Changing the Java language using OpenJDK – a small experiment

Of late, I have been fascinated with languages, compilers, and specifically parsing and code generation. To that end, I started looking into how the Java language itself might be modified. It turns out that Java itself uses a handwritten lexer and parser with a very well-defined tree structure for the AST. That being said, good documentation if hard to come by.

I chanced upon this gem of a blog – Making Java esoteric which shows a simple example of how to add a maybe keyword to Java so that it may randomly execute statements and/or blocks.

I was inspired enough to try out my first hack on OpenJDK. To that end, I decided to implement a when statement that would essentially act as its Common Lisp equivalent.

My attempt at implementing the when keyword didn’t go down so well since it appears that the word has been used in umpteen places in the JDK itself. All right, let’s just use the Russian equivalent, kogda meaning essentially the same.

My basic logic was to define the new keyword, kogda that would work with block statements or simple statements, and then offload the actual work to the if statement. So here were the code changes in order:

First off the lexer:
/langtools/src/jdk.compiler/share/classes/com/sun/tools/javac/parser/Tokens.java):

 public enum TokenKind implements Formattable, Filter<TokenKind> {
    ...
    KOGDA("kogda"),
    ...
}

And then the parser:/langtools/src/jdk.compiler/share/classes/com/sun/tools/javac/parser/JavacParer.java:

 List<JCStatement> blockStatement() {
    ...
    case KOGDA:
    ...
    case ASSERT:
            return List.of(parseSimpleStatement());
    ...
}

and

 public JCStatement parseSimpleStatement() {
   ...
     case KOGDA: {
                nextToken();
                JCExpression cond = parExpression();
                JCStatement body = parseStatementAsBlock();
                return F.at(pos).Kogda(cond, body);
            }
   ...
}

Nothing special here – we simply read in the next token, get the conditional part, and then get the body of the when expression/statement.

And then finally, langtools/src/jdk.compiler/share/classes/com/sun/tools/javac/tree/TreeMaker.java:

public JCIf Kogda(JCExpression cond, JCStatement body) {
        JCIf tree = new JCIf(cond, body, null);
        tree.pos = pos;

        return tree;
    }

This is the actual meat of the logic per se. In this case, I simply offload the work to the reliable if expression and insert null for the else part. That’s all there is to it!

Then we build the new version of the JDK:

Macushla:9dev z0ltan$ pwd
/Users/z0ltan/Projects/Resources/Languages/Java/9dev

Macushla:9dev z0ltan$ make clean && make images
<output elided>

Macushla:9dev z0ltan$ cd build/macosx-x86_64-normal-server-release/jdk/bin/

Macushla:bin z0ltan$ ./javac -version
javac 9-internal

Here’s our little test program to test out:

Macushla:bin z0ltan$ cat HelloKogda.java
public class HelloKogda {
    public static void main(String[] args) {
        kogda (1 < 2) {
            System.out.println("Yes, 1 < 2");
        }

        kogda (2 > 100) {
            System.out.println("2 > 100");
        }

        kogda (1 < 10 && 1 == 1) System.out.println("Booyah!");
    }
}

And finally, compiling and running the test code using our custom-built JDK:

Macushla:bin z0ltan$ ./javac HelloKogda.java
Macushla:bin z0ltan$ ./java -cp . HelloKogda
Yes, 1 < 2
Booyah!

Nice! There was just one thing that I was not completely happy with – JShell. In theory, it should be trivial to modify JShell to reflect these same changes, but I had immense trouble trying to get the jdk.jshell project pick up the changes from my modified jdk.compiler project instead of the default JDK (which, of course, does not contain my changes). Maybe when I work my head around the internals of the whole ecosystem, I will post an update here. For now, this was a fun experiment!

Changing the Java language using OpenJDK – a small experiment

A parser using Regular Expressions?

The oft-quoted joke is that Regular Expressions are such a useful tool that it is inevitably abused beyond its abilities. Case in point – Dmitry Olshansky’s Dconf 2017 talk where he mentions the tonnes of questions on StackOverflow about how to write a full-fledged HTML parser using only regular expressions (not quite possible!).

That amused me to such an extent that I decided to experiment a bit with regular expressions and to see if I could use a regular expression engine (which, of course, do not conform to strict regular languages (think backreferences, etc.). A modern regex engine (such as Java’s java.util.regex package) is suitably powerful enough to be a perfect candidate for such wanton abuse!

All right, so what’s the scope here? For the purpose of this experiment, I will restrict the scope of the parser to the following:

  1. Allow List initialisation of the form
              List<Double> specialNos = { 3.14519, 2.71828, 1.818181 }
             
  2. Allow similar syntactic sugar for SetS:
              Set<String> languages = { "Java", "Rust", "Haskell", "Common Lisp", "C++" }
             
  3. A Rubyesque Map initialisation syntax:
              Map<Integer, String> nos = { 1 => "One"; 2 => "Two"; 3 => "Three" }
              

So here’s how a sample file might look like:

Macushla:basic-regex-parser z0ltan$ cat samples/com/z0ltan/regexparser/test/HelloParser.javax
package com.z0ltan.regexparser.test;

import java.util.Arrays;
import java.util.HashMap;
import java.util.HashSet;
import java.util.List;
import java.util.Map;
import java.util.Set;

public class HelloParser {
    public static void main(String[] args) {
        listDemo();

        setDemo();

        mapDemo();
    }

    private static void listDemo() {
        // normal list
        List<String> greetings = Arrays.asList("Hello", "Hallo", "Hola");
        System.out.printf("greetings = %s\n", greetings);

        // custom list syntax
        List<Integer> myNums =
                    { 1, 3, 5, 7, 9, 11 }

        System.out.printf("myNums = %s\n", myNums);

    }

    private static void setDemo() {
        // normal set
        Set<Integer> evensBelow10 = new HashSet<>();
        evensBelow10.addAll(Arrays.asList(2, 4, 6, 8));

        System.out.printf("evensBelow10 = %s\n", evensBelow10);

        // custom set syntax
        Set<String> mammals = { "Cat", "Dog", "Lion", "Beaver", "Raccoon" }

        System.out.printf("mammals = %s\n", mammals);
    }

    private static void mapDemo() {
        // normal map
        Map<Integer, String> numberToString = new HashMap<>();
        numberToString.put(1,  "One");
        numberToString.put(2, "Two");
        numberToString.put(3,  "Three");

        System.out.printf("numberToString = %s\n",  numberToString);

        // custom map syntax
        Map<Integer, Foo> numToFoo =
                {
                    1 => new Foo(1, "Ein");

                    2 => new Foo(2, "Zwei");

                    3 => new Foo(3, "Drei");

                    4 => new Foo(4, "Vier");

                    5 => new Foo(5, "Fuenf");
                }

        System.out.printf("numToFoo = %s\n", numToFoo);
    }

    static class Foo {
        private int id;
        private String name;

        public Foo(int id, String name) {
            this.id = id;
            this.name = name;
        }

        @Override
        public String toString() {
            return "Foo { id = " + id + ", name = " + name + " }";
        }
    }
}

So this sample input file includes examples for all three variants. So how would the parser for this look like? Let’s split the parser into three variants as well:

The parser for List:

/**
	 * Handle custom List syntax.
	 * 
	 * @param contents
	 * @return
	 */
	private static String parseListSyntax(String contents) {
		Matcher m = Patterns.LIST_PATTERN.matcher(contents);

		while (m.find()) {
			if (m.groupCount() == 3) {
				String type = m.group(1);
				String var = m.group(2);
				String val = m.group(3).trim();

				String replacement = "List<" + type + ">" + var + " = new ArrayList<" + type + ">() {{ ";
				String[] vals = val.split(",");

				for (String rep : vals) {
					String v = rep.trim();

					if (!v.isEmpty()) {
						replacement += "add(" + v + ");\n\t\t\t\t\t ";
					}
				}
				replacement += "}};";

				contents = m.replaceFirst(replacement);

				m = Patterns.LIST_PATTERN.matcher(contents);
			}
		}
		return contents;
	}

The parser for Set:

/**
	 * Handle custom Set syntax
	 * 
	 * @param contents
	 * @return
	 */
	private static String parseSetSyntax(String contents) {
		Matcher m = Patterns.SET_PATTERN.matcher(contents);

		while (m.find()) {
			if (m.groupCount() == 3) {
				String type = m.group(1);
				String var = m.group(2);
				String val = m.group(3).trim();

				String replacement = "Set<" + type + "> " + var + " = new HashSet<" + type + ">() {{ ";
				String[] vals = val.split(",");

				for (String rep : vals) {
					String v = rep.trim();

					if (!v.isEmpty()) {
						replacement += "add(" + v + ");\n\t\t\t\t\t";
					}
				}

				replacement += " }};";

				contents = m.replaceFirst(replacement);

				m = Patterns.SET_PATTERN.matcher(contents);
			}
		}

		return contents;
	}

And finally, the parser for Map:

/**
	 * Handle custom Map syntax
	 * 
	 * @param contents
	 * @return
	 */
	private static String parseMapSyntax(String contents) {
		Matcher m = Patterns.MAP_PATTERN.matcher(contents);

		while (m.find()) {
			if (m.groupCount() == 3) {
				String type = m.group(1);
				String var = m.group(2);
				String val = m.group(3).trim();

				String replacement = "Map<" + type + "> " + var + " = new HashMap<" + type + ">() {{";
				String[] vals = val.split(";");

				for (String rep : vals) {
					String v = rep.trim();

					if (!rep.isEmpty()) {
						String[] keyVal = v.split("=>");

						if (keyVal == null || keyVal.length != 2 || keyVal[0] == null || keyVal[1] == null) {
							throw new RuntimeException("invalid map syntax");
						}
						replacement += "put(" + keyVal[0].trim() + ", " + keyVal[1].trim() + ");\n\t\t\t\t\t";
					}
				}

				replacement += " }};";

				contents = m.replaceFirst(replacement);

				m = Patterns.MAP_PATTERN.matcher(contents);
			}
		}

		return contents;
	}

The code is quite straightforward and similar for all three variants – we simply to see if the relevant pattern is found in the current code contents (the whole source file is read in in one go), and if so, the relevant bit is rewritten, and the updated String is then passed to the next stage of the pipeline for further processing.

One thing to observe here is that in every parsing method, the find operation is performed in a loop. This is so that every occurrence can be updated correctly, and that is also the reason why the Matcher is updated at the end of every iteration of the loop – m = Patterns.LIST_PATTERN.matcher(contents), for instance. If this is not done so, the result will be an infinite loop since the while (m.find()) will never fail!

Here’s how the main parsing method looks like:

public static void parse(final String fileName) {
		try (BufferedReader reader = new BufferedReader(new FileReader(fileName))) {
			String ext = fileName.substring(fileName.lastIndexOf('.') + 1);

			if (!ext.equalsIgnoreCase("javax")) {
				throw new RuntimeException("input files must have a .javax extension");
			}

			File srcFile = new File(fileName);

			if (!srcFile.exists()) {
				throw new RuntimeException("source file " + fileName + " does not exist!");
			}

			final String basePath = srcFile.getAbsolutePath().substring(0,
					srcFile.getAbsolutePath().lastIndexOf(File.separator));
			final String baseFileName = srcFile.getName().substring(0, srcFile.getName().lastIndexOf('.'));
			final String outputFile = basePath + File.separator + baseFileName
					+ ".java";

			String contents = reader.lines().reduce("", (acc, s) -> {
				acc += s;
				acc += "\n";
				return acc;
			});

			contents = insertImports(contents);
			contents = parseListSyntax(contents);
			contents = parseSetSyntax(contents);
			contents = parseMapSyntax(contents);

			try (BufferedWriter writer = new BufferedWriter(new FileWriter(outputFile))) {
				writer.write(contents);
			} catch (Exception ex) {
				throw ex;
			}

		} catch (Exception ex) {
			throw new RuntimeException("Error during parsing: " + ex.getLocalizedMessage());
		}
	}

A final subtlety is in the way the importS are handled. For the full source code as well as more explanation on the way the whole tool works, visit the Github page.

Let’s take it for a spin:

Macushla:basic-regex-parser z0ltan$ ls
README.md pom.xml   samples   src       target

Macushla:basic-regex-parser z0ltan$ mvn package 1> /dev/null

Macushla:basic-regex-parser z0ltan$ java -jar target/basic-regex-parser-1.0-SNAPSHOT.jar samples/Default.javax samples/com/z0ltan/test/Package.javax samples/com/z0ltan/regexparser/test/HelloParser.javax

Let’s see how the sample file shown earlier has been rewritten:

Macushla:basic-regex-parser z0ltan$ cat samples/com/z0ltan/regexparser/test/HelloParser.java
package com.z0ltan.regexparser.test;

import java.util.*;

import java.util.Arrays;
import java.util.HashMap;
import java.util.HashSet;
import java.util.List;
import java.util.Map;
import java.util.Set;

public class HelloParser {
    public static void main(String[] args) {
        listDemo();

        setDemo();

        mapDemo();
    }

    private static void listDemo() {
        // normal list
        List<String> greetings = Arrays.asList("Hello", "Hallo", "Hola");
        System.out.printf("greetings = %s\n", greetings);

        // custom list syntax
        List<Integer>myNums = new ArrayList<Integer>() {{ add(1);
					 add(3);
					 add(5);
					 add(7);
					 add(9);
					 add(11);
					 }};

        System.out.printf("myNums = %s\n", myNums);

    }

    private static void setDemo() {
        // normal set
        Set<Integer> evensBelow10 = new HashSet<>();
        evensBelow10.addAll(Arrays.asList(2, 4, 6, 8));

        System.out.printf("evensBelow10 = %s\n", evensBelow10);

        // custom set syntax
        Set<String> mammals = new HashSet<String>() {{ add("Cat");
					add("Dog");
					add("Lion");
					add("Beaver");
					add("Raccoon");
					 }};

        System.out.printf("mammals = %s\n", mammals);
    }

    private static void mapDemo() {
        // normal map
        Map<Integer, String> numberToString = new HashMap<>();
        numberToString.put(1,  "One");
        numberToString.put(2, "Two");
        numberToString.put(3,  "Three");

        System.out.printf("numberToString = %s\n",  numberToString);

        // custom map syntax
        Map<Integer, Foo> numToFoo = new HashMap<Integer, Foo>() {{put(1, new Foo(1, "Ein"));
					put(2, new Foo(2, "Zwei"));
					put(3, new Foo(3, "Drei"));
					put(4, new Foo(4, "Vier"));
					put(5, new Foo(5, "Fuenf"));
					 }};

        System.out.printf("numToFoo = %s\n", numToFoo);
    }

    static class Foo {
        private int id;
        private String name;

        public Foo(int id, String name) {
            this.id = id;
            this.name = name;
        }

        @Override
        public String toString() {
            return "Foo { id = " + id + ", name = " + name + " }";
        }
    }
}

Excellent! Now let’s try and run all the parsed files to make sure they are all working as expected:

Macushla:basic-regex-parser z0ltan$ cd samples/

Macushla:samples z0ltan$ javac Default.java
Macushla:samples z0ltan$ java -cp . Default
users = [Cornelius, Marcus, Linnaeus, Gaius, Augustus]

Macushla:samples z0ltan$ javac com/z0ltan/test/Package.java
Macushla:samples z0ltan$ java -cp . com.z0ltan.test.Package
map = {One=1, Two=2, Three=3}
normalMap = {Hello=100, Again=300, World=200}

Macushla:samples z0ltan$ javac com/z0ltan/regexparser/test/HelloParser.java
Macushla:samples z0ltan$ java -cp . com.z0ltan.regexparser.test.HelloParser
greetings = [Hello, Hallo, Hola]
myNums = [1, 3, 5, 7, 9, 11]
evensBelow10 = [2, 4, 6, 8]
mammals = [Cat, Lion, Beaver, Raccoon, Dog]
numberToString = {1=One, 2=Two, 3=Three}
numToFoo = {1=Foo { id = 1, name = Ein }, 2=Foo { id = 2, name = Zwei }, 3=Foo { id = 3, name = Drei }, 4=Foo { id = 4, name = Vier }, 5=Foo { id = 5, name = Fuenf }}

Nice! Of course, this is just a toy tool that does trivial rewriting. If Java had a proper macro system (or even a macro system to begin with!), all this would have been trivial at best without so much redundant code. Nevertheless, it does illustrate the regular expression engines are indeed quite powerful well beyond their original intent.

EDIT: Just in case you are curious as to what the patterns look like, here they are:

public class Patterns {
	public static final Pattern LIST_PATTERN = Pattern
			.compile("List<([a-zA-Z]+)>\\s+([a-zA-Z0-9]+)\\s*=\\s*\\{\\s*([a-zA-Z0-9,\" ]+)\\s*\\}");

	public static final Pattern SET_PATTERN = Pattern
			.compile("Set<([a-zA-Z]+)>\\s+([a-zA-Z0-9]+)\\s*=\\s*\\{\\s*([a-zA-Z0-9,\" ]+)\\s*\\}");

	public static final Pattern MAP_PATTERN = Pattern
			.compile("Map<([a-zA-Z, ]+)>\\s+([a-zA-Z0-9]+)\\s*=\\s*\\{\\s*([a-zA-Z0-9\"\\s=>();, ]+)\\s*\\}");
	
	public static final Pattern PACKAGE_PATTERN = Pattern
			.compile("package\\s+([a-zA-Z0-9._;]+)");
	
	public static final Pattern IMPORT_PATTERN = Pattern
			.compile("import\\s+([a-zA-Z0-9.;]+)");
	
	public static final Pattern PUBLIC_CLASS_PATTERN = Pattern
			.compile("public class\\s+([a-zA-Z0-9_]+)");
	
	public static final Pattern NORMAL_CLASS_PATTERN = Pattern
			.compile("class\\s+([a-zA-Z0-9_]+)");
}

You may safely ignore the patterns after the first three patterns – they are used to insert the import java.util.*; declaration at the first valid location. Again, for more details, visit the Github location!

A parser using Regular Expressions?

Creating custom Java 8 Stream collectors

The streams feature (which makes heavy use of Functional Interfaces) is arguably the strongest feature in Java 8 (and above). The introduction of lambdas in Java in the same release to target Functional Interfaces also meant that creating chained operations in a functional style has never been easier in Java.

That being said, there are plenty of examples in the official docs that show how streams can be “collected” (effectively reduced/folded depending on your preference of the terminology) into Collection types – List, Set, or even Map. Now, I must make it absolutely clear at this stage that I love Java streams, and barring the absence (or rather, deprecation) of the zip iterator, it’s almost comprehensive. However, there are no examples in the docs to show how we might a series of intermediate operations into a custom type. Of course the helper class, Collectors has several helper methods such as groupingBy, partitioningBy, filtering, and reducing, but they either return a Map, or expect a reducible expression which may not always be the case as explained next.

Recently, I did a project in which I needed to process a stream of integers (the lack of zip forced me to take quite the peripatetic approach to finally make things work. Perhaps more on that in a later blogpost) acting as indices into a simple wrapper around a List of integers, and then ultimately collect the updated values of the list into a new instance of the custom type. It was quite an interesting experience that sparked interest in exploring how much more we could push the collect mechanism. (If you are interested in checking out the code for the mentioned example, you can find it here – Functional Nim.

For some more examples of custom Collector implementations, you can check out my Github page.

Use Case

For the purposes of this blog, to keep things simple, let us consider a a hypothetical example. Suppose we have a Point class with the following structure:

package com.z0ltan.custom.collectors.types;

public class Point {
	private int x;
	private int y;

	public Point(final int x, final int y) {
		this.x = x;
		this.y = y;
	}

	public int x() {
		return this.x;
	}

	public int y() {
		return this.y;
	}

	@Override
	public int hashCode() {
		return (this.x + this.y) % 31;
	}

	@Override
	public boolean equals(Object other) {
		if (other == null || !(other instanceof Point)) {
			return false;
		}

		Point p = (Point) other;

		return p.x == this.x && p.y == this.y;
	}

	@Override
	public String toString() {
		return "Point { x = " + x + ", y = " + y + " }";
	}
}

and we have a List of such points. Now suppose we wish to collate the information into a custom Points object which has the following structure:

package com.z0ltan.custom.collectors.types;

import java.util.Set;

public class Points {
	private Set<Integer> xs;
	private Set<Integer> ys;

	public Points(final Set<Integer> xs, final Set<Integer> ys) {
		this.xs = xs;
		this.ys = ys;
	}

	@Override
	public int hashCode() {
		return (this.xs.hashCode() + this.ys.hashCode()) % 31;
	}

	@Override
	public boolean equals(Object other) {
		if (other == null || !(other instanceof Points)) {
			return false;
		}

		Points p = (Points) other;

		return p.xs.equals(this.xs) && p.ys.equals(this.ys);
	}

	@Override
	public String toString() {
		return "Points { xs = " + xs + ", ys = " + ys + " }";
	}
}

As can be seen, we wish to retain only the unique x and y coordinate values into our final object.

A simple and logical way would be to collect the items from this stream (collect being a “terminal operation” defined in the Stream interface) into our Points object using a custom collector. Before we can do that, let us first understand what is involved in implementing a custom collector.

How to implement a custom collector

In brief, there are two forms of operations involved in Java streams – non-terminal or intermediate operations, which produce streams of their own, and terminal operations, which effectively stop the pipeline resulting in a final result of some sort.

As mentioned before, in most cases, the helper class, Collectors, provides enough functionality to cater to almost any requirement imaginable. However, in cases such as these, where we want to collect data into a custom type, we might be better off defining our own custom collector.

To do that, let us examine the signature of the collect method in the Stream interface. In fact, we will find that there are two versions of this method available to us:

 R collect​(Collector collector)

and

 R collect​(Supplier supplier,
              BiConsumer accumulator,
              BiConsumer combiner)

So which one do we use? Well, we can actually use either one for our purposes. In fact, the former is preferable if we have some involved logic, and we wish to encapsulate all of that in a nice class. However, functionally speaking, the latter is exactly the same. This can be further clarified by examining the Collector interface (showing only the abstract methods):

public interface Collector<T, A, R> {
        Supplier<A>	supplier​()
        BiConsumer<A,T>	accumulator​();	
        BinaryOperator<A>	combiner​()	
        Function<A,R>	finisher​()
        Set<Collector.Characteristics>	characteristics​()	
 }

As can be seen, if we do implement the Collector interface, we will have to implement essentially the same methods as that used by the second version of the collect method. In addition, we have a couple of extra methods which are not only interesting, but quite vital if we wish to implement the interface:

  • finisher: This is the main method that we need to implement as per our collection logic. This is the actual part where the accumulated values of the stream are massaged together into the final return value. The type parameters give a big hint in this regard – the return type,
    R is the same as that returned by the overall collect
    method.
  • characteristics: This is where we need to be careful. The enum has three variants – CONCURRENT, IDENTITY_FINISH, and UNORDERED. The bottomline is this – always use CONCURRENT for your custom types if the final value depends on the ordering of the values in the stream, or use UNORDERED if they do not. In the case of collecting values into a custom non-collection type, I don’t see any scenario where you would want to use IDENTITY_FINISH (unless you are a big fan of unsolicited ClassCastExceptionS).

    In short, this variant indicates that the finisher function is essentially an identity function, meaning that it can be skipped, and the currently accumulated value returned as the overall result (which is precisely what we wish to avoid).

One final comment to understand the collect method once and for all – what all those terms mean!

  • Supplier: This is the mechanism by which the input values are supplied to the collect method.
  • Accumulator: This is where the elements of the stream are combined with a running accumulator (which may be of a different type from the elements themselves), “reduced”, or “folded” in Functional terms.
  • Combiner: Similar to the accumulator, but the elements being combined together are of the same type. In most cases, this type would be a collection type, and finally,
  • Finisher: This is the meat of the whole collector. This is where the actual custom logic goes into to take the values produced by the combiner into the final result of the given return type.

Now that we’ve analysed the signature of the collect method, we must be in a position to realise that we can actually create custom collectors in multiple ways:

  • Using the static of methods in the Collector interface by supplying the correct supplier, accumulator, combiner, finisher, and Collector characteristics,
  • By creating a class that implements the Collector interface itself, and thereby providing implementations of the same supplier, accumulator, combiner, finisher, and Collector characteristics,
  • By simply creating any anonymous class conforming to the Collector interface, and providing the same inputs as in the previous two cases, or
  • Using any combination of the above.

To keep matters simple, let us create a custom class that implements the Collector interface. This will not only make things easier to understand, but also allow us to maintain code cleanliness.

Now let’s proceed with the implementation of the given use case to solidify these concepts.

Implementation and Demo

Let’s create a simple Maven project called custom_stream_collectors:

Macushla:Blog z0ltan$ mvn archetype:generate -DgroupId=com.z0ltan.custom.collectors -DartifactId=custom-collector -DarchetypeArtifactId=maven-archetype-quickstart -DinteractiveMode=false
[INFO] Scanning for projects...
[INFO]
[INFO] ------------------------------------------------------------------------
[INFO] Building Maven Stub Project (No POM) 1
[INFO] ------------------------------------------------------------------------

               <elided>

[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 30.967 s
[INFO] Finished at: 2017-07-11T20:57:45+05:30
[INFO] Final Memory: 18M/62M
[INFO] ------------------------------------------------------------------------

After customising the project to our heart’s desire, let’s fill in our custom collector class:

package com.z0ltan.custom.collectors.collectors;

import java.util.ArrayList;
import java.util.Collections;
import java.util.EnumSet;
import java.util.HashSet;
import java.util.List;
import java.util.Set;
import java.util.function.BiConsumer;
import java.util.function.BinaryOperator;
import java.util.function.Function;
import java.util.function.Supplier;
import java.util.stream.Collector;

import com.z0ltan.custom.collectors.types.Point;
import com.z0ltan.custom.collectors.types.Points;

public class PointToPointsCollector implements Collector<Point, List<Point>, Points> {
	@Override
	public Supplier<List<Point>> supplier() {
		return ArrayList::new;
	}

	@Override
	public BiConsumer<List<Point>, Point> accumulator() {
		return List::add;
	}

	@Override
	public BinaryOperator<List<Point>> combiner() {
		return (acc, ps) -> {
			acc.addAll(ps);
			return acc;
		};
	}

	@Override
	public Function<List<Point>, Points> finisher() {
		return (points) -> {
			final Set<Integer> xs = new HashSet<>();
			final Set<Integer> ys = new HashSet<>();
			
			for (Point p : points) {
				xs.add(p.x());
				ys.add(p.y());
			}
			
			return new Points(xs, ys);
		};
	}

	@Override
	public Set<java.util.stream.Collector.Characteristics> characteristics() {
		return Collections.unmodifiableSet(EnumSet.of(Collector.Characteristics.UNORDERED));
	}
}

We use ArrayList::new (method references are another excellent feature in Java 8 and beyond) for our Supplier since we start off with a blank slate, and for the Accumulator, we use List::add since the last section made it clear that the accumulator’s only job is to keep collecting items into running value of another type (a List in this case).

Then we have the Combiner which is implemented by the little lambda expression:

   (acc, ps) -> { acc.addAll(ps); return acc; }

As mentioned in the previous section, the combiner simply flattens the collections together into a single collection. In case of confusion, always look to the type signature for clarity.

And finally, we have the Finisher:

return (points) -> {
			final Set<Integer> xs = new HashSet<>();
			final Set<Integer> ys = new HashSet<>();
			
			for (Point p : points) {
				xs.add(p.x());
				ys.add(p.y());
			}
			
			return new Points(xs, ys);
		};

At this point of the stream pipeline, the points variable holds the list of accumulated Point objects. All we do then is to create an instance of the Points class by using the data available in the points variable. The whole point (if you will forgive the pun) is that this method will have logic peculiar to your specific use case, so the implementation will vary tremendously (which is more than can be said about the others – supplier, accumulator, and combiner).

And finally, here is our main class:

package com.z0ltan.custom.collectors;

import java.util.Arrays;
import java.util.List;

import com.z0ltan.custom.collectors.collectors.PointToPointsCollector;
import com.z0ltan.custom.collectors.types.Point;
import com.z0ltan.custom.collectors.types.Points;

public class Main {
	public static void main(String[] args) {
		final List<Point> points = Arrays.asList(new Point(1, 2), new Point(1, 2), new Point(3, 4), new Point(4, 3),
				new Point(2, 5), new Point(2, 5));

		// the result of our custom collector
		final Points pointsData = points.stream().collect(new PointToPointsCollector());

		System.out.printf("\npoints = %s\n", points);
		System.out.printf("\npoints data = %s\n", pointsData);
	}
}

Well, let’s run it and see the output!

Macushla:custom-collector z0ltan$ mvn package && java -jar target/custom-collector-1.0-SNAPSHOT.jar
[INFO] Scanning for projects...
[INFO]
[INFO] ------------------------------------------------------------------------
[INFO] Building custom-collector 1.0-SNAPSHOT
        <elided>
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 2.627 s
[INFO] Finished at: 2017-07-11T21:31:19+05:30
[INFO] Final Memory: 16M/55M
[INFO] ------------------------------------------------------------------------

points = [Point { x = 1, y = 2 }, Point { x = 1, y = 2 }, Point { x = 3, y = 4 }, Point { x = 4, y = 3 }, Point { x = 2, y = 5 }, Point { x = 2, y = 5 }]

points data = Points { xs = [1, 2, 3, 4], ys = [2, 3, 4, 5] }

Success!

Creating custom Java 8 Stream collectors

Implementing Nat in Rust (a la Haskell)

The power of Algebraic Data Types comes to the fore when dealing with complex problem domains. Most modern languages tend towards supporting some form of ADTs, but the complexity of expressing such paradigms is not completely lost when compared to languages that were designed with such support in mind.

Case in point – defining the prototypical example of an algebraic data type, defining the Natural Number type (Nat). Let’s compare the naive implementation in two languages that do have varying support for ADTs – Haskell, and Rust.

First, the Haskell version. Haskell, just like ML and other members of this family, is eminently suited for describing such types almost effortlessly:

module NatDemo where

import Test.HUnit

data Nat = Zero | Succ Nat deriving (Show, Eq)

nat2int :: Nat -> Int
nat2int Zero = 0
nat2int (Succ n) = 1 + nat2int n

int2nat :: Int -> Nat
int2nat 0 = Zero
int2nat n = Succ (int2nat (n-1))

add :: Nat -> Nat -> Nat
add Zero n = n
add (Succ m) n = Succ (add m n)


-- test cases
test1 = TestCase (assertEqual "nat to int" (nat2int (Succ (Succ Zero))) 2)
test2 = TestCase (assertEqual "int to nat" (int2nat 5) (Succ (Succ (Succ (Succ (Succ Zero))))))
test3 = TestCase (assertEqual "add two nats" (add (int2nat 5) (int2nat 6)) (int2nat 11))

tests = TestList [test1, test2, test3]

run = runTestTT tests

Running it to confirm:


Macushla:Nat-in-Rust z0ltan$ ghci
GHCi, version 8.0.2: http://www.haskell.org/ghc/  😕 for help
Loaded GHCi configuration from /Users/z0ltan/.ghci

Prelude> :l Nat.hs
[1 of 1] Compiling NatDemo          ( Nat.hs, interpreted )
Ok, modules loaded: NatDemo.

*NatDemo> run
Cases: 3  Tried: 3  Errors: 0  Failures: 0
Counts {cases = 3, tried = 3, errors = 0, failures = 0}

As can be seen, it’s almost trivial to define the Nat type in Haskell, as well as write functions that operate on this type. It reads almost exactly like the mathematical version of the concept. Simply beautiful.

Now let’s show the Rust version first, and then perhaps a comment or two on the implementation.

#[derive(Debug, PartialEq)]
enum Nat {
    Zero,
    Succ(Box<Nat>),
}

/// open it up for code readability
use Nat::{Zero, Succ};

fn nat_to_int(n: Nat) -> i32 {
    match n {
        Zero => 0,
        Succ(n) => 1 + nat_to_int(*n),
    }
}


fn int_to_nat(n: i32) -> Nat {
    match n {
        0 => Zero,
        n => Succ(Box::new(int_to_nat(n-1))),
    }
}


fn add(m: Nat, n: Nat) -> Nat {
    match (m, n) {
        (Zero, n) => n,
        (Succ(m), n) => Succ(Box::new(add(*m, n))),
    }
}

fn main() {
    assert_eq!(nat_to_int(Succ(Box::new(Succ(Box::new(Succ(Box::new(Zero))))))), 3);
    assert_eq!(int_to_nat(5), Succ(Box::new(Succ(Box::new(Succ(Box::new(Succ(Box::new(Succ(Box::new(Zero)))))))))));
    assert_eq!(add(int_to_nat(5), int_to_nat(11)), int_to_nat(16));
}

Let’s make sure that it’s all in working condition:

Macushla:Nat-in-Rust z0ltan$ rustc nat.rs && ./nat
Macushla:Nat-in-Rust z0ltan$

Excellent! It all just works. However, a couple of points to notice – first of all, because Rust doesn’t really have a runtime, we need to wrap the recursive part in a Box (amongst other choices) so that the compiler knows what size the data is (a box, being a pointer, is of a fixed size determinable at compile time). Of course, the test code is positively ghastly because we have to wrap all calls with Box::new, not to mention to remember to dereference the boxed value inside the functions.

Secondly, the code is surprisingly clean for the most part. However, writing the code in not nearly as intuitive as writing the same in Haskell (or ML for that matter). I can only imagine how much more daunting the task would be in a language without pattern matching (amongst other higher-level features) such as Java. Of course we would make it work, but the distance between the intent of the code and the code itself would be much much higher. Syntax does matter, folks!

Well, that was a nice little experiment of an evening.

Implementing Nat in Rust (a la Haskell)