ChatterBot 1.2.10__tar.gz → 1.2.12__tar.gz

{chatterbot-1.2.10 → chatterbot-1.2.12}/ChatterBot.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ChatterBot
-Version: 1.2.10
+Version: 1.2.12
 Summary: ChatterBot is a machine learning, conversational dialog engine
 Author: Gunther Cox
 License-Expression: BSD-3-Clause
@@ -39,7 +39,7 @@ Requires-Dist: tqdm
 Provides-Extra: test
 Requires-Dist: flake8; extra == "test"
 Requires-Dist: coverage; extra == "test"
-Requires-Dist: sphinx<8.2,>=5.3; extra == "test"
+Requires-Dist: sphinx<9.2,>=5.3; extra == "test"
 Requires-Dist: sphinx-sitemap>=2.6.0; extra == "test"
 Requires-Dist: huggingface_hub; extra == "test"
 Requires-Dist: django<6.0,<=4.1; extra == "test"
@@ -50,13 +50,13 @@ Requires-Dist: chatterbot-corpus<1.3.0,>=1.2.2; extra == "dev"
 Requires-Dist: ollama<1.0,>=0.6.0; extra == "dev"
 Requires-Dist: openai; extra == "dev"
 Provides-Extra: redis
-Requires-Dist: redis[hiredis]<7.0; extra == "redis"
-Requires-Dist: langchain-redis<=0.2.5; extra == "redis"
-Requires-Dist: langchain-huggingface<=0.1.2; extra == "redis"
-Requires-Dist: accelerate<=1.6.0; extra == "redis"
-Requires-Dist: sentence-transformers<=4.0.2; extra == "redis"
+Requires-Dist: redis[hiredis]<7.2,>=7.0; extra == "redis"
+Requires-Dist: langchain-redis<0.3.0; extra == "redis"
+Requires-Dist: langchain-huggingface<1.3.0,>=0.1.2; extra == "redis"
+Requires-Dist: accelerate<1.13,>=1.6.0; extra == "redis"
+Requires-Dist: sentence-transformers<5.3.0,>=4.0.2; extra == "redis"
 Provides-Extra: mongodb
-Requires-Dist: pymongo<4.16,>=4.11; extra == "mongodb"
+Requires-Dist: pymongo<4.17,>=4.11; extra == "mongodb"
 Dynamic: license-file
 
 ![ChatterBot: Machine learning in Python](https://i.imgur.com/b3SCmGT.png)
@@ -162,9 +162,9 @@ https://docs.chatterbot.us/contributing/
 
 ChatterBot is sponsored by:
 
-<p style="font-size:21px; color:black;">Browser testing via 
+<p>
    <a href="https://www.lambdatest.com/?utm_source=chatterbot&utm_medium=sponsor" target="_blank">
-      <img src="https://www.lambdatest.com/blue-logo.png" style="vertical-align: middle;" width="250" height="45" />
+      <img src="docs/_static/testmu-ai-white-logo.png" style="vertical-align: middle;" width="250" height="80" />
    </a>
 </p>
 

{chatterbot-1.2.10 → chatterbot-1.2.12}/ChatterBot.egg-info/SOURCES.txt

@@ -56,6 +56,7 @@ chatterbot/ext/django_chatterbot/migrations/0017_tags_unique.py
 chatterbot/ext/django_chatterbot/migrations/0018_text_max_length.py
 chatterbot/ext/django_chatterbot/migrations/0019_alter_statement_id_alter_tag_id_and_more.py
 chatterbot/ext/django_chatterbot/migrations/0020_alter_statement_conversation_and_more.py
+chatterbot/ext/django_chatterbot/migrations/0021_increase_text_max_length_to_1100.py
 chatterbot/ext/django_chatterbot/migrations/__init__.py
 chatterbot/ext/sqlalchemy_app/__init__.py
 chatterbot/ext/sqlalchemy_app/models.py

{chatterbot-1.2.10 → chatterbot-1.2.12}/ChatterBot.egg-info/requires.txt

@@ -12,19 +12,19 @@ ollama<1.0,>=0.6.0
 openai
 
 [mongodb]
-pymongo<4.16,>=4.11
+pymongo<4.17,>=4.11
 
 [redis]
-redis[hiredis]<7.0
-langchain-redis<=0.2.5
-langchain-huggingface<=0.1.2
-accelerate<=1.6.0
-sentence-transformers<=4.0.2
+redis[hiredis]<7.2,>=7.0
+langchain-redis<0.3.0
+langchain-huggingface<1.3.0,>=0.1.2
+accelerate<1.13,>=1.6.0
+sentence-transformers<5.3.0,>=4.0.2
 
 [test]
 flake8
 coverage
-sphinx<8.2,>=5.3
+sphinx<9.2,>=5.3
 sphinx-sitemap>=2.6.0
 huggingface_hub
 django<6.0,<=4.1

{chatterbot-1.2.10 → chatterbot-1.2.12}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ChatterBot
-Version: 1.2.10
+Version: 1.2.12
 Summary: ChatterBot is a machine learning, conversational dialog engine
 Author: Gunther Cox
 License-Expression: BSD-3-Clause
@@ -39,7 +39,7 @@ Requires-Dist: tqdm
 Provides-Extra: test
 Requires-Dist: flake8; extra == "test"
 Requires-Dist: coverage; extra == "test"
-Requires-Dist: sphinx<8.2,>=5.3; extra == "test"
+Requires-Dist: sphinx<9.2,>=5.3; extra == "test"
 Requires-Dist: sphinx-sitemap>=2.6.0; extra == "test"
 Requires-Dist: huggingface_hub; extra == "test"
 Requires-Dist: django<6.0,<=4.1; extra == "test"
@@ -50,13 +50,13 @@ Requires-Dist: chatterbot-corpus<1.3.0,>=1.2.2; extra == "dev"
 Requires-Dist: ollama<1.0,>=0.6.0; extra == "dev"
 Requires-Dist: openai; extra == "dev"
 Provides-Extra: redis
-Requires-Dist: redis[hiredis]<7.0; extra == "redis"
-Requires-Dist: langchain-redis<=0.2.5; extra == "redis"
-Requires-Dist: langchain-huggingface<=0.1.2; extra == "redis"
-Requires-Dist: accelerate<=1.6.0; extra == "redis"
-Requires-Dist: sentence-transformers<=4.0.2; extra == "redis"
+Requires-Dist: redis[hiredis]<7.2,>=7.0; extra == "redis"
+Requires-Dist: langchain-redis<0.3.0; extra == "redis"
+Requires-Dist: langchain-huggingface<1.3.0,>=0.1.2; extra == "redis"
+Requires-Dist: accelerate<1.13,>=1.6.0; extra == "redis"
+Requires-Dist: sentence-transformers<5.3.0,>=4.0.2; extra == "redis"
 Provides-Extra: mongodb
-Requires-Dist: pymongo<4.16,>=4.11; extra == "mongodb"
+Requires-Dist: pymongo<4.17,>=4.11; extra == "mongodb"
 Dynamic: license-file
 
 ![ChatterBot: Machine learning in Python](https://i.imgur.com/b3SCmGT.png)
@@ -162,9 +162,9 @@ https://docs.chatterbot.us/contributing/
 
 ChatterBot is sponsored by:
 
-<p style="font-size:21px; color:black;">Browser testing via 
+<p>
    <a href="https://www.lambdatest.com/?utm_source=chatterbot&utm_medium=sponsor" target="_blank">
-      <img src="https://www.lambdatest.com/blue-logo.png" style="vertical-align: middle;" width="250" height="45" />
+      <img src="docs/_static/testmu-ai-white-logo.png" style="vertical-align: middle;" width="250" height="80" />
    </a>
 </p>
 

{chatterbot-1.2.10 → chatterbot-1.2.12}/README.md

@@ -101,9 +101,9 @@ https://docs.chatterbot.us/contributing/
 
 ChatterBot is sponsored by:
 
-<p style="font-size:21px; color:black;">Browser testing via 
+<p>
    <a href="https://www.lambdatest.com/?utm_source=chatterbot&utm_medium=sponsor" target="_blank">
-      <img src="https://www.lambdatest.com/blue-logo.png" style="vertical-align: middle;" width="250" height="45" />
+      <img src="docs/_static/testmu-ai-white-logo.png" style="vertical-align: middle;" width="250" height="80" />
    </a>
 </p>
 

{chatterbot-1.2.10 → chatterbot-1.2.12}/chatterbot/__init__.py

@@ -4,7 +4,7 @@ ChatterBot is a machine learning, conversational dialog engine.
 from .chatterbot import ChatBot
 
 
-__version__ = '1.2.10'
+__version__ = '1.2.12'
 
 __all__ = (
     'ChatBot',

{chatterbot-1.2.10 → chatterbot-1.2.12}/chatterbot/constants.py

@@ -5,11 +5,11 @@ from chatterbot import languages
 
 '''
 The maximum length of characters that the text of a statement can contain.
-The number 255 is used because that is the maximum length of a char field
-in most databases. This value should be enforced on a per-model basis by
-the data model for each storage adapter.
+The number 1100 is used to support longer conversational statements while
+remaining within VARCHAR limits for most databases. This value should be
+enforced on a per-model basis by the data model for each storage adapter.
 '''
-STATEMENT_TEXT_MAX_LENGTH = 255
+STATEMENT_TEXT_MAX_LENGTH = 1100
 
 '''
 The maximum length of characters that the text label of a conversation can contain.

chatterbot-1.2.12/chatterbot/ext/django_chatterbot/migrations/0021_increase_text_max_length_to_1100.py

@@ -0,0 +1,55 @@
+"""
+Django migration to increase text field max_length from 255 to 1100.
+
+This migration alters all text-related fields in the Statement model:
+- text
+- search_text
+- in_response_to
+- search_in_response_to
+
+This change supports longer conversational statements while remaining
+within VARCHAR limits for most databases.
+"""
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('django_chatterbot', '0020_alter_statement_conversation_and_more'),
+    ]
+
+    operations = [
+        migrations.AlterField(
+            model_name='statement',
+            name='text',
+            field=models.CharField(max_length=1100, help_text='The text of the statement.'),
+        ),
+        migrations.AlterField(
+            model_name='statement',
+            name='search_text',
+            field=models.CharField(
+                blank=True,
+                max_length=1100,
+                help_text='A modified version of the statement text optimized for searching.'
+            ),
+        ),
+        migrations.AlterField(
+            model_name='statement',
+            name='in_response_to',
+            field=models.CharField(
+                max_length=1100,
+                null=True,
+                help_text='The text of the statement that this statement is in response to.'
+            ),
+        ),
+        migrations.AlterField(
+            model_name='statement',
+            name='search_in_response_to',
+            field=models.CharField(
+                blank=True,
+                max_length=1100,
+                help_text='A modified version of the in_response_to text optimized for searching.'
+            ),
+        ),
+    ]

{chatterbot-1.2.10 → chatterbot-1.2.12}/chatterbot/storage/redis.py

@@ -65,6 +65,7 @@ class RedisVectorStorageAdapter(StorageAdapter):
             index_name='chatterbot',
             redis_url=self.database_uri,
             content_field='in_response_to',
+            legacy_key_format=False,
             metadata_schema=[
                 {
                     'name': 'conversation',
@@ -212,12 +213,16 @@ class RedisVectorStorageAdapter(StorageAdapter):
             - search_in_response_to_contains
             - order_by
         """
-        from redisvl.query import VectorQuery
         from redisvl.query.filter import Tag, Text
 
         # https://redis.io/docs/latest/develop/interact/search-and-query/advanced-concepts/query_syntax/
         filter_condition = None
 
+        ordering = kwargs.get('order_by', None)
+
+        if ordering:
+            ordering = ','.join(ordering)
+
         if 'in_response_to' in kwargs:
             filter_condition = Text('in_response_to') == kwargs['in_response_to']
 
@@ -255,7 +260,7 @@ class RedisVectorStorageAdapter(StorageAdapter):
             _query = '|'.join([
                 f'%%{text}%%' for text in kwargs['exclude_text_words']
             ])
-            query = Text('text') % f'-({ _query })'
+            query = Text('text') % f'-({_query})'
             if filter_condition:
                 filter_condition &= query
             else:
@@ -290,67 +295,20 @@ class RedisVectorStorageAdapter(StorageAdapter):
             # similar responses.
             _search_query = kwargs['search_text_contains']
 
-            # Use vector similarity to find statements responding to similar contexts
-            embedding = self.vector_store.embeddings.embed_query(_search_query)
-
-            return_fields = [
-                'text', 'in_response_to', 'conversation', 'persona', 'tags', 'created_at'
-            ]
-
-            query = VectorQuery(
-                vector=embedding,
-                vector_field_name='embedding',
-                return_fields=return_fields,
-                num_results=page_size,
-                filter_expression=filter_condition
+            documents = self.vector_store.similarity_search(
+                _search_query,
+                k=page_size,  # The number of results to return
+                return_all=True,  # Include the full document with IDs
+                filter=filter_condition,
+                sort_by=ordering
             )
 
-            results = self.vector_store.index.query(query)
-
-            Document = self.get_statement_model()
-            documents = []
-
-            # Calculate confidence from vector distances
+            # Add confidence scores based on similarity ordering
             # Results are ordered by similarity (best match first)
-            for idx, result in enumerate(results):
-                in_response_to = result.get('in_response_to', '')
-
-                # Redis vector_score is cosine distance (lower is better)
-                # Convert to confidence: confidence = 1 - distance
-                # If vector_score not available, use result order
-                vector_score = result.get('vector_score')
-                if vector_score is not None:
-                    # Cosine distance ranges from 0 (identical) to 2 (opposite)
-                    # Normalize to confidence: 1.0 (identical) to 0.0 (opposite)
-                    confidence = max(0.0, 1.0 - (float(vector_score) / 2.0))
-                else:
-                    # Fallback: use result order (first result = highest confidence)
-                    # Start at 0.95 for first result, decay by 0.05 per position
-                    confidence = max(0.0, 0.95 - (idx * 0.05))
-
-                # Parse timestamp
-                created_at_value = result.get('created_at', 0)
-                if isinstance(created_at_value, str):
-                    created_at = datetime.fromtimestamp(float(created_at_value))
-                elif created_at_value:
-                    created_at = datetime.fromtimestamp(float(created_at_value))
-                else:
-                    created_at = datetime.now()
-
-                metadata = {
-                    'text': result.get('text', ''),
-                    'conversation': result.get('conversation', ''),
-                    'persona': result.get('persona', ''),
-                    'tags': result.get('tags', ''),
-                    'created_at': created_at,
-                    'confidence': confidence,
-                }
-                doc = Document(
-                    page_content=in_response_to,
-                    metadata=metadata,
-                    id=result['id']
-                )
-                documents.append(doc)
+            for idx, doc in enumerate(documents):
+                # Start at 0.95 for first result, decay by 0.05 per position
+                confidence = max(0.0, 0.95 - (idx * 0.05))
+                doc.metadata['confidence'] = confidence
 
             return [self.model_to_object(document) for document in documents]
 
@@ -379,74 +337,20 @@ class RedisVectorStorageAdapter(StorageAdapter):
         if 'search_in_response_to_contains' in kwargs:
             _search_text = kwargs.get('search_in_response_to_contains', '')
 
-            # Get embedding for the search text
-            embedding = self.vector_store.embeddings.embed_query(_search_text)
-
-            # Build return fields from metadata schema
-            return_fields = [
-                'text', 'in_response_to', 'conversation', 'persona', 'tags', 'created_at'
-            ]
-
-            # Use direct index query via RedisVL
-            # langchain's similarity_search has issues with filters in v0.2.4
-            # and may not work properly with existing indexes
-            # TODO: Look into similarity_search_with_score implementation
-            query = VectorQuery(
-                vector=embedding,
-                vector_field_name='embedding',
-                return_fields=return_fields,
-                num_results=page_size,
-                filter_expression=filter_condition
+            documents = self.vector_store.similarity_search(
+                _search_text,
+                k=page_size,  # The number of results to return
+                return_all=True,  # Include the full document with IDs
+                filter=filter_condition,
+                sort_by=ordering
             )
 
-            # Execute query
-            results = self.vector_store.index.query(query)
-
-            # Convert results to Document objects
-            Document = self.get_statement_model()
-            documents = []
-
-            # Calculate confidence from vector distances
+            # Add confidence scores based on similarity ordering
             # Results are ordered by similarity (best match first)
-            for idx, result in enumerate(results):
-                # Extract metadata and content
-                in_response_to = result.get('in_response_to', '')
-
-                # Redis vector_score is cosine distance (lower is better)
-                # Convert to confidence: confidence = 1 - distance
-                # If vector_score not available, use result order
-                vector_score = result.get('vector_score')
-                if vector_score is not None:
-                    # Cosine distance ranges from 0 (identical) to 2 (opposite)
-                    # Normalize to confidence: 1.0 (identical) to 0.0 (opposite)
-                    confidence = max(0.0, 1.0 - (float(vector_score) / 2.0))
-                else:
-                    # Fallback: use result order (first result = highest confidence)
-                    # Start at 0.95 for first result, decay by 0.05 per position
-                    confidence = max(0.0, 0.95 - (idx * 0.05))
-
-                # Convert Unix timestamp back to datetime
-                # Redis returns numeric fields as strings
-                created_at_timestamp = result.get('created_at', '0')
-                if created_at_timestamp and created_at_timestamp != '0':
-                    created_at = datetime.fromtimestamp(float(created_at_timestamp))
-                else:
-                    created_at = datetime.now()
-
-                metadata = {
-                    'text': result.get('text', ''),
-                    'conversation': result.get('conversation', ''),
-                    'persona': result.get('persona', ''),
-                    'tags': result.get('tags', ''),
-                    'created_at': created_at,
-                    'confidence': confidence,
-                }
-                doc = Document(
-                    page_content=in_response_to,
-                    metadata=metadata,
-                    id=result['id']
-                )
-                documents.append(doc)
+            for idx, doc in enumerate(documents):
+                # Start at 0.95 for first result, decay by 0.05 per position
+                confidence = max(0.0, 0.95 - (idx * 0.05))
+                doc.metadata['confidence'] = confidence
         else:
             documents = self.vector_store.query_search(
                 k=page_size,
@@ -551,11 +455,8 @@ class RedisVectorStorageAdapter(StorageAdapter):
             client = self.vector_store.index.client
             client.delete(statement.id)
 
-            # NOTE: langchain-redis has an inconsistency - it uses :: for auto-generated
-            # IDs but : (single colon) when keys are explicitly provided
-            if '::' in statement.id:
-                key = statement.id.split('::', 1)[1]
-            elif ':' in statement.id:
+            # Extract the key from the full ID (format: prefix:key)
+            if ':' in statement.id:
                 key = statement.id.split(':', 1)[1]
             else:
                 # If no delimiter found, use the entire ID as the key
@@ -564,13 +465,6 @@ class RedisVectorStorageAdapter(StorageAdapter):
             ids = self.vector_store.add_texts(
                 [document.page_content], [metadata], keys=[key]
             )
-
-            # Normalize the ID to use :: delimiter (if langchain-redis returned single colon)
-            if ids and ':' in ids[0] and '::' not in ids[0]:
-                # Replace first occurrence of single colon with double colon
-                normalized_id = ids[0].replace(':', '::', 1)
-                # Update the key in Redis to use the correct format
-                client.rename(ids[0], normalized_id)
         else:
             self.vector_store.add_documents([document])
 

{chatterbot-1.2.10 → chatterbot-1.2.12}/chatterbot/storage/sql_storage.py

@@ -23,7 +23,7 @@ class SQLStorageAdapter(StorageAdapter):
         from sqlalchemy import create_engine, inspect, event
         from sqlalchemy import Index
         from sqlalchemy.engine import Engine
-        from sqlalchemy.orm import sessionmaker
+        from sqlalchemy.orm import sessionmaker, scoped_session
 
         self.database_uri = kwargs.get('database_uri', False)
 
@@ -35,7 +35,10 @@ class SQLStorageAdapter(StorageAdapter):
         if not self.database_uri:
             self.database_uri = 'sqlite:///db.sqlite3'
 
-        self.engine = create_engine(self.database_uri)
+        # Configure connection pool with safe defaults to prevent exhaustion
+        # Note: SQLite uses SingletonThreadPool which doesn't support these params
+        # PostgreSQL, MySQL, etc. use QueuePool which does support them
+        pool_config = {}
 
         if self.database_uri.startswith('sqlite://'):
 
@@ -66,6 +69,23 @@ class SQLStorageAdapter(StorageAdapter):
                 cursor.execute('PRAGMA synchronous=NORMAL')
                 cursor.close()
 
+        else:
+            # Only apply pool configuration for databases that support QueuePool
+            # pool_size: Maximum persistent connections (10)
+            # max_overflow: Additional connections during peak load (20)
+            # pool_timeout: Seconds to wait for connection before error (30)
+            # pool_recycle: Recycle connections after 1 hour to prevent stale connections
+            # pool_pre_ping: Test connections before using to detect disconnects
+            pool_config = {
+                'pool_size': kwargs.get('pool_size', 10),
+                'max_overflow': kwargs.get('max_overflow', 20),
+                'pool_timeout': kwargs.get('pool_timeout', 30),
+                'pool_recycle': kwargs.get('pool_recycle', 3600),
+                'pool_pre_ping': kwargs.get('pool_pre_ping', True),
+            }
+
+        self.engine = create_engine(self.database_uri, **pool_config)
+
         if not inspect(self.engine).has_table('statement'):
             self.create_database()
 
@@ -91,7 +111,10 @@ class SQLStorageAdapter(StorageAdapter):
 
             search_in_response_to_index.create(bind=self.engine)
 
-        self.Session = sessionmaker(bind=self.engine, expire_on_commit=True)
+        # Use a scoped session for thread-safe session management
+        # This provides thread-local session storage to prevent session sharing across threads
+        session_factory = sessionmaker(bind=self.engine, expire_on_commit=True)
+        self.Session = scoped_session(session_factory)
 
     def get_statement_model(self):
         """
@@ -119,9 +142,11 @@ class SQLStorageAdapter(StorageAdapter):
         Statement = self.get_model('statement')
 
         session = self.Session()
-        statement_count = session.query(Statement).count()
-        session.close()
-        return statement_count
+        try:
+            statement_count = session.query(Statement).count()
+            return statement_count
+        finally:
+            session.close()
 
     def remove(self, statement_text):
         """
@@ -131,13 +156,14 @@ class SQLStorageAdapter(StorageAdapter):
         """
         Statement = self.get_model('statement')
         session = self.Session()
+        try:
+            query = session.query(Statement).filter_by(text=statement_text)
+            record = query.first()
 
-        query = session.query(Statement).filter_by(text=statement_text)
-        record = query.first()
-
-        session.delete(record)
-        session.commit()
-        session.close()
+            session.delete(record)
+            session.commit()
+        finally:
+            session.close()
 
     def filter(self, **kwargs):
         """
@@ -152,8 +178,6 @@ class SQLStorageAdapter(StorageAdapter):
         Statement = self.get_model('statement')
         Tag = self.get_model('tag')
 
-        session = self.Session()
-
         page_size = kwargs.pop('page_size', 1000)
         order_by = kwargs.pop('order_by', None)
         tags = kwargs.pop('tags', [])
@@ -167,65 +191,69 @@ class SQLStorageAdapter(StorageAdapter):
         if isinstance(tags, str):
             tags = [tags]
 
-        if len(kwargs) == 0:
-            statements = session.query(Statement).filter()
-        else:
-            statements = session.query(Statement).filter_by(**kwargs)
-
-        if tags:
-            statements = statements.join(Statement.tags).filter(
-                Tag.name.in_(tags)
-            )
-
-        if exclude_text:
-            statements = statements.filter(
-                ~Statement.text.in_(exclude_text)
-            )
+        # Use context manager to ensure session cleanup even if generator is partially consumed
+        session = self.Session()
+        try:
+            if len(kwargs) == 0:
+                statements = session.query(Statement).filter()
+            else:
+                statements = session.query(Statement).filter_by(**kwargs)
+
+            if tags:
+                statements = statements.join(Statement.tags).filter(
+                    Tag.name.in_(tags)
+                )
 
-        if exclude_text_words:
-            or_word_query = [
-                Statement.text.ilike('%' + word + '%') for word in exclude_text_words
-            ]
-            statements = statements.filter(
-                ~or_(*or_word_query)
-            )
+            if exclude_text:
+                statements = statements.filter(
+                    ~Statement.text.in_(exclude_text)
+                )
 
-        if persona_not_startswith:
-            statements = statements.filter(
-                ~Statement.persona.startswith('bot:')
-            )
+            if exclude_text_words:
+                or_word_query = [
+                    Statement.text.ilike('%' + word + '%') for word in exclude_text_words
+                ]
+                statements = statements.filter(
+                    ~or_(*or_word_query)
+                )
 
-        if search_text_contains:
-            or_query = [
-                Statement.search_text.contains(word) for word in search_text_contains.split(' ')
-            ]
-            statements = statements.filter(
-                or_(*or_query)
-            )
+            if persona_not_startswith:
+                statements = statements.filter(
+                    ~Statement.persona.startswith('bot:')
+                )
 
-        if search_in_response_to_contains:
-            or_query = [
-                Statement.search_in_response_to.contains(word) for word in search_in_response_to_contains.split(' ')
-            ]
-            statements = statements.filter(
-                or_(*or_query)
-            )
+            if search_text_contains:
+                or_query = [
+                    Statement.search_text.contains(word) for word in search_text_contains.split(' ')
+                ]
+                statements = statements.filter(
+                    or_(*or_query)
+                )
 
-        if order_by:
+            if search_in_response_to_contains:
+                or_query = [
+                    Statement.search_in_response_to.contains(word) for word in search_in_response_to_contains.split(' ')
+                ]
+                statements = statements.filter(
+                    or_(*or_query)
+                )
 
-            if 'created_at' in order_by:
-                index = order_by.index('created_at')
-                order_by[index] = Statement.created_at.asc()
+            if order_by:
 
-            statements = statements.order_by(*order_by)
+                if 'created_at' in order_by:
+                    index = order_by.index('created_at')
+                    order_by[index] = Statement.created_at.asc()
 
-        total_statements = statements.count()
+                statements = statements.order_by(*order_by)
 
-        for start_index in range(0, total_statements, page_size):
-            for statement in statements.slice(start_index, start_index + page_size):
-                yield self.model_to_object(statement)
+            total_statements = statements.count()
 
-        session.close()
+            for start_index in range(0, total_statements, page_size):
+                for statement in statements.slice(start_index, start_index + page_size):
+                    yield self.model_to_object(statement)
+        finally:
+            # Always close session, even if generator is abandoned or exception occurs
+            session.close()
 
     def create(
         self,
@@ -336,8 +364,11 @@ class SQLStorageAdapter(StorageAdapter):
                 statement_model_object.tags.append(tag)
             create_statements.append(statement_model_object)
 
-        session.add_all(create_statements)
-        session.commit()
+        try:
+            session.add_all(create_statements)
+            session.commit()
+        finally:
+            session.close()
 
     def update(self, statement):
         """
@@ -348,49 +379,51 @@ class SQLStorageAdapter(StorageAdapter):
         Tag = self.get_model('tag')
 
         session = self.Session()
-        record = None
+        try:
+            record = None
 
-        if hasattr(statement, 'id') and statement.id is not None:
-            record = session.get(Statement, statement.id)
-        else:
-            record = session.query(Statement).filter(
-                Statement.text == statement.text,
-                Statement.conversation == statement.conversation,
-            ).first()
-
-            # Create a new statement entry if one does not already exist
-            if not record:
-                record = Statement(
-                    text=statement.text,
-                    conversation=statement.conversation,
-                    persona=statement.persona
-                )
+            if hasattr(statement, 'id') and statement.id is not None:
+                record = session.get(Statement, statement.id)
+            else:
+                record = session.query(Statement).filter(
+                    Statement.text == statement.text,
+                    Statement.conversation == statement.conversation,
+                ).first()
 
-        # Update the response value
-        record.in_response_to = statement.in_response_to
+                # Create a new statement entry if one does not already exist
+                if not record:
+                    record = Statement(
+                        text=statement.text,
+                        conversation=statement.conversation,
+                        persona=statement.persona
+                    )
 
-        record.created_at = statement.created_at
+            # Update the response value
+            record.in_response_to = statement.in_response_to
 
-        if not statement.search_text:
-            if self.raise_on_missing_search_text:
-                raise Exception('update issued without search_text value')
+            record.created_at = statement.created_at
 
-        if statement.in_response_to and not statement.search_in_response_to:
-            if self.raise_on_missing_search_text:
-                raise Exception('update issued without search_in_response_to value')
+            if not statement.search_text:
+                if self.raise_on_missing_search_text:
+                    raise Exception('update issued without search_text value')
 
-        for tag_name in statement.get_tags():
-            tag = session.query(Tag).filter_by(name=tag_name).first()
+            if statement.in_response_to and not statement.search_in_response_to:
+                if self.raise_on_missing_search_text:
+                    raise Exception('update issued without search_in_response_to value')
 
-            if not tag:
-                # Create the record
-                tag = Tag(name=tag_name)
+            for tag_name in statement.get_tags():
+                tag = session.query(Tag).filter_by(name=tag_name).first()
 
-            record.tags.append(tag)
+                if not tag:
+                    # Create the record
+                    tag = Tag(name=tag_name)
 
-        session.add(record)
-        session.commit()
-        session.close()
+                record.tags.append(tag)
+
+            session.add(record)
+            session.commit()
+        finally:
+            session.close()
 
     def get_random(self):
         """
@@ -399,17 +432,19 @@ class SQLStorageAdapter(StorageAdapter):
         Statement = self.get_model('statement')
 
         session = self.Session()
-        count = self.count()
-        if count < 1:
-            raise self.EmptyDatabaseException()
+        try:
+            count = self.count()
+            if count < 1:
+                raise self.EmptyDatabaseException()
 
-        random_index = random.randrange(0, count)
-        random_statement = session.query(Statement)[random_index]
+            random_index = random.randrange(0, count)
+            random_statement = session.query(Statement)[random_index]
 
-        statement = self.model_to_object(random_statement)
+            statement = self.model_to_object(random_statement)
 
-        session.close()
-        return statement
+            return statement
+        finally:
+            session.close()
 
     def drop(self):
         """
@@ -419,12 +454,13 @@ class SQLStorageAdapter(StorageAdapter):
         Tag = self.get_model('tag')
 
         session = self.Session()
+        try:
+            session.query(Statement).delete()
+            session.query(Tag).delete()
 
-        session.query(Statement).delete()
-        session.query(Tag).delete()
-
-        session.commit()
-        session.close()
+            session.commit()
+        finally:
+            session.close()
 
     def create_database(self):
         """
@@ -438,5 +474,10 @@ class SQLStorageAdapter(StorageAdapter):
         Close the database connection and dispose of the engine.
         This ensures proper cleanup of resources.
         """
+        # Remove thread-local sessions from scoped_session registry
+        if hasattr(self, 'Session'):
+            self.Session.remove()
+
+        # Dispose of the connection pool
         if hasattr(self, 'engine'):
             self.engine.dispose()

{chatterbot-1.2.10 → chatterbot-1.2.12}/pyproject.toml

@@ -71,7 +71,7 @@ dependencies = [
 test = [
     "flake8",
     "coverage",
-    "sphinx>=5.3,<8.2",
+    "sphinx>=5.3,<9.2",
     "sphinx-sitemap>=2.6.0",
     "huggingface_hub",
     "django<=4.1,<6.0"
@@ -84,12 +84,12 @@ dev = [
     "openai"
 ]
 redis = [
-    "redis[hiredis]<7.0",
-    "langchain-redis<=0.2.5",
-    "langchain-huggingface<=0.1.2",
-    "accelerate<=1.6.0",
-    "sentence-transformers<=4.0.2",
+    "redis[hiredis]>=7.0,<7.2",
+    "langchain-redis<0.3.0",
+    "langchain-huggingface>=0.1.2,<1.3.0",
+    "accelerate>=1.6.0,<1.13",
+    "sentence-transformers>=4.0.2,<5.3.0",
 ]
 mongodb = [
-    "pymongo>=4.11,<4.16",
+    "pymongo>=4.11,<4.17",
 ]