Bug report, \frac as redefined in m-newmat.tex
The following simple file works fine if I comment out the \usemodule[newmat] line, but crashes if I put it in. \usemodule[newmat] \starttext \placeformula \startformula \eqalign{\frac{1}{2} \cr} \stopformula \stoptext - Brooks
Hi Brooks
The following simple file works fine if I comment out the \usemodule[newmat] line, but crashes if I put it in.
\usemodule[newmat] \starttext \placeformula \startformula \eqalign{\frac{1}{2} \cr} \stopformula \stoptext
ah ... the alignment lookahead problem, i think we can safely patch \mathematics to catch lookahead as well as make frac more robust for unwanted expansion \usemodule[newmat] \unexpanded\def\frac#1#2{\mathematics{\genfrac{}{}{}\donothing{#1}{#2}}} \def\mathematics#1{\relax\ifmmode#1\else$#1$\fi} \starttext \placeformula \startformula \eqalign{\frac{1}{2}} \stopformula \stoptext ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
At 01:40 PM 8/2/2004, you wrote:
ah ... the alignment lookahead problem, i think we can safely patch \mathematics to catch lookahead as well as make frac more robust for unwanted expansion
\usemodule[newmat]
\unexpanded\def\frac#1#2{\mathematics{\genfrac{}{}{}\donothing{#1}{#2}}}
\def\mathematics#1{\relax\ifmmode#1\else$#1$\fi} [...]
That fixes the problem on my end; thanks! Will you be updating the defintions of \tfrac, \dfrac, and so forth in m-newmat to match? I admit to not being too sure whether I think it's a good idea to support using math constructs such as \frac outside of math mode, but I do think that if one's going to do that, it's good to be consistent and do it with the variants too. - Brooks
Brooks Moses wrote:
At 01:40 PM 8/2/2004, you wrote:
ah ... the alignment lookahead problem, i think we can safely patch \mathematics to catch lookahead as well as make frac more robust for unwanted expansion
\usemodule[newmat]
\unexpanded\def\frac#1#2{\mathematics{\genfrac{}{}{}\donothing{#1}{#2}}}
\def\mathematics#1{\relax\ifmmode#1\else$#1$\fi}
[...]
That fixes the problem on my end; thanks!
Will you be updating the defintions of \tfrac, \dfrac, and so forth in m-newmat to match? I admit to not being too sure whether I think it's a good idea to support using math constructs such as \frac
do they need patching? they don't use \mathematics afaik Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
On Wed, Aug 04, 2004 at 12:14:29AM +0200, Hans Hagen Outside wrote:
Brooks Moses wrote:
That fixes the problem on my end; thanks!
Will you be updating the defintions of \tfrac, \dfrac, and so forth in m-newmat to match? I admit to not being too sure whether I think it's a good idea to support using math constructs such as \frac
do they need patching? they don't use \mathematics afaik
They don't need patching in the sense of fixing a bug, no -- it's more a matter of consistency in behavior. What I see is that there's a list of definitions for \frac, \dfrac, \tfrac, and various other similar forms, all of which can be used in math mode only, and do not use \mathematics. Then, a couple of lines after that, there's the second definition for \frac, which is the one that needed patching. This one, unlike the first definition, is wrapped in \mathematics so that it can work outside math mode. So I'm wondering why only \frac has this second definition that allows it to work outside of math mode. I would think that things should be consistent -- either that all of the definitions in the first list should be changed to use \mathematics, or that \frac should be returned to the simpler version -- so that \tfrac is still simply "\frac typeset in text mode" as one would expect it to be. Even if they're all left in the current form, I'd think that the unused definition of \frac in the first list should be removed. My argument for simplifying \frac to take the \mathematics out of it is that it is a math-mode construct that typesets its arguments in math mode, and allowing it to be used outside of math mode encourages sloppy TeXing. It also hides the point of transition to math mode, meaning that x and \frac{1}{x} will typeset x differently, which strikes me as confusing. I recognize that this is a very debatable position, though, and that there are at the very least arguments for backwards compatiblity that contradict it. - Brooks
Hi Brooks,
Then, a couple of lines after that, there's the second definition for \frac, which is the one that needed patching. This one, unlike the first definition, is wrapped in \mathematics so that it can work outside math mode.
So I'm wondering why only \frac has this second definition that allows it to work outside of math mode. I would think that things should be consistent -- either that all of the definitions in the first list should be changed to use \mathematics, or that \frac should be returned to the simpler version -- so that \tfrac is still simply "\frac typeset in text mode" as one would expect it to be.
this has a historical reason; it was added because latex users expect \frac to be there and to work in text mode as well;
Even if they're all left in the current form, I'd think that the unused definition of \frac in the first list should be removed.
well, just consider it not to be there; it should definitely not be mentioned in manuals -)
My argument for simplifying \frac to take the \mathematics out of it is that it is a math-mode construct that typesets its arguments in math mode, and allowing it to be used outside of math mode encourages sloppy TeXing. It also hides the point of transition to math mode, meaning that x and \frac{1}{x} will typeset x differently, which strikes me as confusing. I recognize that this is a very debatable position, though, and that there are at the very least arguments for backwards compatiblity that contradict it.
if i remember right (maybe tobias burnus remembers more) it was mainly meant to be used in section titles; in such cases, when one sets a font explicitly, math is often not set up accordingly and frac permits one to do at least fractions Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
participants (3)
-
Brooks Moses -
Hans Hagen -
Hans Hagen Outside