I looks like not resetting plotsetup to default at the end of the function (i.e. returning the plot strucutre) seems to work.

so this does save a file

savePlot:=proc(v_p, v_fileName)  # lets define a function to do this
  plotsetup(ps, plotoutput=v_fileName, plotoptions=`portrait,noborder`);
  display(v_p);   # also tried without "display" 
end proc;

savePlot(p, "test.eps");

however if one wants to print a plot to the screen, one has to call plotsetup(default) manually after calling savePlot... which is not good.

I am on linux using maple 14. thanks

thanks for your answers.

good news about shift+tab in maple 15... i ll wait for that.

it is however pretty sad that key customization is not possible - this in principle should be fairly simple to implement and would be useful for many people from what I understand.

thanks for your answers.

good news about shift+tab in maple 15... i ll wait for that.

it is however pretty sad that key customization is not possible - this in principle should be fairly simple to implement and would be useful for many people from what I understand.

thanks for responding... yes - in fact i tried doing everythign in one go before.

I am using maple 13 and from your code i get the following error:

Error, (in Setup) multiple definition of algebra rule; received values for %Commutator(Dagger(b), c) are {0, rule}

does that code work for you?? what version of maple?

i now get around things by just defining aD, bD, etc... as operators.. and after the algebra just sub {aD=Dagger(a), bD=Dagger(b), etc}.

I am still finding however that the usefulness (as is) of this package for me is very limited as simplify and algsub/sub cannot not really simplify my terms far enough if they're operators (i.e commutators are not taken into account as much as i'd like.

thanks for responding... yes - in fact i tried doing everythign in one go before.

I am using maple 13 and from your code i get the following error:

Error, (in Setup) multiple definition of algebra rule; received values for %Commutator(Dagger(b), c) are {0, rule}

does that code work for you?? what version of maple?

i now get around things by just defining aD, bD, etc... as operators.. and after the algebra just sub {aD=Dagger(a), bD=Dagger(b), etc}.

I am still finding however that the usefulness (as is) of this package for me is very limited as simplify and algsub/sub cannot not really simplify my terms far enough if they're operators (i.e commutators are not taken into account as much as i'd like.

thanks to all for suggestions.

In my case the matrices usually do have a full set of linearly independent eigenvectors, and hence I have been simply diagonalizing before exponentiating (just as Robert suggested - although his code snippet manages to look much cleaner that mine!)...  

I am still curious however if Mathematica manages to do something clever(er) inside their equivalent function...
 

hi Jakubi... I am aware of the '~' operator, however it is a new development and only works on maple 13.. (i need my worksheets to work on older versions, so use map instead).

... the problem is that (IMO) Im should behave in the same way as Re... and hence if one can operate on a matrix, so should the other.

... I will try to minizmie my worksheet to show a case where Im(Matrix) always crashes the kernel (and not just returns an error).

getting the imaginary part of a matrix is borken for me (on both maple 12 and 13 - maybe someone can verify?)

Here's a simple example:

m:=Matrix([[1,2],[3,4+4*I]]);

Re(m); # works!

Im(m);
Error, invalid input: simpl/Im expects its 1st argument, x, to be of type {boolean, algebraic}, but received Matrix(2, 2, [[...],[...]], datatype
= anything)

When i try to do this on a bigger matrix (in a large worksheet) the kernel crashes.

This is a workaround for me, which works:
map(Im, m);

NOTE: Perhaps Im was not meant be able to operate on a matrix - but it should still not kill the kernel.

thanks.