my_chat_interface.py

"""
This file defines a useful high-level abstraction to build Gradio chatbots: ChatInterface.
"""

from __future__ import annotations

import inspect
import re
import traceback
from typing import AsyncGenerator, Callable, Literal, Union, cast

import anyio
from gradio.blocks import Blocks
from gradio.components import (
    Button,
    Chatbot,
    Component,
    Markdown,
    MultimodalTextbox,
    State,
    Textbox,
    get_component_instance,
)
from gradio.events import Dependency, on
from gradio.helpers import create_examples as Examples  # noqa: N812
from gradio.helpers import special_args
from gradio.layouts import Accordion, Group, Row
from gradio.routes import Request
from gradio.themes import ThemeClass as Theme
from gradio.utils import SyncToAsyncIterator, async_iteration, async_lambda
from gradio_client.documentation import document
from langchain_experimental.tools.python.tool import PythonREPLTool

from utils.output_parser import parse_code_action


def replace_image_path(markdown_text):
    pattern = r"!\[.*?\]\((.*?)\)"
    # 替换图片链接并添加"file/"前缀
    modified_text = re.sub(pattern, r"![\g<1>](file/\g<1>)", markdown_text)
    return modified_text


def execute_code(code_str: str, tool: PythonREPLTool):
    """Execute python code and return the execution result.

    Args:
        code_str (str): The code to be executed.
        tool (PythonREPLTool): Python AST REPL tool.

    Returns:
        str: Code execution result.
    """
    try:
        result = tool.invoke(code_str)
        result = str(result)
        return result
    except:
        return traceback.format_exc()


def extract_code(response):
    """Extract code from the chatbot response.

    Args:
        response (str): The chatbot response.

    Returns:
        str: Extracted code.
    """
    try:
        code = response.split("```")[-2]
        code = code.replace("python", "")
    except:
        code = None
    return code


@document()
class MyChatInterface(Blocks):
    """
    ChatInterface is Gradio's high-level abstraction for creating chatbot UIs, and allows you to create
    a web-based demo around a chatbot model in a few lines of code. Only one parameter is required: fn, which
    takes a function that governs the response of the chatbot based on the user input and chat history. Additional
    parameters can be used to control the appearance and behavior of the demo.

    Example:
        import gradio as gr

        def echo(message, history):
            return message

        demo = gr.ChatInterface(fn=echo, examples=["hello", "hola", "merhaba"], title="Echo Bot")
        demo.launch()
    Demos: chatinterface_multimodal, chatinterface_random_response, chatinterface_streaming_echo
    Guides: creating-a-chatbot-fast, sharing-your-app
    """

    def __init__(
        self,
        fn: Callable,
        *,
        multimodal: bool = False,
        chatbot: Chatbot | None = None,
        textbox: Textbox | MultimodalTextbox | None = None,
        additional_inputs: str | Component | list[str | Component] | None = None,
        additional_inputs_accordion_name: str | None = None,
        additional_inputs_accordion: str | Accordion | None = None,
        examples: list[str] | list[dict[str, str | list]] | list[list] | None = None,
        cache_examples: bool | Literal["lazy"] | None = None,
        examples_per_page: int = 10,
        title: str | None = None,
        description: str | None = None,
        theme: Theme | str | None = None,
        css: str | None = None,
        js: str | None = None,
        head: str | None = None,
        analytics_enabled: bool | None = None,
        submit_btn: str | None | Button = "Submit",
        stop_btn: str | None | Button = "Stop",
        retry_btn: str | None | Button = "🔄  Retry",
        undo_btn: str | None | Button = "↩️ Undo",
        clear_btn: str | None | Button = "🗑️  Clear",
        autofocus: bool = True,
        concurrency_limit: int | None | Literal["default"] = "default",
        fill_height: bool = True,
        delete_cache: tuple[int, int] | None = None,
        mode: str = "prompt",
        code_start_token: str = "<|execute_start|>\n```python\n",
        code_end_token: str = "```\n<|execute_end|>",
        tool_call_token: str = "<|tool_call|>",
    ):
        """
        Parameters:
            fn: The function to wrap the chat interface around. Should accept two parameters: a string input message and list of two-element lists of the form [[user_message, bot_message], ...] representing the chat history, and return a string response. See the Chatbot documentation for more information on the chat history format.
            multimodal: If True, the chat interface will use a gr.MultimodalTextbox component for the input, which allows for the uploading of multimedia files. If False, the chat interface will use a gr.Textbox component for the input.
            chatbot: An instance of the gr.Chatbot component to use for the chat interface, if you would like to customize the chatbot properties. If not provided, a default gr.Chatbot component will be created.
            textbox: An instance of the gr.Textbox or gr.MultimodalTextbox component to use for the chat interface, if you would like to customize the textbox properties. If not provided, a default gr.Textbox or gr.MultimodalTextbox component will be created.
            additional_inputs: An instance or list of instances of gradio components (or their string shortcuts) to use as additional inputs to the chatbot. If components are not already rendered in a surrounding Blocks, then the components will be displayed under the chatbot, in an accordion.
            additional_inputs_accordion_name: Deprecated. Will be removed in a future version of Gradio. Use the `additional_inputs_accordion` parameter instead.
            additional_inputs_accordion: If a string is provided, this is the label of the `gr.Accordion` to use to contain additional inputs. A `gr.Accordion` object can be provided as well to configure other properties of the container holding the additional inputs. Defaults to a `gr.Accordion(label="Additional Inputs", open=False)`. This parameter is only used if `additional_inputs` is provided.
            examples: Sample inputs for the function; if provided, appear below the chatbot and can be clicked to populate the chatbot input. Should be a list of strings if `multimodal` is False, and a list of dictionaries (with keys `text` and `files`) if `multimodal` is True.
            cache_examples: If True, caches examples in the server for fast runtime in examples. The default option in HuggingFace Spaces is True. The default option elsewhere is False.
            examples_per_page: If examples are provided, how many to display per page.
            title: a title for the interface; if provided, appears above chatbot in large font. Also used as the tab title when opened in a browser window.
            description: a description for the interface; if provided, appears above the chatbot and beneath the title in regular font. Accepts Markdown and HTML content.
            theme: Theme to use, loaded from gradio.themes.
            css: Custom css as a string or path to a css file. This css will be included in the demo webpage.
            js: Custom js as a string or path to a js file. The custom js should be in the form of a single js function. This function will automatically be executed when the page loads. For more flexibility, use the head parameter to insert js inside <script> tags.
            head: Custom html to insert into the head of the demo webpage. This can be used to add custom meta tags, multiple scripts, stylesheets, etc. to the page.
            analytics_enabled: Whether to allow basic telemetry. If None, will use GRADIO_ANALYTICS_ENABLED environment variable if defined, or default to True.
            submit_btn: Text to display on the submit button. If None, no button will be displayed. If a Button object, that button will be used.
            stop_btn: Text to display on the stop button, which replaces the submit_btn when the submit_btn or retry_btn is clicked and response is streaming. Clicking on the stop_btn will halt the chatbot response. If set to None, stop button functionality does not appear in the chatbot. If a Button object, that button will be used as the stop button.
            retry_btn: Text to display on the retry button. If None, no button will be displayed. If a Button object, that button will be used.
            undo_btn: Text to display on the delete last button. If None, no button will be displayed. If a Button object, that button will be used.
            clear_btn: Text to display on the clear button. If None, no button will be displayed. If a Button object, that button will be used.
            autofocus: If True, autofocuses to the textbox when the page loads.
            concurrency_limit: If set, this is the maximum number of chatbot submissions that can be running simultaneously. Can be set to None to mean no limit (any number of chatbot submissions can be running simultaneously). Set to "default" to use the default concurrency limit (defined by the `default_concurrency_limit` parameter in `.queue()`, which is 1 by default).
            fill_height: If True, the chat interface will expand to the height of window.
            delete_cache: A tuple corresponding [frequency, age] both expressed in number of seconds. Every `frequency` seconds, the temporary files created by this Blocks instance will be deleted if more than `age` seconds have passed since the file was created. For example, setting this to (86400, 86400) will delete temporary files every day. The cache will be deleted entirely when the server restarts. If None, no cache deletion will occur.
        """
        super().__init__(
            analytics_enabled=analytics_enabled,
            mode="chat_interface",
            css=css,
            title=title or "Gradio",
            theme=theme,
            js=js,
            head=head,
            fill_height=fill_height,
            delete_cache=delete_cache,
        )
        self.multimodal = multimodal
        self.concurrency_limit = concurrency_limit
        self.fn = fn
        self.is_async = inspect.iscoroutinefunction(
            self.fn
        ) or inspect.isasyncgenfunction(self.fn)
        self.is_generator = inspect.isgeneratorfunction(
            self.fn
        ) or inspect.isasyncgenfunction(self.fn)
        self.buttons: list[Button | None] = []

        self.examples = examples
        self.cache_examples = cache_examples

        if additional_inputs:
            if not isinstance(additional_inputs, list):
                additional_inputs = [additional_inputs]
            self.additional_inputs = [
                get_component_instance(i)
                for i in additional_inputs  # type: ignore
            ]
        else:
            self.additional_inputs = []
        if additional_inputs_accordion_name is not None:
            print(
                "The `additional_inputs_accordion_name` parameter is deprecated and will be removed in a future version of Gradio. Use the `additional_inputs_accordion` parameter instead."
            )
            self.additional_inputs_accordion_params = {
                "label": additional_inputs_accordion_name
            }
        if additional_inputs_accordion is None:
            self.additional_inputs_accordion_params = {
                "label": "Additional Inputs",
                "open": False,
            }
        elif isinstance(additional_inputs_accordion, str):
            self.additional_inputs_accordion_params = {
                "label": additional_inputs_accordion
            }
        elif isinstance(additional_inputs_accordion, Accordion):
            self.additional_inputs_accordion_params = (
                additional_inputs_accordion.recover_kwargs(
                    additional_inputs_accordion.get_config()
                )
            )
        else:
            raise ValueError(
                f"The `additional_inputs_accordion` parameter must be a string or gr.Accordion, not {type(additional_inputs_accordion)}"
            )

        with self:
            if title:
                Markdown(
                    f"<h1 style='text-align: center; margin-bottom: 1rem'>{self.title}</h1>"
                )
            if description:
                Markdown(description)

            if chatbot:
                self.chatbot = chatbot.render()
            else:
                self.chatbot = Chatbot(
                    label="Chatbot", scale=1, height=200 if fill_height else None
                )

            with Row():
                for btn in [retry_btn, undo_btn, clear_btn]:
                    if btn is not None:
                        if isinstance(btn, Button):
                            btn.render()
                        elif isinstance(btn, str):
                            btn = Button(
                                btn, variant="secondary", size="sm", min_width=60
                            )
                        else:
                            raise ValueError(
                                f"All the _btn parameters must be a gr.Button, string, or None, not {type(btn)}"
                            )
                    self.buttons.append(btn)  # type: ignore

            with Group():
                with Row():
                    if textbox:
                        if self.multimodal:
                            submit_btn = None
                        else:
                            textbox.container = False
                        textbox.show_label = False
                        textbox_ = textbox.render()
                        if not isinstance(textbox_, (Textbox, MultimodalTextbox)):
                            raise TypeError(
                                f"Expected a gr.Textbox or gr.MultimodalTextbox component, but got {type(textbox_)}"
                            )
                        self.textbox = textbox_
                    elif self.multimodal:
                        submit_btn = None
                        self.textbox = MultimodalTextbox(
                            show_label=False,
                            label="Message",
                            placeholder="Type a message...",
                            scale=7,
                            autofocus=autofocus,
                        )
                    else:
                        self.textbox = Textbox(
                            container=False,
                            show_label=False,
                            label="Message",
                            placeholder="Type a message...",
                            scale=7,
                            autofocus=autofocus,
                        )
                    if submit_btn is not None and not multimodal:
                        if isinstance(submit_btn, Button):
                            submit_btn.render()
                        elif isinstance(submit_btn, str):
                            submit_btn = Button(
                                submit_btn,
                                variant="primary",
                                scale=1,
                                min_width=150,
                            )
                        else:
                            raise ValueError(
                                f"The submit_btn parameter must be a gr.Button, string, or None, not {type(submit_btn)}"
                            )
                    if stop_btn is not None:
                        if isinstance(stop_btn, Button):
                            stop_btn.visible = False
                            stop_btn.render()
                        elif isinstance(stop_btn, str):
                            stop_btn = Button(
                                stop_btn,
                                variant="stop",
                                visible=False,
                                scale=1,
                                min_width=150,
                            )
                        else:
                            raise ValueError(
                                f"The stop_btn parameter must be a gr.Button, string, or None, not {type(stop_btn)}"
                            )
                    self.buttons.extend([submit_btn, stop_btn])  # type: ignore

                self.fake_api_btn = Button("Fake API", visible=False)
                self.fake_response_textbox = Textbox(label="Response", visible=False)
                (
                    self.retry_btn,
                    self.undo_btn,
                    self.clear_btn,
                    self.submit_btn,
                    self.stop_btn,
                ) = self.buttons

            if examples:
                if self.is_generator:
                    examples_fn = self._examples_stream_fn
                else:
                    examples_fn = self._examples_fn

                self.examples_handler = Examples(
                    examples=examples,
                    inputs=[self.textbox] + self.additional_inputs,
                    outputs=self.chatbot,
                    fn=examples_fn,
                    cache_examples=self.cache_examples,
                    _defer_caching=True,
                    examples_per_page=examples_per_page,
                )

            any_unrendered_inputs = any(
                not inp.is_rendered for inp in self.additional_inputs
            )
            if self.additional_inputs and any_unrendered_inputs:
                with Accordion(**self.additional_inputs_accordion_params):  # type: ignore
                    for input_component in self.additional_inputs:
                        if not input_component.is_rendered:
                            input_component.render()

            # The example caching must happen after the input components have rendered
            if examples:
                self.examples_handler._start_caching()

            self.saved_input = State()
            self.chatbot_state = (
                State(self.chatbot.value) if self.chatbot.value else State([])
            )

            self._setup_events()
            self._setup_api()
        self.mode = mode
        self.code_start_token = code_start_token
        self.code_end_token = code_end_token
        self.tool_call_token = tool_call_token
        self.tool = PythonREPLTool()

    def _setup_events(self) -> None:
        submit_fn = self._stream_fn if self.is_generator else self._submit_fn
        submit_triggers = (
            [self.textbox.submit, self.submit_btn.click]
            if self.submit_btn
            else [self.textbox.submit]
        )
        submit_event = (
            on(
                submit_triggers,
                self._clear_and_save_textbox,
                [self.textbox],
                [self.textbox, self.saved_input],
                show_api=False,
                queue=False,
            )
            .then(
                self._display_input,
                [self.saved_input, self.chatbot_state],
                [self.chatbot, self.chatbot_state],
                show_api=False,
                queue=False,
            )
            .then(
                submit_fn,
                [self.saved_input, self.chatbot_state] + self.additional_inputs,
                [self.chatbot, self.chatbot_state],
                show_api=False,
                concurrency_limit=cast(
                    Union[int, Literal["default"], None], self.concurrency_limit
                ),
            )
        )
        self._setup_stop_events(submit_triggers, submit_event)

        if self.retry_btn:
            retry_event = (
                self.retry_btn.click(
                    self._delete_prev_fn,
                    [self.saved_input, self.chatbot_state],
                    [self.chatbot, self.saved_input, self.chatbot_state],
                    show_api=False,
                    queue=False,
                )
                .then(
                    self._display_input,
                    [self.saved_input, self.chatbot_state],
                    [self.chatbot, self.chatbot_state],
                    show_api=False,
                    queue=False,
                )
                .then(
                    submit_fn,
                    [self.saved_input, self.chatbot_state] + self.additional_inputs,
                    [self.chatbot, self.chatbot_state],
                    show_api=False,
                    concurrency_limit=cast(
                        Union[int, Literal["default"], None], self.concurrency_limit
                    ),
                )
            )
            self._setup_stop_events([self.retry_btn.click], retry_event)

        if self.undo_btn:
            self.undo_btn.click(
                self._delete_prev_fn,
                [self.saved_input, self.chatbot_state],
                [self.chatbot, self.saved_input, self.chatbot_state],
                show_api=False,
                queue=False,
            ).then(
                async_lambda(lambda x: x),
                [self.saved_input],
                [self.textbox],
                show_api=False,
                queue=False,
            )

        if self.clear_btn:
            self.clear_btn.click(
                async_lambda(lambda: ([], [], None)),
                None,
                [self.chatbot, self.chatbot_state, self.saved_input],
                queue=False,
                show_api=False,
            )

    def _setup_stop_events(
        self, event_triggers: list[Callable], event_to_cancel: Dependency
    ) -> None:
        if self.stop_btn and self.is_generator:
            if self.submit_btn:
                for event_trigger in event_triggers:
                    event_trigger(
                        async_lambda(
                            lambda: (
                                Button(visible=False),
                                Button(visible=True),
                            )
                        ),
                        None,
                        [self.submit_btn, self.stop_btn],
                        show_api=False,
                        queue=False,
                    )
                event_to_cancel.then(
                    async_lambda(lambda: (Button(visible=True), Button(visible=False))),
                    None,
                    [self.submit_btn, self.stop_btn],
                    show_api=False,
                    queue=False,
                )
            else:
                for event_trigger in event_triggers:
                    event_trigger(
                        async_lambda(lambda: Button(visible=True)),
                        None,
                        [self.stop_btn],
                        show_api=False,
                        queue=False,
                    )
                event_to_cancel.then(
                    async_lambda(lambda: Button(visible=False)),
                    None,
                    [self.stop_btn],
                    show_api=False,
                    queue=False,
                )
            self.stop_btn.click(
                None,
                None,
                None,
                cancels=event_to_cancel,
                show_api=False,
            )

    def _setup_api(self) -> None:
        api_fn = self._api_stream_fn if self.is_generator else self._api_submit_fn

        self.fake_api_btn.click(
            api_fn,
            [self.textbox, self.chatbot_state] + self.additional_inputs,
            [self.textbox, self.chatbot_state],
            api_name="chat",
            concurrency_limit=cast(
                Union[int, Literal["default"], None], self.concurrency_limit
            ),
        )

    def _clear_and_save_textbox(self, message: str) -> tuple[str | dict, str]:
        if self.multimodal:
            return {"text": "", "files": []}, message
        else:
            return "", message

    def _append_multimodal_history(
        self,
        message: dict[str, list],
        response: str | None,
        history: list[list[str | tuple | None]],
    ):
        for x in message["files"]:
            history.append([(x,), None])
        if message["text"] is None or not isinstance(message["text"], str):
            return
        elif message["text"] == "" and message["files"] != []:
            history.append([None, response])
        else:
            history.append([message["text"], response])

    async def _display_input(
        self, message: str | dict[str, list], history: list[list[str | tuple | None]]
    ) -> tuple[list[list[str | tuple | None]], list[list[str | tuple | None]]]:
        if self.multimodal and isinstance(message, dict):
            self._append_multimodal_history(message, None, history)
        elif isinstance(message, str):
            history.append([message, None])
        print("history after display input:", history)
        return history, history

    async def _submit_fn(
        self,
        message: str | dict[str, list],
        history_with_input: list[list[str | tuple | None]],
        request: Request,
        *args,
    ) -> tuple[list[list[str | tuple | None]], list[list[str | tuple | None]]]:
        print("calling custom submit fn")
        print("message:", message)
        print("history_with_input:", history_with_input)
        if self.multimodal and isinstance(message, dict):
            remove_input = (
                len(message["files"]) + 1
                if message["text"] is not None
                else len(message["files"])
            )
            history = history_with_input[:-remove_input]
            if len(message["files"]) > 0:
                history.append(
                    (
                        f"[INFO]The data is uploaded to {','.join(message['files'])}",
                        None,
                    )
                )
            message = message["text"]
        else:
            history = history_with_input[:-1]
        while True:
            inputs, _, _ = special_args(
                self.fn, inputs=[message, history, *args], request=request
            )
            print("history:", inputs[1])

            if self.is_async:
                response = await self.fn(*inputs)
            else:
                response = await anyio.to_thread.run_sync(
                    self.fn, *inputs, limiter=self.limiter
                )
            response = replace_image_path(response)
            reasoning, code_script = parse_code_action(
                response,
                mode="prompt",
                code_start_token="```python\n",
                code_end_token="```",
            )
            if code_script is None or code_script.strip() == "":
                print("no code action")
                response = reasoning
            else:
                print("have code action")
                response = f"{reasoning}\n{self.code_start_token}\n{code_script}\n{self.code_end_token}"
            if self.multimodal and isinstance(message, dict):
                self._append_multimodal_history(message, response, history)
            else:
                history.append((message, response))
            message = None
            if code_script is None or code_script.strip() == "":
                break

            code_response = execute_code(code_script, self.tool)
            history.append((None, code_response))

        return history, history

    async def _stream_fn(
        self,
        message: str | dict[str, list],
        history_with_input: list[list[str | tuple | None]],
        request: Request,
        *args,
    ) -> AsyncGenerator:
        if self.multimodal and isinstance(message, dict):
            remove_input = (
                len(message["files"]) + 1
                if message["text"] is not None
                else len(message["files"])
            )
            history = history_with_input[:-remove_input]
        else:
            history = history_with_input[:-1]
        inputs, _, _ = special_args(
            self.fn, inputs=[message, history, *args], request=request
        )

        if self.is_async:
            generator = self.fn(*inputs)
        else:
            generator = await anyio.to_thread.run_sync(
                self.fn, *inputs, limiter=self.limiter
            )
            generator = SyncToAsyncIterator(generator, self.limiter)
        try:
            first_response = await async_iteration(generator)
            if self.multimodal and isinstance(message, dict):
                for x in message["files"]:
                    history.append([(x,), None])
                update = history + [[message["text"], first_response]]
                yield update, update
            else:
                update = history + [[message, first_response]]
                yield update, update
        except StopIteration:
            if self.multimodal and isinstance(message, dict):
                self._append_multimodal_history(message, None, history)
                yield history, history
            else:
                update = history + [[message, None]]
                yield update, update
        async for response in generator:
            if self.multimodal and isinstance(message, dict):
                update = history + [[message["text"], response]]
                yield update, update
            else:
                update = history + [[message, response]]
                yield update, update

    async def _api_submit_fn(
        self, message: str, history: list[list[str | None]], request: Request, *args
    ) -> tuple[str, list[list[str | None]]]:
        inputs, _, _ = special_args(
            self.fn, inputs=[message, history, *args], request=request
        )

        if self.is_async:
            response = await self.fn(*inputs)
        else:
            response = await anyio.to_thread.run_sync(
                self.fn, *inputs, limiter=self.limiter
            )
        history.append([message, response])
        return response, history

    async def _api_stream_fn(
        self, message: str, history: list[list[str | None]], request: Request, *args
    ) -> AsyncGenerator:
        inputs, _, _ = special_args(
            self.fn, inputs=[message, history, *args], request=request
        )

        if self.is_async:
            generator = self.fn(*inputs)
        else:
            generator = await anyio.to_thread.run_sync(
                self.fn, *inputs, limiter=self.limiter
            )
            generator = SyncToAsyncIterator(generator, self.limiter)
        try:
            first_response = await async_iteration(generator)
            yield first_response, history + [[message, first_response]]
        except StopIteration:
            yield None, history + [[message, None]]
        async for response in generator:
            yield response, history + [[message, response]]

    async def _examples_fn(self, message: str, *args) -> list[list[str | None]]:
        inputs, _, _ = special_args(self.fn, inputs=[message, [], *args], request=None)

        if self.is_async:
            response = await self.fn(*inputs)
        else:
            response = await anyio.to_thread.run_sync(
                self.fn, *inputs, limiter=self.limiter
            )
        return [[message, response]]

    async def _examples_stream_fn(
        self,
        message: str,
        *args,
    ) -> AsyncGenerator:
        inputs, _, _ = special_args(self.fn, inputs=[message, [], *args], request=None)

        if self.is_async:
            generator = self.fn(*inputs)
        else:
            generator = await anyio.to_thread.run_sync(
                self.fn, *inputs, limiter=self.limiter
            )
            generator = SyncToAsyncIterator(generator, self.limiter)
        async for response in generator:
            yield [[message, response]]

    async def _delete_prev_fn(
        self,
        message: str | dict[str, list],
        history: list[list[str | tuple | None]],
    ) -> tuple[
        list[list[str | tuple | None]],
        str | dict[str, list],
        list[list[str | tuple | None]],
    ]:
        if self.multimodal and isinstance(message, dict):
            remove_input = (
                len(message["files"]) + 1
                if message["text"] is not None
                else len(message["files"])
            )
            history = history[:-remove_input]
        else:
            history = history[:-1]
        return history, message or "", history