@icegood I uploaded a new version that might fix this issue on Windows.  I've tested it on Windows 7.  If that doesn't work you can now assign the 'emacs' option to a procedure which will then be used to launch Emacs.  So you can put whatever code is necessary in that procedure to do the job.

I just noticed that I cannot paste stuff into the text window (the box I'm typing in right now) using the Chrome browser.  At least I cannot cut from Emacs and paste into here, on a Linux system.  It worked fine with Firefox, but I can no longer log in with Firefox. Any ideas?

@icegood This part of the debugger (launching emacs when needed) has not been tested much because (1) I always have Emacs running and don't need to launch it, and (2) I don't use Windows much. I thought I had it working, though, on Windows 7.  It doesn't help that each Windows release seems to have different, incompatible, methods for launching programs.  If you can recommend a Maple call that launches Emacs and returns control to Maple, on a Windows system, let me know and I'll make the changes.  Do let me know what version of Windows you have used it with.

@icegood What about 1/x?  Should that go to 1/x^1.0?  One way to do this is

evalindets( 1/x, rational, evalf);

@icegood What about 1/x?  Should that go to 1/x^1.0?  One way to do this is

evalindets( 1/x, rational, evalf);

What are you using to query for input?  A maplet? Is this being run from the Standard GUI?

@icegood Try going through the instructions here, specifically sections 3.4, 3.5, and 3.5.1. Sorry this is more painful than it should be, installing and configuring Emacs on Windows invariably is that way.

The single forward quote, in the elisp command (autoload 'mds "mds" "Restart the Maple Debugger server" t) is correct.

A useful method is to check what Emacs thinks your HOME directory is.  Run

M-x getenv

then type HOME when it queries you for an environmental variable.  The result is displayed.  That is the directory where Emacs will look for the .emacs file.

@icegood Thanks for posting that zip file with your .emacs file, it makes debugging this easier.

I believe that Emacs won't work with backslashes in the path, even if doubled.  Try the following (replace the backslashes in the DOS path with forward slashes):

(setq load-path (cons ("C:/Users/ice/AppData/Roaming/.emacs.d/maple/") load-path))

After modifying that, restart Emacs and execute

M-x mds

M-x means "depress the "Alt" key and type "x", then type "mds".   The "M" really means Meta, but on PC keyboards it is mapped to the Alt key.

Actually, you probably want to add the following line to your .emacs file

(mds)

after the setq command (above).  That will launch the debugger server when Emacs is started, so you won't have to type M-x mds.

@pagan Thanks for the response.  That sounds exactly like what has happened here.  Strange problem...

1/(s+1) is expressible as a single diff-equation because there are no zeros (in the numerator).  That isn't the case for s/(s+1).

You can use ?addtable to define a laplace transform:

with(inttrans):

addtable(invlaplace, Y(s), y(t), s, t):
addtable(invlaplace, U(s), u(t), s, t):

eq := (s^2+s+1)*Y(s) = U(s):
y(0) := 0:
D(y)(0) := 0:

invlaplace(eq,s,t);
                     / 2      \
                     |d       |   /d      \
                     |--- y(t)| + |-- y(t)| + y(t) = u(t)
                     |  2     |   \dt     /
                     \dt      /

1/(s+1) is expressible as a single diff-equation because there are no zeros (in the numerator).  That isn't the case for s/(s+1).

You can use ?addtable to define a laplace transform:

with(inttrans):

addtable(invlaplace, Y(s), y(t), s, t):
addtable(invlaplace, U(s), u(t), s, t):

eq := (s^2+s+1)*Y(s) = U(s):
y(0) := 0:
D(y)(0) := 0:

invlaplace(eq,s,t);
                     / 2      \
                     |d       |   /d      \
                     |--- y(t)| + |-- y(t)| + y(t) = u(t)
                     |  2     |   \dt     /
                     \dt      /

equn2 doesn't include beta2, so just solve equn1 for beta1, then substitute equn2 into the solution:

sol := solve(equn1,{beta2});
subs(equn2, sol[1]);

equn2 doesn't include beta2, so just solve equn1 for beta1, then substitute equn2 into the solution:

sol := solve(equn1,{beta2});
subs(equn2, sol[1]);

@acer I agree---the Maple GUI really shouldn't be creating invalid worksheets. My limited observation is that the corrupted ones primarily originate from users with a non-ascii keyboard.  I modified the above routine to print the invalid characters.  Does the ^U come from inserting an umlaut?  What about ^V?

@acer I agree---the Maple GUI really shouldn't be creating invalid worksheets. My limited observation is that the corrupted ones primarily originate from users with a non-ascii keyboard.  I modified the above routine to print the invalid characters.  Does the ^U come from inserting an umlaut?  What about ^V?