Enhanced Monitoring and Stability

.github/pull_request_template.md

@@ -0,0 +1,5 @@
+### What is the purpose of this change?
+
+### How is this accomplished?
+
+### Anything reviews should focus on/be aware of?

.github/workflows/ci.yml

@@ -17,9 +17,15 @@ jobs:
     uses: SolaceDev/solace-public-workflows/.github/workflows/hatch_ci.yml@main
     with:
       min-python-version: "3.10"
+      whitesource_product_name: "solaceai"
+      whitesource_config_file: "wss-unified-agent.config"
+
     secrets:
       SONAR_TOKEN: ${{ secrets.SONAR_TOKEN }}
       SONAR_HOST_URL: ${{ vars.SONAR_HOST_URL }}
+      WHITESOURCE_API_KEY: ${{ secrets.WHITESOURCE_API_KEY }}
+      MANIFEST_AWS_ACCESS_KEY_ID: ${{ secrets.MANIFEST_READ_ONLY_AWS_ACCESS_KEY_ID }}
+      MANIFEST_AWS_SECRET_ACCESS_KEY: ${{ secrets.MANIFEST_READ_ONLY_AWS_SECRET_ACCESS_KEY }}
   structure-test:
     name: Test Docker Image Structure
     runs-on: ubuntu-latest

docs/components/broker_input.md

@@ -8,11 +8,6 @@ Connect to a messaging broker and receive messages from it. The component will o
 component_name: <user-supplied-name>
 component_module: broker_input
 component_config:
-  broker_type: <string>
-  broker_url: <string>
-  broker_username: <string>
-  broker_password: <string>
-  broker_vpn: <string>
   broker_queue_name: <string>
   temporary_queue: <string>
   broker_subscriptions: <string>
@@ -22,11 +17,6 @@ component_config:
 
 | Parameter | Required | Default | Description |
 | --- | --- | --- | --- |
-| broker_type | True |  | Type of broker (Solace, MQTT, etc.) |
-| broker_url | True |  | Broker URL (e.g. tcp://localhost:55555) |
-| broker_username | True |  | Client username for broker |
-| broker_password | True |  | Client password for broker |
-| broker_vpn | True |  | Client VPN for broker |
 | broker_queue_name | False |  | Queue name for broker, if not provided it will use a temporary queue |
 | temporary_queue | False | False | Whether to create a temporary queue that will be deleted after disconnection, defaulted to True if broker_queue_name is not provided |
 | broker_subscriptions | True |  | Subscriptions for broker |

docs/components/broker_output.md

@@ -8,11 +8,6 @@ Connect to a messaging broker and send messages to it. Note that this component
 component_name: <user-supplied-name>
 component_module: broker_output
 component_config:
-  broker_type: <string>
-  broker_url: <string>
-  broker_username: <string>
-  broker_password: <string>
-  broker_vpn: <string>
   payload_encoding: <string>
   payload_format: <string>
   propagate_acknowledgements: <string>
@@ -23,11 +18,6 @@ component_config:
 
 | Parameter | Required | Default | Description |
 | --- | --- | --- | --- |
-| broker_type | True |  | Type of broker (Solace, MQTT, etc.) |
-| broker_url | True |  | Broker URL (e.g. tcp://localhost:55555) |
-| broker_username | True |  | Client username for broker |
-| broker_password | True |  | Client password for broker |
-| broker_vpn | True |  | Client VPN for broker |
 | payload_encoding | False | utf-8 | Encoding for the payload (utf-8, base64, gzip, none) |
 | payload_format | False | json | Format for the payload (json, yaml, text) |
 | propagate_acknowledgements | False | True | Propagate acknowledgements from the broker to the previous components |

docs/components/broker_request_response.md

@@ -17,10 +17,17 @@ component_config:
   payload_format: <string>
   response_topic_prefix: <string>
   response_topic_suffix: <string>
+  response_topic_insertion_expression: <string>
   response_queue_prefix: <string>
+  user_properties_reply_topic_key: <string>
+  user_properties_reply_metadata_key: <string>
   request_expiry_ms: <integer>
   streaming: <string>
   streaming_complete_expression: <string>
+  streaming: <string>
+  streaming_complete_expression: <string>
+  streaming: <string>
+  streaming_complete_expression: <string>
 ```
 
 | Parameter | Required | Default | Description |
@@ -34,10 +41,17 @@ component_config:
 | payload_format | False | json | Format for the payload (json, yaml, text) |
 | response_topic_prefix | False | reply | Prefix for reply topics |
 | response_topic_suffix | False |  | Suffix for reply topics |
+| response_topic_insertion_expression | False |  | Expression to insert the reply topic into the request message. If not set, the reply topic will only be added to the request_response_metadata. The expression uses the same format as other data expressions: (e.g input.payload:myObj.replyTopic). If there is no object type in the expression, it will default to 'input.payload'. |
 | response_queue_prefix | False | reply-queue | Prefix for reply queues |
+| user_properties_reply_topic_key | False | __solace_ai_connector_broker_request_response_topic__ | Key to store the reply topic in the user properties. Start with : for nested object |
+| user_properties_reply_metadata_key | False | __solace_ai_connector_broker_request_reply_metadata__ | Key to store the reply metadata in the user properties. Start with : for nested object |
 | request_expiry_ms | False | 60000 | Expiry time for cached requests in milliseconds |
 | streaming | False |  | The response will arrive in multiple pieces. If True, the streaming_complete_expression must be set and will be used to determine when the last piece has arrived. |
 | streaming_complete_expression | False |  | The source expression to determine when the last piece of a streaming response has arrived. |
+| streaming | False |  | The response will arrive in multiple pieces. If True, the streaming_complete_expression must be set and will be used to determine when the last piece has arrived. |
+| streaming_complete_expression | False |  | The source expression to determine when the last piece of a streaming response has arrived. |
+| streaming | False |  | The response will arrive in multiple pieces. If True, the streaming_complete_expression must be set and will be used to determine when the last piece has arrived. |
+| streaming_complete_expression | False |  | The source expression to determine when the last piece of a streaming response has arrived. |
 
 
 ## Component Input Schema

docs/components/error_input.md

@@ -9,11 +9,13 @@ component_name: <user-supplied-name>
 component_module: error_input
 component_config:
   max_rate: <string>
+  max_queue_depth: <string>
 ```
 
 | Parameter | Required | Default | Description |
 | --- | --- | --- | --- |
 | max_rate | False | None | Maximum rate of errors to process per second. Any errors above this rate will be dropped. If not set, all errors will be processed. |
+| max_queue_depth | False | 1000 | Maximum number of messages that can be queued in the input queue.If the queue is full, the new message is dropped. |
 
 
 

docs/components/index.md

@@ -14,13 +14,17 @@
 | [langchain_chat_model](langchain_chat_model.md) | Provide access to all the LangChain chat models via configuration |
 | [langchain_chat_model_with_history](langchain_chat_model_with_history.md) | A chat model based on LangChain that includes keeping per-session history of the conversation. Note that this component will only take the first system message and the first human message in the messages array. |
 | [langchain_embeddings](langchain_embeddings.md) | Provide access to all the LangChain Text Embeddings components via configuration |
+| [langchain_split_text](langchain_split_text.md) | Split a long text into smaller parts using the LangChain text splitter module |
 | [langchain_vector_store_delete](langchain_vector_store_delete.md) | This component allows for entries in a LangChain Vector Store to be deleted. This is needed for the continued maintenance of the vector store. Due to the nature of langchain vector stores, you need to specify an embedding component even though it is not used in this component. |
 | [langchain_vector_store_embedding_index](langchain_vector_store_embedding_index.md) | Use LangChain Vector Stores to index text for later semantic searches. This will take text, run it through an embedding model and then store it in a vector database. |
 | [langchain_vector_store_embedding_search](langchain_vector_store_embedding_search.md) | Use LangChain Vector Stores to search a vector store with a semantic search. This will take text, run it through an embedding model with a query embedding and then find the closest matches in the store. |
 | [litellm_chat_model](litellm_chat_model.md) | LiteLLM chat component |
 | [litellm_chat_model_with_history](litellm_chat_model_with_history.md) | LiteLLM model handler component with conversation history |
 | [litellm_embeddings](litellm_embeddings.md) | Embed text using a LiteLLM model |
 | [message_filter](message_filter.md) | A filtering component. This will apply a user configurable expression. If the expression evaluates to True, the message will be passed on. If the expression evaluates to False, the message will be discarded. If the message is discarded, any previous components that require an acknowledgement will be acknowledged. |
+| [mongo_base](mongo_base.md) | Base MongoDB database component |
+| [mongo_insert](mongo_insert.md) | Inserts data into a MongoDB database. |
+| [mongo_search](mongo_search.md) | Searches a MongoDB database. |
 | [openai_chat_model](openai_chat_model.md) | OpenAI chat model component |
 | [openai_chat_model_with_history](openai_chat_model_with_history.md) | OpenAI chat model component with conversation history |
 | [parser](parser.md) | Parse input from the given type to output type. |

docs/components/langchain_chat_model.md

@@ -11,6 +11,9 @@ component_config:
   langchain_module: <string>
   langchain_class: <string>
   langchain_component_config: <string>
+  llm_mode: <string>
+  stream_to_flow: <string>
+  stream_batch_size: <string>
   llm_response_format: <string>
 ```
 
@@ -19,6 +22,9 @@ component_config:
 | langchain_module | True |  | The chat model module - e.g. 'langchain_openai.chat_models' |
 | langchain_class | True |  | The chat model class to use - e.g. ChatOpenAI |
 | langchain_component_config | True |  | Model specific configuration for the chat model. See documentation for valid parameter names. |
+| llm_mode | False |  | The mode for streaming results: 'none' or 'stream'. 'stream' will just stream the results to the named flow. 'none' will wait for the full response. |
+| stream_to_flow | False |  | Name the flow to stream the output to - this must be configured for llm_mode='stream'. |
+| stream_batch_size | False | 15 | The minimum number of words in a single streaming result. Default: 15. |
 | llm_response_format | False |  | The response format for this LLM request. This can be 'json', 'yaml', or 'text'. If set to 'json' or 'yaml', the response will be parsed by the appropriate parser and the fields will be available in the response object. If set to 'text', the response will be returned as a string. |
 
 

docs/components/langchain_chat_model_with_history.md

@@ -11,6 +11,9 @@ component_config:
   langchain_module: <string>
   langchain_class: <string>
   langchain_component_config: <string>
+  llm_mode: <string>
+  stream_to_flow: <string>
+  stream_batch_size: <string>
   llm_response_format: <string>
   history_max_turns: <string>
   history_max_message_size: <string>
@@ -27,6 +30,9 @@ component_config:
 | langchain_module | True |  | The chat model module - e.g. 'langchain_openai.chat_models' |
 | langchain_class | True |  | The chat model class to use - e.g. ChatOpenAI |
 | langchain_component_config | True |  | Model specific configuration for the chat model. See documentation for valid parameter names. |
+| llm_mode | False |  | The mode for streaming results: 'none' or 'stream'. 'stream' will just stream the results to the named flow. 'none' will wait for the full response. |
+| stream_to_flow | False |  | Name the flow to stream the output to - this must be configured for llm_mode='stream'. |
+| stream_batch_size | False | 15 | The minimum number of words in a single streaming result. Default: 15. |
 | llm_response_format | False |  | The response format for this LLM request. This can be 'json', 'yaml', or 'text'. If set to 'json' or 'yaml', the response will be parsed by the appropriate parser and the fields will be available in the response object. If set to 'text', the response will be returned as a string. |
 | history_max_turns | False | 20 | The maximum number of turns to keep in the history. If not set, the history will be limited to 20 turns. |
 | history_max_message_size | False | 1000 | The maximum amount of characters to keep in a single message in the history.  |

docs/components/langchain_split_text.md

@@ -0,0 +1,42 @@
+# LangChainTextSplitter
+
+Split a long text into smaller parts using the LangChain text splitter module
+
+## Configuration Parameters
+
+```yaml
+component_name: <user-supplied-name>
+component_module: langchain_split_text
+component_config:
+  langchain_module: <string>
+  langchain_class: <string>
+  langchain_component_config: <string>
+```
+
+| Parameter | Required | Default | Description |
+| --- | --- | --- | --- |
+| langchain_module | True |  | The text split module - e.g. 'langchain_text_splitters' |
+| langchain_class | True |  | The text split class to use - e.g. TokenTextSplitter |
+| langchain_component_config | True |  | Model specific configuration for the text splitting. See documentation for valid parameter names.https://python.langchain.com/docs/how_to/split_by_token/#nltk |
+
+
+## Component Input Schema
+
+```
+{
+  text:   <string>
+}
+```
+| Field | Required | Description |
+| --- | --- | --- |
+| text | True |  |
+
+
+## Component Output Schema
+
+```
+[
+  <string>,
+  ...
+]
+```

docs/components/litellm_chat_model.md

@@ -11,13 +11,14 @@ component_config:
   load_balancer: <string>
   embedding_params: <string>
   temperature: <string>
+  set_response_uuid_in_user_properties: <boolean>
+  timeout: <string>
+  retry_policy: <string>
+  allowed_fails_policy: <string>
   stream_to_flow: <string>
   stream_to_next_component: <string>
   llm_mode: <string>
   stream_batch_size: <string>
-  set_response_uuid_in_user_properties: <boolean>
-  history_max_turns: <string>
-  history_max_time: <string>
   history_max_turns: <string>
   history_max_time: <string>
   stream_to_flow: <string>
@@ -31,13 +32,14 @@ component_config:
 | load_balancer | False |  | Add a list of models to load balancer. |
 | embedding_params | False |  | LiteLLM model parameters. The model, api_key and base_url are mandatory.find more models at https://docs.litellm.ai/docs/providersfind more parameters at https://docs.litellm.ai/docs/completion/input |
 | temperature | False | 0.7 | Sampling temperature to use |
+| set_response_uuid_in_user_properties | False | False | Whether to set the response_uuid in the user_properties of the input_message. This will allow other components to correlate streaming chunks with the full response. |
+| timeout | False | 60 | Request timeout in seconds |
+| retry_policy | False |  | Retry policy for the load balancer. Find more at https://docs.litellm.ai/docs/routing#cooldowns |
+| allowed_fails_policy | False |  | Allowed fails policy for the load balancer. Find more at https://docs.litellm.ai/docs/routing#cooldowns |
 | stream_to_flow | False |  | Name the flow to stream the output to - this must be configured for llm_mode='stream'. This is mutually exclusive with stream_to_next_component. |
 | stream_to_next_component | False | False | Whether to stream the output to the next component in the flow. This is mutually exclusive with stream_to_flow. |
-| llm_mode | False | none | The mode for streaming results: 'sync' or 'stream'. 'stream' will just stream the results to the named flow. 'none' will wait for the full response. |
+| llm_mode | False | none | The mode for streaming results: 'none' or 'stream'. 'stream' will just stream the results to the named flow. 'none' will wait for the full response. |
 | stream_batch_size | False | 15 | The minimum number of words in a single streaming result. Default: 15. |
-| set_response_uuid_in_user_properties | False | False | Whether to set the response_uuid in the user_properties of the input_message. This will allow other components to correlate streaming chunks with the full response. |
-| history_max_turns | False | 10 | Maximum number of conversation turns to keep in history |
-| history_max_time | False | 3600 | Maximum time to keep conversation history (in seconds) |
 | history_max_turns | False | 10 | Maximum number of conversation turns to keep in history |
 | history_max_time | False | 3600 | Maximum time to keep conversation history (in seconds) |
 | stream_to_flow | False |  | Name the flow to stream the output to - this must be configured for llm_mode='stream'. This is mutually exclusive with stream_to_next_component. |
@@ -57,6 +59,7 @@ component_config:
     },
     ...
   ],
+  stream:   <boolean>,
   clear_history_but_keep_depth:   <integer>
 }
 ```
@@ -65,6 +68,7 @@ component_config:
 | messages | True |  |
 | messages[].role | True |  |
 | messages[].content | True |  |
+| stream | False | Whether to stream the response - overwrites llm_mode |
 | clear_history_but_keep_depth | False | Clear history but keep the last N messages. If 0, clear all history. If not set, do not clear history. |
 
 

docs/components/litellm_chat_model_with_history.md

@@ -11,13 +11,14 @@ component_config:
   load_balancer: <string>
   embedding_params: <string>
   temperature: <string>
+  set_response_uuid_in_user_properties: <boolean>
+  timeout: <string>
+  retry_policy: <string>
+  allowed_fails_policy: <string>
   stream_to_flow: <string>
   stream_to_next_component: <string>
   llm_mode: <string>
   stream_batch_size: <string>
-  set_response_uuid_in_user_properties: <boolean>
-  history_max_turns: <string>
-  history_max_time: <string>
   history_max_turns: <string>
   history_max_time: <string>
 ```
@@ -27,13 +28,14 @@ component_config:
 | load_balancer | False |  | Add a list of models to load balancer. |
 | embedding_params | False |  | LiteLLM model parameters. The model, api_key and base_url are mandatory.find more models at https://docs.litellm.ai/docs/providersfind more parameters at https://docs.litellm.ai/docs/completion/input |
 | temperature | False | 0.7 | Sampling temperature to use |
+| set_response_uuid_in_user_properties | False | False | Whether to set the response_uuid in the user_properties of the input_message. This will allow other components to correlate streaming chunks with the full response. |
+| timeout | False | 60 | Request timeout in seconds |
+| retry_policy | False |  | Retry policy for the load balancer. Find more at https://docs.litellm.ai/docs/routing#cooldowns |
+| allowed_fails_policy | False |  | Allowed fails policy for the load balancer. Find more at https://docs.litellm.ai/docs/routing#cooldowns |
 | stream_to_flow | False |  | Name the flow to stream the output to - this must be configured for llm_mode='stream'. This is mutually exclusive with stream_to_next_component. |
 | stream_to_next_component | False | False | Whether to stream the output to the next component in the flow. This is mutually exclusive with stream_to_flow. |
-| llm_mode | False | none | The mode for streaming results: 'sync' or 'stream'. 'stream' will just stream the results to the named flow. 'none' will wait for the full response. |
+| llm_mode | False | none | The mode for streaming results: 'none' or 'stream'. 'stream' will just stream the results to the named flow. 'none' will wait for the full response. |
 | stream_batch_size | False | 15 | The minimum number of words in a single streaming result. Default: 15. |
-| set_response_uuid_in_user_properties | False | False | Whether to set the response_uuid in the user_properties of the input_message. This will allow other components to correlate streaming chunks with the full response. |
-| history_max_turns | False | 10 | Maximum number of conversation turns to keep in history |
-| history_max_time | False | 3600 | Maximum time to keep conversation history (in seconds) |
 | history_max_turns | False | 10 | Maximum number of conversation turns to keep in history |
 | history_max_time | False | 3600 | Maximum time to keep conversation history (in seconds) |
 
@@ -49,6 +51,7 @@ component_config:
     },
     ...
   ],
+  stream:   <boolean>,
   clear_history_but_keep_depth:   <integer>
 }
 ```
@@ -57,6 +60,7 @@ component_config:
 | messages | True |  |
 | messages[].role | True |  |
 | messages[].content | True |  |
+| stream | False | Whether to stream the response - overwrites llm_mode |
 | clear_history_but_keep_depth | False | Clear history but keep the last N messages. If 0, clear all history. If not set, do not clear history. |