I have Caffeine cache with Key->Value mapping. There are multiple implementations of Key interface with different equals methods. In order to delete value from cache based on someOtherVal, I had to use code like cache.asMap().keySet().removeIf(comp::isSame) which is super slow.
Is there any other solution for this kind of many keys to single value mapping in cache? One thing that comes to my mind is to have 2 Cache instances, one with Cache<Key, String> and other with Cache<someOtherVal, Key>, and whenever I want to delete a value I locate Key using this other cache.
Then only question is how to keep this 2 caches in sync? Are there already solutions for this?
import java.time.Duration;
import java.util.Objects;
import java.util.UUID;
import com.github.benmanes.caffeine.cache.Cache;
import com.github.benmanes.caffeine.cache.Caffeine;
import com.google.common.base.Stopwatch;
public class Removal {
private static final int MAX = 1_000_000;
interface Key{
String getSomeOtherVal();
default boolean isSame(Key k){
return Objects.equals(k.getSomeOtherVal(),getSomeOtherVal());
}
}
static class KeyImpl implements Key{
int id;
String someOtherVal;
public KeyImpl(int id, String someOtherVal) {
this.id = id;
this.someOtherVal = someOtherVal;
}
public int getId() {
return id;
}
#Override
public String getSomeOtherVal() {
return someOtherVal;
}
#Override
public boolean equals(Object o) {
if (this == o)
return true;
if (o == null || getClass() != o.getClass())
return false;
KeyImpl key = (KeyImpl)o;
return id == key.id;
}
#Override
public int hashCode() {
return Objects.hash(id);
}
}
Cache<Key, String> cache = Caffeine.newBuilder().build();
public static void main(String[] args) {
Removal s = new Removal();
s.fill();
Duration sRem = s.slowRemovalFirst100();
Duration fRem = s.fastRemoval100To200();
System.out.println("Slow removal in " + sRem);
System.out.println("Fast removal in " + fRem);
}
private Duration slowRemovalFirst100(){
Stopwatch sw = Stopwatch.createStarted();
for(int i=0; i<100; i++){
Key comp = new KeyImpl(i, String.valueOf(i));
cache.asMap().keySet().removeIf(comp::isSame); //Finds a key by some other property and then removes it (SLOW)
//System.out.println("Removed " + i);
}
return sw.stop().elapsed();
}
private Duration fastRemoval100To200(){
Stopwatch sw = Stopwatch.createStarted();
for(int i=100; i<200; i++){
Key comp = new KeyImpl(i, String.valueOf(i));
cache.invalidate(comp); //Uses direct access to map by key (FAST)
//System.out.println("Removed " + i);
}
return sw.stop().elapsed();
}
private void fill(){
for(int i=0; i<MAX; i++){
cache.put(new KeyImpl(i, String.valueOf(i)), UUID.randomUUID().toString());
}
}
}
Result of running this code on my machine:
Slow removal in PT2.807105177S
Fast removal in PT0.000126183S
where you can see such a big difference...
Ok, I managed to solve this:
public class IndexedCache<K,V> implements Cache<K,V> {
#Delegate
private Cache<K, V> cache;
private Map<Class<?>, Map<Object, Set<K>>> indexes;
private IndexedCache(Builder<K, V> bldr){
this.indexes = bldr.indexes;
cache = bldr.caf.build();
}
public <R> void invalidateAllWithIndex(Class<R> clazz, R value) {
cache.invalidateAll(indexes.get(clazz).getOrDefault(value, new HashSet<>()));
}
public static class Builder<K, V>{
Map<Class<?>, Function<K, ?>> functions = new HashMap<>();
Map<Class<?>, Map<Object, Set<K>>> indexes = new ConcurrentHashMap<>();
Caffeine<K,V> caf;
public <R> Builder<K,V> withIndex(Class<R> clazz, Function<K, R> function){
functions.put(clazz, function);
indexes.put(clazz, new ConcurrentHashMap<>());
return this;
}
public IndexedCache<K, V> buildFromCaffeine(Caffeine<Object, Object> caffeine) {
caf = caffeine.writer(new CacheWriter<K, V>() {
#Override
public void write( K k, V v) {
for(Map.Entry<Class<?>, Map<Object, Set<K>>> indexesEntry : indexes.entrySet()){
indexesEntry.getValue().computeIfAbsent(functions.get(indexesEntry.getKey()).apply(k), (ky)-> new HashSet<>())
.add(k);
}
}
#Override
public void delete( K k, V v, RemovalCause removalCause) {
for(Map.Entry<Class<?>, Map<Object, Set<K>>> indexesEntry : indexes.entrySet()){
indexesEntry.getValue().remove(functions.get(indexesEntry.getKey()).apply(k));
}
}
});
return new IndexedCache<>(this);
}
}
}
and this is use-case:
#AllArgsConstructor
#Data
#EqualsAndHashCode(onlyExplicitlyIncluded = true)
static class CompositeKey{
#EqualsAndHashCode.Include
Integer k1;
String k2;
Long k3;
}
public static void main(String[] args) {
Caffeine<Object, Object> cfein = Caffeine.newBuilder().softValues().maximumSize(200_000);
IndexedCache<CompositeKey, String> cache = new IndexedCache.Builder<CompositeKey, String>()
.withIndex(Long.class, ck -> ck.getK3())
.withIndex(String.class, ck -> ck.getK2())
.buildFromCaffeine(cfein);
for(int i=0; i<100; i++){
cache.put(new CompositeKey(i, String.valueOf(i), Long.valueOf(i)), "sdfsdf");
}
for(int i=0; i<10; i++){
//use equals method of CompositeKey to do equals comp.
cache.invalidate(new CompositeKey(i, String.valueOf(i), Long.valueOf(i)));
}
for(int i=10; i<20; i++){
//use Long index
cache.invalidateAllWithIndex(Long.class, Long.valueOf(i));
}
for(int i=20; i<30; i++){
//use String index
cache.invalidateAllWithIndex(String.class, String.valueOf(i));
}
int y = 4;
}
here is link to discussion I had: https://github.com/ben-manes/caffeine/issues/279
Let's say that I want my application to determine user's fitness level based on some criteria.
The criteria could be something like: age, currently taking medication?, 400m run
At first I though I could create a Map where the value is the fitness level and the key is an object that has all the criteria, but since the criteria are ranges this wouldn't work.
For example:
if age is between 18 and 22 and onMedication = false and run400m = [70, 80]
fitness level = GOOD
Now if only one of the parameters is in a different range the fitness level would be different. How could I achieve this?
You can use a TreeMap class for this. There are very useful methods to deal with ranges of key values. For example:
TreeMap<Integer, String> myTreeMap = new TreeMap<>();
myTreeMap.put(10, "A");
myTreeMap.put(20, "B");
myTreeMap.put(30, "C");
myTreeMap.put(40, "D");
System.out.println(myTreeMap.floorEntry(25));
Will be print the second option (20=B). I recommend that you check the TreeMap and all its methods for this case.
Maybe you could use OOP and do something like this:
public class FitnessApp {
public static void main(String[] args) {
Map<String, Object> params = new HashMap<>();
params.put("age", 17);
params.put("onMedication", false);
System.out.printf(new FitnessLevelCalculator().calculateFor(params).name());
}
}
class FitnessLevelCalculator {
private LinkedList<FitnessLevel> fitnessLevels = new LinkedList<>();
public FitnessLevelCalculator() {
fitnessLevels.add(new FitnessLevel(FitnessLevelEnum.ATHLETIC, Arrays.asList(new RangeCriteria("age", 18, 25), new BooleanCriteria("onMedication", false))));
fitnessLevels.add(new FitnessLevel(FitnessLevelEnum.GOOD, Arrays.asList(new RangeCriteria("age", 14, 17), new BooleanCriteria("onMedication", false))));
fitnessLevels.add(new FitnessLevel(FitnessLevelEnum.ILL, Arrays.asList(new RangeCriteria("age", 16, 17))));
}
public FitnessLevelEnum calculateFor(Map<String, Object> params) {
ListIterator<FitnessLevel> listIterator = fitnessLevels.listIterator();
while (listIterator.hasNext()) {
FitnessLevel fitnessLevel = listIterator.next();
if (fitnessLevel.accept(params)) {
return fitnessLevel.getLevel();
}
}
return FitnessLevelEnum.NOT_CLASSIFIED;
}
}
enum FitnessLevelEnum {
ILL, GOOD, ATHLETIC, NOT_CLASSIFIED
}
class FitnessLevel {
private List<Criteria> criteriaList = new ArrayList<>();
private FitnessLevelEnum level;
public FitnessLevel(FitnessLevelEnum level, List<Criteria> criteriaList) {
this.criteriaList = criteriaList;
this.level = level;
}
public boolean accept(Map<String, Object> params) {
for (Criteria criteria : criteriaList) {
if (!params.containsKey(criteria.getName())) {
return false;
}
if (!criteria.satisfies(params.get(criteria.getName()))) {
return false;
}
}
return true;
}
public FitnessLevelEnum getLevel() {
return level;
}
}
abstract class Criteria<T> {
private String name;
public Criteria(String name) {
this.name = name;
}
public abstract boolean satisfies(T param);
public String getName() {
return name;
}
}
class RangeCriteria extends Criteria<Integer> {
private int min;
private int max;
public RangeCriteria(String name, int min, int max) {
super(name);
this.min = min;
this.max = max;
}
#Override
public boolean satisfies(Integer param) {
return param >= min && param <= max;
}
}
class BooleanCriteria extends Criteria<Boolean> {
private Boolean expectedValue;
public BooleanCriteria(String name, Boolean expectedValue) {
super(name);
this.expectedValue = expectedValue;
}
#Override
public boolean satisfies(Boolean param) {
return param == expectedValue;
}
}
In your specific case, I don't think it's good to use as a key in Map. There maybe a way to put the Object with all conditions as your business but it's quite complex and not worth to do that. The Interpreter Pattern may help you on this.
Hope this help.
I have a distributed map and I want to find the lowest or highest key (an object implementing compareable). What is the most efficient way to get those keys? I mean something like every node provides his lowest key and in the end the lowest key is the lowest of every node.
So I think:
MyObj max = Collections.max(map.keySet());
is not the most efficient way. And if I want to use
new DistributedTask<>(new Max(input), key);
I would need to now the key and therefore fetch all Keys over wire. I think in that case I could do Collections.max(map.keySet()); as well.
Hmm ... any ideas?
You could use EntryProcessor.executeOnEntries - with a stateful EntryProcessor - and then let the it do all the work for you; have each key map to a sentinel MIN and MAX enum if they are the min and max.
If you have some idea of the bounds, you could attach a filter Predicate as well to speed it up that way, too.
This map reduce solution seems to have a lot of overhead but it is the best way I could get the job done. Any better ideas are still welcome.
public static void main(String[] args) throws ExecutionException, InterruptedException {
IMap<String, Integer> map = instance.getMap("test");
JobTracker jobTracker = instance.getJobTracker( "default" );
KeyValueSource<String, Integer> source = KeyValueSource.fromMap( map );
Job<String, Integer> job = jobTracker.newJob(source);
JobCompletableFuture<Map<String, String>> future = job
.mapper(new MaxMapper())
.reducer(new MaxReducerFactory())
.submit();
System.out.println("mr max: " + future.get());
}
public static class MaxMapper implements Mapper<String, Integer, String, String> {
private volatile String max = null;
#Override
public void map(String s, Integer integer, Context<String, String> ctx) {
if (max == null || s.compareTo(max)>0) {
max = s;
ctx.emit("max", max);
}
}
}
public static class MaxReducerFactory implements ReducerFactory<String,String,String> {
#Override
public Reducer<String, String> newReducer(String s) {
return new MaxReducer();
}
private class MaxReducer extends Reducer<String, String> {
private volatile String max = null;
#Override
public void reduce(String s) {
if (max == null || s.compareTo(max)>0) max = s;
}
#Override
public String finalizeReduce() {
return max; // == null ? "" : max;
}
}
}
Mapper:
import com.hazelcast.mapreduce.Context;
import com.hazelcast.mapreduce.Mapper;
import stock.Stock;
public class MinMaxMapper implements Mapper<String, Stock, String, Double> {
static final String MIN = "min";
static final String MAX = "max";
#Override
public void map(String key, Stock value, Context<String, Double> context) {
context.emit(MIN, value.getPrice());
context.emit(MAX, value.getPrice());
}
}
Combiner:
import com.hazelcast.mapreduce.Combiner;
import com.hazelcast.mapreduce.CombinerFactory;
public class MinMaxCombinerFactory implements CombinerFactory<String, Double, Double> {
#Override
public Combiner<Double, Double> newCombiner(String key) {
return new MinMaxCombiner(MinMaxMapper.MAX.equals(key) ? true : false);
}
private static class MinMaxCombiner extends Combiner<Double, Double> {
private final boolean maxCombiner;
private double value;
private MinMaxCombiner(boolean maxCombiner) {
this.maxCombiner = maxCombiner;
this.value = maxCombiner ? -Double.MAX_VALUE : Double.MAX_VALUE;
}
#Override
public void combine(Double value) {
if (maxCombiner) {
this.value = Math.max(value, this.value);
} else {
this.value = Math.min(value, this.value);
}
}
#Override
public Double finalizeChunk() {
return value;
}
#Override
public void reset() {
this.value = maxCombiner ? -Double.MAX_VALUE : Double.MAX_VALUE;
}
}
}
Reducer:
import com.hazelcast.mapreduce.Reducer;
import com.hazelcast.mapreduce.ReducerFactory;
public class MinMaxReducerFactory implements ReducerFactory<String, Double, Double> {
#Override
public Reducer<Double, Double> newReducer(String key) {
return new MinMaxReducer(MinMaxMapper.MAX.equals(key) ? true : false);
}
private static class MinMaxReducer extends Reducer<Double, Double> {
private final boolean maxReducer;
private volatile double value;
private MinMaxReducer(boolean maxReducer) {
this.maxReducer = maxReducer;
this.value = maxReducer ? -Double.MAX_VALUE : Double.MAX_VALUE;
}
#Override
public void reduce(Double value) {
if (maxReducer) {
this.value = Math.max(value, this.value);
} else {
this.value = Math.min(value, this.value);
}
}
#Override
public Double finalizeReduce() {
return value;
}
}
}
Returns two elements map with min and max:
ICompletableFuture<Map<String, Double>> future =
job.mapper(new MinMaxMapper())
.combiner(new MinMaxCombinerFactory())
.reducer(new MinMaxReducerFactory())
.submit();
Map<String, Double> result = future.get();
Why don't you create an ordered index? Although I'm not quite sure if it currently is possible to find a maximum value using a predicate and once found, abort the evaluation of the predicate.
I learned how to use the comparable but I'm having difficulty with the Comparator. I am having a error in my code:
Exception in thread "main" java.lang.ClassCastException: New.People cannot be cast to java.lang.Comparable
at java.util.Arrays.mergeSort(Unknown Source)
at java.util.Arrays.sort(Unknown Source)
at java.util.Collections.sort(Unknown Source)
at New.TestPeople.main(TestPeople.java:18)
Here is my code:
import java.util.Comparator;
public class People implements Comparator {
private int id;
private String info;
private double price;
public People(int newid, String newinfo, double newprice) {
setid(newid);
setinfo(newinfo);
setprice(newprice);
}
public int getid() {
return id;
}
public void setid(int id) {
this.id = id;
}
public String getinfo() {
return info;
}
public void setinfo(String info) {
this.info = info;
}
public double getprice() {
return price;
}
public void setprice(double price) {
this.price = price;
}
public int compare(Object obj1, Object obj2) {
Integer p1 = ((People) obj1).getid();
Integer p2 = ((People) obj2).getid();
if (p1 > p2) {
return 1;
} else if (p1 < p2){
return -1;
} else {
return 0;
}
}
}
import java.util.ArrayList;
import java.util.Collections;
public class TestPeople {
public static void main(String[] args) {
ArrayList peps = new ArrayList();
peps.add(new People(123, "M", 14.25));
peps.add(new People(234, "M", 6.21));
peps.add(new People(362, "F", 9.23));
peps.add(new People(111, "M", 65.99));
peps.add(new People(535, "F", 9.23));
Collections.sort(peps);
for (int i = 0; i < peps.size(); i++){
System.out.println(peps.get(i));
}
}
}
I believe it has to do something with the casting in the compare method but I was playing around with it and still could not find the solution
There are a couple of awkward things with your example class:
it's called People while it has a price and info (more something for objects, not people);
when naming a class as a plural of something, it suggests it is an abstraction of more than one thing.
Anyway, here's a demo of how to use a Comparator<T>:
public class ComparatorDemo {
public static void main(String[] args) {
List<Person> people = Arrays.asList(
new Person("Joe", 24),
new Person("Pete", 18),
new Person("Chris", 21)
);
Collections.sort(people, new LexicographicComparator());
System.out.println(people);
Collections.sort(people, new AgeComparator());
System.out.println(people);
}
}
class LexicographicComparator implements Comparator<Person> {
#Override
public int compare(Person a, Person b) {
return a.name.compareToIgnoreCase(b.name);
}
}
class AgeComparator implements Comparator<Person> {
#Override
public int compare(Person a, Person b) {
return a.age < b.age ? -1 : a.age == b.age ? 0 : 1;
}
}
class Person {
String name;
int age;
Person(String n, int a) {
name = n;
age = a;
}
#Override
public String toString() {
return String.format("{name=%s, age=%d}", name, age);
}
}
EDIT
And an equivalent Java 8 demo would look like this:
public class ComparatorDemo {
public static void main(String[] args) {
List<Person> people = Arrays.asList(
new Person("Joe", 24),
new Person("Pete", 18),
new Person("Chris", 21)
);
Collections.sort(people, (a, b) -> a.name.compareToIgnoreCase(b.name));
System.out.println(people);
Collections.sort(people, (a, b) -> a.age < b.age ? -1 : a.age == b.age ? 0 : 1);
System.out.println(people);
}
}
Here's a super short template to do the sorting right away :
Collections.sort(people, new Comparator<Person>() {
#Override
public int compare(final Person lhs, Person rhs) {
// TODO return 1 if rhs should be before lhs
// return -1 if lhs should be before rhs
// return 0 otherwise (meaning the order stays the same)
}
});
If it's hard to remember, try to just remember that it's similar (in terms of the sign of the number) to:
lhs-rhs
That's in case you want to sort in ascending order : from smallest number to largest number.
Use People implements Comparable<People> instead; this defines the natural ordering for People.
A Comparator<People> can also be defined in addition, but People implements Comparator<People> is not the right way of doing things.
The two overloads for Collections.sort are different:
<T extends Comparable<? super T>> void sort(List<T> list)
Sorts Comparable objects using their natural ordering
<T> void sort(List<T> list, Comparator<? super T> c)
Sorts whatever using a compatible Comparator
You're confusing the two by trying to sort a Comparator (which is again why it doesn't make sense that Person implements Comparator<Person>). Again, to use Collections.sort, you need one of these to be true:
The type must be Comparable (use the 1-arg sort)
A Comparator for the type must be provided (use the 2-args sort)
Related questions
When to use Comparable vs Comparator
Sorting an ArrayList of Contacts
Also, do not use raw types in new code. Raw types are unsafe, and it's provided only for compatibility.
That is, instead of this:
ArrayList peps = new ArrayList(); // BAD!!! No generic safety!
you should've used the typesafe generic declaration like this:
List<People> peps = new ArrayList<People>(); // GOOD!!!
You will then find that your code doesn't even compile!! That would be a good thing, because there IS something wrong with the code (Person does not implements Comparable<Person>), but because you used raw type, the compiler didn't check for this, and instead you get a ClassCastException at run-time!!!
This should convince you to always use typesafe generic types in new code. Always.
See also
What is a raw type and why shouldn't we use it?
For the sake of completeness, here's a simple one-liner compare method:
Collections.sort(people, new Comparator<Person>() {
#Override
public int compare(Person lhs, Person rhs) {
return Integer.signum(lhs.getId() - rhs.getId());
}
});
Java 8 added a new way of making Comparators that reduces the amount of code you have to write, Comparator.comparing. Also check out Comparator.reversed
Here's a sample
import org.junit.Test;
import java.util.ArrayList;
import java.util.Comparator;
import java.util.List;
import static org.junit.Assert.assertTrue;
public class ComparatorTest {
#Test
public void test() {
List<Person> peopleList = new ArrayList<>();
peopleList.add(new Person("A", 1000));
peopleList.add(new Person("B", 1));
peopleList.add(new Person("C", 50));
peopleList.add(new Person("Z", 500));
//sort by name, ascending
peopleList.sort(Comparator.comparing(Person::getName));
assertTrue(peopleList.get(0).getName().equals("A"));
assertTrue(peopleList.get(peopleList.size() - 1).getName().equals("Z"));
//sort by name, descending
peopleList.sort(Comparator.comparing(Person::getName).reversed());
assertTrue(peopleList.get(0).getName().equals("Z"));
assertTrue(peopleList.get(peopleList.size() - 1).getName().equals("A"));
//sort by age, ascending
peopleList.sort(Comparator.comparing(Person::getAge));
assertTrue(peopleList.get(0).getAge() == 1);
assertTrue(peopleList.get(peopleList.size() - 1).getAge() == 1000);
//sort by age, descending
peopleList.sort(Comparator.comparing(Person::getAge).reversed());
assertTrue(peopleList.get(0).getAge() == 1000);
assertTrue(peopleList.get(peopleList.size() - 1).getAge() == 1);
}
class Person {
String name;
int age;
Person(String n, int a) {
name = n;
age = a;
}
public String getName() {
return name;
}
public int getAge() {
return age;
}
public void setName(String name) {
this.name = name;
}
public void setAge(int age) {
this.age = age;
}
}
}
For the sake of completeness.
Using Java8
people.sort(Comparator.comparingInt(People::getId));
if you want in descending order
people.sort(Comparator.comparingInt(People::getId).reversed());
You want to implement Comparable, not Comparator. You need to implement the compareTo method. You're close though. Comparator is a "3rd party" comparison routine. Comparable is that this object can be compared with another.
public int compareTo(Object obj1) {
People that = (People)obj1;
Integer p1 = this.getId();
Integer p2 = that.getid();
if (p1 > p2 ){
return 1;
}
else if (p1 < p2){
return -1;
}
else
return 0;
}
Note, you may want to check for nulls in here for getId..just in case.
Two corrections:
You have to make an ArrayList of People objects:
ArrayList<People> preps = new ArrayList<People>();
After adding the objects to the preps, use:
Collections.sort(preps, new CompareId());
Also, add a CompareId class as:
class CompareId implements Comparator {
public int compare(Object obj1, Object obj2) {
People t1 = (People)obj1;
People t2 = (People)obj2;
if (t1.marks > t2.marks)
return 1;
else
return -1;
}
}
Here's an example of a Comparator that will work for any zero arg method that returns a Comparable. Does something like this exist in a jdk or library?
import java.lang.reflect.Method;
import java.util.Comparator;
public class NamedMethodComparator implements Comparator<Object> {
//
// instance variables
//
private String methodName;
private boolean isAsc;
//
// constructor
//
public NamedMethodComparator(String methodName, boolean isAsc) {
this.methodName = methodName;
this.isAsc = isAsc;
}
/**
* Method to compare two objects using the method named in the constructor.
*/
#Override
public int compare(Object obj1, Object obj2) {
Comparable comp1 = getValue(obj1, methodName);
Comparable comp2 = getValue(obj2, methodName);
if (isAsc) {
return comp1.compareTo(comp2);
} else {
return comp2.compareTo(comp1);
}
}
//
// implementation
//
private Comparable getValue(Object obj, String methodName) {
Method method = getMethod(obj, methodName);
Comparable comp = getValue(obj, method);
return comp;
}
private Method getMethod(Object obj, String methodName) {
try {
Class[] signature = {};
Method method = obj.getClass().getMethod(methodName, signature);
return method;
} catch (Exception exp) {
throw new RuntimeException(exp);
}
}
private Comparable getValue(Object obj, Method method) {
Object[] args = {};
try {
Object rtn = method.invoke(obj, args);
Comparable comp = (Comparable) rtn;
return comp;
} catch (Exception exp) {
throw new RuntimeException(exp);
}
}
}
public static Comparator<JobSet> JobEndTimeComparator = new Comparator<JobSet>() {
public int compare(JobSet j1, JobSet j2) {
int cost1 = j1.cost;
int cost2 = j2.cost;
return cost1-cost2;
}
};
The solution can be optimized in following way:
Firstly, use a private inner class as the scope for the fields is to be the enclosing class TestPeople so as the implementation of class People won't get exposed to outer world. This can be understood in terms of creating an APIthat expects a sorted list of people
Secondly, using the Lamba expression(java 8) which reduces the code, hence development effort
Hence code would be as below:
import java.util.ArrayList;
import java.util.Collections;
import java.util.Comparator;
public class TestPeople {
public static void main(String[] args) {
ArrayList<People> peps = new ArrayList<>();// Be specific, to avoid
// classCast Exception
TestPeople test = new TestPeople();
peps.add(test.new People(123, "M", 14.25));
peps.add(test.new People(234, "M", 6.21));
peps.add(test.new People(362, "F", 9.23));
peps.add(test.new People(111, "M", 65.99));
peps.add(test.new People(535, "F", 9.23));
/*
* Collections.sort(peps);
*
* for (int i = 0; i < peps.size(); i++){
* System.out.println(peps.get(i)); }
*/
// The above code can be replaced by followin:
peps.sort((People p1, People p2) -> p1.getid() - p2.getid());
peps.forEach((p) -> System.out.println(" " + p.toString()));
}
private class People {
private int id;
#Override
public String toString() {
return "People [id=" + id + ", info=" + info + ", price=" + price + "]";
}
private String info;
private double price;
public People(int newid, String newinfo, double newprice) {
setid(newid);
setinfo(newinfo);
setprice(newprice);
}
public int getid() {
return id;
}
public void setid(int id) {
this.id = id;
}
public String getinfo() {
return info;
}
public void setinfo(String info) {
this.info = info;
}
public double getprice() {
return price;
}
public void setprice(double price) {
this.price = price;
}
}
}
Here is a lambda version of comparator. This will sort a string list according to length.
Collections.sort(str, (str1, str2) -> {
if(str1.length() < str2.length())
return 1;
else if(str2.length() < str1.length())
return -1;
else
return 0;
});
You should use the overloaded sort(peps, new People()) method
import java.util.ArrayList;
import java.util.Collections;
import java.util.Comparator;
import java.util.List;
public class Test
{
public static void main(String[] args)
{
List<People> peps = new ArrayList<>();
peps.add(new People(123, "M", 14.25));
peps.add(new People(234, "M", 6.21));
peps.add(new People(362, "F", 9.23));
peps.add(new People(111, "M", 65.99));
peps.add(new People(535, "F", 9.23));
Collections.sort(peps, new People().new ComparatorId());
for (int i = 0; i < peps.size(); i++)
{
System.out.println(peps.get(i));
}
}
}
class People
{
private int id;
private String info;
private double price;
public People()
{
}
public People(int newid, String newinfo, double newprice) {
setid(newid);
setinfo(newinfo);
setprice(newprice);
}
public int getid() {
return id;
}
public void setid(int id) {
this.id = id;
}
public String getinfo() {
return info;
}
public void setinfo(String info) {
this.info = info;
}
public double getprice() {
return price;
}
public void setprice(double price) {
this.price = price;
}
class ComparatorId implements Comparator<People>
{
#Override
public int compare(People obj1, People obj2) {
Integer p1 = obj1.getid();
Integer p2 = obj2.getid();
if (p1 > p2) {
return 1;
} else if (p1 < p2){
return -1;
} else {
return 0;
}
}
}
}
Here is my answer for a simple comparator tool
public class Comparator {
public boolean isComparatorRunning = false;
public void compareTableColumns(List<String> tableNames) {
if(!isComparatorRunning) {
isComparatorRunning = true;
try {
for (String schTableName : tableNames) {
Map<String, String> schemaTableMap = ComparatorUtil.getSchemaTableMap(schTableName);
Map<String, ColumnInfo> primaryColMap = ComparatorUtil.getColumnMetadataMap(DbConnectionRepository.getConnectionOne(), schemaTableMap);
Map<String, ColumnInfo> secondaryColMap = ComparatorUtil.getColumnMetadataMap(DbConnectionRepository.getConnectionTwo(), schemaTableMap);
ComparatorUtil.publishColumnInfoOutput("Comparing table : "+ schemaTableMap.get(CompConstants.TABLE_NAME));
compareColumns(primaryColMap, secondaryColMap);
}
} catch (Exception e) {
ComparatorUtil.publishColumnInfoOutput("ERROR"+e.getMessage());
}
isComparatorRunning = false;
}
}
public void compareColumns(Map<String, ColumnInfo> primaryColMap, Map<String, ColumnInfo> secondaryColMap) {
try {
boolean isEqual = true;
for(Map.Entry<String, ColumnInfo> entry : primaryColMap.entrySet()) {
String columnName = entry.getKey();
ColumnInfo primaryColInfo = entry.getValue();
ColumnInfo secondaryColInfo = secondaryColMap.remove(columnName);
if(secondaryColInfo == null) {
// column is not present in Secondary Environment
ComparatorUtil.publishColumnInfoOutput("ALTER", primaryColInfo);
isEqual = false;
continue;
}
if(!primaryColInfo.equals(secondaryColInfo)) {
isEqual = false;
// Column not equal in secondary env
ComparatorUtil.publishColumnInfoOutput("MODIFY", primaryColInfo);
}
}
if(!secondaryColMap.isEmpty()) {
isEqual = false;
for(Map.Entry<String, ColumnInfo> entry : secondaryColMap.entrySet()) {
// column is not present in Primary Environment
ComparatorUtil.publishColumnInfoOutput("DROP", entry.getValue());
}
}
if(isEqual) {
ComparatorUtil.publishColumnInfoOutput("--Exact Match");
}
} catch (Exception e) {
ComparatorUtil.publishColumnInfoOutput("ERROR"+e.getMessage());
}
}
public void compareTableColumnsValues(String primaryTableName, String primaryColumnNames, String primaryCondition, String primaryKeyColumn,
String secTableName, String secColumnNames, String secCondition, String secKeyColumn) {
if(!isComparatorRunning) {
isComparatorRunning = true;
Connection conn1 = DbConnectionRepository.getConnectionOne();
Connection conn2 = DbConnectionRepository.getConnectionTwo();
String query1 = buildQuery(primaryTableName, primaryColumnNames, primaryCondition, primaryKeyColumn);
String query2 = buildQuery(secTableName, secColumnNames, secCondition, secKeyColumn);
try {
Map<String,Map<String, Object>> query1Data = executeAndRefactorData(conn1, query1, primaryKeyColumn);
Map<String,Map<String, Object>> query2Data = executeAndRefactorData(conn2, query2, secKeyColumn);
for(Map.Entry<String,Map<String, Object>> entry : query1Data.entrySet()) {
String key = entry.getKey();
Map<String, Object> value = entry.getValue();
Map<String, Object> secondaryValue = query2Data.remove(key);
if(secondaryValue == null) {
ComparatorUtil.publishColumnValuesInfoOutput("NO SUCH VALUE AVAILABLE IN SECONDARY DB "+ value.toString());
continue;
}
compareMap(value, secondaryValue, key);
}
if(!query2Data.isEmpty()) {
ComparatorUtil.publishColumnValuesInfoOutput("Extra Values in Secondary table "+ ((Map)query2Data.values()).values().toString());
}
} catch (Exception e) {
ComparatorUtil.publishColumnValuesInfoOutput("ERROR"+e.getMessage());
}
isComparatorRunning = false;
}
}
private void compareMap(Map<String, Object> primaryValues, Map<String, Object> secondaryValues, String columnIdentification) {
for(Map.Entry<String, Object> entry : primaryValues.entrySet()) {
String key = entry.getKey();
Object value = entry.getValue();
Object secValue = secondaryValues.get(key);
if(value!=null && secValue!=null && !String.valueOf(value).equalsIgnoreCase(String.valueOf(secValue))) {
ComparatorUtil.publishColumnValuesInfoOutput(columnIdentification+" : Secondary Table does not match value ("+ value +") for column ("+ key+")");
}
if(value==null && secValue!=null) {
ComparatorUtil.publishColumnValuesInfoOutput(columnIdentification+" : Values not available in primary table for column "+ key);
}
if(value!=null && secValue==null) {
ComparatorUtil.publishColumnValuesInfoOutput(columnIdentification+" : Values not available in Secondary table for column "+ key);
}
}
}
private String buildQuery(String tableName, String column, String condition, String keyCol) {
if(!"*".equalsIgnoreCase(column)) {
String[] keyColArr = keyCol.split(",");
for(String key: keyColArr) {
if(!column.contains(key.trim())) {
column+=","+key.trim();
}
}
}
StringBuilder queryBuilder = new StringBuilder();
queryBuilder.append("select "+column+" from "+ tableName);
if(!ComparatorUtil.isNullorEmpty(condition)) {
queryBuilder.append(" where 1=1 and "+condition);
}
return queryBuilder.toString();
}
private Map<String,Map<String, Object>> executeAndRefactorData(Connection connection, String query, String keyColumn) {
Map<String,Map<String, Object>> result = new HashMap<String, Map<String,Object>>();
try {
PreparedStatement preparedStatement = connection.prepareStatement(query);
ResultSet resultSet = preparedStatement.executeQuery();
resultSet.setFetchSize(1000);
if (resultSet != null && !resultSet.isClosed()) {
while (resultSet.next()) {
Map<String, Object> columnValueDetails = new HashMap<String, Object>();
int columnCount = resultSet.getMetaData().getColumnCount();
for (int i=1; i<=columnCount; i++) {
String columnName = String.valueOf(resultSet.getMetaData().getColumnName(i));
Object columnValue = resultSet.getObject(columnName);
columnValueDetails.put(columnName, columnValue);
}
String[] keys = keyColumn.split(",");
String newKey = "";
for(int j=0; j<keys.length; j++) {
newKey += String.valueOf(columnValueDetails.get(keys[j]));
}
result.put(newKey , columnValueDetails);
}
}
} catch (SQLException e) {
ComparatorUtil.publishColumnValuesInfoOutput("ERROR"+e.getMessage());
}
return result;
}
}
Utility Tool for the same
public class ComparatorUtil {
public static Map<String, String> getSchemaTableMap(String tableNameWithSchema) {
if(isNullorEmpty(tableNameWithSchema)) {
return null;
}
Map<String, String> result = new LinkedHashMap<>();
int index = tableNameWithSchema.indexOf(".");
String schemaName = tableNameWithSchema.substring(0, index);
String tableName = tableNameWithSchema.substring(index+1);
result.put(CompConstants.SCHEMA_NAME, schemaName);
result.put(CompConstants.TABLE_NAME, tableName);
return result;
}
public static Map<String, ColumnInfo> getColumnMetadataMap(Connection conn, Map<String, String> schemaTableMap) {
try {
String schemaName = schemaTableMap.get(CompConstants.SCHEMA_NAME);
String tableName = schemaTableMap.get(CompConstants.TABLE_NAME);
ResultSet resultSetConnOne = conn.getMetaData().getColumns(null, schemaName, tableName, null);
Map<String, ColumnInfo> resultSetTwoColInfo = getColumnInfo(schemaName, tableName, resultSetConnOne);
return resultSetTwoColInfo;
} catch (SQLException e) {
e.printStackTrace();
}
return null;
}
/* Number Type mapping
* 12-----VARCHAR
* 3-----DECIMAL
* 93-----TIMESTAMP
* 1111-----OTHER
*/
public static Map<String, ColumnInfo> getColumnInfo(String schemaName, String tableName, ResultSet columns) {
try {
Map<String, ColumnInfo> tableColumnInfo = new LinkedHashMap<String, ColumnInfo>();
while (columns.next()) {
ColumnInfo columnInfo = new ColumnInfo();
columnInfo.setSchemaName(schemaName);
columnInfo.setTableName(tableName);
columnInfo.setColumnName(columns.getString("COLUMN_NAME"));
columnInfo.setDatatype(columns.getString("DATA_TYPE"));
columnInfo.setColumnsize(columns.getString("COLUMN_SIZE"));
columnInfo.setDecimaldigits(columns.getString("DECIMAL_DIGITS"));
columnInfo.setIsNullable(columns.getString("IS_NULLABLE"));
tableColumnInfo.put(columnInfo.getColumnName(), columnInfo);
}
return tableColumnInfo;
} catch (Exception e) {
e.printStackTrace();
}
return null;
}
public static boolean isNullOrEmpty(Object obj) {
if (obj == null)
return true;
if (String.valueOf(obj).equalsIgnoreCase("NULL"))
return true;
if (obj.toString().trim().length() == 0)
return true;
return false;
}
public static boolean isNullorEmpty(String str) {
if(str == null)
return true;
if(str.trim().length() == 0)
return true;
return false;
}
public static void publishColumnInfoOutput(String type, ColumnInfo columnInfo) {
String str = "ALTER TABLE "+columnInfo.getSchemaName()+"."+columnInfo.getTableName();
switch(type.toUpperCase()) {
case "ALTER":
if("NUMBER".equalsIgnoreCase(columnInfo.getDatatype()) || "DATE".equalsIgnoreCase(columnInfo.getDatatype())) {
str += " ADD ("+columnInfo.getColumnName()+" "+ columnInfo.getDatatype()+");";
} else {
str += " ADD ("+columnInfo.getColumnName()+" "+ columnInfo.getDatatype() +"("+columnInfo.getColumnsize()+"));";
}
break;
case "DROP":
str += " DROP ("+columnInfo.getColumnName()+");";
break;
case "MODIFY":
if("NUMBER".equalsIgnoreCase(columnInfo.getDatatype()) || "DATE".equalsIgnoreCase(columnInfo.getDatatype())) {
str += " MODIFY ("+columnInfo.getColumnName()+" "+ columnInfo.getDatatype()+");";
} else {
str += " MODIFY ("+columnInfo.getColumnName()+" "+ columnInfo.getDatatype() +"("+columnInfo.getColumnsize()+"));";
}
break;
}
publishColumnInfoOutput(str);
}
public static Map<Integer, String> allJdbcTypeName = null;
public static Map<Integer, String> getAllJdbcTypeNames() {
Map<Integer, String> result = new HashMap<Integer, String>();
if(allJdbcTypeName != null)
return allJdbcTypeName;
try {
for (Field field : java.sql.Types.class.getFields()) {
result.put((Integer) field.get(null), field.getName());
}
} catch (Exception e) {
e.printStackTrace();
}
return allJdbcTypeName=result;
}
public static String getStringPlaces(String[] attribs) {
String params = "";
for(int i=0; i<attribs.length; i++) { params += "?,"; }
return params.substring(0, params.length()-1);
}
}
Column Info Class
public class ColumnInfo {
private String schemaName;
private String tableName;
private String columnName;
private String datatype;
private String columnsize;
private String decimaldigits;
private String isNullable;
If you are using Java 8 then it's better to use below code like this:
Comparator<People> comparator = Comparator.comparing(People::getName);
And then simply use:
Collections.sort(list, comparator);
If you are using Java 7 or below then you can use a comparator for customized sorting order by implementing compare method.
For example:
import java.util.Comparator;
public class PeopleNameComparator implements Comparator<People> {
#Override
public int compare(People people1, People people2) {
return people1.getName().compareTo(people2.getName());
}
}
And then simply use like this:
Collections.sort(list, new PeopleNameComparator);
Do not waste time implementing Sorting Algorithm by your own. Instead use Collections.sort() to sort data.