# Adding Components to `pydase` This guide provides a step-by-step process for adding new components to the `pydase` package. Components in `pydase` consist of both backend (Python) and frontend (React) parts. They work together to create interactive and dynamic data services. ## Overview A component in `pydase` is a unique combination of a backend class (e.g., `Image`) and its corresponding frontend React component. The backend class stores the attributes needed for the component, and possibly methods for setting those in the backend, while the frontend part is responsible for rendering and interacting with the component. ## Adding a Backend Component to `pydase` Backend components belong in the `src/pydase/components` directory. ### Step 1: Create a New Python File in the Components Directory Navigate to the `src/pydase/components` directory and create a new Python file for your component. The name of the file should be descriptive of the component's functionality. For example, for a `Image` component, create a file named `image.py`. ### Step 2: Define the Backend Class Within the newly created file, define a Python class representing the component. This class should inherit from `DataService` and contains the attributes that the frontend needs to render the component. Every public attribute defined in this class will synchronise across the clients. It can also contain methods which can be used to interact with the component from the backend. For the `Image` component, the class may look like this: ```python # file: pydase/components/image.py from pydase.data_service.data_service import DataService class Image(DataService): def __init__( self, image_representation: bytes = b"", ) -> None: self.image_representation = image_representation super().__init__() # need to decode the bytes def __setattr__(self, __name: str, __value: Any) -> None: if __name == "value": if isinstance(__value, bytes): __value = __value.decode() return super().__setattr__(__name, __value) ``` So, changing the `image_representation` will push the updated value to the browsers connected to the service. ### Step 3: Register the Backend Class The component should be added to the `__init__.py` file to ensure `pydase` handles them properly: ```python # file: pydase/components/__init__.py from pydase.components.image import Image from pydase.components.number_slider import NumberSlider __all__ = [ "NumberSlider", "Image", # add the new components here ] ``` ### Step 4: Implement Necessary Methods (Optional) If your component requires specific logic or methods, implement them within the class. Document any public methods or attributes to ensure that other developers understand their purpose and usage. ### Step 5: Write Tests for the Component (Recommended) Consider writing unit tests for the component to verify its behavior. Place the tests in the appropriate directory within the `tests` folder. For example, a test for the `Image` component could look like this: ```python from pytest import CaptureFixture from pydase.components.image import Image from pydase.data_service.data_service import DataService def test_Image(capsys: CaptureFixture) -> None: class ServiceClass(DataService): image = Image() service = ServiceClass() # ... ``` ## Adding a Frontend Component to `pydase` Frontend components in `pydase` live in the `frontend/src/components/` directory. Follow these steps to create and add a new frontend component: ### Step 1: Create a New React Component File in the Components Directory Navigate to the `frontend/src/components/` directory and create a new React component file for your component. The name of the file should be descriptive of the component's functionality and reflect the naming conventions used in your project. For example, for an `Image` component, create a file named `ImageComponent.tsx`. ### Step 2: Write the React Component Code Write the React component code, following the structure and patterns used in existing components. Make sure to import necessary libraries and dependencies. For example, for the `Image` component, a template could look like this: ```tsx import { emit_update } from '../socket'; // use this when your component should update values in the backend import { DocStringComponent } from './DocStringComponent'; import React, { useEffect, useRef, useState } from 'react'; import { Card, Collapse, Image } from 'react-bootstrap'; import { DocStringComponent } from './DocStringComponent'; import { ChevronDown, ChevronRight } from 'react-bootstrap-icons'; interface ImageComponentProps { name: string; parentPath: string; readOnly: boolean; docString: string; addNotification: (string) => void; // Define your component specific props here value: string; format: string; } export const ImageComponent = React.memo((props: ImageComponentProps) => { const { name, parentPath, value, docString, format, addNotification } = props; const renderCount = useRef(0); const [open, setOpen] = useState(true); // add this if you want to expand/collapse your component useEffect(() => { renderCount.current++; }); // This will trigger a notification if notifications are enabled. useEffect(() => { addNotification(`${parentPath}.${name} changed to ${value}.`); }, [props.value]); // Your component logic here return (
{/* Add the Card and Collapse components here if you want to be able to expand and collapse your component. */} setOpen(!open)} style={{ cursor: 'pointer' }} // Change cursor style on hover > {name} {open ? : } {process.env.NODE_ENV === 'development' && (

Render count: {renderCount.current}

)} {/* Your component TSX here */}
); }); ``` ### Step 3: Emitting Updates to the Backend Often, React components in the frontend will need to send updates to the backend, especially when user interactions result in a change of state or data. In `pydase`, we use `socketio` to seamlessly communicate these changes. Here's a detailed guide on how to emit update events from your frontend component: 1. **Setting Up Emission**: Ensure you've imported the required functions and methods for emission. The main function we'll use for this is `emit_update` from the `socket` module: ```tsx import { emit_update } from '../socket'; ``` 2. **Understanding the Emission Parameters**: When emitting an update, we send three main pieces of data: - `parentPath`: This is the access path for the parent object of the attribute to be updated. This forms the basis to create the full access path for the attribute. For instance, for the attribute access path `attr1.list_attr[0].attr2`, `attr1.list_attr[0]` would be the `parentPath`. - `name`: This represents the name of the attribute to be updated within the `DataService` instance. If the attribute is part of a nested structure, this would be the name of the attribute in the last nested object. So, for `attr1.list_attr[0].attr2`, `attr2` would be the name. - `value`: This is the new value intended for the attribute. Ensure that the type of this value matches the type of the attribute in the backend. 3. **Implementing the Emission**: To illustrate the emission process, let's consider the `ButtonComponent`. When the button state changes, we want to send this update to the backend: ```tsx // ... (other imports) export const ButtonComponent = React.memo((props: ButtonComponentProps) => { // ... const { name, parentPath, value } = props; const setChecked = (checked: boolean) => { emit_update(name, parentPath, checked); }; return ( setChecked(e.currentTarget.checked)}>

{name}

); }); ``` In this example, whenever the button's checked state changes (`onChange` event), we invoke the `setChecked` method, which in turn emits the new state to the backend using `emit_update`. ### Step 4: Add the New Component to the GenericComponent The `GenericComponent` is responsible for rendering different types of components based on the attribute type. You can add the new `ImageComponent` to the `GenericComponent` by following these sub-steps: #### 1. Import the New Component At the beginning of the `GenericComponent` file, import the newly created `ImageComponent`: ```tsx // file: frontend/src/components/GenericComponent.tsx import { ImageComponent } from './ImageComponent'; ``` #### 2. Update the AttributeType Update the `AttributeType` type definition to include the new type for the `ImageComponent`. For example, if the new attribute type is `'Image'` (which should correspond to the name of the backend component class), you can add it to the union: ```tsx type AttributeType = | 'str' | 'bool' | 'float' | 'int' | 'Quantity' | 'list' | 'method' | 'DataService' | 'Enum' | 'NumberSlider' | 'Image'; // Add the name of the backend component class here ``` #### 3. Add a Conditional Branch for the New Component Inside the `GenericComponent` function, add a new conditional branch to render the `ImageComponent` when the attribute type is `'Image'`: ```tsx } else if (attribute.type === 'Image') { return ( ); } else { // other code ``` Make sure to update the props passed to the `ImageComponent` based on its specific requirements. ### Step 5: Adding Custom Notification Message (Optional) In some cases, you may want to provide a custom notification message to the user when an attribute of a specific type is updated. This can be useful for enhancing user experience and providing contextual information about the changes. For example, updating an `Image` component corresponds to setting a very long string. We don't want to display the whole string in the notification but just notify the user that the image was updated (and maybe also the format). To create a custom notification message, you can update the message passed to the `addNotification` method in the `useEffect` hook in the component file file. For the `ImageComponent`, this could look like this: ```tsx useEffect(() => { addNotification(`${parentPath}.${name} changed.`); }, [props.value]); ``` However, you might want to use the `addNotification` at different places. For an example, see the [MethodComponent](../../frontend/src/components/MethodComponent.tsx). ### Step 6: Write Tests for the Component (TODO) Test the frontend component to ensure that it renders correctly and interacts seamlessly with the backend. Consider writing unit tests using a testing library like Jest or React Testing Library, and manually test the component in the browser.