Better solution for TOC

[ibreathebsb]

Hi guys!

I have been using marked for an doc site recently, marked is a good library but when I realize that I need a TOC, it does not work very well. There are existing solutions like #545 ,while reading marked docs, I found a better(👀maybe?) solution, that is:

marked.lexer

for example, suppose we have the following markdown text

# A
> please ignore me 
## B
# C
## D

then we use marked.lexer(markdownText) to get the tokens:

[
  {
    "type": "heading",
    "depth": 1,
    "text": "A"
  },
  {
    "type": "blockquote_start"
  },
  {
    "type": "paragraph",
    "text": "please ignore me"
  },
  {
    "type": "blockquote_end"
  },
  {
    "type": "heading",
    "depth": 2,
    "text": "B"
  },
  {
    "type": "heading",
    "depth": 1,
    "text": "C"
  },
  {
    "type": "heading",
    "depth": 2,
    "text": "D"
  }
]

wow! that's all we need, right? now we can filter these headings and do whatever we want!

//  tokens is the result returned by marked.lexer()
const headings = tokens.filter(token => token.type === 'heading')
//
DO_WHATEVER_U_WANT(headings)

👏 That's all, wish this could help, thanks!

[UziTech]

This would probably be better as a comment on #545

We are trying to consolidate the issues so I am going to close this issue.