Efficient way to change a struct of (mixed-type) arrays into a structure array?

12 Ansichten (letzte 30 Tage)
Hi! Resulting from a DB query, I have a 6-field scalar struct consisting of 1522-element arrays, some cell, some numeric (if that matters); I want to convert it into a 1522x1 struct array with the same six fields--is there a non-for-loop way to do this?
  1 Kommentar
Jan
Jan am 18 Aug. 2012
Please provide example data in valid Matlab syntax. "6-field scalar struct consisting of 1522-element arrays, some cell, some numeric" does neither define the inputs nor the wanted output unequivocally. As you see in the given answers, forcing contributors to guess the details is no efficient method to get your problem solved.

Melden Sie sich an, um zu kommentieren.

Antworten (2)

Walter Roberson
Walter Roberson am 17 Aug. 2012
Bearbeitet: Walter Roberson am 17 Aug. 2012
I am not sure if there is a loop-less way to handle the situation if the field names are not known in advance. But perhaps the below is good enough, when "A" is your scalar struct:
B = [fieldnames(A).';cellfun(@num2cell,struct2cell(A).','un',0)]
C = struct(B{:}};
  2 Kommentare
Walter Roberson
Walter Roberson am 18 Aug. 2012
In my tests, the code works with cells as well. I have not tried with nested structures or objects, though.
Looking at the code for num2cell(), I can see that the code should work for any field which can be handle numel() and linear subscript referencing.

Melden Sie sich an, um zu kommentieren.


Azzi Abdelmalek
Azzi Abdelmalek am 17 Aug. 2012
Bearbeitet: Azzi Abdelmalek am 17 Aug. 2012
% A contains multiple array (num, str,...)
eval(sprintf('a.field%d=A(:,%d);',[1 :6;1:6]))
  5 Kommentare
Azzi Abdelmalek
Azzi Abdelmalek am 17 Aug. 2012
tell us what is the size of your cell array: size(A)
Jan
Jan am 18 Aug. 2012
No, Azzi, do not use EVAL to access structs. "Dynamic fieldnames" are much more efficient and cleaner: a.(sprintf('field%d', i)). The topic EVAL has been discussed hundreds of times in this forum.

Melden Sie sich an, um zu kommentieren.

Kategorien

Mehr zu Data Type Conversion finden Sie in Help Center und File Exchange

Produkte

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by