cellfun() is a powerful function in MATLAB which mirrors the idea of applying a ‘map’ operation (as in functional programming).<\/p>\n
One common use is to identify which cells are empty so you can ignore them. Typically, you can do<\/p>\n
index.emptyCells = cellfun(@isempty, C);<\/pre>\nAccording to the help (H1) page,\u00a0there is a limited set of strings that you can use in place of the function handle (functor) for backward compatibility:<\/p>\n
\nA = CELLFUN('fun', C), where 'fun' is one of the following strings,\nreturns a logical or double array A the elements of which are computed \nfrom those of C as follows:\n \n 'isreal' -- true for cells containing a real array, false\n otherwise \n 'isempty' -- true for cells containing an empty array, false \n otherwise \n 'islogical' -- true for cells containing a logical array, false \n otherwise \n 'length' -- the length of the contents of each cell \n 'ndims' -- the number of dimensions of the contents of each cell\n 'prodofsize' -- the number of elements of the contents of each cell\n \nA = CELLFUN('size', C, K) returns the size along the K-th dimension of\nthe contents of each cell of C.\n \nA = CELLFUN('isclass', C, CLASSNAME) returns true for each cell of C\nthat contains an array of class CLASSNAME. Unlike the ISA function, \n'isclass' of a subclass of CLASSNAME returns false.<\/pre>\nTurns out these functions have their native implementation and runs super-fast. But I got burned once when I tried to use this on cells containing table() objects:<\/p>\n
index.emptyCells = cellfun('isempty', cellsOfTables);<\/pre>\nI meant to find out if I have zero-row or zero-column (empty) table objects with that. It gave\u00a0me all false even when my cells have these 0-by-0 tables. What a Terrible Failure?! Turns out these ‘backward compatibility’ native\u00a0implementations (I guess they already have cellfun() before having function handles) looks at the raw data stream like PODs (Plain Old Datatypes) as a C program would do.<\/p>\n
A table() object has lots of stuff stored in it like variable (column) names, so there’s no way a program looking at an arbitrary\u00a0binary stream without accounting for such data type will consider that object\u00a0empty. It’s up to the overloaded isempty() or numel() of the class to tell what is empty and what’s not, but it needs to be called by the function handle to establish which method to call.<\/p>\n
Lesson learned: don’t use those\u00a0special string based functor in cellfun() unless you know for sure it’s a POD. Otherwise it will get you the wrong answer at the speed of light.<\/p>\n
<\/div>\n