Create file-tree from nested dict | Python 2.7

Question

I have a nested dictionary like this:

Main = {
    'root' : {
        'object': 'data'
        'users': {
            'otherobj': 'data'
        }
    }
}

I need to create a file tree from it in the local dir /tree/ where the name of each dict is the name of the folder
All values within each dict belong to it
All values that are not dict's are ignored. (I'll handle those separately)
The depth of the tree will be unknown.
Each dir can(and usually will) have multiple sub dirs.

@ZdaR A hand full of google false-magic that didn't help me at all. Truth is i don't know what to try. That's why i'm here. — Kieee
– Kieee, Commented Feb 18, 2017 at 18:49
Useful functions for manipulating files and directories are in the os, os.path and shutil modules. You could use them from a recursive function that walks through your dict of dicts. — Paul Panzer
– Paul Panzer, Commented Feb 18, 2017 at 18:54

adgon92 · Accepted Answer · 2017-02-18 18:52:27Z

1

This code snippet should help. Here I assume that the python script will be run in target local directory.

import os

# dir_definition = <your 'Main'>

def create_dirs(dir_definition):
    for dir_dame, sub_dir in dir_definition.items():
        os.mkdir(dir_dame)
        if isinstance(sub_dir, dict):
            create_dirs(sub_dir)

answered Feb 18, 2017 at 18:52

adgon92

2171 silver badge7 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Elmex80s Over a year ago

The problem is indeed begging for a recursive solution.

hoyland · Accepted Answer · 2017-02-18 19:04:03Z

0

If I'm understanding properly, you want to end up with something like the following:

root/object containing data
root/users/otherobj containing data
and so on

So let's break this down a bit. You basically need the ability to do two things:

determine whether the value of a key in your dictionary is "data" or another dictionary
given some data, write it out to a file with a given path

I'm going to leave it to you to write these functions, but I'll assume they have the following signatures:

def is_data(obj) returning True/False (this could be not isinstanceof(obj, dict) unless your data objects could be dicts with some special properties)
def write_data(directory, filename, obj)

At this point, we're ready to write a function to walk the tree. I'm going to assume you pull the first key and dictionary out of Main. For each (key, value) pair, we need to check if the value is "data" or if it's another dict. If it's data, write it out. If it's another dict, we add the key to our path and call our function on that dict.

def walk(root_directory, obj_dict):
    for k, v in obj_dict.iteritems():
        if is_data(v):
            # write the file
            write_data(root_directory, k, v)
        else: # it's another dict, so recurse
            # add the key to the path
            new_root = os.path.join(root_directory, k) # you'll need to import os
            walk(new_root, v)

answered Feb 18, 2017 at 19:04

hoyland

1,82415 silver badges14 bronze badges

5 Comments

Kieee Over a year ago

If the objects in the dict are not dict's (eg: 'something': 'that isn't a dict') it needs to be ignored.

Kieee Over a year ago

Actually I can just impliment that here. But will this create the directories if no files exist within the dict's?

Kieee Over a year ago

(by that, i mean ignore my first comment)

Kieee Over a year ago

I need to create the entire file-tree regardless of if the files exist or not. And each dir can(and usually will) have multiple sub dirs.

Kieee Over a year ago

With a minor edit, this solved my problem. I just added if not os.path.exists(new_root): os.makedirs(new_root) right under new_root = os.path.join(root_directory, k)

Collectives™ on Stack Overflow

Create file-tree from nested dict | Python 2.7

2 Answers 2

1 Comment

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related