feat(crawler): fixing crawl functionality (#5)

## Motive

this MR to fix issue with the crawling functionallity of the bot. the issue was with default fetcher that was using the `http.Client` that and it will block if the request takes too long to respond. the fix was to use `http.Client` with a timeout and a custom transport that will limit the number of concurrent requests.

## Changes

- removed `OnError` Method.
- added zerolog logger.
- refactor metrics monitoring and add new metrics
- reverted to simple queue.
- timeout to HTTP client.
- update gitignore.

Author: twiny

.gitignore

@@ -1,4 +1,5 @@
-cmd/tests
-makefile
-logs
-tmp
+bin/
+tests/
+.idea.md
+*.*prof
+.vscode/

Makefile

@@ -0,0 +1,6 @@
+clean:
+start:
+init:
+build:
+usage:
+.PHONY: clean start init build usage

README.md

@@ -1,22 +1,67 @@
-## WBot
+# WBot
 
 A configurable, thread-safe web crawler, provides a minimal interface for crawling and downloading web pages.
 
-### Features:
+## Features
+
 - Clean minimal API.
 - Configurable: MaxDepth, MaxBodySize, Rate Limit, Parrallelism,  User Agent & Proxy rotation.
 - Memory-efficient, thread-safe.
 - Provides built-in interface: Fetcher, Store, Queue & a Logger.
 
+## API
+
+WBot provides a minimal API for crawling web pages.
+
+```go
+Run(links ...string) error
+OnReponse(fn func(*wbot.Response))
+Metrics() map[string]int64
+Shutdown()
+```
+
+## Usage
+
+```go
+package main
+
+import (
+ "fmt"
+ "log"
 
-### [Examples & API](https://github.com/twiny/wbot/wiki)
+ "github.com/rs/zerolog"
+ "github.com/twiny/wbot"
+ "github.com/twiny/wbot/crawler"
+)
 
-### TODO
-- [ ] Add support for robots.txt.
-- [ ] Add test cases.
-- [ ] Implement `Fetch` using Chromedp.
-- [ ] Add more examples.
-- [ ] Add documentation.
+func main() {
+ bot := crawler.New(
+  crawler.WithParallel(50),
+  crawler.WithMaxDepth(5),
+  crawler.WithRateLimit(&wbot.RateLimit{
+   Hostname: "*",
+   Rate:     "10/1s",
+  }),
+  crawler.WithLogLevel(zerolog.DebugLevel),
+ )
+ defer bot.Shutdown()
+
+ // read responses
+ bot.OnReponse(func(resp *wbot.Response) {
+  fmt.Printf("crawled: %s\n", resp.URL.String())
+ })
+
+ if err := bot.Run(
+  "https://crawler-test.com/",
+ ); err != nil {
+  log.Fatal(err)
+ }
+
+ log.Printf("finished crawling\n")
+}
+
+```
 
 ### Bugs
+
 Bugs or suggestions? Please visit the [issue tracker](https://github.com/twiny/wbot/issues).

config.go

@@ -0,0 +1,66 @@
+package crawler
+
+import (
+	"runtime"
+	"time"
+
+	"github.com/twiny/poxa"
+)
+
+const (
+	defaultReferrer    = "https://www.google.com/search"
+	defaultUserAgent   = "WBot/v0.2.0 (+https://github.com/twiny/wbot)"
+	defaultTimeout     = 10 * time.Second
+	defaultMaxBodySize = int64(1024 * 1024 * 5) // 5MB
+)
+
+type (
+	config struct {
+		parallel    int
+		maxDepth    int32
+		maxBodySize int64
+		timeout     time.Duration
+		userAgents  poxa.Spinner[string]
+		referrers   poxa.Spinner[string]
+		proxies     poxa.Spinner[string]
+	}
+)
+
+func newConfig(maxDepth int32, userAgents, referrers, proxies []string) *config {
+	if maxDepth <= 0 {
+		maxDepth = 10
+	}
+
+	var conf = &config{
+		parallel:    runtime.NumCPU(),
+		maxDepth:    maxDepth,
+		maxBodySize: defaultMaxBodySize,
+		timeout:     defaultTimeout,
+		userAgents:  poxa.NewSpinner(defaultUserAgent),
+		referrers:   poxa.NewSpinner(defaultReferrer),
+		proxies:     nil,
+	}
+
+	if len(userAgents) > 0 {
+		uaList := poxa.NewSpinner(userAgents...)
+		if uaList != nil {
+			conf.userAgents = uaList
+		}
+	}
+
+	if len(referrers) > 0 {
+		refList := poxa.NewSpinner(referrers...)
+		if refList != nil {
+			conf.referrers = refList
+		}
+	}
+
+	if len(proxies) > 0 {
+		proxyList := poxa.NewSpinner(proxies...)
+		if proxyList != nil {
+			conf.proxies = proxyList
+		}
+	}
+
+	return conf
+}