Implementation of Tree Traversal in java

[xam4lor]

No description provided.

[zsparal]

The same argument as in the C# case. I don't think we need separate implementations just to include in the chapters. The final code, especially with the explicit this makes it incredibly clear what the reader should expect. In Java you would normally write the parameterless version, so just stick to that one.

Additionally, I'd prefer to keep merging stuff into the archive that actually compiles.

[Butt4cak3]

Yup. What @Gustorn already said. Show the entire code at the bottom of the chapter and import parts of it for the smaller snippets. No need to have two versions of the same code.

[xam4lor]

Allright !

[xam4lor]

The problem is that if I do that, I have to options :

• Create useless methods in the Tree class who are only use on the website to explain part by part of the function (like DFS_recursive_postorder())
• Don't add this methods and let a 'hole' which says 'This has not been implemented in your chosen language, so here is the Julia code'

That's why I made two different codes.

[Butt4cak3]

I'd just go for unused methods. They don't hurt and people can take the code and play around with them if they want.

[june128]

Currently the C# code is split into MdAdditional (for in-text code) and normal example code as well. Both compile, but one is the actual example.
With the most recent PR (since things and how to do them changed a bit) they now get merged into one file.
So I support and strongly recommend one implementation instead of two and of course ones that actually compile. After all people should be able to run the code and be able use it for learning.

[zsparal]

You should really just remove these versions. They are not any more clear and they take up a bunch of space for no reason. People who'll want to use the Java implementation will know what this.root means in the other implementations.

[xam4lor]

But we need this method to have a parameter of type Node.
I have two solutions : one is to create a function with a same name but no parameters like DFSRecursive() who just call DFSRecursive(this.root).
Or I call DFSRecursive(null) and check if the node is null, the node equals this.root.
In java we cannot create functions within other functions.

[zsparal]

You could create a private function under the public one which does the recursive part and then you can include both in the code example

[Butt4cak3]

I was about to suggest method overloading as well. Is there a reason not to?

[xam4lor]

It was one of my solutions, I think It's the better thing to do. (overloading)

[zsparal]

One more thing, the methods do not follow Java's naming convention: they should be camelCase, e.g.: dfsStack.

[zsparal]

I'm a bit unsure about the excessive whitespace use to separate the different functions. I see why you did it but maybe change it to 2 newlines.

[zsparal]

My last nitpicks, sorry for doing them piecemeal, I did the earlier ones during the day when I had a little free time.

[zsparal]

This indentation is not quite right

[zsparal]

You can move this inside the loop

[zsparal]

Any opinion on this one?

[xam4lor]

What do you mean by 'inside a loop' ?

[zsparal]

You only use tmp inside the following while loop, so you could write this instead:

while(stack.size() != 0) {
    System.out.println(stack.peek().id);
    Node tmp = stack.pop();

    for (Node c : tmp.children) {
        stack.push(c);
    }
}

[zsparal]

The standard Java coding style would place the else on the same line as the closing brace:

if (...) {

} else {

}

[zsparal]

I think the de-facto standard is to have spaces around brackets in Java, so

if (condition) {

}
// Instead of
if(condition) {

}

Similarly for while, for, etc. loops

[xam4lor]

You are right ! That's a bad practice that I have :/

[zsparal]

With the newest whitespace/formatting changes the line ranges in the chapter are out of date

[xam4lor]

And I think that's done !

[zsparal]

Thank you for the contribution, sorry again for giving the feedback piece-by-piece, I know it's more annoying to fix it this way :)

[xam4lor]

You do what you can ! No problem ! :)