summaryrefslogtreecommitdiff
path: root/tex/context/base/math-stc.mkvi
diff options
context:
space:
mode:
Diffstat (limited to 'tex/context/base/math-stc.mkvi')
-rw-r--r--tex/context/base/math-stc.mkvi18
1 files changed, 17 insertions, 1 deletions
diff --git a/tex/context/base/math-stc.mkvi b/tex/context/base/math-stc.mkvi
index 2dc2b2c22..8d5b1d18f 100644
--- a/tex/context/base/math-stc.mkvi
+++ b/tex/context/base/math-stc.mkvi
@@ -568,7 +568,7 @@
\definemathstackers
[\v!bottom]
[\v!mathematics]
- [\c!location=\v!top,
+ [\c!location=\v!top, % ?
\c!strut=\v!no,
\c!middlecommand=\mathematics,
\c!hoffset=\zeropoint]
@@ -667,6 +667,22 @@
\definemathunderextensible [\v!bottom] [undertwoheadleftarrow] ["27F8]
\definemathunderextensible [\v!bottom] [undertwoheadrightarrow] ["27F9]
+% We don't use overline and underline. This is one of the overlooked aspects of
+% unicode cq. opentype math: why treat rules different than e.g. arrows and
+% accents. It is a bit unfortunate that the opportunity to move math to new
+% technologies happened outside the tex domain (and/or some aspects were kept
+% while in fact they were side effects of limitations of traditional fonts).
+% From the unicode aware tex engines' implementation point of view things
+% could have been done a bit nicer but then: the community didn't seem to care
+% too much and just has to follow now.
+%
+% Anyhow, we use a character based approach so that at least we get unicode
+% stuff in the backend (okay, we still need to deal with some cut and paste
+% issues but at least we now know what we deal with.
+
+\definemathunderextensible [\v!bottom] [underbar] ["203E]
+\definemathoverextensible [\v!top] [overbar] ["203E]
+
%D Some bonus ones (for the moment here):
\definemathstackers