Repository navigation
[css-text-decor][css-lists][css-counter-styles] text-emphasis is similar to (symbolic) counters #13335
Description
text-emphasis-style is remindful of list-style-type at first impression, but upon further inspection their predefined values hardly overlap. Even where they should use the same character, their keywords and terms mismatch:
list-style-type: disc≈text-emphasis-style: filled dot: •list-style-type: circle≈text-emphasis-style: open dot: ◦
Nevertheless, I think it might be worthwhile to consider unifying them. That means,
- current emphasis marks would need to be added to predefined symbolic counter styles and
- the
text-emphasis-stylevalue would need to accept<counter-style>.
I can see educational and linguistic use cases where it makes sense to add ordinals to characters within a sequence/word.
Background
<'text-emphasis-style'>=none|<string>|filled | open||dot | circle | double-circle | triangle | sesame
<'list-style-type'>=none|<string>|<counter-style>=<symbols()>|<counter-style-name>=decimalSimple Predefined Counter Styles: Numeric|disc | square | circle | disclosure-open | disclosure-closedSimple Predefined Counter Styles: Symbolic|<custom-ident>
| Shape | Symbolic Counter | Char | Emphasis Mark Value | Char |
|---|---|---|---|---|
| black/filled small circle | disc |
• U+2022 | filled dot |
• U+2022 |
| white/hollow small circle | circle |
◦ U+25E6 | open dot |
◦ U+25E6 |
| black/filled large circle | filled circle |
● U+25CF | ||
| white/hollow large circle | open circle |
○ U+25CB | ||
| black/filled double circle | filled double-circle |
◉ U+25C9 | ||
| white/hollow double circle | open double-circle |
◎ U+25CE | ||
| black/filled equilateral triangle pointing upwards | filled triangle |
▲ U+25B2 | ||
| white/hollow equilateral triangle pointing upwards | open triangle |
△ U+25B3 | ||
| black/filled equilateral triangle pointing downwards | disclosure-open |
▾ U+25BE | ||
| white/hollow equilateral triangle pointing downwards | ||||
| black/filled equilateral triangle pointing sideways | disclosure-closed |
▸ U+25B8 / ◂ U+25C2 | ||
| white/hollow equilateral triangle pointing sideways | ||||
| black/filled square | square |
▪ U+25AA | ||
| white/hollow square | ||||
| black/filled sesame | filled sesame |
﹅ U+FE45 | ||
| white/hollow sesame | open sesame |
﹆ U+FE46 |
Activity
jfkthame commentedon Jan 12, 2026
This sounds like it would add substantial implementation complexity for what seems to be a hypothetical use case. I'd like to see evidence of a genuine author need before considering something like this.
Crissov commentedon Jan 12, 2026
It probably would add some complexity indeed, although my motivation was the reuse of existing definitions.
I do not feel strongly about this suggestion, it just came to my mind when commenting on #13329.
Loirooriol commentedon Jan 12, 2026
I'm not sure how you are expecting e.g.
text-emphasis-style: decimalto work. Would the first character get 1, the second 2, etc? What if the counter would produce a text that doesn't fit?Crissov commentedon Jan 12, 2026
Yes, of course, and it would fail in the same way as if an author provided too long a
<string>with the existing possibilities.