I want to map math characters in my PDF output to Unicode characters. XeLaTeX already does this somewhat, but with very unsatisfactory results.
Just for comparison, consider the very simple example $\mathbb E[X] = \sum_x x p(x)$.
- As baseline, Plain PDFLaTeX gives:
E[X] =
P
x xp(x)
- PDFLaTeX +
\usepackage[noTeX]{mmap}gives:
[] =
Σ︀
()
- Plain XeLaTeX gives:
E[X] =
Σ
x xp(x)
4 Finally, XeLaTeX + \usepackage{unicode-math} gives:
[] = Σ ()
However, I want to avoid unicode-math due to its font management issues (see here and its related questions). Is there any way I can get behavior in 2 and 4 in XeLaTeX without changing math fonts?
unicode-math; or do you know how to switch off its main functionalities (font mappings) and only keep the character mapping? – Mygod Apr 06 '20 at 00:20