Integrating Cloud Natural Language API in Android

anchit

10 August 2022

In this blog, we will learn how to integrate Cloud Natural Language API in Android application.

Natural language processing (NLP) describes the interaction between human language and computers.

It’s a technology that many people use daily and has been around for years, but is often taken for granted.

While NLP has uncountable use cases and benefits, some of the cases where NLP can be used are :

Language translation
Data analysis
Text analytics
Search results
Predictive text

Well still if you want to learn more about NLP, then you can start understanding it by reading from internet/books and there is a lot of material available.

Now let’s have a look at what exactly is “Cloud Natural Language API”?

Natural Language uses machine learning to reveal the structure and meaning of text. You can extract information about people, places, and events, and better understand social media sentiment and customer conversations. Natural Language enables you to analyze text and also integrate it with your document storage on Google Cloud Storage.

–> Google Cloud NLP Documentation

Now, let’s see how we can use this Powerful API and include it in our android application.

Before we begin:

You will need to create your Google Application Credentials in a JSON file format.

If you are new to using Google Cloud Platform, simply login with your Gmail account over here –> https://console.cloud.google.com.

Create a new project.
Enable the Google Natural Language API for that project.
Create a service account.
Download a private key as JSON.

After this, keep this Credential JSON file with you, we will be needing it very soon.

Approach:

The approach to complete this integration is quite simple and easy to understand.

Firstly, save the Credential JSON file in your android project under the res directory in a folder named raw like shown in the screenshot below :
Now, we will create AccessTokenLoader.java class that is an extension of AsyncTaskLoader.java class. this class will help us to extract the credentials from our JSON file and set it in the Cloud Natural Language API request that we will be making.
After that, we will create a very simple page with an edit text, a button, and a text view.
In onCreate method, we will simply prepare Cloud Natural Language API with the credentials and the request to be made.
On Click of the button, we will simply call the Cloud Natural Language API to get the results.
The obtained results will be simply displayed in the Text view.

Let’s quickly start with the code.

CODE :

Firstly, add the lines below in your module-level build.gradle file (Path to the file –> Project–>app–> build.gradle)

// Dependencies for Google API Client Libraries
    api("com.google.http-client:google-http-client:1.31.0") {
        exclude module: 'httpclient'
        exclude module: 'jsr305'
    }
    api ("com.google.api-client:google-api-client-android:1.30.2"){
        exclude module: 'httpclient'
        exclude module: 'jsr305'
    }
    api("com.google.apis:google-api-services-language:v1-rev477-1.25.0") {
        exclude module: 'httpclient'
        exclude module: 'jsr305'
    }

// Dependencies for Google API Client Libraries

api("com.google.http-client:google-http-client:1.31.0") {

exclude module: 'httpclient'

exclude module: 'jsr305'

}

api ("com.google.api-client:google-api-client-android:1.30.2"){

exclude module: 'httpclient'

exclude module: 'jsr305'

}

api("com.google.apis:google-api-services-language:v1-rev477-1.25.0") {

exclude module: 'httpclient'

exclude module: 'jsr305'

}

AccessTokenLoader.java class –>

public class AccessTokenLoader extends AsyncTaskLoader<String> {

    private static final String TAG = "AccessTokenLoader";

    private static final String PREFS = "AccessTokenLoader";
    private static final String PREF_ACCESS_TOKEN = "access_token";

    public AccessTokenLoader(Context context) {
        super(context);
    }

    @Override
    protected void onStartLoading() {
        forceLoad();
    }

    @Override
    public String loadInBackground() {

        final SharedPreferences prefs =
                getContext().getSharedPreferences(PREFS, Context.MODE_PRIVATE);
        String currentToken = prefs.getString(PREF_ACCESS_TOKEN, null);

        // Check if the current token is still valid for a while
        if (currentToken != null) {
            final GoogleCredential credential = new GoogleCredential()
                    .setAccessToken(currentToken)
                    .createScoped(CloudNaturalLanguageScopes.all());
            final Long seconds = credential.getExpiresInSeconds();
            if (seconds != null && seconds > 3600) {
                return currentToken;
            }
        }

        // ***** WARNING *****
        // In this sample, we load the credential from a JSON file stored in a raw resource folder
        // of this client app. You should never do this in your app. Instead, store the file in your
        // server and obtain an access token from there.
        // *******************
        final InputStream stream = getContext().getResources().openRawResource(R.raw.credential);
        try {
            final GoogleCredential credential = GoogleCredential.fromStream(stream)
                    .createScoped(CloudNaturalLanguageScopes.all());
            credential.refreshToken();
            final String accessToken = credential.getAccessToken();
            prefs.edit().putString(PREF_ACCESS_TOKEN, accessToken).apply();
            return accessToken;
        } catch (IOException e) {
            Log.e(TAG, "Failed to obtain access token.", e);
        }
        return null;
    }

}

public class AccessTokenLoader extends AsyncTaskLoader<String> {

private static final String TAG = "AccessTokenLoader";

private static final String PREFS = "AccessTokenLoader";

private static final String PREF_ACCESS_TOKEN = "access_token";

public AccessTokenLoader(Context context) {

super(context);

}

@Override

protected void onStartLoading() {

forceLoad();

}

@Override

public String loadInBackground() {

final SharedPreferences prefs =

getContext().getSharedPreferences(PREFS, Context.MODE_PRIVATE);

String currentToken = prefs.getString(PREF_ACCESS_TOKEN, null);

// Check if the current token is still valid for a while

if (currentToken != null) {

final GoogleCredential credential = new GoogleCredential()

.setAccessToken(currentToken)

.createScoped(CloudNaturalLanguageScopes.all());

final Long seconds = credential.getExpiresInSeconds();

if (seconds != null && seconds > 3600) {

return currentToken;

}

// ***** WARNING *****

// In this sample, we load the credential from a JSON file stored in a raw resource folder

// of this client app. You should never do this in your app. Instead, store the file in your

// server and obtain an access token from there.

// *******************

final InputStream stream = getContext().getResources().openRawResource(R.raw.credential);

try {

final GoogleCredential credential = GoogleCredential.fromStream(stream)

.createScoped(CloudNaturalLanguageScopes.all());

credential.refreshToken();

final String accessToken = credential.getAccessToken();

prefs.edit().putString(PREF_ACCESS_TOKEN, accessToken).apply();

return accessToken;

} catch (IOException e) {

Log.e(TAG, "Failed to obtain access token.", e);

}

return null;

}

main_activity.xml file –>

<?xml version="1.0" encoding="utf-8"?>
<LinearLayout xmlns:android="http://schemas.android.com/apk/res/android"
    xmlns:app="http://schemas.android.com/apk/res-auto"
    xmlns:tools="http://schemas.android.com/tools"
    android:layout_width="match_parent"
    android:layout_height="match_parent"
    android:orientation="vertical"
    tools:context=".MainActivity">

    <TextView
        android:id="@+id/label_tv"
        android:layout_width="match_parent"
        android:layout_height="wrap_content"
        android:layout_gravity="center"
        android:gravity="center"
        android:text="@string/enter_the_text_to_analyze_below" />

    <androidx.appcompat.widget.AppCompatEditText
        android:id="@+id/text_et"
        android:layout_width="match_parent"
        android:layout_height="wrap_content"
        android:layout_gravity="center"
        android:layout_margin="8dp"
        android:hint="@string/your_text_goes_here" />

    <Button
        android:layout_width="wrap_content"
        android:layout_height="wrap_content"
        android:layout_gravity="center"
        android:onClick="startAnalysis"
        android:text="@string/start_analysis" />

    <androidx.core.widget.NestedScrollView
        android:id="@+id/nsv"
        android:layout_width="match_parent"
        android:layout_height="wrap_content"
        android:layout_gravity="center"
        android:visibility="gone"
        tools:visibility="visible"
        >

        <LinearLayout
            android:layout_width="match_parent"
            android:layout_height="wrap_content"
            android:orientation="vertical"
            android:layout_margin="8dp"
            android:padding="4dp"
            >

            <TextView
                android:layout_width="match_parent"
                android:layout_height="wrap_content"
                android:text="@string/result_will_be_displayed_below"
                android:layout_marginBottom="8dp"
                />


            <TextView
                android:id="@+id/result_tv"
                android:layout_width="match_parent"
                android:layout_height="wrap_content"
                android:text="@string/result"

                />

        </LinearLayout>


    </androidx.core.widget.NestedScrollView>


</LinearLayout>

<?xml version="1.0" encoding="utf-8"?>

<LinearLayout xmlns:android="http://schemas.android.com/apk/res/android"

xmlns:app="http://schemas.android.com/apk/res-auto"

xmlns:tools="http://schemas.android.com/tools"

android:layout_width="match_parent"

android:layout_height="match_parent"

android:orientation="vertical"

tools:context=".MainActivity">

<TextView

android:id="@+id/label_tv"

android:layout_width="match_parent"

android:layout_height="wrap_content"

android:layout_gravity="center"

android:gravity="center"

android:text="@string/enter_the_text_to_analyze_below" />

<androidx.appcompat.widget.AppCompatEditText

android:id="@+id/text_et"

android:layout_width="match_parent"

android:layout_height="wrap_content"

android:layout_gravity="center"

android:layout_margin="8dp"

android:hint="@string/your_text_goes_here" />

<Button

android:layout_width="wrap_content"

android:layout_height="wrap_content"

android:layout_gravity="center"

android:onClick="startAnalysis"

android:text="@string/start_analysis" />

<androidx.core.widget.NestedScrollView

android:id="@+id/nsv"

android:layout_width="match_parent"

android:layout_height="wrap_content"

android:layout_gravity="center"

android:visibility="gone"

tools:visibility="visible"

<LinearLayout

android:layout_width="match_parent"

android:layout_height="wrap_content"

android:orientation="vertical"

android:layout_margin="8dp"

android:padding="4dp"

<TextView

android:layout_width="match_parent"

android:layout_height="wrap_content"

android:text="@string/result_will_be_displayed_below"

android:layout_marginBottom="8dp"

<TextView

android:id="@+id/result_tv"

android:layout_width="match_parent"

android:layout_height="wrap_content"

android:text="@string/result"

</LinearLayout>

</androidx.core.widget.NestedScrollView>

</LinearLayout>

MainActivity.java file –>

public class MainActivity extends AppCompatActivity {
    EditText editTextView; // View to get the text to be analyzed
    TextView resultTextView; // View to display the results obtained after analysis
    NestedScrollView nestedScrollView; // Wrapper view so that if the results are getting out of bounds from screen so the user can scroll and see complete results
    private static final int LOADER_ACCESS_TOKEN = 1; // Token used to initiate the request loader
    private GoogleCredential mCredential = null; //GoogleCredential object so that the requests for NLP Api could be made

    // A  Thread on which the Api request will be made and results will be delivered. As network calls cannot be made on the amin thread, so we are creating a separate thread for the network calls
    private Thread mThread;

    // Google Request for the NLP Api. This actually is acting like a Http client queue that will process each request and response from the Google Cloud server
    private final BlockingQueue<CloudNaturalLanguageRequest<? extends GenericJson>> mRequests
            = new ArrayBlockingQueue<>(3);


    // Api for CloudNaturalLanguage from the Google Client library, this is the instance of our request that we will make to analyze the text.
    private CloudNaturalLanguage mApi = new CloudNaturalLanguage.Builder(
            new NetHttpTransport(),
            JacksonFactory.getDefaultInstance(),
            new HttpRequestInitializer() {
                @Override
                public void initialize(HttpRequest request) throws IOException {
                    mCredential.initialize(request);
                }
            }).build();


    @Override
    protected void onCreate(Bundle savedInstanceState) {
        super.onCreate(savedInstanceState);
        setContentView(R.layout.activity_main);
        editTextView = (EditText) findViewById(R.id.text_et);
        resultTextView = (TextView) findViewById(R.id.result_tv);
        nestedScrollView = (NestedScrollView) findViewById(R.id.nsv);
        prepareApi();
    }

    /**
     * Method called on the click of the Button
     * @param view -> the view which is clicked
     */
    public void startAnalysis(View view) {
        String textToAnalyze = editTextView.getText().toString();
        if (TextUtils.isEmpty(textToAnalyze)) {
            editTextView.setError(getString(R.string.empty_text_error_msg));
        } else {
            editTextView.setError(null);
            analyzeTextUsingCloudNLPApi(textToAnalyze);
        }
    }

    /**
     * This function will send the text to Cloud Api for analysis
     * @param text -> String to be analyzed
     */
    public void analyzeTextUsingCloudNLPApi(String text) {
        try {
            mRequests.add(mApi
                    .documents()
                    .annotateText(new AnnotateTextRequest()
                            .setDocument(new Document()
                                    .setContent(text)
                                    .setType("PLAIN_TEXT"))
                            .setFeatures(new Features()
                                    .setExtractSyntax(true)
                                    .setExtractEntities(true)
                            )));
        } catch (IOException e) {
            Log.e("TAG", "Failed to create analyze request.", e);
        }
    }


    /**
     * Preparing the Cloud Api before maiking the actual request.
     * This method will actually initiate the AccessTokenLoader async task on completion
     * of which we will recieve the token that should be set in our request for Cloud NLP Api.
     */
    private void prepareApi() {
        // Initiate token refresh
        getSupportLoaderManager().initLoader(LOADER_ACCESS_TOKEN, null,
                new LoaderManager.LoaderCallbacks<String>() {
                    @Override
                    public Loader<String> onCreateLoader(int id, Bundle args) {
                        return new AccessTokenLoader(MainActivity.this);
                    }

                    @Override
                    public void onLoadFinished(Loader<String> loader, String token) {
                        setAccessToken(token);
                    }

                    @Override
                    public void onLoaderReset(Loader<String> loader) {
                    }
                });
    }


    /**
     * This method will set the token from the Credentials.json file to the Google credential object.
     * @param token -> token recieved from the Credentials.json file.
     */
    public void setAccessToken(String token) {
        mCredential = new GoogleCredential()
                .setAccessToken(token)
                .createScoped(CloudNaturalLanguageScopes.all());
        startWorkerThread();
    }


    /**
     * This method will actually initiate a Thread and on this thread we will execute our Api request
     * and responses.
     *
     * Responses recieved will be delivered from here.
     */
    private void startWorkerThread() {
        if (mThread != null) {
            return;
        }
        mThread = new Thread(new Runnable() {
            @Override
            public void run() {
                while (true) {
                    if (mThread == null) {
                        break;
                    }
                    try {
                        // API calls are executed here in this worker thread
                        deliverResponse(mRequests.take().execute());
                    } catch (InterruptedException e) {
                        Log.e("TAG", "Interrupted.", e);
                        break;
                    } catch (IOException e) {
                        Log.e("TAG", "Failed to execute a request.", e);
                    }
                }
            }
        });
        mThread.start();
    }


    /**
     * this method will handle the response recieved from the Cloud NLP request.
     * The response is a JSON object only.
     * This has been casted to GenericJson from Google Cloud so that the developers can easily parse through the same and can understand the response.
     *
     *
     * @param response --> the JSON object recieved as a response for the cloud NLP Api request
     */
    private void deliverResponse(final GenericJson response) {
        Log.d("TAG", "Generic Response --> " + response);
        runOnUiThread(new Runnable() {
            @Override
            public void run() {
                Toast.makeText(MainActivity.this, "Response Recieved from Cloud NLP API", Toast.LENGTH_SHORT).show();
                try {
                    resultTextView.setText(response.toPrettyString());
                    nestedScrollView.setVisibility(View.VISIBLE);
                } catch (IOException e) {
                    e.printStackTrace();
                }


            }
        });


    }
}

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

public class MainActivity extends AppCompatActivity {

EditText editTextView; // View to get the text to be analyzed

TextView resultTextView; // View to display the results obtained after analysis

NestedScrollView nestedScrollView; // Wrapper view so that if the results are getting out of bounds from screen so the user can scroll and see complete results

private static final int LOADER_ACCESS_TOKEN = 1; // Token used to initiate the request loader

private GoogleCredential mCredential = null; //GoogleCredential object so that the requests for NLP Api could be made

// A Thread on which the Api request will be made and results will be delivered. As network calls cannot be made on the amin thread, so we are creating a separate thread for the network calls

private Thread mThread;

// Google Request for the NLP Api. This actually is acting like a Http client queue that will process each request and response from the Google Cloud server

private final BlockingQueue<CloudNaturalLanguageRequest<? extends GenericJson>> mRequests

= new ArrayBlockingQueue<>(3);

// Api for CloudNaturalLanguage from the Google Client library, this is the instance of our request that we will make to analyze the text.

private CloudNaturalLanguage mApi = new CloudNaturalLanguage.Builder(

new NetHttpTransport(),

JacksonFactory.getDefaultInstance(),

new HttpRequestInitializer() {

@Override

public void initialize(HttpRequest request) throws IOException {

mCredential.initialize(request);

}

}).build();

@Override

protected void onCreate(Bundle savedInstanceState) {

super.onCreate(savedInstanceState);

setContentView(R.layout.activity_main);

editTextView = (EditText) findViewById(R.id.text_et);

resultTextView = (TextView) findViewById(R.id.result_tv);

nestedScrollView = (NestedScrollView) findViewById(R.id.nsv);

prepareApi();

}

/**

* Method called on the click of the Button

* @param view -> the view which is clicked

public void startAnalysis(View view) {

String textToAnalyze = editTextView.getText().toString();

if (TextUtils.isEmpty(textToAnalyze)) {

editTextView.setError(getString(R.string.empty_text_error_msg));

} else {

editTextView.setError(null);

analyzeTextUsingCloudNLPApi(textToAnalyze);

}

/**

* This function will send the text to Cloud Api for analysis

* @param text -> String to be analyzed

public void analyzeTextUsingCloudNLPApi(String text) {

try {

mRequests.add(mApi

.documents()

.annotateText(new AnnotateTextRequest()

.setDocument(new Document()

.setContent(text)

.setType("PLAIN_TEXT"))

.setFeatures(new Features()

.setExtractSyntax(true)

.setExtractEntities(true)

)));

} catch (IOException e) {

Log.e("TAG", "Failed to create analyze request.", e);

}

/**

* Preparing the Cloud Api before maiking the actual request.

* This method will actually initiate the AccessTokenLoader async task on completion

* of which we will recieve the token that should be set in our request for Cloud NLP Api.

private void prepareApi() {

// Initiate token refresh

getSupportLoaderManager().initLoader(LOADER_ACCESS_TOKEN, null,

new LoaderManager.LoaderCallbacks<String>() {

@Override

public Loader<String> onCreateLoader(int id, Bundle args) {

return new AccessTokenLoader(MainActivity.this);

}

@Override

public void onLoadFinished(Loader<String> loader, String token) {

setAccessToken(token);

}

@Override

public void onLoaderReset(Loader<String> loader) {

}

});

}

/**

* This method will set the token from the Credentials.json file to the Google credential object.

* @param token -> token recieved from the Credentials.json file.

public void setAccessToken(String token) {

mCredential = new GoogleCredential()

.setAccessToken(token)

.createScoped(CloudNaturalLanguageScopes.all());

startWorkerThread();

}

/**

* This method will actually initiate a Thread and on this thread we will execute our Api request

* and responses.

* Responses recieved will be delivered from here.

private void startWorkerThread() {

if (mThread != null) {

return;

}

mThread = new Thread(new Runnable() {

@Override

public void run() {

while (true) {

if (mThread == null) {

break;

}

try {

// API calls are executed here in this worker thread

deliverResponse(mRequests.take().execute());

} catch (InterruptedException e) {

Log.e("TAG", "Interrupted.", e);

break;

} catch (IOException e) {

Log.e("TAG", "Failed to execute a request.", e);

}

});

mThread.start();

}

/**

* this method will handle the response recieved from the Cloud NLP request.

* The response is a JSON object only.

* This has been casted to GenericJson from Google Cloud so that the developers can easily parse through the same and can understand the response.

* @param response --> the JSON object recieved as a response for the cloud NLP Api request

private void deliverResponse(final GenericJson response) {

Log.d("TAG", "Generic Response --> " + response);

runOnUiThread(new Runnable() {

@Override

public void run() {

Toast.makeText(MainActivity.this, "Response Recieved from Cloud NLP API", Toast.LENGTH_SHORT).show();

try {

resultTextView.setText(response.toPrettyString());

nestedScrollView.setVisibility(View.VISIBLE);

} catch (IOException e) {

e.printStackTrace();

}

});

}

We have successfully integrated the Cloud Natural Language API in our Android application.

Hope this article helps you.

Keep coding and Keep Sharing

For any queries, please feel free to add a comment in the comments section.

References –>