Documentation ¶
Overview ¶
Package schema implements a shared core set of data types for use in langchaingo.
The primary interface through which end users interact with LLMs is a chat interface. For this reason, some model providers have started providing access to the underlying API in a way that expects chat messages. These messages have a content field (which is usually text) and are associated with a user (or role). Right now the supported users are System, Human, AI, and a generic/arbitrary user.
Index ¶
- Variables
- func GetBufferString(messages []ChatMessage, humanPrefix string, aiPrefix string) (string, error)
- type AIChatMessage
- type AgentAction
- type AgentFinish
- type AgentStep
- type ChatMessage
- type ChatMessageHistory
- type ChatMessageType
- type Document
- type FunctionCall
- type FunctionChatMessage
- type GenericChatMessage
- type HumanChatMessage
- type Memory
- type Named
- type OutputParser
- type PromptValue
- type Retriever
- type SystemChatMessage
Constants ¶
This section is empty.
Variables ¶
var ErrUnexpectedChatMessageType = errors.New("unexpected chat message type")
ErrUnexpectedChatMessageType is returned when a chat message is of an unexpected type.
Functions ¶
func GetBufferString ¶
func GetBufferString(messages []ChatMessage, humanPrefix string, aiPrefix string) (string, error)
GetBufferString gets the buffer string of messages.
Types ¶
type AIChatMessage ¶
type AIChatMessage struct { // Content is the content of the message. Content string // FunctionCall represents the model choosing to call a function. FunctionCall *FunctionCall `json:"function_call,omitempty"` }
AIChatMessage is a message sent by an AI.
func (AIChatMessage) GetContent ¶
func (m AIChatMessage) GetContent() string
func (AIChatMessage) GetFunctionCall ¶
func (m AIChatMessage) GetFunctionCall() *FunctionCall
func (AIChatMessage) GetType ¶
func (m AIChatMessage) GetType() ChatMessageType
type AgentAction ¶
AgentAction is the agent's action to take.
type AgentFinish ¶
AgentFinish is the agent's return value.
type AgentStep ¶
type AgentStep struct { Action AgentAction Observation string }
AgentStep is a step of the agent.
type ChatMessage ¶
type ChatMessage interface { // GetType gets the type of the message. GetType() ChatMessageType // GetContent gets the content of the message. GetContent() string }
ChatMessage represents a message in a chat.
type ChatMessageHistory ¶
type ChatMessageHistory interface { // AddMessage adds a message to the store. AddMessage(ctx context.Context, message ChatMessage) error // AddUserMessage is a convenience method for adding a human message string // to the store. AddUserMessage(ctx context.Context, message string) error // AddAIMessage is a convenience method for adding an AI message string to // the store. AddAIMessage(ctx context.Context, message string) error // Clear removes all messages from the store. Clear(ctx context.Context) error // Messages retrieves all messages from the store Messages(ctx context.Context) ([]ChatMessage, error) // SetMessages replaces existing messages in the store SetMessages(ctx context.Context, messages []ChatMessage) error }
ChatMessageHistory is the interface for chat history in memory/store.
type ChatMessageType ¶
type ChatMessageType string
ChatMessageType is the type of chat message.
const ( // ChatMessageTypeAI is a message sent by an AI. ChatMessageTypeAI ChatMessageType = "ai" // ChatMessageTypeHuman is a message sent by a human. ChatMessageTypeHuman ChatMessageType = "human" // ChatMessageTypeSystem is a message sent by the system. ChatMessageTypeSystem ChatMessageType = "system" // ChatMessageTypeGeneric is a message sent by a generic user. ChatMessageTypeGeneric ChatMessageType = "generic" // ChatMessageTypeFunction is a message sent by a function. ChatMessageTypeFunction ChatMessageType = "function" )
type FunctionCall ¶
FunctionCall is the name and arguments of a function call.
type FunctionChatMessage ¶
FunctionChatMessage is a chat message representing the result of a function call.
func (FunctionChatMessage) GetContent ¶
func (m FunctionChatMessage) GetContent() string
func (FunctionChatMessage) GetName ¶
func (m FunctionChatMessage) GetName() string
func (FunctionChatMessage) GetType ¶
func (m FunctionChatMessage) GetType() ChatMessageType
type GenericChatMessage ¶
GenericChatMessage is a chat message with an arbitrary speaker.
func (GenericChatMessage) GetContent ¶
func (m GenericChatMessage) GetContent() string
func (GenericChatMessage) GetName ¶
func (m GenericChatMessage) GetName() string
func (GenericChatMessage) GetType ¶
func (m GenericChatMessage) GetType() ChatMessageType
type HumanChatMessage ¶
type HumanChatMessage struct {
Content string
}
HumanChatMessage is a message sent by a human.
func (HumanChatMessage) GetContent ¶
func (m HumanChatMessage) GetContent() string
func (HumanChatMessage) GetType ¶
func (m HumanChatMessage) GetType() ChatMessageType
type Memory ¶
type Memory interface { // GetMemoryKey getter for memory key. GetMemoryKey(ctx context.Context) string // MemoryVariables Input keys this memory class will load dynamically. MemoryVariables(ctx context.Context) []string // LoadMemoryVariables Return key-value pairs given the text input to the chain. // If None, return all memories LoadMemoryVariables(ctx context.Context, inputs map[string]any) (map[string]any, error) // SaveContext Save the context of this model run to memory. SaveContext(ctx context.Context, inputs map[string]any, outputs map[string]any) error // Clear memory contents. Clear(ctx context.Context) error }
Memory is the interface for memory in chains.
type Named ¶
type Named interface {
GetName() string
}
Named is an interface for objects that have a name.
type OutputParser ¶
type OutputParser[T any] interface { // Parse parses the output of an LLM call. Parse(text string) (T, error) // ParseWithPrompt parses the output of an LLM call with the prompt used. ParseWithPrompt(text string, prompt PromptValue) (T, error) // GetFormatInstructions returns a string describing the format of the output. GetFormatInstructions() string // Type returns the string type key uniquely identifying this class of parser Type() string }
OutputParser is an interface for parsing the output of an LLM call.
type PromptValue ¶
type PromptValue interface { String() string Messages() []ChatMessage }
PromptValue is the interface that all prompt values must implement.
type Retriever ¶
type Retriever interface {
GetRelevantDocuments(ctx context.Context, query string) ([]Document, error)
}
Retriever is an interface that defines the behavior of a retriever.
type SystemChatMessage ¶
type SystemChatMessage struct {
Content string
}
SystemChatMessage is a chat message representing information that should be instructions to the AI system.
func (SystemChatMessage) GetContent ¶
func (m SystemChatMessage) GetContent() string
func (SystemChatMessage) GetType ¶
func (m SystemChatMessage) GetType() ChatMessageType