Merge pull request #7 from gooberinc/rewrite/spacy

merge it

assets/locales/en.json

@@ -1,4 +1,7 @@
 {
+    "active_users:": "Active users:",
+    "spacy_initialized": "spaCy and spacytextblob are ready.",
+    "spacy_model_not_found": "The spaCy model was not found! Downloading it....`",
     "env_file_not_found": "The .env file was not found! Please create one with the required variables.",
     "error_fetching_active_users": "Error fetching active users: {error}",
     "error_sending_alive_ping": "Error sending alive ping: {error}",

assets/locales/it.json

@@ -1,4 +1,7 @@
 {
+    "active_users:": "Utenti attivi:",
+    "spacy_initialized": "spaCy e spacytextblob sono pronti.",
+    "spacy_model_not_found": "Il modello spaCy non è stato trovato! Lo sto scaricando...",
     "env_file_not_found": "Il file .env non è stato trovato! Crea un file con le variabili richieste.",
     "error fetching_active_users": "Errore nel recupero degli utenti attivi:",
     "error_sending_alive_ping": "Errore nell'invio di aliveping:",

bot.py

@@ -109,11 +109,11 @@ async def on_ready():
         print(f"{GREEN}{get_translation(LOCALE, 'synced_commands')} {len(synced)} {get_translation(LOCALE, 'synced_commands2')} {RESET}")
         slash_commands_enabled = True
         ping_server()  # ping_server from modules/central.py
-        # --- Mostra utenti attivi ---
+        # I FORGOT TO REMOVE THE ITALIAN VERSION FUCKKKKKKKKK
         active_users = await fetch_active_users()
-        print(f"{GREEN}Utenti attivi: {active_users}{RESET}")
+        print(f"{GREEN}{get_translation(LOCALE, 'active_users:')} {active_users}{RESET}")
         print(f"{GREEN}{get_translation(LOCALE, 'started').format(name=NAME)}{RESET}")
-        # --- Avvia il task periodico ---
+
         bot.loop.create_task(send_alive_ping_periodically())
     except discord.errors.Forbidden as perm_error:
         print(f"{RED}Permission error while syncing commands: {perm_error}{RESET}")

example.env

@@ -2,9 +2,6 @@ DISCORD_BOT_TOKEN=token
 BOT_PREFIX="g."
 PING_LINE="The Beretta fires fast and won't make you feel any better!"
 BLACKLISTED_USERS=
-cooldown=10800
-hourlyspeak=1318263176134918246
-ownerid=542701119948849163
 USERTRAIN_ENABLED="true"
 showmemenabled="true"
 NAME="an instance of goober"

modules/globalvars.py

@@ -39,5 +39,5 @@ arch = platform.machine()
 slash_commands_enabled = False
 launched = False
 latest_version = "0.0.0"
-local_version = "1.0.6"
+local_version = "1.0.7"
 os.environ['gooberlocal_version'] = local_version

modules/image.py

@@ -3,7 +3,7 @@ import re
 from PIL import Image, ImageDraw, ImageFont
 from modules.markovmemory import load_markov_model
 from modules.sentenceprocessing import improve_sentence_coherence, rephrase_for_coherence
-
+# add comments l8r
 generated_sentences = set()
 
 async def gen_image(input_image_path, sentence_size=5, max_attempts=10):

modules/sentenceprocessing.py

@@ -2,59 +2,43 @@ import re
 from modules.globalvars import *
 from modules.translations import *
 
-import nltk
-import nltk.data
+import spacy
+from spacy.tokens import Doc
+from spacytextblob.spacytextblob import SpacyTextBlob
+nlp = spacy.load("en_core_web_sm")
+nlp.add_pipe("spacytextblob")
+Doc.set_extension("polarity", getter=lambda doc: doc._.blob.polarity)
 
-# Ensure required NLTK resources are available
 def check_resources():
-    # Check for required NLTK resources and download if missing
-    resources = {
-        'vader_lexicon': 'sentiment/vader_lexicon',
-        'punkt_tab': 'tokenizers/punkt',
-    }
-    for resource, path in resources.items():
     try:
-            nltk.data.find(path) 
-            logger.info(f"{resource} is already installed.")
-        except Exception:
-            nltk.download(str(resource))
+        nlp = spacy.load("en_core_web_sm")
+    except OSError:
+        print(get_translation(LOCALE, 'spacy_model_not_found'))
+        spacy.cli.download("en_core_web_sm")
+        nlp = spacy.load("en_core_web_sm")
+    if "spacytextblob" not in nlp.pipe_names:
+        nlp.add_pipe("spacytextblob")
+    print(get_translation(LOCALE, 'spacy_initialized'))
 
 check_resources()
 
-from nltk.sentiment.vader import SentimentIntensityAnalyzer
-from nltk.tokenize import word_tokenize
-
-# Initialize the sentiment analyzer
-analyzer = SentimentIntensityAnalyzer()
-
 def is_positive(sentence):
-    """
-    Determines if the sentiment of the sentence is positive.
-    logger.infos debug information and returns True if sentiment score > 0.1.
-    """
-    scores = analyzer.polarity_scores(sentence)
-    sentiment_score = scores['compound']
+    doc = nlp(sentence)
+    sentiment_score = doc._.polarity  # from spacytextblob
 
-    # logger.info debug message with sentiment score
     debug_message = f"{DEBUG}{get_translation(LOCALE, 'sentence_positivity')} {sentiment_score}{RESET}"
-    logger.info(debug_message) 
+    print(debug_message)
 
     return sentiment_score > 0.1
 
 async def send_message(ctx, message=None, embed=None, file=None, edit=False, message_reference=None):
-    """
-    Sends or edits a message in a Discord context.
-    Handles both slash command and regular command contexts.
-    """
     if edit and message_reference:
         try:
-            # Editing the existing message
             await message_reference.edit(content=message, embed=embed)
         except Exception as e:
             await ctx.send(f"{RED}{get_translation(LOCALE, 'edit_fail')} {e}{RESET}")
     else:
         if hasattr(ctx, "respond"):
-            # For slash command contexts
             sent_message = None
             if embed:
                 sent_message = await ctx.respond(embed=embed, ephemeral=False)
@@ -63,7 +47,6 @@ async def send_message(ctx, message=None, embed=None, file=None, edit=False, mes
             if file:
                 sent_message = await ctx.respond(file=file, ephemeral=False)
         else:
-            # For regular command contexts
             sent_message = None
             if embed:
                 sent_message = await ctx.send(embed=embed)
@@ -74,34 +57,19 @@ async def send_message(ctx, message=None, embed=None, file=None, edit=False, mes
         return sent_message
 
 def append_mentions_to_18digit_integer(message):
-    """
-    Removes 18-digit integers from the message (commonly used for Discord user IDs).
-    """
     pattern = r'\b\d{18}\b'
-    return re.sub(pattern, lambda match: f"", message)
+    return re.sub(pattern, lambda match: "", message)
 
 def preprocess_message(message):
-    """
-    Preprocesses the message by removing 18-digit integers and non-alphanumeric tokens.
-    Returns the cleaned message as a string.
-    """
     message = append_mentions_to_18digit_integer(message)
-    tokens = word_tokenize(message)
-    tokens = [token for token in tokens if token.isalnum()]
+    doc = nlp(message)
+    tokens = [token.text for token in doc if token.is_alpha or token.is_digit]
     return " ".join(tokens)
 
 def improve_sentence_coherence(sentence):
-    """
-    Improves sentence coherence by capitalizing isolated 'i' pronouns.
-    """
-    sentence = sentence.replace(" i ", " I ")  
-    return sentence
+    return re.sub(r'\bi\b', 'I', sentence)
 
 def rephrase_for_coherence(sentence):
-    """
-    Rephrases the sentence for coherence by joining words with spaces.
-    (Currently a placeholder function.)
-    """
     words = sentence.split()
     coherent_sentence = " ".join(words)
     return coherent_sentence

requirements.txt

@@ -1,6 +1,7 @@
 discord.py
 markovify
-nltk
+spacy
+spacytextblob
 requests
 psutil
 better_profanity