dntrplytch's picture
Upload folder using huggingface_hub
7d134e4 verified
"""gr.Chatbot() component."""
from __future__ import annotations
import inspect
from dataclasses import dataclass, field
from pathlib import Path
from typing import (
TYPE_CHECKING,
Any,
Callable,
Dict,
List,
Literal,
Optional,
Sequence,
Tuple,
Type,
TypedDict,
Union,
cast,
)
from gradio_client import utils as client_utils
from gradio_client.documentation import document
from pydantic import Field
from typing_extensions import NotRequired
from gradio import utils
from gradio.component_meta import ComponentMeta
from gradio.components import (
Component as GradioComponent,
)
from gradio.components.base import Component
from gradio.data_classes import FileData, GradioModel, GradioRootModel
from gradio.events import Events
from gradio.exceptions import Error
from gradio.processing_utils import move_resource_to_block_cache
class MetadataDict(TypedDict):
title: Union[str, None]
class FileDataDict(TypedDict):
path: str # server filepath
url: NotRequired[Optional[str]] # normalised server url
size: NotRequired[Optional[int]] # size in bytes
orig_name: NotRequired[Optional[str]] # original filename
mime_type: NotRequired[Optional[str]]
is_stream: NotRequired[bool]
meta: dict[Literal["_type"], Literal["gradio.FileData"]]
class MessageDict(TypedDict):
content: str | FileDataDict | tuple | Component
role: Literal["user", "assistant", "system"]
metadata: NotRequired[MetadataDict]
class FileMessage(GradioModel):
file: FileData
alt_text: Optional[str] = None
class ComponentMessage(GradioModel):
component: str
value: Any
constructor_args: Dict[str, Any]
props: Dict[str, Any]
class ChatbotDataTuples(GradioRootModel):
root: List[
Tuple[
Union[str, FileMessage, ComponentMessage, None],
Union[str, FileMessage, ComponentMessage, None],
]
]
class Metadata(GradioModel):
title: Optional[str] = None
class Message(GradioModel):
role: str
metadata: Metadata = Field(default_factory=Metadata)
content: Union[str, FileMessage, ComponentMessage]
@dataclass
class ChatMessage:
role: Literal["user", "assistant", "system"]
content: str | FileData | Component | FileDataDict | tuple | list
metadata: MetadataDict | Metadata = field(default_factory=Metadata)
class ChatbotDataMessages(GradioRootModel):
root: List[Message]
TupleFormat = List[List[Union[str, Tuple[str], Tuple[str, str], None]]]
if TYPE_CHECKING:
from gradio.components import Timer
def import_component_and_data(
component_name: str,
) -> GradioComponent | ComponentMeta | Any | None:
try:
for component in utils.get_all_components():
if component_name == component.__name__ and isinstance(
component, ComponentMeta
):
return component
except ModuleNotFoundError as e:
raise ValueError(f"Error importing {component_name}: {e}") from e
except AttributeError:
pass
@document()
class Chatbot(Component):
"""
Creates a chatbot that displays user-submitted messages and responses. Supports a subset of Markdown including bold, italics, code, tables.
Also supports audio/video/image files, which are displayed in the Chatbot, and other kinds of files which are displayed as links. This
component is usually used as an output component.
Demos: chatbot_simple, chatbot_core_components_simple
Guides: creating-a-chatbot
"""
EVENTS = [Events.change, Events.select, Events.like]
def __init__(
self,
value: (
Sequence[
Sequence[
str | GradioComponent | tuple[str] | tuple[str | Path, str] | None
]
]
| Callable
| None
) = None,
*,
type: Literal["messages", "tuples"] = "tuples",
label: str | None = None,
every: Timer | float | None = None,
inputs: Component | Sequence[Component] | set[Component] | None = None,
show_label: bool | None = None,
container: bool = True,
scale: int | None = None,
min_width: int = 160,
visible: bool = True,
elem_id: str | None = None,
elem_classes: list[str] | str | None = None,
render: bool = True,
key: int | str | None = None,
height: int | str | None = None,
latex_delimiters: list[dict[str, str | bool]] | None = None,
rtl: bool = False,
show_share_button: bool | None = None,
show_copy_button: bool = False,
avatar_images: tuple[str | Path | None, str | Path | None] | None = None,
sanitize_html: bool = True,
render_markdown: bool = True,
bubble_full_width: bool = True,
line_breaks: bool = True,
likeable: bool = False,
layout: Literal["panel", "bubble"] | None = None,
placeholder: str | None = None,
show_copy_all_button=False,
):
"""
Parameters:
value: Default value to show in chatbot. If callable, the function will be called whenever the app loads to set the initial value of the component.
type: The format of the messages. If 'tuples', expects a `list[list[str | None | tuple]]`, i.e. a list of lists. The inner list should have 2 elements: the user message and the response message. The individual messages can be (1) strings in valid Markdown, (2) tuples if sending files: (a filepath or URL to a file, [optional string alt text]) -- if the file is image/video/audio, it is displayed in the Chatbot, or (3) None, in which case the message is not displayed. If 'messages', passes the value as a list of dictionaries with 'role' and 'content' keys. The `content' key's value supports everything the 'tuples' format supports. The 'role' key should be one of 'user' or 'assistant'. Any other roles will not be displayed in the output.
label: The label for this component. Appears above the component and is also used as the header if there are a table of examples for this component. If None and used in a `gr.Interface`, the label will be the name of the parameter this component is assigned to.
every: Continously calls `value` to recalculate it if `value` is a function (has no effect otherwise). Can provide a Timer whose tick resets `value`, or a float that provides the regular interval for the reset Timer.
inputs: Components that are used as inputs to calculate `value` if `value` is a function (has no effect otherwise). `value` is recalculated any time the inputs change.
show_label: if True, will display label.
container: If True, will place the component in a container - providing some extra padding around the border.
scale: relative size compared to adjacent Components. For example if Components A and B are in a Row, and A has scale=2, and B has scale=1, A will be twice as wide as B. Should be an integer. scale applies in Rows, and to top-level Components in Blocks where fill_height=True.
min_width: minimum pixel width, will wrap if not sufficient screen space to satisfy this value. If a certain scale value results in this Component being narrower than min_width, the min_width parameter will be respected first.
visible: If False, component will be hidden.
elem_id: An optional string that is assigned as the id of this component in the HTML DOM. Can be used for targeting CSS styles.
elem_classes: An optional list of strings that are assigned as the classes of this component in the HTML DOM. Can be used for targeting CSS styles.
render: If False, component will not render be rendered in the Blocks context. Should be used if the intention is to assign event listeners now but render the component later.
key: if assigned, will be used to assume identity across a re-render. Components that have the same key across a re-render will have their value preserved.
height: The height of the component, specified in pixels if a number is passed, or in CSS units if a string is passed.
latex_delimiters: A list of dicts of the form {"left": open delimiter (str), "right": close delimiter (str), "display": whether to display in newline (bool)} that will be used to render LaTeX expressions. If not provided, `latex_delimiters` is set to `[{ "left": "$$", "right": "$$", "display": True }]`, so only expressions enclosed in $$ delimiters will be rendered as LaTeX, and in a new line. Pass in an empty list to disable LaTeX rendering. For more information, see the [KaTeX documentation](https://katex.org/docs/autorender.html).
rtl: If True, sets the direction of the rendered text to right-to-left. Default is False, which renders text left-to-right.
show_share_button: If True, will show a share icon in the corner of the component that allows user to share outputs to Hugging Face Spaces Discussions. If False, icon does not appear. If set to None (default behavior), then the icon appears if this Gradio app is launched on Spaces, but not otherwise.
show_copy_button: If True, will show a copy button for each chatbot message.
avatar_images: Tuple of two avatar image paths or URLs for user and bot (in that order). Pass None for either the user or bot image to skip. Must be within the working directory of the Gradio app or an external URL.
sanitize_html: If False, will disable HTML sanitization for chatbot messages. This is not recommended, as it can lead to security vulnerabilities.
render_markdown: If False, will disable Markdown rendering for chatbot messages.
bubble_full_width: If False, the chat bubble will fit to the content of the message. If True (default), the chat bubble will be the full width of the component.
line_breaks: If True (default), will enable Github-flavored Markdown line breaks in chatbot messages. If False, single new lines will be ignored. Only applies if `render_markdown` is True.
likeable: Whether the chat messages display a like or dislike button. Set automatically by the .like method but has to be present in the signature for it to show up in the config.
layout: If "panel", will display the chatbot in a llm style layout. If "bubble", will display the chatbot with message bubbles, with the user and bot messages on alterating sides. Will default to "bubble".
placeholder: a placeholder message to display in the chatbot when it is empty. Centered vertically and horizontally in the Chatbot. Supports Markdown and HTML. If None, no placeholder is displayed.
show_copy_all_button: If True, will show a copy all button that copies all chatbot messages to the clipboard.
"""
self.likeable = likeable
if type not in ["messages", "tuples"]:
raise ValueError("type must be 'messages' or 'tuples', received: {type}")
self.type: Literal["tuples", "messages"] = type
if type == "messages":
self.data_model = ChatbotDataMessages
else:
self.data_model = ChatbotDataTuples
self.height = height
self.rtl = rtl
if latex_delimiters is None:
latex_delimiters = [{"left": "$$", "right": "$$", "display": True}]
self.latex_delimiters = latex_delimiters
self.show_share_button = (
(utils.get_space() is not None)
if show_share_button is None
else show_share_button
)
self.render_markdown = render_markdown
self.show_copy_button = show_copy_button
self.sanitize_html = sanitize_html
self.bubble_full_width = bubble_full_width
self.line_breaks = line_breaks
self.layout = layout
self.show_copy_all_button = show_copy_all_button
super().__init__(
label=label,
every=every,
inputs=inputs,
show_label=show_label,
container=container,
scale=scale,
min_width=min_width,
visible=visible,
elem_id=elem_id,
elem_classes=elem_classes,
render=render,
key=key,
value=value,
)
self.avatar_images: list[dict | None] = [None, None]
if avatar_images is None:
pass
else:
self.avatar_images = [
self.serve_static_file(avatar_images[0]),
self.serve_static_file(avatar_images[1]),
]
self.placeholder = placeholder
@staticmethod
def _check_format(messages: list[Any], type: Literal["messages", "tuples"]):
if type == "messages":
all_valid = all(
isinstance(message, dict)
and "role" in message
and "content" in message
or isinstance(message, ChatMessage)
for message in messages
)
if not all_valid:
raise Error(
"Data incompatible with messages format. Each message should be a dictionary with 'role' and 'content' keys or a ChatMessage object."
)
elif not all(
isinstance(message, (tuple, list)) and len(message) == 2
for message in messages
):
raise Error(
"Data incompatible with tuples format. Each message should be a list of length 2."
)
def _preprocess_content(
self,
chat_message: str | FileMessage | ComponentMessage | None,
) -> str | GradioComponent | tuple[str | None] | tuple[str | None, str] | None:
if chat_message is None:
return None
elif isinstance(chat_message, FileMessage):
if chat_message.alt_text is not None:
return (chat_message.file.path, chat_message.alt_text)
else:
return (chat_message.file.path,)
elif isinstance(chat_message, str):
return chat_message
elif isinstance(chat_message, ComponentMessage):
capitalized_component = (
chat_message.component.upper()
if chat_message.component in ("json", "html")
else chat_message.component.capitalize()
)
component = import_component_and_data(capitalized_component)
if component is not None:
instance = component() # type: ignore
if not instance.data_model:
payload = chat_message.value
elif issubclass(instance.data_model, GradioModel):
payload = instance.data_model(**chat_message.value)
elif issubclass(instance.data_model, GradioRootModel):
payload = instance.data_model(root=chat_message.value)
else:
payload = chat_message.value
value = instance.preprocess(payload)
return component(value=value, **chat_message.constructor_args) # type: ignore
else:
raise ValueError(
f"Invalid component for Chatbot component: {chat_message.component}"
)
else:
raise ValueError(f"Invalid message for Chatbot component: {chat_message}")
def _preprocess_messages_tuples(
self, payload: ChatbotDataTuples
) -> list[list[str | tuple[str] | tuple[str, str] | None]]:
processed_messages = []
for message_pair in payload.root:
if not isinstance(message_pair, (tuple, list)):
raise TypeError(
f"Expected a list of lists or list of tuples. Received: {message_pair}"
)
if len(message_pair) != 2:
raise TypeError(
f"Expected a list of lists of length 2 or list of tuples of length 2. Received: {message_pair}"
)
processed_messages.append(
[
self._preprocess_content(message_pair[0]),
self._preprocess_content(message_pair[1]),
]
)
return processed_messages
def preprocess(
self,
payload: ChatbotDataTuples | ChatbotDataMessages | None,
) -> (
list[list[str | tuple[str] | tuple[str, str] | None]] | list[MessageDict] | None
):
"""
Parameters:
payload: data as a ChatbotData object
Returns:
If type is 'tuples', passes the messages in the chatbot as a `list[list[str | None | tuple]]`, i.e. a list of lists. The inner list has 2 elements: the user message and the response message. Each message can be (1) a string in valid Markdown, (2) a tuple if there are displayed files: (a filepath or URL to a file, [optional string alt text]), or (3) None, if there is no message displayed. If type is 'messages', passes the value as a list of dictionaries with 'role' and 'content' keys. The `content` key's value supports everything the `tuples` format supports.
"""
if payload is None:
return payload
if self.type == "tuples":
if not isinstance(payload, ChatbotDataTuples):
raise Error("Data incompatible with the tuples format")
return self._preprocess_messages_tuples(cast(ChatbotDataTuples, payload))
if not isinstance(payload, ChatbotDataMessages):
raise Error("Data incompatible with the messages format")
message_dicts = []
for message in payload.root:
message_dict = cast(MessageDict, message.model_dump())
message_dict["content"] = self._preprocess_content(message.content)
message_dicts.append(message_dict)
return message_dicts
@staticmethod
def _get_alt_text(chat_message: dict | list | tuple | GradioComponent):
if isinstance(chat_message, dict):
return chat_message.get("alt_text")
elif not isinstance(chat_message, GradioComponent) and len(chat_message) > 1:
return chat_message[1]
@staticmethod
def _create_file_message(chat_message, filepath):
mime_type = client_utils.get_mimetype(filepath)
return FileMessage(
file=FileData(path=filepath, mime_type=mime_type),
alt_text=Chatbot._get_alt_text(chat_message),
)
def _postprocess_content(
self,
chat_message: str
| tuple
| list
| FileDataDict
| FileData
| GradioComponent
| None,
) -> str | FileMessage | ComponentMessage | None:
if chat_message is None:
return None
elif isinstance(chat_message, FileMessage):
return chat_message
elif isinstance(chat_message, FileData):
return FileMessage(file=chat_message)
elif isinstance(chat_message, GradioComponent):
component = import_component_and_data(type(chat_message).__name__)
if component:
component = chat_message.__class__(**chat_message.constructor_args)
chat_message.constructor_args.pop("value", None)
config = component.get_config()
return ComponentMessage(
component=type(chat_message).__name__.lower(),
value=config.get("value", None),
constructor_args=chat_message.constructor_args,
props=config,
)
elif isinstance(chat_message, dict) and "path" in chat_message:
filepath = chat_message["path"]
return self._create_file_message(chat_message, filepath)
elif isinstance(chat_message, (tuple, list)):
filepath = str(chat_message[0])
return self._create_file_message(chat_message, filepath)
elif isinstance(chat_message, str):
chat_message = inspect.cleandoc(chat_message)
return chat_message
else:
raise ValueError(f"Invalid message for Chatbot component: {chat_message}")
def _postprocess_messages_tuples(self, value: TupleFormat) -> ChatbotDataTuples:
processed_messages = []
for message_pair in value:
processed_messages.append(
[
self._postprocess_content(message_pair[0]),
self._postprocess_content(message_pair[1]),
]
)
return ChatbotDataTuples(root=processed_messages)
def _postprocess_message_messages(
self, message: MessageDict | ChatMessage
) -> list[Message]:
if isinstance(message, dict):
message["content"] = self._postprocess_content(message["content"])
msg = Message(**message) # type: ignore
elif isinstance(message, ChatMessage):
message.content = self._postprocess_content(message.content) # type: ignore
msg = Message(
role=message.role,
content=message.content, # type: ignore
metadata=message.metadata, # type: ignore
)
else:
raise Error(
f"Invalid message for Chatbot component: {message}", visible=False
)
# extract file path from message
new_messages = []
if isinstance(msg.content, str):
for word in msg.content.split(" "):
filepath = Path(word)
try:
is_file = filepath.is_file() and filepath.exists()
except OSError:
is_file = False
if is_file:
filepath = cast(
str, move_resource_to_block_cache(filepath, block=self)
)
mime_type = client_utils.get_mimetype(filepath)
new_messages.append(
Message(
role=msg.role,
metadata=msg.metadata,
content=FileMessage(
file=FileData(path=filepath, mime_type=mime_type)
),
),
)
return [msg, *new_messages]
def postprocess(
self,
value: TupleFormat | list[MessageDict | Message] | None,
) -> ChatbotDataTuples | ChatbotDataMessages:
"""
Parameters:
value: If type is `tuples`, expects a `list[list[str | None | tuple]]`, i.e. a list of lists. The inner list should have 2 elements: the user message and the response message. The individual messages can be (1) strings in valid Markdown, (2) tuples if sending files: (a filepath or URL to a file, [optional string alt text]) -- if the file is image/video/audio, it is displayed in the Chatbot, or (3) None, in which case the message is not displayed. If type is 'messages', passes the value as a list of dictionaries with 'role' and 'content' keys. The `content` key's value supports everything the `tuples` format supports.
Returns:
an object of type ChatbotData
"""
data_model = cast(
Union[Type[ChatbotDataTuples], Type[ChatbotDataMessages]], self.data_model
)
if value is None:
return data_model(root=[])
if self.type == "tuples":
self._check_format(value, "tuples")
return self._postprocess_messages_tuples(cast(TupleFormat, value))
self._check_format(value, "messages")
processed_messages = [
msg
for message in value
for msg in self._postprocess_message_messages(cast(MessageDict, message))
]
return ChatbotDataMessages(root=processed_messages)
def example_payload(self) -> Any:
if self.type == "messages":
return [
Message(role="user", content="Hello!").model_dump(),
Message(role="assistant", content="How can I help you?").model_dump(),
]
return [["Hello!", None]]
def example_value(self) -> Any:
if self.type == "messages":
return [
Message(role="user", content="Hello!").model_dump(),
Message(role="assistant", content="How can I help you?").model_dump(),
]
return [["Hello!", None]]