We Like New Functions

[latot]

Hi again, well this time i have this functions, about the zero functions and related, i try avoid the most posible execute this line y = sym(eye(cell2nosyms(varargin){:}));, because this is fast eye(cell2nosyms(varargin){:}), but transform that to sym takes to much time with big matrix.

well other problem is, although i add the cell2sym function when i try execute it i get an error:

>> cell2sym({sym(1)})
error: 'cell2sym' undefined near line 1 column 1

and actually divisors depends of it, if still don't works maybe at least add it to private.

This Fixes #405 and add some functions of #215
Well. Cya.

[latot]

from travis:

>>>>> processing @sym/cell2sym.m
***** test
 A = {1, 2, 3;4, 5, 6;7, 8, 9};
 B = [1, 2, 3;4, 5, 6;7, 8, 9];
 assert( isequal( cell2sym(A), sym(B)))
!!!!! test failed
'cell2sym' undefined near line 5 column 19

D:

[latot]

moved the problem of cell2sym to #527

[cbm755]

maybe A and x (or var) are better names here? Can you also file an issue for the missing vector output?

[cbm755]

this needs a test, maybe something like:

octave:17> A = dawson([1 2])
A =

   0.53808   0.30134

octave:18> B = double(dawson(sym([1 2])))
B =

   0.53808   0.30134

octave:19> assert(A, B, -eps)

[cbm755]

This looks great! Some minor comments above.

I don't really understand your eye, ones and zeros yet. Are you trying to support calls like:

zeros(sym(3), 2, 'double')

If so, those should be tested (and I don't think it works right now). Or is it for something else?

If easy, you could split off those functions and do a new PR and we can discuss. Then I can merge the other bits!

[latot]

okidokis, all proposed changes applied! zeros and similar to this pr #532.