Concatenate fields of a structure

Hi,

I have a structure inside an app:

app.C1 struct with fields:

DHO: {{1×3 cell} {1×3 cell}}

DHOnw: {{1×3 cell} {1×3 cell}}

with:

app.C1.DHO{1}= {["ID1"]} {["GD1"]} {["ED1"]}

app.C1.DHO{2}= {["ID2"]} {["GD2"]} {["ED2"]}

and

app.C1.DHOnw{1}= {["IDnw1"]} {["GDnw1"]} {["EDnw1"]}

app.C1.DHOnw{2}= {["IDnw2"]} {["GDnw2"]} {["EDnw2"]}

The fieldnames DHO, DHOnw are not fixed, and in other cases I could have different ones.

I want to create a new “quantity” with all these fields concatenated, for example:

new={ ‘ID1’; ‘GD1’; ED1; ‘ID2’; ‘GD2’; ED2; ‘IDnw1’; ‘GDnw1’; EDnw1; ‘IDnw2’; ‘GDnw2’; EDnw2;};

to be then used as a single column of a table.

I tried with cellfun, but I was not able to get reasonable results: do you have some suggestions?

Thank you in advance for your help.

Answer 1

[Stephen23]

Note that using a string array is much more efficient than storing scalar strings in a cell array.

app.C1.DHO = {{"ID1","GD1","ED1"},{"ID2","GD2","ED2"}};
app.C1.DHOnw = {{"IDnw1","GDnw1","EDnw1"},{"IDnw2","GDnw2","EDnw2"}};
app.C1

ans = struct with fields:
      DHO: {{1×3 cell}  {1×3 cell}}
    DHOnw: {{1×3 cell}  {1×3 cell}}

tmp = struct2cell(app.C1);
tmp = vertcat(tmp{:}).';
out = horzcat(tmp{:})

out = 1×12 cell array
    {["ID1"]}    {["GD1"]}    {["ED1"]}    {["ID2"]}    {["GD2"]}    {["ED2"]}    {["IDnw1"]}    {["GDnw1"]}    {["EDnw1"]}    {["IDnw2"]}    {["GDnw2"]}    {["EDnw2"]}

Answer 2

[Ferdi]

Thank you very much!

I will try to follow your suggestion and using a string array.

Answer 3

[Ferdi]

Hi, I tried to follow your suggestion, surely not in the smartest way.

I recover the various function (DHO and so on) in this way (I keep the coding simple here):

function DHOSpinnerValueChanging(app, event)

app.nDHO = event.Value; % numers of DHOs

if app.nDHO ~= 0

app.PD = zeros(3*app.nDHO,1);

par0=[1,0.5,2]; %initialize some parameters

for ii=1:app.nDHO

app.PD(3*ii-2:3*ii)=ii*par0;

app.C1.DHO{ii}={"ID"+ii, "GD"+ii, "ED"+ii}; % scalar strings

app.T1.DHO(ii,:)=["ID"+ii "GD"+ii "ED"+ii]; % alternative: string array

end

end

and similarly for DHOnw.

You showed me how to treat app.C1.

To obtain a similar results witht he string array, I wrote a multiple loop:

A=string([]);

cf = cellfun(@(x) app.T1.(x), F, 'UniformOutput', false);

for j=1:numel(fieldnames(app.T1))

for k=1:height(cf{j})

for l=1:width(cf{j})

A=[A, cf{j}(k,l)];

end

end

end

A =

1×10 string array

"IL1" "GL1" "IL2" "GL2" "ID1" "GD1" "ED1" "ID2" "GD2" "ED2"

Answer 4

[Ferdi]

Hi again, I have a comment about your first answer.

It is working only when the fields have the same dimensions.

In case like:

app.C1 struct with fields:

Lrz: {{1×2 cell}}

DHO: {{1×3 cell} {1×3 cell}}

vertcat is not working anymore.

Is there a way to make it working in this more general case?

thanks!

[Stephen23]

"Is there a way to make it working in this more general case?"

Probably, once you specify the order that should be used for said "general case". The only reason I used VERTCAT was to provide the same order that you specified in your question. If the order is not significant, then you could use HORZCAT instead:

app.C1.Lrz = {{"One","Two"}};
app.C1.DHOnw = {{"IDnw1","GDnw1","EDnw1"},{"IDnw2","GDnw2","EDnw2"}};
app.C1

ans = struct with fields:
      Lrz: {{1×2 cell}}
    DHOnw: {{1×3 cell}  {1×3 cell}}

tmp = struct2cell(app.C1);
tmp = horzcat(tmp{:});
out = horzcat(tmp{:})

out = 1×8 cell array
    {["One"]}    {["Two"]}    {["IDnw1"]}    {["GDnw1"]}    {["EDnw1"]}    {["IDnw2"]}    {["GDnw2"]}    {["EDnw2"]}

[Ferdi]

thanks again, I should have found by myself.... sorry.

For general case I meant a variable number of functions.

I do not understand your reference to the order as it seems however preserved.

thanks