Java interview question: List “contains” method

This question verifies that you know how the “contains” method of List works.
Let’s have a look at this small piece of code:

Person person = new Person("guy");
List<Person> persons = new ArrayList<Person>();
persons.add(person);
System.out.println(persons.contains(person));

The question: Is it possible, that the output of this little piece of code returns: “false”.

At first look, the answer seems to be: NO, meaning, this code will always return “true”. So how can a “Person” object that was just created and added to a list, not be contained in it?

The answer, is in the way the “contains” method works. The “contains” method, is using the “equals” method in order to check if an object is contained in the list. Therefore, the code above can return “false” if the writer of the “Person” object overridden the equals method.

Java interview question: Count the number of “1” bits in a byte

Frankly, I think it is quite an idiotic question. Not because it is too easy or too hard question, but because I think this question doesn’t really imply on the practical programming capabilities of the person being interviewed. Anyway, a friend of mine was asked to solve this question, so I though it might be an interest for more developers looking for a job and having to handle all sort of weird questions.
In general, there are 2 solutions for this question:

• First solution is iterating 8 times (a byte contains 8 bits). Each time the right most bit is extracted from the byte. and then the byte is shifted right one bit.
• Second solution, is a little bit less straight forward, but more efficient: Divide the number by 2, if there is a modulo, count it. Repeat this operation as long as the number we are dividing is greater than zero.

Here is how the first solution looks in Java code:

public class GetBits {
public static int countBits(byte num)
{
  int count = 0;
  for (int i = 0; i < 8; i++)
{
    if ((num & 1) == 1) // check if right most bit is 1
{
count++;
}
num = (byte)(num >>> 1); // shit right 1 bit, including the sign bit
}
  return count;    
}

Here is how the second solution looks like:

public static int countBits(byte num)
{
  int count = 0;
  while (num > 0)
{
    if (num % 2 == 1) // check if number have modulo
{
count++;
}
num /= 2; // divide the number by 2
}
  return count;    
}

Note, that the first solution is a bit less efficient, because it always iterates 8 times. But, it handles both positive and negative values. The second solution will simply won’t work with negative numbers.

Excel Pivot Tables and Dynamic Range

Excel Pivot Table is a great tool for viewing and aggregating data. The data shown in a Pivot Table is taken from some range in one of the sheets of the excel file. The problem with Pivot Tables, is when you add new data. After refreshing the Pivot Table the new data is not shown. To solve this issue we can use a dynamic range. A dynamic range knows to grow automatically according to the rows we add. Suppose we have an excel file that contains 2 sheets: “data” and “pivot table”. We would like to create from the raw data in the sheet: “data” a dynamic range. This will allow us to add more rows to the “data” sheet, and just refresh the pivot table to see our new rows.
This is how our “data” sheet looks like:


And this is how our “pivot_table” sheet looks like:




In order to create a dynamic range from our raw data, we should go to the “Formulas” tab on excel and choose the “Define Name” option. A pop up window will be opened. In the window we should name the range. For example: “pivot_data_source”. In the “Refers to” field we define the dynamic range. The value that should be put there look like:

=OFFSET(data!$A$1,0,0, COUNTA(data!$A:$A), COUNTA(data!$1:$1))

Where the “data” is the name of the sheet in which the dynamic range is located.
The popup window looks like this:



After defining the dynamic data source, we should connect the pivot table with it. We do it by standing on the pivot table and selecting the “Options” tab from the main menu and choosing the “Change Data Source” option. A popup window is opened. In the field: “Table/Range” we should put the name of the dynamic range we defined: “pivot_data_source”.
The popup windows looks like:



Finally we would like the pivot table to be automatically refreshed every time our excel file is being opened. We do it by standing on the pivot table, then selecting “PivotTable Options…”. A popup window will be opened. On the “Data” tab we should check the option: “Refresh data when opening the file”. The popup window looks like:



Note that if you add new rows to the raw data, and you immediately want to see the changes in the pivot table, you should right click on the pivot table and choose the option “Refresh”. Press here to see an example of an excel file with pivot table and dynamic range defined.

Seam and Quartz integration

Seam is a great framework for easily building Internet applications in Java. It integrates with lots of well known existing java technologies and serves as a complement to JSF (Java Server Faces).
Quartz is a job scheduling framework allowing to easily add the capability of running scheduled jobs in your applications.
Quartz can be easily integrated with Seam. By integrating both technologies you will be able to enjoy Seam approach to managing beans on your web application.
Let’s take a look at a simple example:
First make sure you have all the jars needed to run Quartz and Seam.
Then go to your Seam “components.xml” file and add these lines:

<async:quartz-dispatcher/>
  <event type="org.jboss.seam.postInitialization">
  <action execute="#{quartzController.scheduleTimer}"/>
</event>

Now, put in your sources root directory (usually the “src”) the file: “seam.quartz.properties” file. A typical file may look like this:

org.quartz.scheduler.instanceName = Sched1
org.quartz.scheduler.instanceId = 1
org.quartz.scheduler.rmi.export = false
org.quartz.scheduler.rmi.proxy = false
org.quartz.threadPool.class = org.quartz.simpl.SimpleThreadPool
org.quartz.threadPool.threadCount = 3
org.quartz.jobStore.class = org.quartz.simpl.RAMJobStore

Now, to the Java code. The code is constructed from 2 Seam beans. The first bean contains the Quartz trigger class. I looks like:

package com.bashan.blog.job;
import org.jboss.seam.annotations.AutoCreate;
import org.jboss.seam.annotations.In;
import org.jboss.seam.annotations.Name;
import org.jboss.seam.async.QuartzTriggerHandle;
import java.io.Serializable;
import java.util.Date;
/**
* @author Bashan
*/
@Name("quartzController")
@AutoCreate
public class QuartzController implements Serializable {
@In
ScheduleProcessor processor;
private QuartzTriggerHandle quartzTriggerHandleDoJob;
public void scheduleTimer() {
quartzTriggerHandleDoJob = processor.doJob(new Date(), "0 0/1 * * * ?");
}
}

Note, that when calling to “doJob” method, a cron expression is passed. This expression describes the frequency in which the job will run (event minute in the example).

The second class contains the actual task we would like to do:

package com.bashan.blog.job;
import org.jboss.seam.ScopeType;
import org.jboss.seam.annotations.AutoCreate;
import org.jboss.seam.annotations.Logger;
import org.jboss.seam.annotations.Name;
import org.jboss.seam.annotations.Scope;
import org.jboss.seam.annotations.async.Asynchronous;
import org.jboss.seam.annotations.async.Expiration;
import org.jboss.seam.annotations.async.IntervalCron;
import org.jboss.seam.async.QuartzTriggerHandle;
import org.jboss.seam.log.Log;
import java.util.Date;
/**
* @author Bashan
*/
@Name("processor")
@AutoCreate
@Scope(ScopeType.APPLICATION)
public class ScheduleProcessor {
 
  @Logger
  static Log log;
 
  @Asynchronous
  public QuartzTriggerHandle doJob(@Expiration Date when, @IntervalCron String interval) {
    System.out.println("I am a Quartz job.");
    log.debug("Pass some information to the log");
    return null;
  }
}

The method “doJob” is a Seam asynchronous method. This method will be executed according to the data of the cron expression given in the class: QuartzController.

Fixing Facelets/XHTML corrupted page issues in Google Chrome

I had a problem for quite some time on a web site being shown corrupted in Google Chrome. On the original HTML page everything was working great. On the resulted JSF/Seam/Facelets produced page, things were looking corrupted. The problem appeared only on Google Chrome. In other web browsers things look good. When I made a diff on the original working page and the page generated by JSF/Seam/Facelets, there was no difference. But still, page was corrupted on Google Chrome. After some investigation I noticed that Google Chrome was showing the page corrupted, simply because the generated page was not setting content type of: text/html. Google Chrome, unlike the other browsers, was probably sensitive to this issue. The solution to this problem is very simple. Just add to your JSF view tag the property: contentType. If your IDE (for example, IntelliJ) marks the contentType as unrecognized, ignore it. It runs ok. Here is an example of how the contentType is used:

<f:view contentType="text/html" />

Note, that you don’t have to wrap the view tag on all of your page, if you are using Facelets (the view tag is not mandatory in Facelets), so just putting the view tag is enough (like in the example).

Resize to max width and crop to height using ImageMagick

ImageMagik is a great open source tool that allows many conversions and manipulations of images. It supports huge amount of image types and comes as a command line tool, which makes it very easy to be executed from your favorite programming language.

Resizing an image to a maximum width is an easy task using ImageMagik. It is done with a command line tool named “convert”. For example, if we would like to proportionally resize an image named “image.jpg” to the maximum width of 160 we can do:

convert "c:\image.jpg" -resize 160 "c:\image_out.jpg"

Suppose we would like to do something a bit more complex than proportionally resizing and image to a maximum width: we would like to resize an image to a maximum width, but also crop the image to maximum height. We might want to do this kind of resize, since we want to show group of thumbnails at exactly the same size no matter what is the original image proportions are. Achieving this goal with ImageMagik is an easy task, but adds quite a few complexities to the above command.

For example, if we would like to proportionally resize an image maximum size of 100 pixels and crop the image to height of 80 pixels we have to use the “convert” command this way:

convert "c:\image.jpg" -resize 160x -resize "x160<" -resize 50% -gravity center -crop 100x100+0+0 +repage "c:\image_out.jpg"

If you really want to understand what this set of command exactly do, you can the ImageMagik documentation, which is pretty good. Note that the number 160 is exactly twice than 80. So if we would like to crop an image to height of 100 we would use the number 200.

Using Hibernate Transformers

There are times we have a class, we would like to fill with data according the data returned from a query. The class is a simple POJO and not an Hibernate entity, so Hibernate won’t recognize this class.
This can be done in Hibernate by using Transformers. Let’s have a look on a simple example, showing how Transformers can be used. First, let’s have a look at a simple POJO class named: “UserActivityStat”.
This class contains some statistical information. We would like to fill the statistical information of an instance, directly from running an Hibernate HQL.

public class UserActivityStat

{

private int totalPhotos;

private int totalViews;


public UserActivityStat() {

}


public int getTotalPhotos() {

return totalPhotos;

}


public void setTotalPhotos(int totalPhotos) {

this.totalPhotos = totalPhotos;

}


public int getTotalViews() {

return totalViews;

}


public void setTotalViews(int totalViews) {

this.totalViews = totalViews;

}

}

Now, let’s have a look at a simple method, that uses hibernate HQL and the Transformers class to fill “UserActivityStat” instance with data:

public UserActivityStat getUserActivityStat(User user)

{

return (UserActivityStat)hibernateSession.createQuery("select count(*) as totalPhotos, sum(p.views) as totalViews " +

"from Photo p where p.user = :user " +

"p.dateCreated  <= :now").

setParameter("user", user).

setTimestamp("now", new Date()).

setResultTransformer(Transformers.aliasToBean(UserActivityStat.class)).uniqueResult();

}

Note, that each of the 2 columns has an alias. This alias must be the name of the property on the “UserActivityStat” class. Also note for the use of the “setResultTransformer” along the “Transformers” class.

Calculate the distance between 2 Ips in Java using maxmind

Calculating the distance between 2 ips can be done easily by using the great geo service: Maxmind.

Maxmind is an affordable geo service with broad range of solutions like:

• Country
• City
• Organization
• ISP

and more…

Maxmind services comes in 2 flavors:

• Paid service: This is very accurate data, in a reasonable low cost.
• Free service: This is a little less accurate data, but a totally free service.

In order to calculate the distance between 2 IPs, we will use the free version of Maxmind. The free service is called: geoLite. we will use the service named: geoLite city. The geoLite city data contains location information. The location contains latitude and longitude information for ip ranges. Latitude and longitude information is a coordinate method used in may location services such GPS.

In order to use Maxmind we need 2 things:

• Download the Maxmind free open source Java API. Note, that Maxmind has API support for many more commonly used programming languages.
• Download the latest city binary file, which contains IP ranges information used by Maxmind API.

Calculating the distance between 2 IPs is very easy, since Maxmind did all the hard work for us. We just have to get “Location” instance for the 2 IP and use the “distance” method of the “Location” instance. The returned result is the distance between the 2 IPs in kilometers.

Let’s have a look at a sample code that calculates the distance between Google and Apple:

package com.bashan.blog.maxmind;
import com.maxmind.geoip.Location;
import com.maxmind.geoip.LookupService;
import java.io.IOException;
/**
 * @author Bashan
 */
public class MaxmindTest {
  public static void main(String args[]) throws Exception {
    LookupService lookupService = new LookupService("C:\\GeoLiteCity.dat");
    Location locationGoogle = lookupService.getLocation("74.125.39.147");
    Location locationMicrosoft = lookupService.getLocation("17.251.200.70");
    System.out.println("Google is located on: " + locationGoogle.city);
    System.out.println("Apple is located on: " + locationMicrosoft.city);
    System.out.print("Distance: " + locationGoogle.distance(locationMicrosoft) + " kilometers");
  }
}

And the output:

Google is located on: Mountain View
Apple is located on: Cupertino
Distance: 13.218970226605398 kilometers

Not too far as you can see… ;-)

Get real IP from request in Java

Simply getting the IP of the remote client in Java is an easy task. Assuming we have a request instance, we can simply invoke the “getRemoteAddr” method:

request.getRemoteAddr();

The problem, is that the IP we get is not always the correct IP. For example if our server is behind a load balancer, the method “request.getRemoteAddr” returns the IP of the load balancer and not the IP of the remote client. Another common example, is that the client is behind some proxy or even several proxies. The IP we will get will not be the correct IP.

Fortunately, In most of the cases when a request passes in a load balancer or a proxy, the IP of the remote client is passed in the header of the request. The header key most of the time is: x-forwarded-for. The header value can be one or more IP addresses. The first address, is the address of the remote client. The second or any other IP is the IP of the proxy on the way. The IP of the last proxy is the IP returned in: “request.getRemoteAddr”. For example, if we have 3 proxies, the request header will look like:

x-forwarded-for: client1, proxy1, proxy2

The IP of the client if the first IP. The IP of the third proxy will be returned when calling to: “request.getRemoteAddr”.

There is one none-common case that may happen. The first IP can be an IP of a private network (for example, an IP that starts with “192” or “10”). I such case we will want to take the second IP.

Writing a Java code that will get the IP from the request is quite easy. We will use assistance from code of these older posts: Convert IP String to numeric representation and numeric representation to IP String in Java, Test if IPv4 belongs to a private network address. The general logic of the code is:

• Look for “x-forwarded-for” header.
• If header exists, get the first IP.
• Check that:
• IP is valid.
• IP is not a private IP.
• If IP passes these 2 tests. Return this IP. If not move to the next IP and do the same test and so on.
• If header doesn’t exist. Return the IP from calling “request.getRemoteAddr”.

Let’s see how it looks in Java:

package com.bashan.blog.ip;
import org.apache.commons.lang.text.StrTokenizer;
import javax.servlet.http.HttpServletRequest;
import java.util.regex.Pattern;
/**
 * @author Bashan
 */
public class IpUtils {
  public static final String _255 = "(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)";
  public static final Pattern pattern = Pattern.compile("^(?:" + _255 + "\\.){3}" + _255 + "$");
  public static String longToIpV4(long longIp) {
    int octet3 = (int) ((longIp >> 24) % 256);
    int octet2 = (int) ((longIp >> 16) % 256);
    int octet1 = (int) ((longIp >> 8) % 256);
    int octet0 = (int) ((longIp) % 256);
    return octet3 + "." + octet2 + "." + octet1 + "." + octet0;
  }
  public static long ipV4ToLong(String ip) {
    String[] octets = ip.split("\\.");
    return (Long.parseLong(octets[0]) << 24) + (Integer.parseInt(octets[1]) << 16) +
        (Integer.parseInt(octets[2]) << 8) + Integer.parseInt(octets[3]);
  }
  public static boolean isIPv4Private(String ip) {
    long longIp = ipV4ToLong(ip);
    return (longIp >= ipV4ToLong("10.0.0.0") && longIp <= ipV4ToLong("10.255.255.255")) ||
        (longIp >= ipV4ToLong("172.16.0.0") && longIp <= ipV4ToLong("172.31.255.255")) ||
        longIp >= ipV4ToLong("192.168.0.0") && longIp <= ipV4ToLong("192.168.255.255");
  }
  public static boolean isIPv4Valid(String ip) {
    return pattern.matcher(ip).matches();
  }
  public static String getIpFromRequest(HttpServletRequest request) {
    String ip;
    boolean found = false;
    if ((ip = request.getHeader("x-forwarded-for")) != null) {
      StrTokenizer tokenizer = new StrTokenizer(ip, ",");
      while (tokenizer.hasNext()) {
        ip = tokenizer.nextToken().trim();
        if (isIPv4Valid(ip) && !isIPv4Private(ip)) {
          found = true;
          break;
        }
      }
    }
    if (!found) {
      ip = request.getRemoteAddr();
    }
    return ip;
  }
}

Note that the class “StrTokenizer” used to iterate the IPs of the header, is the Apache version from Apache commons lang.

You can download the class here.

Test if IPv4 belongs to a private network address

Sometime we need to check if a given IP belongs to a private network. IP of a private network belong to a special range of IPs. It is most likely, that if you have a small local network at your home or work, the IP in that network is starting with “192” or “10”. For example: “192.168.0.1” or “10.0.0.1”.

The ranges of private network IPs is:

START	END
10.0.0.0	10.255.255.255
172.16.0.0	172.31.255.255
192.168.0.0	192.168.255.255

Since we are dealing here with range of IPs, it will be much easier to convert the IPs to long representation. We will use the code from this blog: Convert IP String to numeric representation and numeric representation to IP String in Java, to easily do the job.

Let’s have a look at the function “isIPv4Private”:

package com.bashan.blog.ip;
/**
 * @author Bashan
 */
public class IpUtils {
  public static String longToIpV4(long longIp) {
    int octet3 = (int) ((longIp >> 24) % 256);
    int octet2 = (int) ((longIp >> 16) % 256);
    int octet1 = (int) ((longIp >> 8) % 256);
    int octet0 = (int) ((longIp) % 256);
    return octet3 + "." + octet2 + "." + octet1 + "." + octet0;
  }
  public static long ipV4ToLong(String ip) {
    String[] octets = ip.split("\\.");
    return (Long.parseLong(octets[0]) << 24) + (Integer.parseInt(octets[1]) << 16) +
        (Integer.parseInt(octets[2]) << 8) + Integer.parseInt(octets[3]);
  }
  public static boolean isIPv4Private(String ip)
  {
    long longIp = ipV4ToLong(ip);
    return (longIp >= ipV4ToLong("10.0.0.0") && longIp <= ipV4ToLong("10.255.255.255")) ||
        (longIp >= ipV4ToLong("172.16.0.0") && longIp <= ipV4ToLong("172.31.255.255")) ||
        longIp >= ipV4ToLong("192.168.0.0") && longIp <= ipV4ToLong("192.168.255.255");
  }
  public static void main(String[] args) {
    System.out.println(isIPv4Private("210.5.80.10"));
    System.out.println(isIPv4Private("192.168.0.1"));
  }
}

You can download the class here.