string matrix navigation with character string

Hi; for my problem, I have a string matrix (DataSour) of size 3 in columns but strong in rows; for which I look to obtain a minimized matrix (DataFind) relative to a character string (ToAssign). Then I try to find the positions of the columns (ColPos) and rows (RowPos) where the character string is positioned.

DataSour = ["Do" "Ri" "Mi"; "AA" "VV" "Fa"; "SS" "BB" "Do"; "Do" "yy" "zz"]
ToAssign = "Do"

for results I have to find the minimized matrix

DataFind =

Do -- --

-- -- Do

Do -- --

and positions of columns

ColPos =

1

3

1

and positions of rows

RowPos =

1

3

4

Answer 1

[Matt J]

DataSour = ["Do" "Ri" "Mi"; "AA" "VV" "Fa"; "SS" "BB" "Do"; "Do" "yy" "zz"];
ToAssign = "Do";

[ColPos,RowPos]=find(DataSour'==ToAssign)

ColPos = 3×1

1 3 1

RowPos = 3×1

1 3 4

[Touts Touts]

thanks, how i get the minimized matrix (DataFind) matrix of 3 columns with the (--) if the character string (ToAssign) not existe

DataFind =

Do -- --

-- -- Do

Do -- --

[Matt J]

DataSour = ["Do" "Ri" "Mi"; "AA" "VV" "Fa"; "SS" "BB" "Do"; "Do" "yy" "zz"];
ToAssign = "Do";

DataFind=DataSour;

DataFind(DataFind~=ToAssign)="--";

DataFind(all(DataFind=="--",2),:)=[]

DataFind = 3×3 string array
    "Do"    "--"    "--"
    "--"    "--"    "Do"
    "Do"    "--"    "--"

[Ferial Assmani]

@ All ; It looks like my concern; I try to import my Excel matrix and retrieve the rows that contain the desired character string and otherwise retrieve the rows that do not contain the desired character string. Why it’s not working for me I have the R2016a version.

[~, ~, MYFile] = xlsread('MYFile.xlsx','Data','A2:D5');
MYFile(cellfun(@(x) ~isempty(x) && isnumeric(x) && isnan(x),MYFile)) = {''};
DataSour = MYFile;
ToAssign = 'DO';

When I use

[RowPos, ColPos] = find(ismember(DataSour, ToAssign))

Error using cell/ismember (line 34)

Input A of class cell and input B of class char must be cell arrays of strings, unless one is a string.

When I use

[ColPos,RowPos]=find(DataSour'==ToAssign) 

Undefined operator '==' for input arguments of type 'cell'.

[Matt J]

You cannot use my solution unless you have a recent Matlab version. If you have upgraded, then convert DataSour to a string array, like in your posted example.

Answer 2

[Walter Roberson]

[RowPos, ColPos] = find(ismember(DataSour, ToAssign))

[Ferial Assmani]

@ All ; It looks like my concern; I try to import my Excel matrix and retrieve the rows that contain the desired character string and otherwise retrieve the rows that do not contain the desired character string. Why it’s not working for me I have the R2016a version.

[~, ~, MYFile] = xlsread('MYFile.xlsx','Data','A2:D5');
MYFile(cellfun(@(x) ~isempty(x) && isnumeric(x) && isnan(x),MYFile)) = {''};
DataSour = MYFile;
ToAssign = 'DO';

When I use

[RowPos, ColPos] = find(ismember(DataSour, ToAssign))

Error using cell/ismember (line 34)

Input A of class cell and input B of class char must be cell arrays of strings, unless one is a string.

When I use

[ColPos,RowPos]=find(DataSour'==ToAssign) 

Undefined operator '==' for input arguments of type 'cell'.

[Walter Roberson]

 DataSour = ["Do" "Ri" "Mi"; "AA" "VV" "Fa"; "SS" "BB" "Do"; "Do" "yy" "zz"]

That code is not compatible with the R2010b release that you marked yourself as using. The release is important for this purpose, as some of the tools most suitable were introduced after r2010b.

[Ferial Assmani]

@ Walter ; I have the R2016a version, but the error is in not in reading DataSour matrix ; is in find function, plz look :

[RowPos, ColPos] = find(ismember(DataSour, ToAssign))

or

[ColPos,RowPos]=find(DataSour'==ToAssign) 

thnks

[Walter Roberson]

The original posted code is incompatible with R2016a. string objects were introduced in r2016b and the use of double-quotes was introduced in r2017a.

[Walter Roberson]

MYFile(cellfun(@(x) ~isempty(x) && isnumeric(x) && isnan(x),MYFile)) = {''};

When you use the third output of xlsread then MATLAB examines every cell individually, and everything that looks like it might be a number gets converted to a number, and date-looking things might get converted to numeric too. If I recall correctly empty cells get converted into nan.

That cellfun you run converts nan to '' but does not convert any other numeric values. I would suggest to you two approaches:

1. Stop using xlsread. readtable() is more robust. In your older version using detectImportOptions can still be mostly advisable, possibly with some setvartypes() to be sure that you get the types right for columns that might be empty 2. If you must use xlsread() change the test in your cellfun to be just @(x)~ischar(x)

[Ferial Assmani]

@Walter Roberson, plz do you mean ?

[~, ~, MYFile] = xlsread('MYFile.xlsx','Data','A2:D5');
MYFile(cellfun(@(x) ~ischar(x) && isnumeric(x) && isnan(x),MYFile)) = {''};
DataSour = MYFile;
ToAssign = 'DO';
[RowPos1, ColPos1] = find(ismember(DataSour, ToAssign))
[ColPos2,RowPos2]=find(DataSour'==ToAssign)

[Ferial Assmani]

@Walter Roberson, plz, i got this error

Error using cell/ismember (line 34)

Input A of class cell and input B of class char must be cell arrays of strings, unless one is a string.

[Walter Roberson]

No, I mean

MYFile(cellfun(@(x) ~ischar(x), MYFile)) = {''};