Python get XML siblings into dictionary

Question

I have an xml that looks like this:

<root>
    <G>
        <G1>1</G1>
        <G2>some text</G2>
        <G3>some text</G3>
        <GP>
            <GP1>1</GP1>
            <GP2>a</GP2>
            <GP3>a</GP3>
        </GP>
        <GP>
            <GP1>2</GP1>
            <GP2>b</GP2>
            <GP3>b</GP3>
        </GP>
        <GP>
            <GP1>3</GP1>
            <GP2>c</GP2>
            <GP3>c</GP3>
        </GP>
    </G>
    <G>
        <G1>2</G1>
        <G2>some text</G2>
        <G3>some text</G3>
        <GP>
            <GP1>1</GP1>
            <GP2>aa</GP2>
            <GP3>aa</GP3>
        </GP>
        <GP>
            <GP1>2</GP1>
            <GP2>bb</GP2>
            <GP3>bb</GP3>
        </GP>
        <GP>
            <GP1>3</GP1>
            <GP2>cc</GP2>
            <GP3>cc</GP3>
        </GP>
    </G>
    <G>
        <G1>3</G1>
        <G2>some text</G2>
        <G3>some text</G3>
        <GP>
            <GP1>1</GP1>
            <GP2>aaa</GP2>
            <GP3>aaa</GP3>
        </GP>
        <GP>
            <GP1>2</GP1>
            <GP2>bbb</GP2>
            <GP3>bbb</GP3>
        </GP>
        <GP>
            <GP1>3</GP1>
            <GP2>ccc</GP2>
            <GP3>ccc</GP3>
        </GP>
    </G>
</root>

Im trying to transform this xml into a nested dictionary called "G":

{ 1: {G1: 1,
      G2: some text,
      G3: some text,
      GP: { 1: {GP1: 1,
                GP2: a,
                GP3: a},
            2: {GP1: 2,
                GP2: b,
                GP3: b},
            3: {GP1: 3,
                GP2: c,
                GP3: c}}
      },
  2: {G1: 2,
      G2: some text,
      G3: some text,
      GP: { 1: {GP1: 1,
                GP2: aa,
                GP3: aa},
            2: {GP1: 2,
                GP2: bb,
                GP3: bb},
            3: {GP1: 3,
                GP2: cc,
                GP3: cc}}
      },
  3: {G1: 3,
      G2: some text,
      G3: some text,
               GP: { 1: {GP1: 1,
                GP2: a,
                GP3: a},
            2: {GP1: 2,
                GP2: bbb,
                GP3: bbb},
            3: {GP1: 3,
                GP2: ccc,
                GP3: ccc}}
      }
    }

My code works fine to get all elements that are straight under "G", so G1, G2 etc, but for GP I either only just get one record, either I get all of them but it duplicates the same thing couple of times either I get all 9 GP elements under one single GP in the dictionary. Here is my code:

    f = 'path to file'
    tree = ET.parse(f)
    root = tree.getroot()
    self.tree = tree
    self.root = root
    gs = len(self.tree.getiterator('G'))
    g = {}
    for i in range(0, gs):
        d = {}
        for elem in self.tree.getiterator('G')[i]:
            if elem.text == "\n      " and elem.tag not in ['GP']:
                    dd = {}
                    for parent in elem:
                        if parent.text == "\n        ":
                            ddd = {}
                            for child in parent:
                                ddd[child.tag] = child.text
                            dd[parent.tag] = ddd
                        else:
                            dd[parent.tag] = parent.text
                    d[elem.tag] = dd
            else:
                d[elem.tag] = elem.text
        g[i+1] = d

    # Build GP
    count = 0
    gp = {}
    for elem in self.tree.getiterator('GP'):
        d = {}
        for parent in elem:
            if parent.text == "\n      ":
                dd = {}
                for child in parent:
                    dd[child.tag] = child.text
                d[parent.tag] = dd
            else:
                d[parent.tag] = parent.text
        count += 1
        gp[count] = d
    g["GP"] = gp

I see handling for GPD named nodes, but I don't see such node in the sample xml. — CristiFati
– CristiFati, Commented Aug 21, 2017 at 16:24

CristiFati · Accepted Answer · 2018-12-30 20:45:57Z

code.py:

#!/usr/bin/env python3

import sys
import xml.etree.ElementTree as ET
from pprint import pprint as pp


FILE_NAME = "data.xml"


def convert_node(node, depth_level=0):
    #print("  " * depth_level + node.tag)
    child_nodes = list(node)
    if not child_nodes:
        return (node.text or "").strip()
    ret_dict = dict()
    child_node_tags = [item.tag for item in child_nodes]
    child_index = 0
    for child_node in child_nodes:
        tag = child_node.tag
        if child_node_tags.count(tag) > 1:
            sub_obj_dict = ret_dict.get(tag, dict())
            child_index += 1
            sub_obj_dict[str(child_index)] = convert_node(child_node, depth_level=depth_level + 1)
            ret_dict[tag] = sub_obj_dict
        else:
            ret_dict[tag] = convert_node(child_node, depth_level=depth_level + 1)
    return ret_dict


def main():
    tree = ET.parse(FILE_NAME)
    root_node = tree.getroot()
    converted_xml = convert_node(root_node)
    print("\nResulting dict(s):\n")
    for key in converted_xml: # converted_xml should be a dictionary having only one key (in our case "G" - we only care about its value, to match the required output)
        pp(converted_xml[key])


if __name__ == "__main__":
    print("Python {:s} on {:s}\n".format(sys.version, sys.platform))
    main()

Notes:

FILE_NAME contains the file name that contains the input xml. Feel free to change it, in order to match yours
The conversion happens in convert_node. It's a recursive function that it's called upon each xml node and returns a Python dictionary (or a string). The algorithm:
- For each node, get a list of its (direct) children. If the node hasn't any (it's a leaf node - like G# or GP# nodes), it will return its text
- If the node has more than one child with a specific tag, then its content will be added under a key representing its index (like G or GP nodes), in a sub dictionary of the current dictionary corresponding to the the child tag key
- All the children with unique tags will have their content placed under a key equal to their tag directly under the current dictionary
- depth_level is not used (you can remove it), I used it to print the xml node tags in a tree form; it's the depth in the xml tree (root - 0, G - 1, G#, GP - 2, GP# - 3, ...)
The code is designed to be:
- General: notice there are no hardcoded key names
- Scalable: if at some point the xml will become ore complex (e.g. under a GP node there will be a GPD node let's say, and that node will have subnodes as well - basically the xml will gain one more depth level), the code will handle it without change
- Python 3 and Python 2 compatible

Output:

(py_064_03.05.04_test0) e:\Work\Dev\StackOverflow\q045799991>"e:\Work\Dev\VEnvs\py_064_03.05.04_test0\Scripts\python.exe" code.py
Python 3.5.4 (v3.5.4:3f56838, Aug  8 2017, 02:17:05) [MSC v.1900 64 bit (AMD64)] on win32

Resulting dict(s):

{'1': {'G1': '1',
       'G2': 'some text',
       'G3': 'some text',
       'GP': {'1': {'GP1': '1', 'GP2': 'a', 'GP3': 'a'},
              '2': {'GP1': '2', 'GP2': 'b', 'GP3': 'b'},
              '3': {'GP1': '3', 'GP2': 'c', 'GP3': 'c'}}},
 '2': {'G1': '2',
       'G2': 'some text',
       'G3': 'some text',
       'GP': {'1': {'GP1': '1', 'GP2': 'aa', 'GP3': 'aa'},
              '2': {'GP1': '2', 'GP2': 'bb', 'GP3': 'bb'},
              '3': {'GP1': '3', 'GP2': 'cc', 'GP3': 'cc'}}},
 '3': {'G1': '3',
       'G2': 'some text',
       'G3': 'some text',
       'GP': {'1': {'GP1': '1', 'GP2': 'aaa', 'GP3': 'aaa'},
              '2': {'GP1': '2', 'GP2': 'bbb', 'GP3': 'bbb'},
              '3': {'GP1': '3', 'GP2': 'ccc', 'GP3': 'ccc'}}}}

Collectives™ on Stack Overflow

Python get XML siblings into dictionary

1 Answer 1

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related