Personally, I just don't use the '​ in LyX at all, especially as it doesn't 
display correctly, and as you show, doesn't combine properly with other 
superscripts/subscripts.
Instead, I always do ^\prime​ which looks much nicer, and lets you control 
things better.

So perhaps the code would work better if it just made '​ autocorrect to 
^\prime​ ?

(In fact I've defined my own \pri​ command which alters the LaTeX height of the 
\prime​ to one I like more; and ones that use the unicode-math​ multi-prime 
commands \dprime​, \tprime​, and \qprime​).

— Isaac Oscar Gariano​
________________________________
From: lyx-devel <lyx-devel-boun...@lists.lyx.org> on behalf of Andrew Parsloe 
<ajpars...@gmail.com>
Sent: Monday, 27 November 2023 10:10 AM
To: LyX Developers <lyx-devel@lists.lyx.org>
Subject: Primes and superscripts in mathed


TeX accepts  S'^{2}  (between $ delimiters) as a valid construction (TeXbook, 
p.130, line above Ex. 16.5) but mathed in LyX converts this to S'{}^{2}. In 
this simple case that is not a problem (in fact the interpolated {} gives 
better spacing) but it becomes a problem with expressions like (real-life 
example)

 S'^{\,-1}_{0i}  or S_{0i}'^{\,-1}

These compile fine in TeX and display as desired but mathed turns them into

 S'{}_{0i}^{\,-1}  and  S_{0i}'{}^{\,-1}

which give very different and clearly wrong results in the pdf.

The combination of a prime followed by a superscript seems a sufficiently basic 
sequence to me that I think mathed should handle it with more finesse than 
immediately and always inserting an empty brace pair between ' and ^.

[[Speculation:  I don't know the code but perhaps the problem is that mathed 
insists on subscript preceding superscript when both are involved in a symbol? 
R^{ij}_{kl} in mathed becomes R_{kl}^{ij}, and inserting a prime as in 
R'_{kl}^{ij} then produces the double superscript error. In contrast, 
R_{kl}'^{ij} does not. TeX accepts the '^ combination and I think mathed should 
do so too.]]

Andrew
-- 
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel

Reply via email to