@@ -1,1 +1,47 @@
 # projectFlock
+
+a twitter botnet with bots replying tweets with text generated by Markov chains
+
+##### generating text with Markov chains
+The algorithm calculates the probabilities of Markov chains, analyzing a considerable amount of text, for the examples, I've done it with the book "The Critique of Pure Reason", by Immanuel Kant (http://www.gutenberg.org/cache/epub/4280/pg4280.txt).
+
+#### Replying tweets with Markov chains
+When the botnet is up working, the bots start streaming all the twitter new tweets containing the configured keywords. Each bot takes a tweet, analyzes the containing words, and generates a reply using the Markov chains previously calculated, and posts the tweet as reply.
+
+In the following examples, the bots ("andreimarkov", "dodecahedron", "projectNSA") are replying some people.
+
+![Argos](https://raw.githubusercontent.com/arnaucode/projectFlock/master/screenshots/01.png "01")
+
+![Argos](https://raw.githubusercontent.com/arnaucode/projectFlock/master/screenshots/02.jpeg "02")
+
+![Argos](https://raw.githubusercontent.com/arnaucode/projectFlock/master/screenshots/03.jpeg "03")
+
+![Argos](https://raw.githubusercontent.com/arnaucode/projectFlock/master/screenshots/04.jpeg "04")
+
+
+configuration file example (flockConfig.json):
+```
+[{
+        "title": "account1",
+        "consumer_key": "xxxxxxxxxxxxx",
+        "consumer_secret": "xxxxxxxxxxxxx",
+        "access_token_key": "xxxxxxxxxxxxx",
+        "access_token_secret": "xxxxxxxxxxxxx"
+    },
+    {
+        "title": "account2",
+        "consumer_key": "xxxxxxxxxxxxx",
+        "consumer_secret": "xxxxxxxxxxxxx",
+        "access_token_key": "xxxxxxxxxxxxx",
+        "access_token_secret": "xxxxxxxxxxxxx"
+    },
+    {
+        "title": "account3",
+        "consumer_key": "xxxxxxxxxxxxx",
+        "consumer_secret": "xxxxxxxxxxxxx",
+        "access_token_key": "xxxxxxxxxxxxx",
+        "access_token_secret": "xxxxxxxxxxxxx"
+    }
+]
+
+```

@@ -11,11 +11,18 @@
 	"github.com/dghubble/go-twitter/twitter"
 )
 
-func isRT(text string) bool {
-	tweetWords := strings.Split(text, " ")
+func isRT(tweet *twitter.Tweet) bool {
+	tweetWords := strings.Split(tweet.Text, " ")
 	for i := 0; i < len(tweetWords); i++ {
 		if tweetWords[i] == "RT" {
-			c.Red(text)
+			return true
+		}
+	}
+	return false
+}
+func isFromBot(flock Flock, tweet *twitter.Tweet) bool {
+	for i := 0; i < len(flock.ScreenNames); i++ {
+		if flock.ScreenNames[i] == tweet.User.ScreenName {
 			return true
 		}
 	}
@@ -40,11 +47,11 @@ fmt.Println(tweet.ID)
 	replyTweet(flockUser, "@"+tweet.User.ScreenName+" "+generatedText, tweet.ID)
 	waitTime(1)
 }
-func startStreaming(states []State, flockUser *twitter.Client, botScreenName string, keywords []string) {
+func startStreaming(states []State, flock Flock, flockUser *twitter.Client, botScreenName string, keywords []string) {
 	// Convenience Demux demultiplexed stream messages
 	demux := twitter.NewSwitchDemux()
 	demux.Tweet = func(tweet *twitter.Tweet) {
-		if isRT(tweet.Text) == false {
+		if isRT(tweet) == false && isFromBot(flock, tweet) == false {
 			processTweet(states, flockUser, botScreenName, keywords, tweet)
 		}
 	}
@@ -101,7 +108,8 @@ switch answer {
 	case "y":
 		fmt.Println("ok, you are sure")
 		for i := 0; i < len(flock.Clients); i++ {
-			go startStreaming(states, flock.Clients[i], flock.ScreenNames[i], keywords)
+			go startStreaming(states, flock, flock.Clients[i], flock.ScreenNames[i], keywords)
+			waitSeconds(35)
 		}
 		break
 	default:

@@ -13,3 +13,10 @@ fmt.Println("waiting " + strconv.Itoa(minutes) + " min to avoid twitter api limitation")
 	fmt.Println(time.Now().Local())
 	time.Sleep(timeToSleep)
 }
+func waitSeconds(seconds int) {
+	//wait to avoid the twitter api limitation
+	timeToSleep := time.Duration(seconds) * time.Second
+	fmt.Println("waiting " + strconv.Itoa(seconds) + " seconds")
+	fmt.Println(time.Now().Local())
+	time.Sleep(timeToSleep)
+}