Heiko Oberdiek <[log in to unmask]> writes:
> what's the unicode code point for \b as accent?
> U+0331 COMBINING MACRON BELOW/NON-SPACING MACRON BELOW
> + opposite of \= (U+0304 COMBINING MACRON/NON-SPACING MACRON)
> - Unicode doesn't have a letter with this accent
Yes, I think it's this.
> +U0332 COMBINING LOW LINE/NON-SPACING UNDERSCORE
> + practical use for underlining (e.g. in bookmarks)
> - it isn't a typical accent
The difference in U-0332 is that the "underline" will connect with
adjacent characters, while a small gap is left with U-0331.
> +U0320 COMBINING MINUS SIGN BELOW/NON-SPACING MINUS SIGN BELOW
> - I don't see use cases, especially, there isn't a `plus sign below'.
Nor I. But that doesn't mean non-existence of use cases.
> My guess would be `macron below', \b seems to be inherited
> from math mode.
Why do you \b has origin in math mode? Doesn't \b make its debut
in section 3.2 of Lamport (2nd edition), "Symbols from Other
Languages"? AFAIK the macron is not used as a standard math accent.
When the original question is raised in connection with html or xml
generation, I think it is better to use precombined glyphs where they
For example, in a UTF-8-enabled xterm (at least in Linux),
the two character sequence [b, U-0331] renders equivalently to
the single character [U-1E07]. The latter (precombined) form
was supported in some web browsers before support for the two
character sequence was provided.
For more see this small png screenshot: