逆向工程字符串格式化过程

Question

I am using slate-react editor for my wysiwyg data entry component that will be displayed on Telegram.

Text input is then formatted as per Telegram's requirement as this example shows:

Slate Object
[
    {
        "type": "paragraph",
        "children": [
            {
                "text": "Dry grains to moisture content of below 12 percent and dress with crop "
            },
            {
                "text": "protection ",
                "bold": true
            },
            {
                "text": "products such as Actellic 2%. Seeds meant for planting should not be stored for more than six "
            },
            {
                "text": "months",
                "underline": true
            },
            {
                "text": ". For more "
            },
            {
                "text": "information",
                "underline": true
            },
            {
                "text": ", please contact your "
            },
            {
                "text": "local ",
                "italic": true
            },
            {
                "text": "Extension Agent."
            }
        ]
    }
]

I then format the string to Telegram specs as follows:

Dry grains to moisture content of below 12 percent and dress with crop **protection **products such as Actellic 2%. Seeds meant for planting should not be stored for more than six months. For more information, please contact your _local _Extension Agent.

I would now like to reverse engineer the process so that I can generate the object above if given such a string so that I'm able to display it on the editor. I would like help in formulating the function that would achieve the intended result.

Answer 1

For those who may need such information in future, I got pointer from comments thanks to @Phil, I was able to search through the net for markdown tokenizers. Prismjs works like a charm.