+	if val := bytes.TrimSpace(b[pos+1:]); len(val) > 0 {
+		return kind, string(val)
 	}
 
-	return kind, string(val)
+	return
 }
 
 func parseRobots(r io.Reader, ua string, t *TXT) (err error) {