Dear Wolfgang,
I am grateful to you for your precious help with derivabbrev : true.
Thanks to your suggestions, I do not have now to load the
modified mactex.lisp any more. (I already added your code
to maxima-init.lisp as you suggested.) Now
(C1) derivabbrev : true$
(C2) tex(diff(u(x,y),x,3,y,3)+'diff(v(x,y),x,1,y,1)=a)$
$$\left(v\left(x,y\right)\right)_{x\,y}+\left(u\left(x,y\right)
\right)_{x\,x\,x\,y\,y\,y}=a$$
After you will have received comments by Richard Fateman,
I understand he is already in principle positive for this change:
"Richard Fateman" <fateman@cs.berkeley.edu> wrote:
> I haven't studied the mactex patch, but it
> seems to me it only affects the derivabbrev:true
> behavior, and if it works for that, install it!
>
> RJF
and by those colleagues with experience in the mactex.lisp
package, naturally I will extremely glad to have the modified
form (addition of your code with respect to derivabbrev:true)
available in advance in the next release of Maxima.
Meanwhile, I am very happy for being already able to correctly
use derivabbrev : true also with TeX output in the way I usually
work with Maxima (i.e. using its TeX ouput after processing it
with LaTeX). Your addition to mactex.lisp is very useful indeed.
Many most sincere thanks again and best regards,
Nikos
> "Wolfgang Jenkner" <wjenkner@inode.at> wrote:
> The lisp system always loads the installed maxima image, which, of
> course, contains the code from the unpatched mactex. If a file is
> changed, you have either to explicitly load it (like you did) or to
> dump a new maxima image and to install it.
>
> However, instead of changing mactex.lisp, you could put the code
> snippet in share/maxima-init.lisp, which is automatically loaded.
> . . .
> Thanks for trying it out! Of course, I'll wait for comments by
> mactex's author Richard Fateman or by other people who know this
> package better than I do. In principle, we could easily add the code.
>
> Wolfgang