Thread: Nested Looping to find if Strings equal

Ok, I'm doing a huge project, but I'll cut it down to my current issue.

This project is designed to write a value to an excel sheet using the SmartXLS library.

I need the most efficient way of going through 2 huge arrays and finding if the strings are the same.

I have an Array (dataArr) with 179 Objects with Name Variables (Strings) and an ArrayList (RevNames) with 279 Strings.

With the exception of 1 String in each array, they are in alphabetical order.

When I get the object from dataArr, I get its Name variable. I then go through the RevNames array to see if that name is in the array. If it is in RevNames, I insert the object's appropriate data in the appropriate cell and continue to collect the next object from dataArr. If the name isnt in RevNames, I have to insert the name in alphebatical order in Excel (not RevNames array) by inserting a new row, setting the name, value, ect.

Now, my current code doesnt work, and I need to check the consistancy of my arrays for skipped data, but I know there has to be a more efficent way of doing this. Currently in my code, I have not dealt with the issue of adding new rows to Excel. As is, this nested loop can take 30 seconds to run and I've got to do the same process to 14 other sheets (just get a different variable from object each time).

I also wanted to know if ArrayList's indexOf(Object elem) method would be useful for me here. Is that more efficent than the nested loop?

Code for Nested Loop:

for(int x=0;x<dataArr.length-1;x++)
{
      String tempRName = dataArr[x].Name;
      for(int y=0;y<RevNames.size()-1;y++)
      {
                  if(tempRName.equalsIgnoreCase(RevNames.get(y)))
                 {
	       workBook2.setNumber(2+y,c,dataArr[y].RevPaxDep);
	       break;
                 }
      }
}

I revised it to this:

for(int x=0;x<dataArr.length-1;x++)
{
	String tempRName = dataArr[x].Name;
	int indexAt = RevNames.indexOf(tempRName);
	if(indexAt == -1)
		System.out.println("--------------Need to Add Row--------------");
	else
		workBook2.setNumber(2+indexAt,c,dataArr[x].ConnPaxDep);
}

And it runs fine now. Doesnt take nearly as long, but still takes some time. Now I've just got to figure out how to add a new entry to my excel. I figure I can deal with adding the rows on all of the sheets during the first loop since they will have to be added to every sheet anyway.

Originally Posted by aussiemcgr

And it runs fine now. Doesnt take nearly as long, but still takes some time

Its up to you, but once again I encourage you to use a Set or Map. This is their purpose. A short demo:

public class Test{
    /**
    *Analyzes the speed of list versus set searches
    */
	public static void main(String[] args){
		final int max = 10000;
		ArrayList<String> list = new ArrayList<String>(max);
		Set<String> set = new HashSet<String>(max);
		//First populate the list and set with identical values
		for ( int i = 0; i < max; i++ ){
			list.add(Integer.toString(i));
			set.add(Integer.toString(i));
		}
		long start = System.currentTimeMillis();
 		//search the list max number of times
		for ( int i = 0; i < max; i++ ){
			int num = (int)(Math.random() * max);
			list.indexOf(Integer.toString(num));
 
		}
		System.out.print("Time to search list: ");
		System.out.println(System.currentTimeMillis() - start);
		start = System.currentTimeMillis();
		//search the set max number of times
		for ( int i = 0; i < max; i++ ){
			int num = (int)(Math.random() * max);
			set.contains(Integer.toString(num));
		}
		System.out.print("Time to search set: ");
		System.out.println(System.currentTimeMillis() - start);
	}
}

On my system:

Time to search list: 531
Time to search set: 0

Just my .02

Originally Posted by copeg

Its up to you, but once again I encourage you to use a Set or Map. This is their purpose. A short demo:

public class Test{
    /**
    *Analyzes the speed of list versus set searches
    */
	public static void main(String[] args){
		final int max = 10000;
		ArrayList<String> list = new ArrayList<String>(max);
		Set<String> set = new HashSet<String>(max);
		//First populate the list and set with identical values
		for ( int i = 0; i < max; i++ ){
			list.add(Integer.toString(i));
			set.add(Integer.toString(i));
		}
		long start = System.currentTimeMillis();
 		//search the list max number of times
		for ( int i = 0; i < max; i++ ){
			int num = (int)(Math.random() * max);
			list.indexOf(Integer.toString(num));
 
		}
		System.out.print("Time to search list: ");
		System.out.println(System.currentTimeMillis() - start);
		start = System.currentTimeMillis();
		//search the set max number of times
		for ( int i = 0; i < max; i++ ){
			int num = (int)(Math.random() * max);
			set.contains(Integer.toString(num));
		}
		System.out.print("Time to search set: ");
		System.out.println(System.currentTimeMillis() - start);
	}
}

On my system:

Time to search list: 531
Time to search set: 0

Just my .02

On my crazy slow work computer (takes about 40 seconds to open a new tab in IE and 40 seconds to open the excel sheet I'm working on) it takes me 313 milliseconds to run the code I have now.

I know its not 10000 x 10000 elements, but the miliisecond difference doesnt justify rediting the code for all of 15 sheet entries again.