Processing ADF format in Python

[Anton Manin]

If you use enterprise insights, you can see different types of multiline-fields storage formats: plain text, HTML or ADF.

We generate reports in Confluence to be able to show all trends/insights on the same page.

I made a python script to convert ADF to html.

Below you can see the result -- I do not parse the 'code' parts as we do not use it in insights for objectives. 

Script itself:

import json
import html
from loguru import logger


def process_atl_format(text, level=0):

    if text is None:
        return None
    res = []
    replacements = {
        'bulletList': 'ul',
        'listItem': 'li',
        'paragraph': 'p',
        'orderedList': 'ol',
        'blockquote': 'blockquote',
        'heading': 'h'

    }
    modifiers = {
        'strong': 'b',
        'em': 'em',
        'underline': 'u',
        'strike': 's',
        'code': 'code'
    }
    if text.get('content') is None:
        if text.get('text'):
            t = html.escape(text.get('text'))
            if text.get('marks'):
                for mark in text.get('marks'):
                    mark_type = mark['type']
                    if mark_type == 'link':
                        link = mark['attrs']['href']
                        t = '<a href="{}">{}</a>'.format(html.escape(link), t)
                    elif mark_type == 'subsup':
                        subsup = mark['attrs']['type']
                        t = '<{tag}>{t}</{tag}>'.format(tag=subsup, t=t)
                    elif modifiers.get(mark_type):
                        t = '<{tag}>{t}</{tag}>'.format(tag=modifiers.get(mark_type), t=t)

            res = [t]
        elif text.get('type') == 'hardBreak':
            res = ['<br/>']
        else:
            logger.debug(f'>>>>> {text}')
    else:
        for x in text['content']:
            tmp = process_atl_format(x, level + 1)
            if replacements.get(x['type']):
                tag = replacements.get(x['type'])
                if tag == 'h':
                    tag = 'h{}'.format(x['attrs']['level'])
                res.append('<{tag}>{txt}</{tag}>'.format(tag=tag, txt=tmp))
            else:
                res.append(tmp)
    res = ''.join(res)
    return res.translate(str.maketrans('', '', ''.join([chr(char) for char in range(1, 32) if char not in (9, 10, 13)])))


if __name__ == '__main__':
    text = json.loads("""{"version":1,"type":"doc","content":[{"type":"heading","attrs":{"level":1},"content":[{"type":"text","text":"Heading"}]},{"type":"paragraph","content":[{"type":"text","text":"Bold","marks":[{"type":"strong"}]}]},{"type":"paragraph","content":[{"type":"text","text":"italic","marks":[{"type":"em"}]}]},{"type":"paragraph","content":[{"type":"text","text":"underline","marks":[{"type":"underline"}]}]},{"type":"paragraph","content":[{"type":"text","text":"strikethrough","marks":[{"type":"strike"}]}]},{"type":"paragraph","content":[{"type":"text","text":"code","marks":[{"type":"code"}]}]},{"type":"paragraph","content":[{"type":"text","text":"subscript"}]},{"type":"paragraph","content":[{"type":"text","text":"superscript"}]},{"type":"paragraph","content":[{"type":"text","text":"list"}]},{"type":"bulletList","content":[{"type":"listItem","content":[{"type":"paragraph","content":[{"type":"text","text":"one"}]}]},{"type":"listItem","content":[{"type":"paragraph","content":[{"type":"text","text":"two"}]}]}]},{"type":"paragraph","content":[{"type":"text","text":"list"}]},{"type":"orderedList","attrs":{"order":1},"content":[{"type":"listItem","content":[{"type":"paragraph","content":[{"type":"text","text":"1"}]}]},{"type":"listItem","content":[{"type":"paragraph","content":[{"type":"text","text":"2"}]}]}]},{"type":"paragraph","content":[{"type":"text","text":"Google","marks":[{"type":"link","attrs":{"href":"https://google.com"}}]}]},{"type":"codeBlock","attrs":{"language":"c"},"content":[{"type":"text","text":"sdfgsdfg"}]},{"type":"blockquote","content":[{"type":"paragraph","content":[{"type":"text","text":"quote"}]}]},{"type":"paragraph","content":[]}]}""")

    result = process_atl_format(text)
    print(result)

It would be great though if Enterprise Insights team will provide options to select, e.g. multiple columns to choose:

• Override Notes -- as it stores in the DB now
• Override Notes HTML -- valid HTML
• Override Notes Plain Text -- just text with correct indents

How do you solve this problem?