From cb33edc3cdce325d945ab6fbf250523b0aa66bf2 Mon Sep 17 00:00:00 2001
From: vylion <volfaria@gmail.com>
Date: Tue, 19 Sep 2017 13:00:56 +0200
Subject: [PATCH] Initial commit. Uploading Velasco v1.1

---
 .gitignore                         |   1 +
 __pycache__/chatlog.cpython-36.pyc | Bin 0 -> 1963 bytes
 __pycache__/markov.cpython-36.pyc  | Bin 0 -> 1723 bytes
 chatlog.py                         |  58 +++++++++
 markov.py                          |  50 ++++++++
 velasco.py                         | 184 +++++++++++++++++++++++++++++
 6 files changed, 293 insertions(+)
 create mode 100644 .gitignore
 create mode 100644 __pycache__/chatlog.cpython-36.pyc
 create mode 100644 __pycache__/markov.cpython-36.pyc
 create mode 100644 chatlog.py
 create mode 100644 markov.py
 create mode 100755 velasco.py

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..6ec0a37
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1 @@
+chatlogs/*
\ No newline at end of file
diff --git a/__pycache__/chatlog.cpython-36.pyc b/__pycache__/chatlog.cpython-36.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..2237ed8fa4dd93abeb497b6e41d9fff8f80299b2
GIT binary patch
literal 1963
zcmZuy&2HO95Z>jI6eUZx>=<dC{-rKZfI$%-iUK_ZMbH%OsX~DUFe;!R80}h-M3M4x
zZP)^JD3m@-pQ2CDL$7=7Nw1Jozge0}Y?uOPhn(Tg%s1bxzV7upfBf_JZ~q32{m0(B
zeCW?{mmfn&CV9qQvY(jrq<F)mP<$<hp7eiZaj=00i@D`bMx-^Dfbtx7Ie<`X$R)!U
z2W|&G6w;T$8#eT1C|g(q8Ob)*P<G@F)|Twb9@a?i$~~-Y*_ZoRJCY;j1KWD>IyG4_
z9*jb4PZzBo?iaYrUmz-9vRPaAE5Bk1%fm6F)$oe1IW({GAVno0{VITOq<@sW@@Ap_
zUV5t__oY|`RRBAzi0Th)FxJJzVsnV?H-L=Aq4nqGxU`})+IkadeRI7~HZT*DDeIqU
zbxuB|%Fb*skt#RCXoS?sOYdL@-6GM2_>1X1EN|nJ*Tr0&TwZ6BB0o7T^lXu(BfL;q
zT8@en`W%%%U0mBJNhbNkBncH?K7?RA;*l5f4iE8z&O_nzW&Z}Qr_Bc=f{HF`(-yX4
zJXCzj)@%iMrd$Fl4%nv9E01&sgg03FpU*C{SlFOkWD{e<bg@vmbj8F%_ppy`%~e^Z
zV`T*pa;4Rx?~JLG3EorArE4Kl?eOJZljx0o#Nj{}5t44#-e*w8Y{joWMQ-c|_WL37
z<&^Ifd9Pj8=iO7#7SDKE34=zvwu2OCr8juXAICur{sZi0{b@1DhvCnuo)wo?j8z`{
zt`6s8`5c{9SLSXVyJKaNx%()=Hy>fEvID`m&%6APN_yCoblY`X8C6KPtLy*@)wISd
ztjK>&`SJl!rBT#J5W_=+QaUxN2~DZRyv~<M^d8dakV<FN*^=-hf-HMYO3rpGlPwph
z3gdEE)5swM&x%atb&v0l)us|h#YJwuq<FL!<IDXf+Sa;dA9>u0SkY+KEJH-@1|#<B
z!e~hD;vbcW=%x(DCg7Isjd5Ecv3<*;Asz_pF<;?O0K>s^$M;$XU(f^S1_#JAs9EJ0
z7v5rAYs<KA!8mW)BE|tRG0-0%g5{k!(rsv(h^7hGdh#RETv6@}_z;iH>;%~)SEYVL
z9(Vc~nj$e*<|~-WT?lGu7r(ya+j6%7yLQ`FPE}Jt$Fi4DsB+9X#~gWA#}#HrvOm2e
znx=!5py{s!v5!zVe{{+)L0rNPDV+2nJz&FSa0IV)fSSX>h5~8g^TrA4?uKBG1KV+j
zMO~W1g4nLfXvHKq7&w|AeN4hZa$o(?nJ(spz&CKanY9Shc7WfzbFOj1qz~d9Iw<pW
zu9C!dl4M@Ui;VPclAK?pS#8n#6qe4TdeA;0?K2XONqj-#35j?8a2g$Ft_J!)DCr3Y
rSw9M*R^&$lzqYetfQMmyP}s0|HB}?y-ru*Kjh@vxen;3iR!83dRnv=8

literal 0
HcmV?d00001

diff --git a/__pycache__/markov.cpython-36.pyc b/__pycache__/markov.cpython-36.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..69fd75edca47bf60a8ff7b0bccff7b3211ef7952
GIT binary patch
literal 1723
zcmb7EPj4GV6rY)${bS>#B7`=fs9J&qF^U{$FI82c2oRMJSV)nYVu4m$&)8mjy=%{I
z65HDC#rY)MxNzry_!b=I$_0Ogo_KH8HdX0`vG&dLn>TOXpWk~Ac6RE2|MtfZUpR#P
zMP6AP_%ryj_kjo^XiPff2@y<KH$+&HF0Bp|^qe&L3N%Q&i*&OrV9wynJ_RC4hYAAP
zZ<*V)V+k(o8`9yz5iabuXo#Aq-_VX@EbeIMyHJhNt9F;ceO$QX_a%JUBOnneNI?2X
zX#wU(o;nb;1b)btMw(j%8PX*M4Gx^WBoqPxhgn;brlaj{*d0i1U#CiB+Ri3%lxr)N
ziLRk2$fAYRo(S{sd6>y2(>#-LPxD;P@?VH*fD<o|PY3B(o?gx4C{0c;QZ<^yVHX54
z4zq50IyPanCUfltL6k&!5I~|ywhM$%2Yw#qv`!mzajOcswbAQR)PaRz<0VXR_6$f4
z;bCBcj)>Y5?3~Qba<-LP2V_aVB0qfs$pF2A9vXc}62C*XyoG$vdJ0LIeslqKg~Zs4
z-~A^tkt)n(=I1J!#IW~NI`;E{^!w43Ongj;pTS|$Z~5n`ANBn7QmrNawN$e1zqXDp
z24U`pew;#H=3rrdbOM?m3%|d9%|Aw!6F)pzD>eNv5%_BIA8W^3r4uyOY;84G2)p(Q
z6Sl(aH7(`BrfeYER$<bIGO9ZMqme0KMRri)uH(H0e=-jUads(Z2XwK!uFP6tTNpHp
zZDKI_EikjMA>m|6Ao-7ev`SjC(kf`ddaUFHVvvp;_0SZBACe&}tR;o?*9)s(m5M)x
ze^3lWDQ$!M`(Rb!8`=pc6PbvnUHQ+hKhx~`eurH@(AG%KUyDZ7;Lct6GK`kceZX0r
z?y)V(Hn3jZhvRLR;Vp!kiRe#Ya`0K8Z%i<N6)PF=R>2I6U>62PjFeW%XAcE0EQnD2
zpIg(`1u3k8gAKcM`lNJcPjg-XydjI<lwRQu?ZU=9fpur(WIq>n;q=fCXXL71b61$t
z@5%H<;lRrto>V%A&<K^?=?loJGprTv5=WNXR)`OUwN`HfS?~`=SMg$ZgMNS>booo_
zH|yE~;D~grUEHH2R|pP+DYb{(yGTrp><M&5uAS~6jk;3vVHzdPT2(M^#tDoNDLOTz
z0S=i?ErE9ot8Sy6J3u-$XbU6D)H^7|QS%B?_cvV=RO2N;Uq1vLGph;Rqwis{={=yF
z?gG7yV}{+Mi@R$mzsUk3j;cej({yoMCE-{GfvyL^I2D&M@{J&vUWRdXqHu^+2u0ke
zBP9R2YB=mHh5d-VV!8@&So0d5XQY5|1gUn?=R?`ejqFV(wD-93;dAWp3=MkhA5Or8
AW&i*H

literal 0
HcmV?d00001

diff --git a/chatlog.py b/chatlog.py
new file mode 100644
index 0000000..f57a380
--- /dev/null
+++ b/chatlog.py
@@ -0,0 +1,58 @@
+#!/usr/bin/env python3
+
+from markov import *
+
+class Chatlog(object):
+    def __init__(self, ident, chattype, title, msgs=None, freq=None):
+        if msgs is not None:
+            self.msgs = msgs
+        else:
+            self.msgs = []
+        self.id = str(ident)
+        self.type = chattype
+        self.title = title
+        if freq is None:
+            if "group" in chattype:
+                freq = 20
+            #elif chattype is "private":
+            else:
+                freq = 5
+        self.freq = freq
+
+    def add_msg(self, message):
+        msg = message.split()
+        msg.append("!kvl")
+        self.msgs.append(msg)
+
+    def get_markov_gen(self):
+        msgs = []
+        for m in self.msgs:
+            msgs.append(' '.join(m))
+        text = ' '.join(msgs)
+        self.gen = Markov(text)
+
+    def speak(self):
+        self.get_markov_gen()
+        return self.gen.generate_markov_text()
+
+    def get_count(self):
+        return len(self.msgs)
+
+    def set_freq(self, freq):
+        self.freq = freq
+
+    def to_txt(self):
+        lines = [self.id]
+        lines.append(self.type)
+        lines.append(self.title)
+        lines.append(str(self.freq))
+        for m in self.msgs:
+            lines.append(' '.join(m))
+        return '\n'.join(lines)
+
+    def from_txt(text):
+        lines = text.splitlines()
+        msgs = []
+        for m in lines[4:]:
+            msgs.append(m.split())
+        return Chatlog(lines[0], lines[1], lines[2], msgs, int(lines[3]))
diff --git a/markov.py b/markov.py
new file mode 100644
index 0000000..bc0725f
--- /dev/null
+++ b/markov.py
@@ -0,0 +1,50 @@
+#!/usr/bin/env python3
+
+import random
+
+class Markov(object):
+    def __init__(self, text=None):
+        self.cache = {}
+        self.words = []
+        if text is None:
+            text = ""
+        self.words = text.split()
+        self.word_size = len(self.words)
+        self.database()
+
+    def triples(self):
+        """ Generates triples from the given data string. So if our string were
+                "What a lovely day", we'd generate (What, a, lovely) and then
+                (a, lovely, day).
+        """
+
+        if len(self.words) < 3:
+            return
+
+        for i in range(len(self.words) - 2):
+            yield (self.words[i], self.words[i+1], self.words[i+2])
+
+    def database(self):
+        for w1, w2, w3 in self.triples():
+            key = (w1, w2)
+            if key in self.cache:
+                self.cache[key].append(w3)
+            else:
+                self.cache[key] = [w3]
+
+    def generate_markov_text(self, size=50):
+        seed = random.randint(0, self.word_size-3)
+        seed_word, next_word = self.words[seed], self.words[seed+1]
+        while "!kvl" in seed_word:
+            seed = random.randint(0, self.word_size-3)
+            seed_word, next_word = self.words[seed], self.words[seed+1]
+        w1, w2 = seed_word, next_word
+        gen_words = []
+        for i in range(size):
+            gen_words.append(w1)
+            if "!kvl" in w2 or not (w1, w2) in self.cache:
+                print("Generated text")
+                break
+            else:
+                w1, w2 = w2, random.choice(self.cache[(w1, w2)])
+        return ' '.join(gen_words)
diff --git a/velasco.py b/velasco.py
new file mode 100755
index 0000000..73deaf7
--- /dev/null
+++ b/velasco.py
@@ -0,0 +1,184 @@
+#!/usr/bin/env python3
+
+import sys, os
+from telegram.ext import Updater, CommandHandler, MessageHandler, Filters
+from chatlog import *
+import logging
+import argparse
+
+# Enable logging
+logging.basicConfig(format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
+                    level=logging.INFO)
+
+logger = logging.getLogger(__name__)
+
+chatlogs = {}
+disabled = {}
+
+GUILLERMO_ID = 8379173
+
+def wake(bot):
+    directory = os.fsencode("chatlogs/")
+
+    for file in os.listdir(directory):
+        filename = os.fsdecode(file)
+        if filename.endswith(".txt"):
+            chat = loadchat("chatlogs/" + filename)
+            chatlogs[chat.id] = chat
+            print("loaded chat " + chat.id)
+            continue
+        else:
+            continue
+"""
+    for c in chatlogs:
+        try:
+            bot.sendMessage(chatlogs[c].id, "Good morning. I just woke up")
+        except:
+            pass
+            #del chatlogs[c]
+"""
+
+def start(bot, update):
+    update.message.reply_text('cowabunga')
+
+def savechat(chatlog):
+    open_file = open('chatlogs/' + chatlog.id + '.txt', 'w')
+    open_file.write(chatlog.to_txt())
+    open_file.close()
+
+def loadchat(path):
+    open_file = open(path, 'r')
+    chat = Chatlog.from_txt(open_file.read())
+    open_file.close()
+    return chat
+
+def help(bot, update):
+    update.message.reply_text("""I answer to the following commands:
+
+/start - I say hi.
+/about - What I'm about.
+/help - I send this message.
+/count - I tell you how many messages from this chat I remember.
+/freq - Change the frequency of both my messages and the times I save my learned vocabulary.
+    """)
+
+def about(bot, update):
+    update.message.reply_text('I am yet another Markov Bot experiment. I read everything you type to me and then spit back nonsensical messages that look like yours')
+
+def echo(bot, update):
+    update.message.reply_text(update.message.text)
+
+def error(bot, update, error):
+    logger.warn('Update "%s" caused error "%s"' % (update, error))
+
+def get_chatname(chat):
+    if chat.title is not None:
+        return chat.title
+    elif chat.first_name is not None:
+        if chat.last_name is not None:
+            return chat.first_name + " " + chat.last_name
+        else:
+            return chat.first_name
+    else:
+        return ""
+
+def read(bot, update):
+    global chatlogs
+    ident = str(update.message.chat.id)
+    if not ident in chatlogs:
+        chat = update.message.chat
+        title = get_chatname(chat)
+        chatlog = Chatlog(chat.id, chat.type, title)
+    else:
+        chatlog = chatlogs[ident]
+    chatlog.add_msg(update.message.text)
+    if chatlog.get_count()%chatlog.freq == 0:
+        msg = chatlog.speak()
+        bot.sendMessage(chatlog.id, msg)
+        savechat(chatlog)
+
+    chatlogs[chatlog.id] = chatlog
+
+def get_chatlogs(bot, update):
+    global GUILLERMO_ID
+    if update.message.chat.id is GUILLERMO_ID:
+        m = "I have these chatlogs:"
+        for c in chatlogs:
+            m += "\n" + chatlogs[c].id + " " + chatlogs[c].title
+        bot.sendMessage(GUILLERMO_ID, m)
+
+def get_count(bot, update):
+    ident = str(update.message.chat.id)
+    reply = "I remember "
+    if ident in chatlogs:
+        reply += str(chatlogs[ident].get_count())
+    else:
+        reply += "no"
+    reply += " messages."
+    update.message.reply_text(reply)
+
+def set_freq(bot, update):
+    ident = str(update.message.chat.id)
+    if not ident in chatlogs:
+        chat = update.message.chat
+        title = get_chatname(chat)
+        chatlog = Chatlog(chat.id, chat.type, title)
+        chatlogs[chatlog.id] = chatlog
+    if not len(update.message.text.split()) > 1:
+        reply = "Current frequency is " + str(chatlogs[ident].freq)
+    else:
+        try:
+            value = update.message.text.split()[1]
+            value = int(value)
+            chatlogs[ident].set_freq(value)
+            reply = "Frequency of speaking set to " + str(value)
+        except:
+            reply = "Format was confusing; requency not changed from " + str(chatlogs[ident].freq)
+    update.message.reply_text(reply)
+
+def stop(bot, update):
+    chatlog = chatlogs[update.message.chat.id]
+    del chatlogs[chatlog.id]
+    os.remove("chatlogs/" + chatlog.id + ".txt")
+    print("I got blocked. Removed user " + chatlog.id)
+
+def main():
+    parser = argparse.ArgumentParser(description='A Telegram markovbot.')
+    parser.add_argument('token', metavar='TOKEN', help='The Bot Token to work with the Telegram Bot API')
+
+    args = parser.parse_args()
+
+    # Create the EventHandler and pass it your bot's token.
+    updater = Updater(args.token)
+
+    # Get the dispatcher to register handlers
+    dp = updater.dispatcher
+
+    # on different commands - answer in Telegram
+    dp.add_handler(CommandHandler("start", start))
+    dp.add_handler(CommandHandler("about", about))
+    dp.add_handler(CommandHandler("help", help))
+    dp.add_handler(CommandHandler("count", get_count))
+    dp.add_handler(CommandHandler("freq", set_freq))
+    dp.add_handler(CommandHandler("list", get_chatlogs))
+    dp.add_handler(CommandHandler("stop", stop))
+
+    # on noncommand i.e message - echo the message on Telegram
+    # dp.add_handler(MessageHandler(Filters.text, echo))
+    dp.add_handler(MessageHandler(Filters.text, read))
+
+    # chatlog all errors
+    dp.add_error_handler(error)
+
+    wake(updater.bot)
+
+    # Start the Bot
+    updater.start_polling()
+
+    # Run the bot until you press Ctrl-C or the process receives SIGINT,
+    # SIGTERM or SIGABRT. This should be used most of the time, since
+    # start_polling() is non-blocking and will stop the bot gracefully.
+    updater.idle()
+
+if __name__ == '__main__':
+    main()