I am trying to reference an element buried within a structure that I did not create (hence I don't know the exact way in which it was built).
Having loaded the structure, if I type:
dataFile.RECORDINGS.eye
I receive the following output:
ans =
2
ans =
2
Both of those variables will always be the same, but they could be at any time 1, 2 or 3. What I'd like to do is check with a switch statement which looks like this:
switch dataFile.RECORDINGS.eye
case {1, 2}
% action A
case 3
% action B
end
Of course, the above throws up an error because 'case' cannot check whether dataFile.RECORDINGS.eye contains a given value since there are two elements stored under that address. So, my question is: how do I reference just one of the elements? I thought it would be as simple as replacing the first line with:
switch dataFile.RECORDINGS.eye(1)
...But, this gives the error:
??? Field reference for multiple structure elements that is followed by more reference blocks is an error.
Similarly, I can't access the element like this:
switch dataFile.RECORDINGS.eye.1
...As I get the following error:
??? Dot name reference on non-scalar structure.
If the values are really always the same, you can try the following to get a scalar that can be used in the switch command:
unique([dataFile.RECORDINGS.eye])
By the way, did you try to index RECORDINGS, i.e.,
dataFile.RECORDINGS(1).eye
dataFile.RECORDINGS(2).eye
Perhaps instead of eye having multiple elements, you have multiple elements of RECORDINGS that each have a single value of eye? You might want dataFile.RECORDINGS(1).eye or dataFile.RECORDINGS(2).eye.
Related
I have large data sets which i want to work with in matlab.
I have a struct called Trail containing serveral structures called trail1, trail2 ...
which then contain several matrices. I now want to add another point to for instance trail1
I can do that with Trail.trail1.a2rotated(i,:) = rotpoint'; the problem is that i have to do it in a loop where the trail number as well as the a2rotated changes to e.g. a3rot...
I tired to do it like that
name ="trail"+num2str(z)+".a2rotated"+"("+i+",:)";
name = convertStringsToChars(name);
Trail.(name) = rotpoint'
But that gives me the error: Invalid field name: 'trail1.a2rotated(1,:)'.
Does someone have a solution?
The name in between brackets after the dot must be the name of a field of the struct. Other indexing operations must be done separately:
Trail.("trail"+z).a2rotated(i,:)
But you might be better off making trail(z) an array instead of separate fields with a number in the name.
In my code, I have a structure and in a field of it, I want to sort its values.
For instance, in the field of File_Neg.name there are the following values, and They should be sorted as the right values.
File_Neg.name --> Sorted File_Neg.name
'-10.000000.dcm' '-10.000000.dcm'
'-102.500000.dcm' '-12.500000.dcm'
'-100.000000.dcm' '-100.000000.dcm'
'-107.500000.dcm' '-102.500000.dcm'
'-112.500000.dcm' '-107.500000.dcm'
'-110.000000.dcm '-110.000000.dcm'
'-12.500000.dcm' '-112.500000.dcm'
There is a folder that there are some pictures with negative labels in it (above example are labels of pictures). I want to get them in the same order as present in the folder(that's mean the Sorted File_Neg.name). But when running the following code the values of Files_Neg.name load as the above example (left: File_Neg.name), while I want the right form.
I have also seen this and that but they didn't help me.
How to sort values of a field in a structure in Matlab?
Files_Neg = dir('D:\Rename-RealN');
File_Neg = dir(strcat('D:\Rename-RealN\', Files_Neg.name, '\', '*.dcm'));
% when running the code the values of Files_Neg.name load as the above example (left: File_Neg.name)
File_Neg.name:
This answer to one of the questions linked in the OP is nearly correct for the problem in the OP. There are two issues:
The first issue is that the answer assumes a scalar value is contained in the field to be sorted, whereas in the OP the values are char arrays (i.e. old-fashioned strings).
This issue can be fixed by adding 'UniformOutput',false to the arrayfun call:
File_Neg = struct('name',{'-10.000000.dcm','-102.500000.dcm','-100.000000.dcm','-107.500000.dcm','-112.500000.dcm','-110.000000.dcm','-12.500000.dcm'},...
'folder',{'a','b','c','d','e1','e2','e3'});
[~,I] = sort(arrayfun(#(x)x.name,File_Neg,'UniformOutput',false));
File_Neg = File_Neg(I);
File_Neg is now sorted according to dictionary sort (using ASCII letter ordering, meaning that uppercase letters come first, and 110 still comes before 12).
The second issue is that OP wants to sort according to the magnitude of the number in the file name, not using dictionary sort. This can be fixed by extracting the value in the anonymous function applied using arrayfun. We use str2double on the file name, minus the last 4 characters '.dcm':
[~,I] = sort(arrayfun(#(x)abs(str2double(x.name(1:end-4))),File_Neg));
File_Neg = File_Neg(I);
Funnily enough, we don't want to use 'UniformOutput',false any more, since the anonymous function now returns a scalar value.
I am trying to run a loop over multiple files, by feeding the file names into a function. I have saved these file names as a structure using:
files=dir('testdata\*.siz');
nrows=size(files,1);
Now my loop is:
for i=1:nrows
filename=files.name{i};
Singapore(filename);
writetable(ans,'file.xls')
end
However, I get the error:
"Field reference for multiple structure elements that is followed by more reference blocks is an error."
I found that the error is in
filename=files.name{1};
but everywhere I've searched tells me to use { } to access fields in a structure. I also have tried other types of brackets in vain.
Additional Information:
'files' is the name of the structure
'name' is the first column field within 'files' containing the file names in inverted commas.
You are referencing the files struct wrong, you need:
files(i).name
The {} is for accessing cell arrays.
You should also use ii (or similar) instead of i as your indexing variable since i is a already Matlab variable (imaginary unit).
I have a <850x1> cell called x. Each of the individual structures has a 'Tag' name and 'Data' cell with <7168x1 double> data values.
(i.e.
x{1,1}.Tag = 'Channel1', x{1,1}.Data= <7168x1 double>)
So, I want to go through the x cell, identify the structures with 'Channel1' Tag names and pull out that structure's data. Then, combine the data into a cell called Ch1. Here is my approach so far:
n=1:850
if x{n,1}.Tag == 'Channel1'
Ch1{:,n} = x{n,1}.Data;
end
However, this gives an error: Bad cell reference operation.
Any ideas what may be going wrong?
There are 2 issues here. First, your if statement will compare each entry in the string x{n,1}.Tag to each entry in the string 'Channel1'. If the dimensions are not the same, you will get an error. To fix this, you could use the string compare function, strcmp. The other issue is that you are assigning n to all values between 1 and 850 at once. This is the issue that is producing the actual error you are seeing. Instead, you want to step through each of these values one at a time with a for loop. I would suggest trying the following code:
for n=1:850
if strcmp(x{n,1}.Tag, 'Channel1')
Ch1{:,n} = x{n,1}.Data;
end
end
I am currently successfully using a CASE expression to update an empty column based on attributes from other columns. For example
UPDATE table SET cat = CASE
WHEN term = '{"Boulder"}' then 'Boulder'
However, I need to do the same but on an text array and particularly when an element is in a specific position within that array.
For example if the data looks like
{"Boulder, Tree, Bush"}
WHEN position('Tree' in term) > 0 then 'Boulder'
But I receive an error message
function pg_catalog.position(character varying[], unknown) does not exist
I have used position in a function before so I am not sure why PostgreSQL does not like it in this situation.
Is there a way to using a CASE expression whilst determining the position of a text element within an array.
Apparently your term column is defined as an array, e.g. varchar[]. The position function only works with scalar values, not with arrays.
If you want to test if an element is contained in an array you need to use a different operator: #>
update foobar
set cat = 'Boulder'
where term #> '{"Boulder"}'
The expression '{"Boulder"}' creates an array with a single element. It's equivalent to array['Boulder'] (which I find more readable). So the above where condition updates all rows where the array term contains all elements of the array on the right hand side of the operator. In this case it's only a single element you are testing for.
More details about the array functions and operators can be found in the manual: http://www.postgresql.org/docs/current/static/functions-array.html
Edit after the requirements have changed
To find and update only those where boulder is in the first, second or third place, you can use this:
update foobar
set cat = 'Boulder'
where 'Boulder' in (term[1], term[2], term[3]);