Animated fade away gifs from image

This Mediajam showcases how to transform an image to gif as a fadeout using cloudinary api's in combination with typescript Nextjs

Codesandbox

The final project can be viewed on Codesandbox.

<CodeSandbox

title="image-giff-anime"

id="image-giff-anime-nwq1w?"

/>

You can find the full source code on my Github repository.

Project initialization and setup

Initiatilization of next js with typescript project

First we initialize an empty next js project. Using the commands below. The command will prompt you to name your project.

npx create-next-app@latest --typescript
# or
yarn create next-app --typescript

Once initialization is complete you can confirm that initial setup is complete by running.

yarn dev
# OR
npm run dev

After running the above command visit localhost port 3000

Dependancies definations and additions

1. html2canvas
2. cloudinary
3. gifshot
4. styled-components

yarn add html2canvas
yarn add gifshot
yarn add cloudinary
yarn add styled-components

Additonal setup for gifshot

Add the following line of code in ./pages/api/decs.d.ts file To enable import and export of the module within the project

declare module "gifshot";

Development steps

• Setup demonstration image in public folder.
• link the image to an image tag in App class renderer
• Retrieve the html image tag and convert the element to canvas using html2canvas library
• Generate 6 other canvases (Due to storage contrainsts) while fading out the pixels in each canvas.
• Take the generated list of canvases convert them into data urls array
• Combine the images to gif using gifshot library
• Upload the generated video/gif to cloudinary for storage.
• Display the generated fadeout/disintegration gif

link the image to an image tag in App class renderer

<Main>
    <div className="inner">
        {loading ? (
        <div>Processing ... </div>
        ) : (
        <Image
            id="world"
            src={gif_image ? gif_image : `/goat.jpg`}
            alt=""
        />
        )}
        <br />
        {url ? <div>{url}</div> : ""}
        {!loading && <button onClick={this.snap.bind(this)}>Snap</button>}
    </div>
</Main>

Retrieve the html image tag and convert the element to canvas using html2canvas library

// convert img tag to canvas
const canvas = await html2canvas(img as HTMLElement);
const ctx = canvas.getContext("2d");
if (!ctx) return;
// Getting image data from the canvas for pixel manupilation
const image_data = ctx.getImageData(0, 0, canvas.width, canvas.height);
if (!image_data) return;
const pixel_arr = image_data.data;

Generate 6 other canvases while fading out the pixels in each canvas

const image_data_array = this.createBlankImageArray(image_data);
//put pixel info to imageDataArray (Weighted Distributed)
for (let i = 0; i < pixel_arr.length; i++) {
    const p = Math.floor((i / pixel_arr.length) * CANVAS_COUNT);

    const dist = Math.round(Math.random() * (CANVAS_COUNT - 1));

    const a = image_data_array[dist];
    a[i] = pixel_arr[i];
    a[i + 1] = pixel_arr[i + 1];
    a[i + 2] = pixel_arr[i + 2];
    a[i + 3] = pixel_arr[i + 3];
}

Take the generated list of canvases convert them into data urls array

// fadeout image list generation and mapping
const images = new Array(CANVAS_COUNT)
    .fill(0)
    .map((_, i) =>
    this.createCanvasFromImageData(
        image_data_array[i],
        canvas.width,
        canvas.height
    ).toDataURL()
    );

Combine the images to gif using gifshot library

gifshot.createGIF(
    {
        images,
        gifWidth: canvas.width,
        gifHeight: canvas.height,
        numFrames: CANVAS_COUNT
    },
    (obj: any) => {
        if (obj.error) {
            console.log(obj.error);
            return;
        }
        console.log(obj.image);
        this.uploadVideoCloudinary(obj.image);
        this.setState({ gif_image: obj.image, loading: false });
    }
);

Upload the generated video/gif to cloudinary for storage

This article will use cloudinary for media upload and storage. Use this link to access cloudinary website and sign up or login to receive your environment variables(Cloud name , API Key and API Secret). The mentioned variables will be found in your dashboard which should look like below: .

Create a file named .env at the root of your project. The file is where we store our environment variables. Paste the following code in the file

CLOUDINARY_NAME = ""

CLOUDINARY_API_KEY =  ""

CLOUDINARY_API_SECRET= ""

Ensure to fill in the blanks with your respective environment variables then restart the project server for the project to update its envs.

Head to the ./pages/api directory and create a file named upload.tsx. In the file we wil access the cloudinary API to uplload our files and receive the cloudinary file URL.

Start by including the necessary imports

import type { NextApiRequest, NextApiResponse } from 'next'
var cloudinary = require('cloudinary').v2

Intergrate your environment variables with the API backend like below:

cloudinary.config({
    cloud_name: process.env.CLOUDINARY_NAME,
    api_key: process.env.CLOUDINARY_API_KEY,
    api_secret: process.env.CLOUDINARY_API_SECRET,
});

Since we are using typescript, we will include a type system to represent the type of value that will be used in our backend. In our case, the value will be a string.

type Data = {
    name: string
}

We can then inroduce a handler function that receives a post request and processes a response feedback for our front end.

// https://rv7py.sse.codesandbox.io/
export default async function handler(
  req: NextApiRequest,
  res: NextApiResponse<Data>
) {
  if (req.method === "POST") {
    let fileStr: string = req.body.data;
    let uploadResponse: any;

    try {
      uploadResponse = await cloudinary.uploader.upload_large(fileStr, {
        resource_type: "auto",
        chunk_size: 6000000,
        timeout: 60000
      });
      console.log(uploadResponse);
    } catch (err) {
      console.log(err);
    }
    res.status(200).json({ name: "" + uploadResponse.secure_url });
  }
}

The code above receives the request body and uploads it to cloudinary. Ith then captures the sent file's cloudinary URL and sends it back to the front end as a response.

Thats it! we've completed our project backend.