Category Archives: Core Java

Java Memory Management

One topic that I have always been curious, is java memory management. How JVM allocates the memory to different objects and frees up the memory when needed? In this post, I will talk about Java Heap Memory and Stack Memory. Heap and Stack are the memories that JVM allocates as per the application requirements.

Java Heap Space

The most basic question that arises during this discussion, is how do you define both of these memories. So I will start with Java Heap Space. When JVM starts, it creates Java Heap Space and it is used by the application till the time the application is running. Java runtime uses heap space to allocate memory to objects and JRE classes.

The heap size is adjusted according to when the application runs. When the heap gets full, garbage collection takes place. During garbage collection, objects that are not being used, get cleaned, in-process making space for new objects.

Java Stack Memory

Stack memory is like a RAM, used by an executing thread for method-specific values or operations. Most of the values or operations in the stack are short-lived. It can also contain references to objects that reside in heap.

Whenever a method is invoked, a block is allocated in the stack for a method to hold local variables. The block gets cleared once the method finishes execution.

From these earlier definitions, it is clear that Stack memory is smaller in size compared to heap space.

Stack Memory and Heap Space

Differences between Heap space and Stack memory

  1. Heap memory is used by the entire application while the stack is used by execution thread only.
  2. When an object is created, it is stored in heap space, while the reference for that object is stored in stack memory.
  3. Since stack memory is thread-specific, it can’t be accessed by multiple threads or other threads than the one thread that created it. Heap space is global.
  4. Heap space is available till the time application is running, stack memory is short-lived.
  5. JVM can throw errors if both memories are full or the application doesn’t have either of memory remaining to continue running the application. StackOverfFowError if JVM is out of stack memory. If the application stops running provided there is no memory to store objects, it will throw OutOfMemoryError: Java Heap Space Error.

Conclusion

In this post, I discussed the differences between Java heap space and stack memory as part of Java memory management.

References

  1. Stack vs Heap
  2. Understanding memory management

 

How to use Streams API

Java 8 added a new feature called Streams. Streams represent a sequence of objects from a source. In this post, I show how to use Streams API.

Previously, using collections API, we would have a collection of objects and then a developer would process this collection to manipulate further to query. With Streams feature, the developer will not have to do any processing operation over the collection of objects.

Streams

Firstly, streams provide a set of elements in a sequential manner. It provides a number of APIs for aggregate operation. Streams take Arrays, Collections, or I/O sources as an input.

How Streams Work

A stream represents a sequence of elements. Stream operations are either intermediate or terminal. Intermediate operations return streams to process further while terminal operations return either void or non-stream results.

List<String> myList =
    Arrays.asList("test1", "sameresult", "netresult", "grossprofit", "test2");

myList
    .stream()
    .filter(s -> s.startsWith("test"))
    .map(String::toUpperCase)
    .sorted()
    .forEach(System.out::println);

As shown above, filter, map, sorted are intermediate operations and forEach is a terminal operation. Javadocs provide the list of all operations on streams.

Most stream operations accept some kind of lambda expression parameter, a functional interface specifying the behavior of the operation.

Instead of using collections, you can also use Stream.Of() operation to create a stream from a bunch of objects.

Intermediate operations have a characteristic of laziness. To look at this, let’s check the example below:

Stream.of("n1", "n2", "n3", "n4", "n5")
    .filter(s -> {
        System.out.println("filter: " + s);
        return true;
    });

It will not print anything on the console. Intermediate operations will work only when there are terminal operations.

Once you call a terminal operation on streams, streams can not be reused.

Operations on Streams

Streams API offers aggregate operations that offer flexibility in using streams. I will show an example here about how to use streams

List<String> listOfStrings = new ArrayList<>();
listOfStrings.add("one");
listOfStrings.add("two");
listOfStrings.add("three");
listOfStrings.add("");
listOfStrings.add("four");

List<String> listOfNotEmptyStrings = listOfStrings.streams().filter(str -> !str.isEmpty()).collect(Collectors.toList());

In the example shown above, I have list of strings that I filter to get only a list containing non-empty strings.

Streams also offer forEach operation which can be used to iterate over the elements of the stream.

Collect is a terminal operation that can transform the elements of a stream into a different kind of result. Example – a List, Map, or a Set.

Map is an operation that allows us to transform objects of a stream into another type of object.

Reduce operation combines all elements of the stream into a single result.

ParallelStreams

Streams also offer something called ParallelStreams . Basically all operations performed over ParallelStreams ,are performed in parallel. Depending on your usage, use these streams carefully as they can cause concurrency issues.

Example – listOfStrings.parallelStream()

ParallelStreams can be used to improve the runtime performance on a large set of input elements.

Conclusion

In this post, I showed how to use Streams API in Java 8. Streams make it handy in many cases where we have a list of objects and we are processing these objects. If you enjoyed this post, subscribe to my blog here.

 

How to add SSLContext with TLSv1.2 in Android Kitkat

Recently, I came across an issue where a server was not responding in my android application running in Kitkat version of Android. Yes, Kitkat is old, and why one should use it when we can always upgrade to a higher version of Android. We can argue, find pros and cons, but that’s not the point. If we have the technology, it should be more flexible and sometimes, a customer will not be able to go for a higher version of Android. When technology plays with time, certain things are not in control. So if websites had made TLSv1.2 compulsory when Android Kitkat was released, Google had no choice but to release a solution. But that wasn’t the case.

Back to the present time, my Android application is trying to contact a server that has TLSv1.2 enabled. In my Android application, I used Android provided DefaultHttpClient.

So the issue is “How do we add a SSLContext in this DefaultHttpClient“?

Solution –

Create a HTTP Client Socket Factory –

We will build a socket factory that will implement LayeredSocketFactory like below:

public class TlsSniSocketFactory implements LayeredSocketFactory {

  private final static HostnameVerifier hostnameVerifier = new StrictHostnameVerifier();

  private final boolean acceptAllCertificates;
  private final String selfSignedCertificateKey;

  public TlsSniSocketFactory() 
        {
    this.acceptAllCertificates = false;
    this.selfSignedCertificateKey = null;
  }

  public TlsSniSocketFactory(String certKey) 
        {
    this.acceptAllCertificates = false;
    this.selfSignedCertificateKey = certKey;
  }

  public TlsSniSocketFactory(boolean acceptAllCertificates) 
        {
    this.acceptAllCertificates = acceptAllCertificates;
    this.selfSignedCertificateKey = null;
  }

  // Plain TCP/IP (layer below TLS)

  @Override
  public Socket connectSocket(Socket s, String host, int port, InetAddress localAddress, int localPort,
                HttpParams params) throws IOException {
    return null;
  }

  @Override
  public Socket createSocket() throws IOException {
    return null;
  }

  @Override
  public boolean isSecure(Socket s) throws IllegalArgumentException {
    if (s instanceof SSLSocket) {
      return s.isConnected();
    }
    return false;
  }

  // TLS layer

  @Override	
  public Socket createSocket(Socket plainSocket, String host, int port, boolean autoClose) throws IOException 
        {
    if (autoClose) 
                {
      plainSocket.close();
    }

    SSLCertificateSocketFactory sslSocketFactory =
        (SSLCertificateSocketFactory) SSLCertificateSocketFactory.getDefault(0);

    // For self-signed certificates use a custom trust manager
    if (acceptAllCertificates) {
      sslSocketFactory.setTrustManagers(new TrustManager[]{new IgnoreSSLTrustManager()});
    } else if (selfSignedCertificateKey != null) {
      sslSocketFactory.setTrustManagers(new TrustManager[]{new SelfSignedTrustManager(selfSignedCertificateKey)});
    }

    // create and connect SSL socket, but don't do hostname/certificate verification yet
    SSLSocket ssl = (SSLSocket) sslSocketFactory.createSocket(InetAddress.getByName(host), port);

    // enable TLSv1.1/1.2 if available
    // ssl.setEnabledProtocols(ssl.getSupportedProtocols());
                // this can be hard coded too
                ssl.setEnabledProtocols(new String[] {"TLSv1.2"});

    // set up SNI before the handshake
    if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.JELLY_BEAN_MR1) {
      sslSocketFactory.setHostname(ssl, host);
    } else {
      try {
        java.lang.reflect.Method setHostnameMethod = ssl.getClass().getMethod("setHostname", String.class);
        setHostnameMethod.invoke(ssl, host);
      } catch (Exception e) {
        Log.d(TlsSniSocketFactory.class.getSimpleName(), "SNI not usable: " + e);
      }
    }

    // verify hostname and certificate
    SSLSession session = ssl.getSession();
    if (!(acceptAllCertificates || selfSignedCertificateKey != null) && !hostnameVerifier.verify(host, session)) {
      throw new SSLPeerUnverifiedException("Cannot verify hostname: " + host);
    }

    return ssl;
  }

}

Register a HTTPS Scheme

We will register a scheme that will use our custom socket factory.

SchemeRegistry schemeRegistry = new SchemeRegistry();
schemeRegistry.register(new Scheme("https", new TlsSniSocketFactory(),443));

ClientConnectionManager ccm = new ThreadSafeClientConnManager(httpparams, schemeRegistry);
DefaultHttpClient defaultHttpClient = new DefaultHttpClient(ccm, httpparams);

Now if use defaultHttpClient to call a GET or POST request, we should be able to connect to a server that is enabled with TLSv1.2.

Conclusion

In this post, we showed how to use DefaultHttpClient in Android Kitkat with TLSv1.2. If you enjoyed this post, subscribe to my blog here.

References

  1. TLS Socket Factory
  2. Default HTTP Client

Custom Twitter Client vs Spring Boot Twitter Plugin

To use twitter data in my saas application, I was going to write my own custom Twitter client by doing a rest call. However, I found out Spring Boot offers a Twitter plugin that can be used to fetch Twitter data. Neat.

In this post, I will show some comparison of these two approaches and why one can choose over another:

Custom Twitter Client

So custom twitter client will be a standalone client which will build an HTTP entity with client secrets that are needed to authenticate with Twitter API. In this client, we will use restOperations to call API endpoint passing HTTP entity and the REST call will respond with Twitter Data Model.

This will look like below:

public TwitterDataModel getTwitterData(long accountId)
{
    String url = buildRestUrl(accountId);
    ParameterizedTypeReference<HashMap<Long, TwitterDataModel>> responseType = new ParameterizedTypeReference<HashMap<Long, TwitterDataModel>>(){};
    HttpEntity entity = buildHttpEntity(CLIENT_ID, CLIENT_SECRET);
    Map<Long, TwitterDataModel> twitterDataModelMap = restOperations.exchange(url, HttpMethod.GET, entity, responseType).getBody();

    Long keyForData = new Long(accountId);
    TwitterDataModel twitterDataModel = twitterDataModelMap.get(keyForData);

    return twitterDataModel;
}

public String buildRestUrl(long accountId)
{
    return TWITTER_REST_ENDPOINT + accountId + TWITTER_REST_API;
}

There is nothing much wrong with this approach, except the fact that we will have to write an extra TwitterDataModel business object. Also, this business model should be created before we do the actual REST call.

Spring Boot Twitter Plugin

To use this plugin, first, we need to add the plugin in Gradle or maven like below:

compile('org.springframework.social:spring-social-twitter')

Once we have this plugin, we can add an object of type Twitter in our code to call REST APIs.

This will look like below:

private final Twitter twitter;

public TwitterDataModel getTwitterData(long accountId)
    {
        String url = buildRestUrl(accountId);
        ParameterizedTypeReference<HashMap<Long, TwitterDataModel>> responseType = new ParameterizedTypeReference<HashMap<Long, TwitterDataModel>>(){};
        HttpEntity entity = buildHttpEntity(CLIENT_ID,CLIENT_SECRET);

        Map<Long, TwitterDataModel> twitterDataModelMap = twitter.restOperations().exchange(url, HttpMethod.GET, entity, responseType).getBody();

        Long keyForData = new Long(accountId);
        TwitterDataModel twitterDataModel = twitterDataModelMap.get(keyForData);

        return twitterDataModel;
    }

    public String buildRestUrl(long accountId)
    {
        return TWITTER_REST_ENDPOINT + accountId + TWITTER_REST_API;
    }

The major advantage of this plugin is that we can get the data in Twitter Data Model that twitter offers. An then we can go on to use to handle our data.

Conclusion

In this post, I showed how we can use a Spring Boot Twitter social plugin to gather Twitter data.

Database and design discussion – Part III

To continue the development of a spring-based web application, this post will discuss using of Twitter API in saas application. If you want to understand, what we are building, you can read the first two posts of this series where we discussed the design of the application we are building:

  1. Database design and discussion – Part I
  2. Database design and discussion – Part II

In the previous post, we discussed the Instagram API that we will be using. With recent events around Facebook, I have decided not to use Facebook API for application development. We will still use Instagram and Twitter API.

Using Twitter API in SAAS Application

Firstly, Twitter offers different APIs for developers to build applications. We will be using Engagement API. You can find more details Twitter API.

Our goal is to use this Twitter API to collect engagement metrics in the Saas application.

Secondly, Engagement API offers us details about account engagement metrics which can help us to design marketing strategy. Sample response of this API looks like the below:

{
  "Tweet metrics": {
    "902301386880286721": {
      "engagements": "433",
      "favorites": "21",
      "impressions": "72218"
    },
    "902731270274166784": {
      "engagements": "61",
      "favorites": "27",
      "impressions": "7827"
    },
    "907022936564838401": {
      "engagements": "187",
      "favorites": "37",
      "impressions": "1916"
    }
  }
}

Therefore this API provides metrics for tweets, which tweet generated more traffic. The key decision that can be devised based on these metrics is what kind of tweet, moment or incident generates the traffic.

What fields we will use in our database?

In conclusion, we will be using the following fields in our database table TwitterData

  • tweet id
  • engagements
  • impressions
  • tweet

Twitter is a viable medium. This data will provide small businesses with a key metric about what tweets have worked with their followers and how they can leverage that pattern. Once the small businesses sort out these patterns, they will be able to create a number of tweets to engage with customers. Eventually, the goal here is to help small businesses to attract customers, and repeat customers.

References

  1. Twitter API documentation – Twitter API