Tree Traversal: Output Standardization

[lazyprop]

This is to standardize language implementations for the tree traversal chapter. It is based on #855.

[lazyprop]

I have made changes to many of the implementations, but there are a few that are
problematic.

• The PHP implementations has a weird way of printing node data.
• The Haskell implementation numbers the nodes in an entire different way from
  the others.
• The Coconut implementation works slightly differently. Shouldn't be hard to
  fix it though.
• OCaml is missing postorder and preorder DFS. Unfortunately I don't know OCaml so I can't write those functions myself.

I'll update this list tomorrow as I'm way too sleepy right now.

[Amaras]

Oh well, apparently I have write access to your repo.
The Coconut output should be okay (it was simply a matter of adding [#] everywhere)

[lazyprop]

@Amaras Coconut does not have any nodes containing 0 which every other language does. This can be fixed just by changing the condition in create_tree to this:

    if num_rows == 0:
        return Node(0, ())

[leios]

Great work with this, maybe we hide Ocaml for now, as it is missing some code snippets?

I think @jiegillet might be willing to fix the haskell?

I don't know about PhP, though. Does it need a rework entirely?

[lazyprop]

Yeah hiding Ocaml would be good for now.

I don't know Haskell but I think it can be fixed by just creating a createTree procedure which numbers nodes appropriately.

As for PHP, I think I have a fix. The only thing that needed changing was the generate_tree function (that seems to be a common problem in the implementations). I also removed an if condition from DFSRecursive so maybe the line numbers need changing. I have zero experience with PHP though so if someone can review the PHP changes that would be nice.

[jiegillet]

I think @jiegillet might be willing to fix the haskell?

Of course, please give me a bit of time.

Yeah sorry, back in the day I used to do things my way, since Haskell programs are structured quite differently anyway... I'll fall into the ranks now.

[Amaras]

Okay, I had to merge, so it might have caused problems with the line numbers, we'll have to make sure we check them before merging.

I did that job because @lazyprop wanted someone to update the PR correctly. Hopefully the history is not too messy (I didn't want to force-push when I was not sure of myself)

[leios]

I think the Java file still needs to be updated (because of the merge), but I think that's the only thing to change and we can then squash and merge this guy.

I think the squash will get rid of the messy history.

[lazyprop]

I think the Java file still needs to be updated

@leios sorry I don't understand what exactly needs to be updated. It seems to be exactly what it should be.

[leios]

I think there are 2 Java files on this branch, then? Tree.java and MainClass.java we recently changed to use Tree.java instead of MainClass, so the [#]\n statements need to be added to Tree.java and MainClass.java needs to be removed.

[lazyprop]

I have removed MainClass.java updated Tree.java with the [#]\n statements.

[leios]

I think we are good to squash and merge this guy, but I wanted to mention that there seem to be 4 languages that are not up-to-date:

• matlab
• smalltalk
• scratch
• emojicode

I think scratch can be scratched off the list, but the other 3 might be standardized in the future? No matter the case, this is a good step and I am happy to merge.

[leios]

The commit history is not clean, but we'll do a squash and merge

[lazyprop]

Thanks for merging. I think Smalltalk was added two days ago here (#453). Maybe we can get Neverik to fix the output format?

As for Matlab, I don't have a license so no way to test the code. Maybe someone who has a license can do it.
MATLAB is Evil