ToolsOutputParser class

A parser that returns the list of tool calls returned by the model.

When streaming, the parser attempts to “auto-complete” the partial json from each chunk into a valid state.

Example:

const tool = ToolSpec(
  name: 'joke',
  description: 'A joke',
  inputJsonSchema: {
    'type': 'object',
    'properties': {
      'setup': {
        'type': 'string',
        'description': 'The setup for the joke',
      },
      'punchline': {
        'type': 'string',
        'description': 'The punchline to the joke',
      },
    },
    'required': ['location', 'punchline'],
  },
);
final promptTemplate = ChatPromptTemplate.fromTemplate(
  'tell me a long joke about {foo}',
);
final chat = ChatOpenAI(
  apiKey: openaiApiKey,
  defaultOptions: ChatOpenAIOptions(
    temperature: 0,
  ),
).bind(
  ChatOpenAIOptions(
    tools: [tool],
    toolChoice: ChatToolChoice.forced(name: 'joke'),
  ),
);
final outputParser = ToolsOutputParser();
final chain = promptTemplate.pipe(chat).pipe(outputParser);
final res = await chain.invoke({'foo': 'bears'});
print(res);
// [ParsedToolCall{
//   id: call_5TU1iYgYO3Z81eAuTe7J23f7,
//   name: joke,
//   arguments: {
//     setup: Why don't bears like fast food restaurants?,
//     punchline: Because they can't bear the wait!
//   },
// }]