Cell array as structure field - why is retrieving contents so difficult?

Hi, im working with Matlab R2012a. I have a structure with a cell array as one of the fields. I would like to simply assign the entire cell array to a variable in the workspace in one line, however this is apparently impossible:

When i try "var=struct.field" it only assigns the first cell of "field" to "var".

When i try "var=struct.field{:}" it throws up an error, apparently any indexing after a field specifier is an error.

When i try "[var{1},var{2},var{3},var{4},var{5}]=struct.field" it assigns all 5 cells of "field" to each cell of "var", leaving me with an unnecessarily large and nested cell array:

[ie "var{1}={field{1},field{2},...,field{5}}, var{2}={field{1},field{2},...,field{5}}, ..., var{5}={field{1},field{2},...,field{5}}". This behaviour seems very odd and for some ungodly reason scales homogenously between "var" and "field", so "[var{1},var{2}]=struct.field" results in "var{1}={field{1},field{2}}, var{2}={field{1},field{2}}"].

I have also tried using the "getfield" command in every way i can think of, but this invariably throws up an error too. The ONLY way i have found to do this is to first do "[a,b,c,d,e]=struct.field" and then assemble the cell array seperately [ie "var={a,b,c,d,e}"]. This seems like a remarkably inelegant way of doing it considering the usual silky smooth programming functionality of Matlab. Is there a way to do this in one line, and i'm just being obtuse? I appreciate that this is a very specific and pernickety little thing, but it's just bugging me.

[Walter Roberson]

Your posting would be easier to read if you broke it up into paragraphs and isolated the code.

[Andy]

Apologies, i did try, but it didnt seem to work. I tried various newlines and none showed up on the preview.

[Andy]

there we go i've at least managed to get it paragraphed.

Answer 1

[Walter Roberson]

[var{1:length(YourStruct)}] = deal(YourStruct.field);

or

var = {YourStruct.field};

[Andy]

face palm it really is that easy. I guess even matlab isn't entirely free of cyptic-ness yet :) Many thanks Walter, apologies for spamming the boards with such trivialities.