Made insultifer bot

8 files changed, 304 insertions, 468 deletions

@@ -1,14 +1,12 @@
 cmake_minimum_required (VERSION 3.1)
-project (chemist)
-
-set(CMAKE_BUILD_TYPE Debug)
+project (insult)
 
 add_subdirectory(vendor/verbly)
 add_subdirectory(vendor/yaml-cpp EXCLUDE_FROM_ALL)
 add_subdirectory(vendor/libtwittercpp)
 
 include_directories(vendor/verbly/lib vendor/yaml-cpp/include vendor/libtwittercpp/src)
-add_executable(chemist chemist.cpp)
-set_property(TARGET chemist PROPERTY CXX_STANDARD 11)
-set_property(TARGET chemist PROPERTY CXX_STANDARD_REQUIRED ON)
-target_link_libraries(chemist verbly twitter++ yaml-cpp)
+add_executable(insult insult.cpp patterner.cpp)
+set_property(TARGET insult PROPERTY CXX_STANDARD 11)
+set_property(TARGET insult PROPERTY CXX_STANDARD_REQUIRED ON)
+target_link_libraries(insult verbly twitter++ yaml-cpp)
@@ -1,183 +0,0 @@
-#include <yaml-cpp/yaml.h>
-#include <iostream>
-#include <sstream>
-#include <verbly.h>
-#include <fstream>
-#include <twitter.h>
-#include <random>
-#include <chrono>
-#include <thread>
-
-int main(int argc, char** argv)
-{
-  if (argc != 2)
-  {
-    std::cout << "usage: chemist [configfile]" << std::endl;
-    return -1;
-  }
-
-  std::string configfile(argv[1]);
-  YAML::Node config = YAML::LoadFile(configfile);
-  
-  twitter::auth auth;
-  auth.setConsumerKey(config["consumer_key"].as<std::string>());
-  auth.setConsumerSecret(config["consumer_secret"].as<std::string>());
-  auth.setAccessKey(config["access_key"].as<std::string>());
-  auth.setAccessSecret(config["access_secret"].as<std::string>());
-  
-  twitter::client client(auth);
-  
-  std::map<std::string, std::vector<std::string>> groups;
-  std::ifstream datafile(config["forms_file"].as<std::string>());
-  if (!datafile.is_open())
-  {
-    std::cout << "Could not find datafile" << std::endl;
-    return 1;
-  }
-  
-  bool newgroup = true;
-  std::string line;
-  std::string curgroup;
-  while (getline(datafile, line))
-  {
-    if (line.back() == '\r')
-    {
-      line.pop_back();
-    }
-    
-    if (newgroup)
-    {
-      curgroup = line;
-      newgroup = false;
-    } else {
-      if (line.empty())
-      {
-        newgroup = true;
-      } else {
-        groups[curgroup].push_back(line);
-      }
-    }
-  }
-  
-  std::random_device random_device;
-  std::mt19937 random_engine{random_device()};
-  
-  verbly::data database {config["verbly_datafile"].as<std::string>()};
-  for (;;)
-  {
-    std::cout << "Generating tweet" << std::endl;
-    std::string action = "{Main}";
-    int tknloc;
-    while ((tknloc = action.find("{")) != std::string::npos)
-    {
-      std::string token = action.substr(tknloc+1, action.find("}")-tknloc-1);
-      std::string modifier;
-      int modloc;
-      if ((modloc = token.find(":")) != std::string::npos)
-      {
-        modifier = token.substr(modloc+1);
-        token = token.substr(0, modloc);
-      }
-      
-      std::string canontkn;
-      std::transform(std::begin(token), std::end(token), std::back_inserter(canontkn), [] (char ch) {
-        return std::toupper(ch);
-      });
-      
-      std::string result;
-      if (canontkn == "NOUN")
-      {
-        result = database.nouns().is_not_proper().random().limit(1).with_complexity(1).run().front().singular_form();
-      } else if (canontkn == "ATTRIBUTE")
-      {
-        result = database.nouns().random().limit(1).full_hyponym_of(database.nouns().with_wnid(100024264).limit(1).run().front()).run().front().singular_form();
-      } else if (canontkn == "ADJECTIVE")
-      {
-        result = database.adjectives().with_complexity(1).random().limit(1).run().front().base_form();
-      } else if (canontkn == "VERBING")
-      {
-        result = database.verbs().random().limit(1).run().front().ing_form();
-      } else if (canontkn == "YEAR")
-      {
-        std::uniform_int_distribution<int> yeardist(1916,2015);
-        int year = yeardist(random_engine);
-        result = std::to_string(year);
-      } else if (canontkn == "REGION")
-      {
-        auto hem1 = database.nouns().with_singular_form("eastern hemisphere").limit(1).run().front();
-        auto hem2 = database.nouns().with_singular_form("western hemisphere").limit(1).run().front();
-        verbly::filter<verbly::noun> region{hem1, hem2};
-        region.set_orlogic(true);
-        
-        result = database.nouns().full_part_holonym_of(region).random().limit(1).run().front().singular_form();
-      } else if (canontkn == "FAMOUSNAME")
-      {
-        auto person = database.nouns().with_singular_form("person").limit(1).run().front();
-        auto ptypes = database.nouns().full_hyponym_of({person}).is_class().random().limit(1).run().front();
-        result = database.nouns().instance_of({ptypes}).random().limit(1).run().front().singular_form();
-      } else if (canontkn == "BODYPART")
-      {
-        auto bp = database.nouns().with_singular_form("body part").limit(1).run().front();
-        result = database.nouns().full_hyponym_of({bp}).with_complexity(1).random().limit(1).run().front().singular_form();
-      } else if (canontkn == "\\N")
-      {
-        result = "\n";
-      } else {
-        auto group = groups[canontkn];
-        std::uniform_int_distribution<int> groupdist(0, group.size()-1);
-        int groupind = groupdist(random_engine);
-        result = group[groupind];
-      }
-      
-      if (modifier == "indefinite")
-      {
-        if ((result.length() > 1) && (isupper(result[0])) && (isupper(result[1])))
-        {
-          result = "an " + result;
-        } else if ((result[0] == 'a') || (result[0] == 'e') || (result[0] == 'i') || (result[0] == 'o') || (result[0] == 'u'))
-        {
-          result = "an " + result;
-        } else {
-          result = "a " + result;
-        }
-      }
-      
-      std::string finalresult;
-      if (islower(token[0]))
-      {
-        std::transform(std::begin(result), std::end(result), std::back_inserter(finalresult), [] (char ch) {
-          return std::tolower(ch);
-        });
-      } else if (isupper(token[0]) && !isupper(token[1]))
-      {
-        auto words = verbly::split<std::list<std::string>>(result, " ");
-        for (auto& word : words)
-        {
-          word[0] = std::toupper(word[0]);
-        }
-        
-        finalresult = verbly::implode(std::begin(words), std::end(words), " ");
-      } else {
-        finalresult = result;
-      }
-      
-      action.replace(tknloc, action.find("}")-tknloc+1, finalresult);
-    }
-    
-    action.resize(140);
-    
-    try
-    {
-      client.updateStatus(action);
-      
-      std::cout << "Tweeted!" << std::endl;
-    } catch (const twitter::twitter_error& e)
-    {
-      std::cout << "Twitter error: " << e.what() << std::endl;
-    }
-    
-    std::cout << "Waiting..." << std::endl;
-    
-    std::this_thread::sleep_for(std::chrono::hours(1));
-  }
-}
@@ -1,279 +1,33 @@
 MAIN
-{NAME} ({CLASS}){\n}{PRIMARY}{\n}{SECONDARY}
-
-PRIMARY
-Used to treat {SYNDROME}
-Treats the {adjective} symptoms of {SYNDROME}
-Cures {SYNDROME}
-Approved to treat {SYNDROME} and {SYNDROME}
-Prescribed for {SYNDROME}
-Used with {EXISTENT} to treat {SYNDROME}
-Used recreationally as {CLASS:indefinite}
-Used recreationally for {verbing}
-
-SECONDARY
-Developed after {year} {Noun} Pandemic
-Often used off-label for {SYNDROME}
-Notable for its {adjective} effect
-Controlled in the US due to its {adjective} effect
-Developed in {year} to replace {EXISTENT}
-May cause {ADJECTIVE} feelings
-Obsoleted by {EXISTENT}
-Contraindicated by {SYNDROME}
-Contraindicated by {EXISTENT}
-Decreases the {adjective} effects of {EXISTENT}
-
-SYNDROME
-irritable {noun} syndrome
-{adjective} {noun} syndrome
-severe {noun}
-major {adjective} disorder
-{adjective} {noun} disorder
-obsessive {noun} disorder
-clinical {noun}
-{adjective} personality disorder
-respiratory {adjective} disease
-{bodypart} cancer
-restless {bodypart}
-{bodypart} failure
-congenital {noun} disease
-{FamousName}'s disease
-{adjective} fever
-hypo{noun}ism
-{EXISTENT} overdose
-{CLASS} overdose
-{CLASS} discontinuation syndrome
-{noun} syndrome
-low {ATTRIBUTE}
-
-CLASSIFIER
-{CLASS:indefinite}
-{NAME}
-{EXISTENT}
-
-CLASS
-analgesic
-painkiller
-anaesthetic
-antihistamine
-anticonvulsant
-antiepileptic
-antidepressant
-antimigraine
-antipsychotic
-benzodiazepine
-antiparkinsonian
-immunosuppressive
-antianaemia
-anticoagulant
-blood thinner
-antiarrhythmic
-antithrombotic
-antifungal
-anti-infective
-anti-inflammatory
-disinfectant
-antiseptic
-antiemetic
-diuretic
-opiod painkiller
-antiulcer
-laxative
-sedative
-hormone
-estrogen
-androgen
-contraceptive
-ovulation inducer
-thyroid stimulant
-antithyroid
-insulin
-vaccine
-oxytocin
-SSRI
-anxiolytic
-antipanic agent
-tricyclic
-tetracyclic
-MAOI
-SNRI
-antiandrogen
-psychedelic
-vitamin
-probiotic
-antibiotic
-antiviral drug
-stimulant
-depressant
-aphrodisiac
-{EXISTENT} prodrug
-
-EXISTENT
-fluoxetine
-Prozac
-sertraline
-Zoloft
-escitalopram
-Lexapro
-venlafaxine
-Effexor
-aripiprazole
-Abilify
-alprazolam
-Xanax
-diazepam
-Valium
-lamotrigine
-Lamictal
-gabapentin
-Neurontin
-acetaminophen
-Tylenol
-ibuprofin
-Advil
-lurasidone
-Latuda
-lithium
-activated charcoal
-estradiol
-AndroGel
-ziprasidone
-Geodon
-risperidone
-Risperdal
-quetiapine
-Seroquel
-Cymbalta
-duloxetine
-bupropion
-Welbutrin
-buspirone
-Buspar
-oxycontin
-Oxycodone
-Concerta
-methylphenidate
-Ritalin
-Vyvanse
-lisdexamfetamine
-Adderall
-epinephrine
-adrenaline
-testosterone gel
-Cialis
-Viagra
-heroin
-morphine
-crystal meth
-mirtazapine
-Remeron
-Luvox
-fluvoxamine
-
-NAME
-{PRENAME}{NAMEMID}{NAMEIFX}
-
-PRENAME
-Oxy
-Ari
-Zi
-Quetia
-Mor
-Her
-Cia
-Via
-He
-Con
-Flu
-Ser
-Es
-Ven
-Al
-Dia
-Lamo
-Gaba
-Aceta
-Ibu
-Lura
-Ris
-Li
-Estra
-Du
-Bus
-Epin
-Co
-Lido
-Pro
-Pri
-Bu
-Levo
-Ro
-Me
-Dibu
-Des
-Ha
-Mir
-
-NAMEMID
-pipra
-pi
-to
-stero
-cer
-oxe
-tra
-cita
-lo
-la
-fa
-pra
-zo
-ze
-tri
-pen
-mino
-pro
-si
-peri
-thi
-loxe
-con
-epher
-piva
-bupi
-va
-piva
-flu
-oxy
-taz
-{NAMEMID}{NAMEMID}
-{NAMEMID}{NAMEMID}
-
-NAMEIFX
-zole
-ne
-tine
-lin
-tamine
-gra
-phine
-ta
-line
-pram
-xine
-lam
-pam
-gine
-tin
-phen
-fin
-done
-um
-diol
-tin
-rone
-ine
-caine
-rane
-ide
-epine
\ No newline at end of file
+{INSULT}
+{INSULT}
+{INSULT}
+{INSULT}
+{Insult}
+{INSULT}!
+{INSULT}!
+{INSULT}!!!!
+{INSULT}!!!!
+{INSULT}?
+{INSULT}????
+{INSULT}?!?!?!?!
+
+INSULT
+{START} {END}
+What the {WORD} did you say to me you little {WORD}?
+
+INSULT,START
+{WORD} off
+{WORD} you
+{WORD} you and the horse you rode in on
+{WORD} you in particular
+go {WORD} yourself
+go to {WORD}
+leave me the {WORD} alone
+you're a piece of {WORD}
+what the {WORD}
+what the {WORD}ing {WORD}
+kindly catch the 9am train to {Word}sville
+
+INSULT,END
+you piece of {WORD}
\ No newline at end of file
@@ -0,0 +1,108 @@
+#include <yaml-cpp/yaml.h>
+#include <iostream>
+#include <sstream>
+#include <verbly.h>
+#include <fstream>
+#include <twitter.h>
+#include <random>
+#include <chrono>
+#include <thread>
+#include "patterner.h"
+
+int main(int argc, char** argv)
+{
+  if (argc != 2)
+  {
+    std::cout << "usage: insult [configfile]" << std::endl;
+    return -1;
+  }
+
+  std::string configfile(argv[1]);
+  YAML::Node config = YAML::LoadFile(configfile);
+
+  twitter::auth auth;
+  auth.setConsumerKey(config["consumer_key"].as<std::string>());
+  auth.setConsumerSecret(config["consumer_secret"].as<std::string>());
+  auth.setAccessKey(config["access_key"].as<std::string>());
+  auth.setAccessSecret(config["access_secret"].as<std::string>());
+
+  twitter::client client(auth);
+
+  std::random_device randomDevice;
+  std::mt19937 rng(randomDevice());
+
+  try
+  {
+    verbly::database database(config["verbly_datafile"].as<std::string>());
+    patterner pgen(config["forms_file"].as<std::string>(), database, rng);
+
+    std::cout << "Starting streaming..." << std::endl;
+
+    twitter::stream userStream(client, [&pgen, &client]
+      (const twitter::notification& n) {
+        if (n.getType() == twitter::notification::type::tweet)
+        {
+          if ((!n.getTweet().isRetweet())
+            && (n.getTweet().getAuthor() != client.getUser()))
+          {
+            std::string original = n.getTweet().getText();
+            std::string canonical;
+
+            std::transform(std::begin(original), std::end(original),
+              std::back_inserter(canonical), [] (char ch)
+            {
+              return std::tolower(ch);
+            });
+
+            if (canonical.find("@teammeanies") != std::string::npos)
+            {
+              std::string doc =
+                n.getTweet().generateReplyPrefill(client.getUser());
+
+              doc += pgen.generate();
+              doc.resize(140);
+
+              try
+              {
+                client.replyToTweet(doc, n.getTweet());
+              } catch (const twitter::twitter_error& error)
+              {
+                std::cout << "Twitter error while tweeting: "
+                  << error.what() << std::endl;
+              }
+            }
+          }
+        }
+      });
+
+    std::this_thread::sleep_for(std::chrono::minutes(1));
+
+    for (;;)
+    {
+      std::cout << "Generating tweet..." << std::endl;
+
+      std::string action = pgen.generate();
+      action.resize(140);
+
+      std::cout << action << std::endl;
+
+      try
+      {
+        client.updateStatus(action);
+
+        std::cout << "Tweeted!" << std::endl;
+      } catch (const twitter::twitter_error& e)
+      {
+        std::cout << "Twitter error: " << e.what() << std::endl;
+      }
+
+      std::cout << "Waiting..." << std::endl;
+
+      std::this_thread::sleep_for(std::chrono::hours(1));
+    }
+  } catch (std::invalid_argument& e)
+  {
+    std::cout << e.what() << std::endl;
+    return -1;
+  }
+}
@@ -0,0 +1,138 @@
+#include "patterner.h"
+#include <fstream>
+#include <stdexcept>
+
+patterner::patterner(
+  std::string datapath,
+  verbly::database& data,
+  std::mt19937& rng) :
+    data_(data),
+    rng_(rng)
+{
+  std::ifstream datafile(datapath);
+  if (!datafile.is_open())
+  {
+    throw std::invalid_argument("Could not find datafile");
+  }
+
+  bool newgroup = true;
+  std::string line;
+  std::list<std::string> curgroups;
+  while (getline(datafile, line))
+  {
+    if (line.back() == '\r')
+    {
+      line.pop_back();
+    }
+
+    if (newgroup)
+    {
+      curgroups = verbly::split<std::list<std::string>>(line, ",");
+      newgroup = false;
+    } else {
+      if (line.empty())
+      {
+        newgroup = true;
+      } else {
+        for (std::string curgroup : curgroups)
+        {
+          groups_[curgroup].push_back(line);
+        }
+      }
+    }
+  }
+}
+
+std::string patterner::generate()
+{
+  std::string action = "{MAIN}";
+  int tknloc;
+  while ((tknloc = action.find("{")) != std::string::npos)
+  {
+    std::string token = action.substr(tknloc+1, action.find("}")-tknloc-1);
+    std::string modifier;
+    int modloc;
+    if ((modloc = token.find(":")) != std::string::npos)
+    {
+      modifier = token.substr(modloc+1);
+      token = token.substr(0, modloc);
+    }
+
+    std::string canontkn;
+    std::transform(std::begin(token), std::end(token),
+        std::back_inserter(canontkn), [] (char ch) {
+      return std::toupper(ch);
+    });
+
+    std::string result;
+    if (canontkn == "WORD")
+    {
+      result = data_.words(
+        (verbly::word::forms(verbly::inflection::base) %=
+          (verbly::form::complexity == 1)
+            && (verbly::form::length == 4)
+            && (verbly::form::proper == false)
+            && (verbly::pronunciation::numOfSyllables == 1))
+        && !(verbly::word::usageDomains %=
+          (verbly::notion::wnid == 106718862))) // Blacklist ethnic slurs
+        .first().getBaseForm().getText();
+    } else if (canontkn == "\\N")
+    {
+      result = "\n";
+    } else {
+      auto group = groups_[canontkn];
+      std::uniform_int_distribution<int> groupdist(0, group.size()-1);
+      int groupind = groupdist(rng_);
+      result = group[groupind];
+    }
+
+    if (modifier == "indefinite")
+    {
+      if ((result.length() > 1) && (isupper(result[0])) && (isupper(result[1])))
+      {
+        result = "an " + result;
+      } else if ((result[0] == 'a') || (result[0] == 'e') || (result[0] == 'i') || (result[0] == 'o') || (result[0] == 'u'))
+      {
+        result = "an " + result;
+      } else {
+        result = "a " + result;
+      }
+    }
+
+    std::string finalresult;
+    if (islower(token[0]))
+    {
+      std::transform(std::begin(result), std::end(result), std::back_inserter(finalresult), [] (char ch) {
+        return std::tolower(ch);
+      });
+    } else if (isupper(token[0]) && !isupper(token[1]))
+    {
+      auto words = verbly::split<std::list<std::string>>(result, " ");
+      for (auto& word : words)
+      {
+        if (word[0] == '{')
+        {
+          word[1] = std::toupper(word[1]);
+
+          for (int k=2; k<word.length(); k++)
+          {
+            if (std::isalpha(word[k]))
+            {
+              word[k] = std::tolower(word[k]);
+            }
+          }
+        } else {
+          word[0] = std::toupper(word[0]);
+        }
+      }
+
+      finalresult = verbly::implode(std::begin(words), std::end(words), " ");
+    } else {
+      finalresult = result;
+    }
+
+    action.replace(tknloc, action.find("}")-tknloc+1, finalresult);
+  }
+
+  return action;
+}
@@ -0,0 +1,21 @@
+#ifndef PATTERNER_H_AB6883F5
+#define PATTERNER_H_AB6883F5
+
+#include <verbly.h>
+#include <random>
+
+class patterner {
+public:
+
+  patterner(std::string datafile, verbly::database& data, std::mt19937& rng);
+
+  std::string generate();
+
+private:
+
+  std::map<std::string, std::vector<std::string>> groups_;
+  verbly::database& data_;
+  std::mt19937& rng_;
+};
+
+#endif /* end of include guard: PATTERNER_H_AB6883F5 */
-Subproject d90a1e74c77ba67f25a812609fd49d479bc464d
+Subproject df906121dd862c0f704e44f28ee079158c431c4
-Subproject 1f898f3bd66c29672275c2c884b17ba662ced62
+Subproject 1fd518d1c2b1d4e88ad88218b606a284b712810