DEV Community

Arnab Sen
Arnab Sen

Posted on • Edited on • Originally published at arnabsen.dev

Using Google Drive API v3 to upload a file to drive (in React)

#tutorial #react #javascript

Google Drive upload w/ Google API in React

Objective

To save text data as a Google Document in your drive.

Main Steps

  1. Authorise your app and enable google api
  2. Writing the code for user authentication
  3. Finally, for the uploading the data in google drive

Authorise your app and enable google api

The google docs for this step is enough to get through.
Here is the link

Some key points to remember:

  1. First create the API key in the credentials and then the Oauth Client Id
  2. Make sure you specify the website from which we will sending the requests, else you will receive CORs issue. Also you cannot use IP, if you working on localhost specify localhost:<PORT>

After this step you will have two things:

  1. API_KEY
  2. Client_ID

Code for user Authentication

We will be using OAuth2.0 cause that's the only thing google allows.
Let's explain what will happen in this step.

  1. User will Sign In
  2. User will be asked if he/she authorises this app
  3. Once user gives a consent we will recieve a token and make further request.

Regarding oauth2.0 there is a lot of theory here

Coming to implementation:

We need to specify some script so that we can use the gapi.client. Now if you are in your local machine using
gapi.client may give you undefined. Instead you should use the window.gapi.client.
There is one github issue regarding this.

Coming back in react we will append this script to the body inside the componentDidMount() function

componentDidMount(){
    var script = document.createElement('script');
    script.onload=this.handleClientLoad;
    script.src="https://apis.google.com/js/api.js";
    document.body.appendChild(script);
  }
Enter fullscreen mode Exit fullscreen mode

Function handleClientLoad will load the gapi.client for us.

handleClientLoad = ()=>{
    window.gapi.load('client:auth2', this.initClient);
  }
Enter fullscreen mode Exit fullscreen mode

As a callback we specify initClient where we initialise the gapi.client
The call to gapi.client.init specifies the following fields:

  1. API_KEY and CLIENT_ID : These specify your application's authorisation credentials. We have got these from the previous step.
  2. Scope: It specifies a space-delimited list of access scopes that correspond to the resources that your application could access on the user's behalf. Here is a list of scopes. For our purpose we would need this https://www.googleapis.com/discovery/v1/apis/drive/v3/rest.
  3. DiscoveryDocs: It identifies a list of API Discovery documents that your application uses. In this example, the code retrieves the discovery document for version 3 of the Google Drive API, https://www.googleapis.com/discovery/v1/apis/drive/v3/rest

Put these at the top

var  SCOPE  =  'https://www.googleapis.com/auth/drive.file';
var  discoveryUrl  =  'https://www.googleapis.com/discovery/v1/apis/drive/v3/rest';
Enter fullscreen mode Exit fullscreen mode

So, now let's define out initClient function.

initClient = () => {
    try{
      window.gapi.client.init({
          'apiKey': "<YOUR API KEY>",
          'clientId': "<YOUR CLIENT ID>",
          'scope': SCOPE,
          'discoveryDocs': [discoveryUrl]
        }).then(() => {
          this.setState({
            googleAuth: window.gapi.auth2.getAuthInstance()
          })
          this.state.googleAuth.isSignedIn.listen(this.updateSigninStatus);  


         document.getElementById('sign
         document.getElementById('signout-btn').addEventListener('click', this.signOutFunction);

      });
    }catch(e){
      console.log(e);
    }
  }
Enter fullscreen mode Exit fullscreen mode

Once the client is initialised we get an Auth Instance and save it in a state variable, googltAuth. The updateSigninStatus function is a listener that listens for changes to the user's authorization status. Also we add some functions to the sign in and sign out buttons. So, before moving on we need to specify our states.

  state = {
    name: '',
    googleAuth: ''
  }
Enter fullscreen mode Exit fullscreen mode

The name variable is for other purpose, we will come later. Now if the user clicks on sign in button the signInFunction will be triggered.

signInFunction =()=>{
    this.state.googleAuth.signIn();
    this.updateSigninStatus()
  }
Enter fullscreen mode Exit fullscreen mode

Since after sign in the state changes we will explicitly call the updateSigninStatus() function. The signout function does something very similar.

signOutFunction =()=>{
    this.state.googleAuth.signOut();
    this.updateSigninStatus()
  }
Enter fullscreen mode Exit fullscreen mode

Now let's come to updateSignStatus(). All it does is fetch some user details(here, the name and this is where we use the name state variable).

updateSignStatus = async ()=>{
    var user = this.state.googleAuth.currentUser.get();
    if (user.wc == null){
      this.setState({
        name: ''
      });
    }
    else{
      var isAuthorized = user.hasGrantedScopes(SCOPE);
      if(isAuthorized){
        this.setState({
          name: user.Ot.Cd
        });
        //we will put the code of the third step here
      }
    }
  }
Enter fullscreen mode Exit fullscreen mode

Important thing to note here is that isAuthorized is true only if the user grants the permissions to the app. Now once we are done upto here, now we can move to the final step of uploading the file.

Uploading the data in google drive

For uploading the data we have various methods. In our case we will use the Multipart method because we will not just create a file but also specify the meta data as well. All the code snippets in this step will be inside the region specified in the above step.

The steps mentioned in the google docs are:

  1. Create a POST request to the method's /upload URI with the query parameter of uploadType=multipart:

    POST https://www.googleapis.com/upload/drive/v3/files?uploadType=multipart

  2. Create the body of the request. Format the body according to the multipart/related content type [RFC 2387], which contains two parts:

    . Metadata. The metadata must come first and must have a Content-Type header set to application/json; charset=UTF-8. Add the file's metadata in JSON format.
    . Media. The media must come second and must have a Content-Type header of any MIME type. Add the file's data to the media part.

    Identify each part with a boundary string, preceded by two hyphens. In addition, add two hyphens after the final boundary string.

  3. Add these top-level HTTP headers:

    . Content-Type. Set to multipart/related and include the boundary string you're using to identify the different parts of the request. For example: Content-Type: multipart/related; boundary=foo_bar_baz
    . Content-Length. Set to the total number of bytes in the request body.

Send the request.

So, let's create the metadata of the file

var fileName='mychat123';
var fileData='this is a sample data';
var contentType='text/plain'
var metadata = {
      'name': fileName,
      'mimeType': contentType
};
Enter fullscreen mode Exit fullscreen mode

You can change the fileName and fileData and also change the contentType accordingly, it will hold the MIME type of the data you will upload to drive.

Now the multipart body. It follows a particular standardisation, you can read more about it here

Without going into much details just copy the following.

const boundary='<ANY RANDOM STRING>'
const delimiter = "\r\n--" + boundary + "\r\n";
const close_delim = "\r\n--" + boundary + "--";
Enter fullscreen mode Exit fullscreen mode

here boundary will differentiate between the various part of the request body.

var multipartRequestBody =
          delimiter +
          'Content-Type: application/json; charset=UTF-8\r\n\r\n' +
          JSON.stringify(metadata) +
          delimiter +
          'Content-Type: ' + contentType + '\r\n\r\n' +
          fileData+'\r\n'+
          close_delim;
Enter fullscreen mode Exit fullscreen mode

This is a format and it needs to be followed. After this all we are left with is sending the request which we will do using the gapi.client.request this will handle the Auth Token automatically.

var request = window.gapi.client.request({
            'path': 'https://www.googleapis.com/upload/drive/v3/files',
            'method': 'POST',
            'params': {'uploadType': 'multipart'},
            'headers': {
              'Content-Type': 'multipart/related; boundary=' + boundary + ''
            },
            'body': multipartRequestBody});
request.execute(callback);
Enter fullscreen mode Exit fullscreen mode

Now we are DONE!!.
To compile it all this was my App.js

import React, { Component } from 'react';

var SCOPE = 'https://www.googleapis.com/auth/drive.file';
var discoveryUrl = 'https://www.googleapis.com/discovery/v1/apis/drive/v3/rest';


class App extends Component {
  state = {
    name: '',
    googleAuth: ''
  }
  componentDidMount(){
    var script = document.createElement('script');
    script.onload=this.handleClientLoad;
    script.src="https://apis.google.com/js/api.js";
    document.body.appendChild(script);
  }


  initClient = () => {
    try{
      window.gapi.client.init({
          'apiKey': "",
          'clientId': "",
          'scope': SCOPE,
          'discoveryDocs': [discoveryUrl]
        }).then(() => {
          this.setState({
            googleAuth: window.gapi.auth2.getAuthInstance()
          })
          this.state.googleAuth.isSignedIn.listen(this.updateSigninStatus);  
         document.getElementById('signin-btn').addEventListener('click', this.signInFunction);
         document.getElementById('signout-btn').addEventListener('click', this.signOutFunction);

      });
    }catch(e){
      console.log(e);
    }
  }


  signInFunction =()=>{
    this.state.googleAuth.signIn();
    this.updateSigninStatus()
  }

  signOutFunction =()=>{
    this.state.googleAuth.signOut();
    this.updateSigninStatus()
  }

  updateSigninStatus = ()=> {
    this.setSigninStatus();
  }


  setSigninStatus= async ()=>{
    var user = this.state.googleAuth.currentUser.get();
    console.log(user)
    if (user.wc == null){
      this.setState({
        name: ''
      });
    }
    else{
      var isAuthorized = user.hasGrantedScopes(SCOPE);
      if(isAuthorized){
        this.setState({
          name: user.Ot.Cd
        });
        const boundary='foo_bar_baz'
        const delimiter = "\r\n--" + boundary + "\r\n";
        const close_delim = "\r\n--" + boundary + "--";
        var fileName='mychat123';
        var fileData='this is a sample data';
        var contentType='text/plain'
        var metadata = {
          'name': fileName,
          'mimeType': contentType
        };

        var multipartRequestBody =
          delimiter +
          'Content-Type: application/json; charset=UTF-8\r\n\r\n' +
          JSON.stringify(metadata) +
          delimiter +
          'Content-Type: ' + contentType + '\r\n\r\n' +
          fileData+'\r\n'+
          close_delim;

          console.log(multipartRequestBody);
          var request = window.gapi.client.request({
            'path': 'https://www.googleapis.com/upload/drive/v3/files',
            'method': 'POST',
            'params': {'uploadType': 'multipart'},
            'headers': {
              'Content-Type': 'multipart/related; boundary=' + boundary + ''
            },
            'body': multipartRequestBody});
        request.execute(function(file) {
          console.log(file)
        });
      }
    }
  }

  handleClientLoad = ()=>{
    window.gapi.load('client:auth2', this.initClient);
  }
  render() {
    return (
      <div className="App">
        <div>UserName: <strong>{ this.state.name}</strong></div>
        <button id="signin-btn">Sign In</button>
        <button id="signout-btn">Sign Out</button>
      </div>
    );
  }
}

export default App;
Enter fullscreen mode Exit fullscreen mode

I had to go through a lot of documentation of Google APIs to be able to come to this. I tried many other approaches but at this was the one that worked for me. If you are stuck at any point do check out the Oauth Playground

Refs

  1. Upload Files Doc
  2. File Create
  3. JSON API multipart
  4. MIME Multipart/Related Content-type
  5. Other MIME types

Top comments (8)

Collapse
 
kevincassidy profile image
Kevin Cassidy •

Hi Arnab, great article and very clear.

Using your code, I'm able to do exactly what you did - create a plain text file with the content 'this is a sample data'.

However, what I'm trying to do is upload an Excel file, and it does create a file but the content is corrupt. Have you managed to upload a file like that?

Collapse
 
taninjs profile image
taninjs •

I managed to upload picture and video by modifying this snippet
gist.github.com/tanaikech/bd53b366...

Hope this help!

Collapse
 
arnabsen1729 profile image
Arnab Sen •

Did you change the mime types accordingly?

Try using the Oauth playground I linked. You will get the exact error message.

Collapse
 
kevincassidy profile image
Kevin Cassidy •

Hi Arnab,

Yes from what I can see I've followed your instructions - and the file does get uploaded - it just seems to be corrupt for some reason.

I've tried with an image and got the same result.

This is how it looks (I've tried making tweaks to what you provided but same result):

--foo_bar_baz
content-disposition: form-data; name="metadata"; filename="test.png"
content-type: application/json

{"name":"test.png","mimeType":"image/png"}
--foo_bar_baz
content-disposition: form-data; name="file"; filename="test.png"
content-type: image/png

data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAlgAAAJYCAMAAACJuGjuAAACdlBMVEUBCQMcHBIZHA8dGw0XGhAVGA4UFwwBBgICDAMDDwQSFgsPGA4eHA8QFAgXGQ0bGAweHBMOEgcXFwscHBAFEQYlHhQLFQsMEgYlHBIoHA4VFwkc.. etc
--foo_bar_baz--

And I get my response like so:
{
"kind": "drive#file",
"id": "13Htxa1Lcfk_VyDDtZPj0yYBrs2_zwO4u",
"name": "test.png",
"mimeType": "image/png"
}

Let me know if you have any ideas.
Thanks,
Kevin

Collapse
 
stevescruz profile image
Steve Cruz •

This is exactly what I needed. I'll try to implement this, let's see what challenges arise as I code this.

Collapse
 
arnabsen1729 profile image
Arnab Sen •

Sure. I found it really challenging and not much resources were available for these. Hope it helps.

Collapse
 
gautham495 profile image
Gautham Vijayan •

Superb post Arnab!

Collapse
 
arnabsen1729 profile image
Arnab Sen •

Thanks a lot @gautham !! ☺️☺️

Read next

prince_beec5ccde00b7c6c73 profile image

👻 Scary Ghost Cursor with Smoke Trail! 💀 code using html5,css3 and javascript

Prince -

shams_arman_510 profile image

BLA ! BLA! BLA! DECEMBER...!

shams Arman siddique -

velan profile image

Mastering Bruno: A Beginner's Guide to API Clients

Velan<> -

danrot90 profile image

Writing high quality tests

Daniel Rotter -