How to disable some specific tokenizers?

[ocavue]

Question:

How to disable some specific tokenizers when using micromark / mdast-util-from-markdown / remark?

Why:

I want to use mdast-util-from-markdown to parse an inline markdown text into some mdast tokens. That means I don't want to use any block-level rule for the text.

My current code is as below:

import mdast from "mdast"
import fromMarkdown from "mdast-util-from-markdown"

function parseInlineMarkdown(text: string): mdast.PhrasingContent[] {
    const root: mdast.Root = fromMarkdown(text)
    if (root.children.length === 0) {
        console.warn("root has not children:", root)
        return []
    }

    const content: mdast.Content = root.children[0]
    if (content.type != "paragraph") {
        console.warn("root has not paragraph children:", root)
        return []
    }

    return content.children
}

The function will fail when parsing * hello since micromark will treat it as a list block instead of a paragraph.

It seems that I can fix my issue by somehow disabling all block-level tokenizers (like what markdown-it does). But I'm not sure I'm in the right direction.

Possible solutions

I found a third party remark plugin that can disable some tokenizers: https://github.com/zestedesavoir/zmarkdown/tree/master/packages/remark-disable-tokenizers. However, they haven't migrated to the latest remark version (with micromark) so this plugin is currently broken.

Related issues:

• remark@next (13) zestedesavoir/zmarkdown#416
• "Turning off a tokenizer" is misleading remarkjs/remark#529

[wooorm]

We had remarkjs/remark#552 about this recently, does that help?

[ocavue]

Oh, I didn't see this issue. Sorry about that.

It didn't help much, but I think it's better to continue the discussion in one place, so I will close this discussion.