XML Serializer

Allows you to convert XML to python dict (with python objects) using a schema.

Examples

We have next xml data (profiles.xml)

<?xml version="1.0" encoding="UTF-8"?>
<payload>
    <MyProfile>
        <record id="1" nickname="eff1c" admin="true">
            <posts topic="something">
                <post name="test post" description="It's my test post." />
                <post name="python xml_serializer" description="It's very useful module!" />
            </posts>
        </record>
    </MyProfile>
</payload>

And we want to turn it into

{
    "payload": {
        "my_profile": {
            "record": {
                "id": 1,
                "nickname": "eff1c",
                "admin": True,
                "posts": {
                    "topic": "something",
                    "post": [
                        {"name": "test post", "description": "It's my test post."},
                        {
                            "name": "python xml_serializer",
                            "description": "It's very useful module!",
                        },
                    ],
                },
            }
        }
    }
}

We will write next schema

from xml_serializer import Tag, TagAttr
from xml_serializer.converter_types import Integer, String, Boolean

profiles_schema = {
    Tag("payload"): {
        Tag("MyProfile", "my_profile"): {
            Tag("record"): {
                TagAttr("id"): Integer(nullable=False),
                TagAttr("nickname"): String(nullable=False),
                TagAttr("admin"): Boolean(),
                Tag("posts"): {
                    TagAttr("topic"): String(nullable=False),
                    Tag("post"): [
                        {
                            TagAttr("name"): String(),
                            TagAttr("description"): String()
                        }
                    ]
                }
            }
        }
    }

Get etree element (tag)

from xml.etree import ElementTree as etree

tree = etree.parse("profiles.xml")
root = tree.getroot()

# you can use root tag or find any else
main_tag = root.find("payload")

And call the method to pass them to

from xml_serializer import xml_serialize

response = xml_serialize(profiles_schema, main_tag)

Schema

Tag/TagAttr

In order to serialize data, you need to describe the scheme of its structure.
To do this, we create a python dict with Tag/TagAttr object keys.
The value for Tag is a set of TagAttr-s or Tag. I think it's clear that TagAttr is an attribute of the current tag, and Tag is actually a nested tag.

from xml_serializer import Tag, TagAttr
from xml_serializer.converter_types import String

schema = {
    Tag("posts"): {
        TagAttr("topic"): String(nullable=False),
        Tag("post"): [
            {
                TagAttr("name"): String(),
                TagAttr("description"): String()
            }
        ]
    }
}

Both Tag and TagAttr have 2 parameters: field_name and name.
field_name - it is name in xml.
name - it is our custom name with which the object will return to us after serialization (by default - field_name).

Field data types

To convert TagAttr values, we use data types: String, Integer, Float, Boolean, NestedType.

String, Integer, Float, Boolean are similar to common data types (as in python). All of them have the nullable parameter (True by default).
When set to False, if the tag does not have this attribute in the input data, an error will be raised

NestedType

It is a data type that allows you to intercept serialization in the middle of a schema and process the resulting data according to your needs.

In order to use it, we have to describe the nested schema separately and create a function that will process the data received from it.
NestedType has 2 attributes for it: schema and data_handling_function

from xml_serializer import Tag, TagAttr
from xml_serializer.converter_types import Boolean, String, NestedType


posts_schema = {
    Tag("post", "posts"): [
        {
            TagAttr("name"): String(),
            TagAttr("description"): String()
        }
    ]
}


def get_post_names(data):
    posts = data["posts"]

    return [post["name"] for post in posts]


schema_with_nesting = {
    Tag("record"): {
        TagAttr("admin"): Boolean(),
        Tag("posts", "post_names"): NestedType(posts_schema, get_post_names)
    }
}

In this example, we return only a list of post-s names to the posts tag, discarding all the information we don't need. This way, NestedType allows you to modify the output data schema without unnecessary iterations.

Custom type

You can create your own field data types. Use AbstractType for it.

from xml_serializer.abstract_type import AbstractType

class Boolean(AbstractType):
    def convert_method(self, value):
        return True if value.lower() == "true" else False