Release/v1.5.3 (#40)

* code grouming, better usage info

README.md

@@ -53,15 +53,15 @@ crawley -headless -delay 0 -depth -1 -dirs only http://some-test.site
 
 # installation
 
-- [binaries](https://github.com/s0rg/crawley/releases) for Linux, FreeBSD, macOS and Windows, just download and run.
+- [binaries / deb / rpm](https://github.com/s0rg/crawley/releases) for Linux, FreeBSD, macOS and Windows.
 - [archlinux](https://aur.archlinux.org/packages/crawley-bin/) you can use your favourite AUR helper to install it, e. g. `paru -S crawley-bin`.
 
 # usage
 
 ```
 crawley [flags] url
 
-possible flags:
+possible flags with default values:
 
 -brute
     scan html comments

cmd/crawley/main.go

@@ -8,7 +8,9 @@ import (
 	"io"
 	"log"
 	"os"
+	"path/filepath"
 	"runtime"
+	"strings"
 	"time"
 
 	"github.com/s0rg/crawley/pkg/crawler"
@@ -17,32 +19,30 @@ import (
 
 const (
 	appName      = "Crawley"
+	appHelp      = "the unix-way web crawler"
 	appSite      = "https://github.com/s0rg/crawley"
 	defaultDelay = 150 * time.Millisecond
 )
 
+// build-time values.
 var (
-	GitHash   string
 	GitTag    string
+	GitHash   string
 	BuildDate string
 	defaultUA = "Mozilla/5.0 (compatible; Win64; x64) Mr." + appName + "/" + GitTag + "-" + GitHash
+)
 
-	cookies, headers values.Smart
-	tags, ignored    values.Simple
-
-	fDepth        = flag.Int("depth", 0, "scan depth (set -1 for unlimited)")
-	fWorkers      = flag.Int("workers", runtime.NumCPU(), "number of workers")
-	fBrute        = flag.Bool("brute", false, "scan html comments")
-	fNoHeads      = flag.Bool("headless", false, "disable pre-flight HEAD requests")
-	fScanJS       = flag.Bool("js", false, "scan js files for endpoints")
-	fSkipSSL      = flag.Bool("skip-ssl", false, "skip ssl verification")
-	fSilent       = flag.Bool("silent", false, "suppress info and error messages in stderr")
-	fVersion      = flag.Bool("version", false, "show version")
-	fDirsPolicy   = flag.String("dirs", "show", "policy for non-resource urls: show / hide / only")
-	fProxyAuth    = flag.String("proxy-auth", "", "credentials for proxy: user:password")
-	fRobotsPolicy = flag.String("robots", "ignore", "policy for robots.txt: ignore / crawl / respect")
-	fUA           = flag.String("user-agent", defaultUA, "user-agent string")
-	fDelay        = flag.Duration("delay", defaultDelay, "per-request delay (0 - disable)")
+// command-line flags.
+var (
+	fDepth, fWorkers        int
+	fSilent, fVersion       bool
+	fBrute, fNoHeads        bool
+	fSkipSSL, fScanJS       bool
+	fDirsPolicy, fProxyAuth string
+	fRobotsPolicy, fUA      string
+	fDelay                  time.Duration
+	cookies, headers        values.Smart
+	tags, ignored           values.List
 )
 
 func version() string {
@@ -56,6 +56,29 @@ 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)
+
+	_, _ = os.Stderr.WriteString(sb.String())
+
+	flag.PrintDefaults()
+}
+
 func puts(s string) {
 	_, _ = os.Stdout.WriteString(s + "\n")
 }
@@ -102,14 +125,14 @@ func loadSmart() (h, c []string, err error) {
 }
 
 func initOptions() (rv []crawler.Option, err error) {
-	robots, err := crawler.ParseRobotsPolicy(*fRobotsPolicy)
+	robots, err := crawler.ParseRobotsPolicy(fRobotsPolicy)
 	if err != nil {
 		err = fmt.Errorf("robots policy: %w", err)
 
 		return
 	}
 
-	dirs, err := crawler.ParseDirsPolicy(*fDirsPolicy)
+	dirs, err := crawler.ParseDirsPolicy(fDirsPolicy)
 	if err != nil {
 		err = fmt.Errorf("dirs policy: %w", err)
 
@@ -124,58 +147,72 @@ func initOptions() (rv []crawler.Option, err error) {
 	}
 
 	rv = []crawler.Option{
-		crawler.WithUserAgent(*fUA),
-		crawler.WithDelay(*fDelay),
-		crawler.WithMaxCrawlDepth(*fDepth),
-		crawler.WithWorkersCount(*fWorkers),
-		crawler.WithSkipSSL(*fSkipSSL),
-		crawler.WithBruteMode(*fBrute),
+		crawler.WithUserAgent(fUA),
+		crawler.WithDelay(fDelay),
+		crawler.WithMaxCrawlDepth(fDepth),
+		crawler.WithWorkersCount(fWorkers),
+		crawler.WithSkipSSL(fSkipSSL),
+		crawler.WithBruteMode(fBrute),
 		crawler.WithDirsPolicy(dirs),
 		crawler.WithRobotsPolicy(robots),
-		crawler.WithoutHeads(*fNoHeads),
-		crawler.WithScanJS(*fScanJS),
+		crawler.WithoutHeads(fNoHeads),
+		crawler.WithScanJS(fScanJS),
 		crawler.WithExtraHeaders(h),
 		crawler.WithExtraCookies(c),
 		crawler.WithTagsFilter(tags.Values),
 		crawler.WithIgnored(ignored.Values),
-		crawler.WithProxyAuth(*fProxyAuth),
+		crawler.WithProxyAuth(fProxyAuth),
 	}
 
 	return rv, nil
 }
 
-func main() {
-	flag.Var(
-		&headers,
-		"header",
+func setupFlags() {
+	flag.Var(&headers, "header",
 		"extra headers for request, can be used multiple times, accept files with '@'-prefix",
 	)
-	flag.Var(
-		&cookies,
-		"cookie",
+	flag.Var(&cookies, "cookie",
 		"extra cookies for request, can be used multiple times, accept files with '@'-prefix",
 	)
-	flag.Var(
-		&tags,
-		"tag",
-		"tags filter, single or comma-separated tag names",
-	)
-	flag.Var(
-		&ignored,
-		"ignore",
-		"patterns (in urls) to be ignored in crawl process",
-	)
+
+	flag.Var(&tags, "tag", "tags filter, single or comma-separated tag names")
+	flag.Var(&ignored, "ignore", "patterns (in urls) to be ignored in crawl process")
+
+	flag.IntVar(&fDepth, "depth", 0, "scan depth (set -1 for unlimited)")
+	flag.IntVar(&fWorkers, "workers", runtime.NumCPU(), "number of workers")
+
+	flag.BoolVar(&fBrute, "brute", false, "scan html comments")
+	flag.BoolVar(&fNoHeads, "headless", false, "disable pre-flight HEAD requests")
+	flag.BoolVar(&fScanJS, "js", false, "scan js files for endpoints")
+	flag.BoolVar(&fSkipSSL, "skip-ssl", false, "skip ssl verification")
+	flag.BoolVar(&fSilent, "silent", false, "suppress info and error messages in stderr")
+	flag.BoolVar(&fVersion, "version", false, "show version")
+
+	flag.StringVar(&fDirsPolicy, "dirs", crawler.DefaultDirsPolicy,
+		"policy for non-resource urls: show / hide / only")
+	flag.StringVar(&fRobotsPolicy, "robots", crawler.DefaultRobotsPolicy,
+		"policy for robots.txt: ignore / crawl / respect")
+	flag.StringVar(&fUA, "user-agent", defaultUA, "user-agent string")
+	flag.StringVar(&fProxyAuth, "proxy-auth", "", "credentials for proxy: user:password")
+
+	flag.DurationVar(&fDelay, "delay", defaultDelay, "per-request delay (0 - disable)")
+
+	flag.Usage = usage
+}
+
+func main() {
+	setupFlags()
 
 	flag.Parse()
 
-	if *fVersion {
+	if fVersion {
 		puts(version())
 
 		return
 	}
 
 	if flag.NArg() != 1 {
-		flag.Usage()
+		usage()
 
 		return
 	}
@@ -185,7 +222,7 @@ func main() {
 		log.Fatal("[-] options:", err)
 	}
 
-	if *fSilent {
+	if fSilent {
 		log.SetOutput(io.Discard)
 	}
 

pkg/crawler/config_test.go

@@ -190,10 +190,12 @@ func TestString(t *testing.T) {
 func TestProxyAuth(t *testing.T) {
 	t.Parallel()
 
+	const creds = "user:pass"
+
 	var (
 		c       = &config{}
-		opts    = []Option{WithProxyAuth("user:pass")}
-		headers = []string{proxyAuthHdr + ": " + proxyAuthTyp + " dXNlcjpwYXNz"}
+		opts    = []Option{WithProxyAuth(creds)}
+		headers = []string{proxyAuthHeader(creds)}
 	)
 
 	for _, o := range opts {

pkg/crawler/crawler.go

@@ -46,6 +46,7 @@ const (
 
 type crawlResult struct {
 	URI  string
+	Hash uint64
 	Flag taskFlag
 }
 
@@ -92,8 +93,8 @@ func (c *Crawler) Run(uri string, fn func(string)) (err error) {
 
 	defer c.close()
 
-	seen := make(set.URI)
-	seen.Add(uri)
+	seen := make(set.Set[uint64])
+	seen.Add(urlhash(uri))
 
 	web := client.New(
 		c.cfg.UserAgent,
@@ -122,7 +123,7 @@ func (c *Crawler) Run(uri string, fn func(string)) (err error) {
 		switch {
 		case t.Flag == TaskDone:
 			w--
-		case seen.Add(t.URI):
+		case seen.TryAdd(t.Hash):
 			if t.Flag == TaskCrawl && c.crawl(base, &t) {
 				w++
 			}
@@ -260,14 +261,6 @@ func (c *Crawler) crawlRobots(host *url.URL) {
 	}
 }
 
-func (c *Crawler) sitemapHandler(s string) {
-	c.linkHandler(atom.A, s)
-}
-
-func (c *Crawler) jsHandler(s string) {
-	c.linkHandler(atom.Link, s)
-}
-
 func (c *Crawler) isIgnored(v string) (yes bool) {
 	if len(c.cfg.Ignored) == 0 {
 		return
@@ -283,7 +276,10 @@ func (c *Crawler) isIgnored(v string) (yes bool) {
 }
 
 func (c *Crawler) linkHandler(a atom.Atom, s string) {
-	r := crawlResult{URI: s}
+	r := crawlResult{
+		URI:  s,
+		Hash: urlhash(s),
+	}
 
 	fetch := (a == atom.A || a == atom.Iframe) ||
 		(c.cfg.ScanJS && a == atom.Script)
@@ -328,9 +324,13 @@ func (c *Crawler) fetch(
 			Handler: c.linkHandler,
 		})
 	case isSitemap(uri):
-		links.ExtractSitemap(body, base, c.sitemapHandler)
+		links.ExtractSitemap(body, base, func(s string) {
+			c.linkHandler(atom.A, s)
+		})
 	case c.cfg.ScanJS && isJS(content, uri):
-		links.ExtractJS(body, c.jsHandler)
+		links.ExtractJS(body, func(s string) {
+			c.linkHandler(atom.Link, s)
+		})
 	}
 
 	client.Discard(body)

pkg/crawler/crawler_test.go

@@ -759,7 +759,7 @@ func TestCrawlerProxyAuth(t *testing.T) {
 	)
 
 	ts := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
-		creds := r.Header.Get(proxyAuthHdr)
+		creds := r.Header.Get(proxyAuthKey)
 		if creds == "" {
 			t.Fatal("auth header empty")
 		}
@@ -769,7 +769,7 @@ func TestCrawlerProxyAuth(t *testing.T) {
 			t.Fatalf("invalid fields count: %d", len(parts))
 		}
 
-		if !strings.EqualFold(parts[0], proxyAuthTyp) {
+		if !strings.EqualFold(parts[0], proxyAuthBasic) {
 			t.Fatalf("invalid auth type: %s", parts[0])
 		}
 

pkg/crawler/options.go

@@ -1,7 +1,6 @@
 package crawler
 
 import (
-	"encoding/base64"
 	"time"
 )
 
@@ -109,9 +108,6 @@ func WithScanJS(v bool) Option {
 // WithProxyAuth enables proxy credentials.
 func WithProxyAuth(v string) Option {
 	return func(c *config) {
-		c.Headers = append(
-			c.Headers,
-			proxyAuthHdr+": "+proxyAuthTyp+" "+base64.StdEncoding.EncodeToString([]byte(v)),
-		)
+		c.Headers = append(c.Headers, proxyAuthHeader(v))
 	}
 }

pkg/crawler/policies.go

@@ -5,6 +5,13 @@ import (
 	"strings"
 )
 
+const (
+	// DefaultRobotsPolicy is a default policy name for robots handling.
+	DefaultRobotsPolicy = "ignore"
+	// DefaultDirsPolicy is a default policy name for non-resource URLs.
+	DefaultDirsPolicy = "show"
+)
+
 // ErrUnknownPolicy is returned when requested policy unknown.
 var ErrUnknownPolicy = errors.New("unknown policy")
 
@@ -33,21 +40,21 @@ const (
 )
 
 // ParseRobotsPolicy parses robots policy from string.
-func ParseRobotsPolicy(s string) (a RobotsPolicy, err error) {
+func ParseRobotsPolicy(s string) (p RobotsPolicy, err error) {
 	switch strings.ToLower(s) {
 	case "ignore":
-		a = RobotsIgnore
+		p = RobotsIgnore
 	case "crawl":
-		a = RobotsCrawl
+		p = RobotsCrawl
 	case "respect":
-		a = RobotsRespect
+		p = RobotsRespect
 	default:
 		err = ErrUnknownPolicy
 
 		return
 	}
 
-	return a, nil
+	return p, nil
 }
 
 // ParseDirsPolicy parses dirs policy from string.