This is a utility that allows someone to write or paste words in a rich-text format, attempts to interpret what is written and converts it into a JSON representation of the text. The JSON will be an array, containing an object representing each paragraph. Each object is of the form:
{
"content": [
{
"style": "normal",
"content": "Hello, world!"
}
]
}
i.e. the object should contain a single key, content, whose value is an array containing 1). the text (value of the child content key), and 2). the style to be applied to that text. The styles may be "normal", "bold", "italic", or "bold-italic".
For example, the text:
Happy 2021. Hope you have a wonderful year ahead!
Hope to see you soon.
Would have the JSON representation:
[
{
"content": [
{
"style": "bold",
"content": "Happy 2021."
},
{
"style": "normal",
"content": " Hope "
},
{
"style": "italic",
"content": "you"
},
{
"style": "normal",
"content": " have a wonderful year ahead!"
}
]
},
{
"content": [
{
"style": "normal",
"content": "Hope to see you soon."
}
]
}
]
We have taken a first pass at parsing input text into this JSON format. But it is quite fragile. Occassionally, when copy/pasting text into the text box, our parsing logic cannot extract the text in the HTML that is copied, which results in skipping over a block of text entirely.
Is there a way to make parsing incoming pasted text more robust? Such that I can copy/paste text written in a Google Doc, or from a website, have the parsing logic be able to extract the text I pasted, along with a reasonable guess as to whether the texts hould be bolded, italicized, etc?
Clone this repository, and install dependencies:
npm i
and run the app locally:
npm start
and visit
http://localhost:8000/
Please fork this repository and develop on your fork. To submit, please email me with a link to the forked repo. You can open a PR on your fork with your submission.
Thank you for participating in this challenge!