2

I have a set of URLs like this:

http://example.com/one/two/three/four/
http://example.com/one/two/three/five/
http://example.com/one/two/three/five/six
http://example.com/one/two/five

And I need to create a tree-like dictionary that looks like this:

{
    'example.com': 
    {
        'one': 
        {
            'two':
            {
                'three':
                {
                    'four': None,
                    'five': 
                    {
                        'six': None
                    }
                },
                'five': None
            }
        }
    }
}

The URLs can have any number of levels, so the depth of the dictionary is not known beforehand.

How can I traverse over this dictionary to add the keys?

My current pseudo-code is:

list_of_links = [
    'http://example.com/one/two/three/four/',
    'http://example.com/one/two/three/five/',
    'http://example.com/one/two/three/five/six',
    'http://example.com/one/two/five',
]

dic = {'example.com':None}

for link in list_of_links: 
    url_path = urlparse.urlparse(link).path
    url_path_parts = url_path.split('/')

    curr_dic_path = 'example.com'
    for segment in url_path_parts:

        # How to do this?
        if segment in dic's current path:
            curr_dic_path = curr_dic_path + segment
        else
            add_segment_to_dic_current_path

Or is there a better way to do this? I have been trying for the past two days and could not get it done.

I've already searched through similar questions such as this one but they don't answer my question.

UnderpoweredNinja
  • 97
  • 4
  • Where would you want to add the keys? What would you expect the outcome to be? – mrCarnivore Jan 04 '18 at 12:09
  • @mrCarnivore: I want to add the keys in the dictionary itself based on the URL I have. Which could have N-number of segments. I have added the outcome to my question itself. – UnderpoweredNinja Jan 04 '18 at 12:12
  • @cᴏʟᴅsᴘᴇᴇᴅ: Okay, please explain why you closed the question as a duplicate. It clearly does not answer my question. – UnderpoweredNinja Jan 04 '18 at 12:12
  • 1
    Um, your question is a textbook TRIE question, just modify the code in the answer and you're good to go. – cs95 Jan 04 '18 at 12:13
  • 1
    @cᴏʟᴅsᴘᴇᴇᴅ **This question was marked as an exact duplicate of an existing question.** -- This question is more about traversing the dictionary. Use a TRIE would have been more appropriate as a comment. It's in no way an exact duplicate. Could you explain why you think otherwise? – UnderpoweredNinja Jan 04 '18 at 12:18
  • 1
    If an answer addresses another question, then, in my opinion, is a suitable duplicate target. If you disagree, you can either 1. Appeal to another badge holder, or 2. Contest the closure on meta. – cs95 Jan 04 '18 at 12:20
  • 4
    It seems to me that the accepted answer on the linked duplicate is really what you need. The only differences are that your `_end` is `None`, and that instead of looping over each character in a word you have to loop over every item in `url_path_parts`, or something similar (cc @cᴏʟᴅsᴘᴇᴇᴅ). – Andras Deak -- Слава Україні Jan 04 '18 at 12:29
  • 3
    Discussed on [meta](https://meta.stackoverflow.com/questions/361421/understanding-why-my-question-was-closed-so-i-can-ask-better-questions-in-the-fu) – BDL Jan 04 '18 at 12:35

0 Answers0