Support Eastern Arabic numerals for lists about micromark

[wooorm]

markdown is an ASCII format, so it doesn’t make sense from the language format to support non-ASCII syntax.

You can solve this with CSS, list-style-type: arabic-indic.

See also the readme, which covers topics of CommonMark and extensions (Comparison).

[mustafa0x]

Thank you for your reply and suggestion.

so it doesn’t make sense from the language format to support non-ASCII syntax

Numerals are part of the text, so in some senses this equates to telling users not to user their own language. I appreciate parsing all numeral systems might not be feasible, but I just wanted to share some perspective. =)

I'm taking a look a the code and will patch necessary files (for my projects). I would appreciate any tips of course of where to look!

Thanks again!

[wooorm]

Numerals are part of the text, so in some senses this equates to telling users not to user their own language. I appreciate parsing all numeral systems might not be feasible, but I just wanted to share some perspective. =)

I understand. I agree that this is a big problem in Markdown. And programming as a whole. But I’m not going to implement non-standard things. See also https://talk.commonmark.org to discuss changes, which includes 6+ years of discussing non-English, too.

[mustafa0x]

You can solve this with CSS, list-style-type: arabic-indic.

Yes, the issue though is when authoring however. And using a numeral system that is different from the rest of the document causes a myriad of issues. They're mostly minor, but they do have far reaching implications. For example, on Macs, if the user want's a list they have to change the language, write 1, change the language back, write the list item's text, change the language, write 2, change the language back, write the list item's text, and so on.

I agree that this is a big problem in Markdown.

It mostly manifests in this situation, since text (i.e., 1, 2, 3) doubles as syntax, unlike all/most other cases, where you only have syntax (e.g. *, -, [], {}).

[mustafa0x]

For what it's worth, I was able to accomplish this by patching the following two files:

• micromark/dist/constructs.js: add 1633: list, // ١ to exports.document
• micromark/dist/character/ascii-digit.js: check(/[\d٠-٩]/)

Thanks again.