sortrows - Sort rows of matrix or table - MATLAB (original) (raw)

Sort rows of matrix or table

Syntax

Description

Array Data

[B](#bt8bwzx-1-B) = sortrows([A](#bt8bwzx-1-A)) sorts the rows of an array based on the elements in the first column. By default, sortrows uses ascending sorted order. When the first column contains repeated elements, sortrows sorts according to the values in the next column and repeats this behavior for succeeding equal values.

example

[B](#bt8bwzx-1-B) = sortrows([A](#bt8bwzx-1-A),[column](#bt8bwzx-1-column)) specifies the columns by which to sort the rows. For example,sortrows(A,[4 6]) first sorts the rows ofA based on elements in the fourth column, then based on elements in the sixth column to break ties.

example

[B](#bt8bwzx-1-B) = sortrows([A](#bt8bwzx-1-A),___,[direction](#bt8bwzx-1-direction)) specifies the sorting direction for any of the previous syntaxes. For example,sortrows(A,'descend') uses descending sorting order.

example

[B](#bt8bwzx-1-B) = sortrows([A](#bt8bwzx-1-A),___,[Name,Value](#namevaluepairarguments)) specifies additional parameters for sorting rows. For example,sortrows(A,'ComparisonMethod','abs') sorts the elements in A by magnitude.

example

[[B](#bt8bwzx-1-B),[index](#bt8bwzx-1-index)] = sortrows([A](#bt8bwzx-1-A),___) also returns an index vector that describes the rearrangement of rows such that B = A(index,:).

example

Table Data

[tblB](#bt8bwzx-1-tblB) = sortrows([tblA](#bt8bwzx-1-tblA)) sorts the rows of a table or timetable.

If tblA is a table, thensortrows sorts tblA in ascending order based on the values in the first variable. If elements in the first variable are repeated, then sortrows sorts by the elements in the second variable, and so on.
If tblA is a timetable, thensortrows sorts the rows oftblA in ascending order based on its row times. However, the rows are sorted only with respect to the row times. If row times are repeated, then sortrows does not sort by the elements in the timetable variables.

example

[tblB](#bt8bwzx-1-tblB) = sortrows([tblA](#bt8bwzx-1-tblA),'RowNames') sorts a table based on its row names. Row names of a table label the rows along the first dimension of the table. If tblA does not have row names, that is, if tblA.Properties.RowNames is empty, thensortrows returns tblA.

This syntax is not supported when tblA is a timetable.

example

[tblB](#bt8bwzx-1-tblB) = sortrows([tblA](#bt8bwzx-1-tblA),[rowDimName](#bt8bwzx-1-rowDimName)) sorts based on row labels along the first dimension.

If tblA is a table, then row labels are row names.
If tblA is a timetable, then row labels are row times.

example

[tblB](#bt8bwzx-1-tblB) = sortrows([tblA](#bt8bwzx-1-tblA),[vars](#bt8bwzx-1-vars)) sorts based on the specified table variables. For example,sortrows(tblA,{'Var1','Var2'}) first sorts the rows oftblA based on the elements in Var1, then by the elements in Var2.

If tblA is a table and it has row names, thenvars can include the row names.
If tblA is a timetable, thenvars can include the row times.

example

[tblB](#bt8bwzx-1-tblB) = sortrows([tblA](#bt8bwzx-1-tblA),___,[direction](#bt8bwzx-1-direction)) sorts the rows in tblA in specified direction for any of the previous table syntaxes.

example

[tblB](#bt8bwzx-1-tblB) = sortrows([tblA](#bt8bwzx-1-tblA),___,[Name,Value](#namevaluepairarguments)) specifies additional parameters for sorting rows of a table or timetable. For example, sortrows(tblA,'Var1','MissingPlacement','first') sorts based on the elements in Var1, ordering missing elements such as NaN at the beginning of the table.

example

[[tblB](#bt8bwzx-1-tblB),[index](#bt8bwzx-1-index)] = sortrows([tblA](#bt8bwzx-1-tblA),___) also returns an index vector such that tblB = tblA(index,:).

example

Examples

collapse all

Create a matrix and sort its rows in ascending order based on the elements in the first column. When the first column contains repeated elements, sortrows looks to the elements in the second column to break the tie. For repeated elements in the second column, sortrows looks to the third column, and so on.

rng default; A = floor(rand([6 7])*100); A(1:4,1) = 95; A(5:6,1) = 76; A(2:4,2) = 7; A(3,3) = 48

A = 6×7

95    27    95    79    67    70    69
95     7    48    95    75     3    31
95     7    48    65    74    27    95
95     7    14     3    39     4     3
76    15    42    84    65     9    43
76    97    91    93    17    82    38

B = 6×7

76    15    42    84    65     9    43
76    97    91    93    17    82    38
95     7    14     3    39     4     3
95     7    48    65    74    27    95
95     7    48    95    75     3    31
95    27    95    79    67    70    69

Sort the rows of A based on the values in the second column. When the specified column has repeated elements, the corresponding rows maintain their original order.

C = 6×7

95     7    48    95    75     3    31
95     7    48    65    74    27    95
95     7    14     3    39     4     3
76    15    42    84    65     9    43
95    27    95    79    67    70    69
76    97    91    93    17    82    38

Sort the rows of A based on the elements in the first column, and look to the seventh column to break any ties.

D = 6×7

76    97    91    93    17    82    38
76    15    42    84    65     9    43
95     7    14     3    39     4     3
95     7    48    95    75     3    31
95    27    95    79    67    70    69
95     7    48    65    74    27    95

Sort the rows of A in descending order based on the elements in the fourth column, and display the output vector index to see how the rows were rearranged.

[E,index] = sortrows(A,4,'descend')

E = 6×7

95     7    48    95    75     3    31
76    97    91    93    17    82    38
76    15    42    84    65     9    43
95    27    95    79    67    70    69
95     7    48    65    74    27    95
95     7    14     3    39     4     3

Create a matrix containing complex numbers, and sort the rows of the matrix in ascending order based on the elements in the first column. Since the magnitudes of A(1,1) and A(3,1) are equal, sortrows computes their angles to break the tie.

A = [1+2i 3+i i; 2+10i 6i 2+5i; 2+i 4 3+3i]

A = 3×3 complex

1.0000 + 2.0000i 3.0000 + 1.0000i 0.0000 + 1.0000i 2.0000 +10.0000i 0.0000 + 6.0000i 2.0000 + 5.0000i 2.0000 + 1.0000i 4.0000 + 0.0000i 3.0000 + 3.0000i

B = 3×3 complex

2.0000 + 1.0000i 4.0000 + 0.0000i 3.0000 + 3.0000i 1.0000 + 2.0000i 3.0000 + 1.0000i 0.0000 + 1.0000i 2.0000 +10.0000i 0.0000 + 6.0000i 2.0000 + 5.0000i

Use the 'real' option to sort the rows of A by their real part. Since A(2,1) and A(3,1) have equal real parts, sortrows uses the imaginary part to break the tie.

C = sortrows(A,'ComparisonMethod','real')

C = 3×3 complex

1.0000 + 2.0000i 3.0000 + 1.0000i 0.0000 + 1.0000i 2.0000 + 1.0000i 4.0000 + 0.0000i 3.0000 + 3.0000i 2.0000 +10.0000i 0.0000 + 6.0000i 2.0000 + 5.0000i

Create a 6-by-2 string array. Sort its rows in ascending alphabetical order based on the elements in the first column. When the first column contains repeated elements, use the second column to break the tie.

A = ["B" "Y"; "B" "Z"; "A" "Z"; ... "B" "X"; "A" "Y"; "A" "X"]; B1 = sortrows(A)

B1 = 6×2 string "A" "X" "A" "Y" "A" "Z" "B" "X" "B" "Y" "B" "Z"

Sort the rows again. This time, break the tie for elements in the first column by sorting the elements in the second column in descending alphabetical order.

B2 = sortrows(A,[1 2],["ascend" "descend"])

B2 = 6×2 string "A" "Z" "A" "Y" "A" "X" "B" "Z" "B" "Y" "B" "X"

Sort the rows of a table by variable values.

Create a table with four variables listing patient information for five people.

LastName = {'Smith';'Johnson';'Williams';'Jones';'Brown'}; Age = [38;43;38;40;49]; Height = [71;69;64;67;64]; Weight = [176;163;131;133;119]; BloodPressure = [124 93; 109 77; 125 83; 117 75; 122 80];

tblA = table(Age,Height,Weight,BloodPressure,'RowNames',LastName)

tblA=5×4 table Age Height Weight BloodPressure ___ ______ ______ _____________

Smith       38       71       176       124     93  
Johnson     43       69       163       109     77  
Williams    38       64       131       125     83  
Jones       40       67       133       117     75  
Brown       49       64       119       122     80

Sort the rows of the table. The sortrows function sorts the rows in ascending order first by the variable Age, and then by the variable Height to break the tie between the two rows with equal ages.

tblB=5×4 table Age Height Weight BloodPressure ___ ______ ______ _____________

Williams    38       64       131       125     83  
Smith       38       71       176       124     93  
Jones       40       67       133       117     75  
Johnson     43       69       163       109     77  
Brown       49       64       119       122     80

Create a table with four variables listing patient information for five people.

LastName = {'Smith';'Johnson';'Williams';'Jones';'Brown'}; Age = [38;43;38;40;49]; Height = [71;69;64;67;64]; Weight = [176;163;131;133;119]; BloodPressure = [124 93; 109 77; 125 83; 117 75; 122 80];

tblA = table(Age,Height,Weight,BloodPressure,'RowNames',LastName)

tblA=5×4 table Age Height Weight BloodPressure ___ ______ ______ _____________

Smith       38       71       176       124     93  
Johnson     43       69       163       109     77  
Williams    38       64       131       125     83  
Jones       40       67       133       117     75  
Brown       49       64       119       122     80

Sort the rows of the table in ascending order based on the row names, and return the index vector that describes how the rows were rearranged.

[tblB,index] = sortrows(tblA,'RowNames')

tblB=5×4 table Age Height Weight BloodPressure ___ ______ ______ _____________

Brown       49       64       119       122     80  
Johnson     43       69       163       109     77  
Jones       40       67       133       117     75  
Smith       38       71       176       124     93  
Williams    38       64       131       125     83

Create a table with four variables listing patient information for five people.

LastName = {'Sweet';'Jacobson';'Wang';'Joiner';'Berger'}; Age = [38;43;38;40;49]; Height = [71;69;64;67;64]; Weight = [176;163;131;133;119]; BloodPressure = [124 93; 109 77; 125 83; 117 75; 122 80];

tblA = table(Age,Height,Weight,BloodPressure,'RowNames',LastName)

tblA=5×4 table Age Height Weight BloodPressure ___ ______ ______ _____________

Sweet       38       71       176       124     93  
Jacobson    43       69       163       109     77  
Wang        38       64       131       125     83  
Joiner      40       67       133       117     75  
Berger      49       64       119       122     80

Sort the rows of the table in ascending order by Height, and then in descending order by Weight.

tblB = sortrows(tblA,{'Height','Weight'},{'ascend','descend'})

tblB=5×4 table Age Height Weight BloodPressure ___ ______ ______ _____________

Wang        38       64       131       125     83  
Berger      49       64       119       122     80  
Joiner      40       67       133       117     75  
Jacobson    43       69       163       109     77  
Sweet       38       71       176       124     93

Create a table with four variables listing patient information for five people. The Weight variable contains missing values.

LastName = {'Sweet';'Jacobson';'Wang';'Joiner';'Berger'}; Age = [38;43;38;40;49]; Height = [71;69;64;67;64]; Weight = [176;NaN;131;133;NaN]; BloodPressure = [124 93; 109 77; 125 83; 117 75; 122 80]; tblA = table(Age,Height,Weight,BloodPressure,'RowNames',LastName)

tblA=5×4 table Age Height Weight BloodPressure ___ ______ ______ _____________

Sweet       38       71       176       124     93  
Jacobson    43       69       NaN       109     77  
Wang        38       64       131       125     83  
Joiner      40       67       133       117     75  
Berger      49       64       NaN       122     80

Sort the rows of the table in ascending order by Weight, placing the rows containing NaN first.

tblB = sortrows(tblA,'Weight','MissingPlacement','first')

tblB=5×4 table Age Height Weight BloodPressure ___ ______ ______ _____________

Jacobson    43       69       NaN       109     77  
Berger      49       64       NaN       122     80  
Wang        38       64       131       125     83  
Joiner      40       67       133       117     75  
Sweet       38       71       176       124     93

Create a timetable, and sort the rows by row times.

TimeDuration = [hours(3) hours(2) hours(1) hours(5) hours(6)]'; TT = timetable(TimeDuration,[98;97.5;97.9;98.1;101],[120;111;119;117;118]);

B = sortrows(TT,'TimeDuration')

B=5×2 timetable TimeDuration Var1 Var2 ____________ ____ ____

1 hr            97.9    119 
2 hr            97.5    111 
3 hr              98    120 
5 hr            98.1    117 
6 hr             101    118

Input Arguments

collapse all

Input array, specified as a column vector or matrix.

Column sorting vector, specified as a nonzero integer scalar or vector of nonzero integers. Each specified integer value indicates a column to sort by. Negative integers indicate that the sort order is descending.

Sorting direction, specified as either 'ascend','descend', or a string array or cell array of character vectors that specifies some combination of these values.

If direction is 'ascend' or'descend', then the sorting direction applies for all column sorting vectors or sorting variables. If direction is a string array or cell array of character vectors, then each entry corresponds to a column sorting vector or a sorting variable.

If you specify both column anddirection, then the direction determines the sorting direction and sortrows ignores the signs of the elements in column.

Example: sortrows(X,[2 3],{'ascend' 'descend'}) first sorts rows in ascending order according to column 2 of X. Then, rows with equal entries in column 2 are sorted in descending order according to column 3.

Input table, specified as a table or timetable. Each variable intblA must be a valid input tosort or sortrows.

Name of the first dimension of the input table or timetable, specified as a string scalar or character vector.

If tblA is a table with row names, thenrowDimName is the name of the first dimension of the table. By default, the name of the first dimension is"Row". Dimension names are a property of tables. You can access the dimension names oftblA usingtblA.Properties.DimensionNames.
If tblA is a timetable, thenrowDimName is the name of the vector of row times. You can specify its name when you create a timetable, such asTime or Date. You can also access the dimension names usingtblA.Properties.DimensionNames.

Example: If a table T has row names, and you changed the name of the first dimension using T.Properties.DimensionName{1} = "Name", then sortrows(T,"Name") sorts the table by row name.

Example: If a timetable TT has a time vector namedDate, then sortrows(TT,"Date") sorts the timetable on the dates and times that Date contains.

Data Types: string | char

Sorting variables, specified as a scalar integer, vector of integers, variable name, string array of variable names, cell array of variable names,pattern scalar, or logical vector. vars indicates the table variables to sort by.

If you do not specify the direction argument, the sign of the elements in vars determines the sorting direction of the variables. If an element is a positive integer, thensortrows sorts the corresponding variable intblA in ascending order. If an element is a negative integer, then sortrows sorts the corresponding variable in tblA in descending order. Otherwise, if you specify the direction argument, sortrows ignores the sign of vars and sorts the specified variables according to direction.

Example: sortrows(tblA,["Height","Weight"]) sorts the rows of tblA in ascending order, first by the variableHeight, then by the variableWeight to break ties.

Example: sortrows(tblA,"X" + wildcardPattern) sorts the rows of tblA are in ascending order by the table variables whose names begin with the letter "X", using a wildcard pattern to match the remaining letters in their names.

Example: sortrows(tblA,[1 4]) sorts by the first variable of tblA in ascending order, then sorts by the fourth variable to break ties.

Example: sortrows(TT,["Time","X"]) sorts the row times of timetable TT in ascending order first, then sorts by the table variable X to break ties.

Name-Value Arguments

collapse all

Specify optional pairs of arguments asName1=Value1,...,NameN=ValueN, where Name is the argument name and Value is the corresponding value. Name-value arguments must appear after other arguments, but the order of the pairs does not matter.

Before R2021a, use commas to separate each name and value, and enclose Name in quotes.

Example: sortrows(A,'MissingPlacement','last')

Placement of missing values (NaN,NaT, <undefined>, andmissing) specified as the comma-separated pair consisting of 'MissingPlacement' and one of the following:

'auto' — Missing elements are placed last for ascending order and first for descending order.
'first' — Missing elements are placed first.
'last' — Missing elements are placed last.

Element comparison method for numeric input, specified as'ComparisonMethod' and one of these values:

'auto' — Sort rows ofA by real(A) whenA is real, and sort byabs(A) when A is complex.
'real' — Sort rows ofA by real(A) whenA is real or complex. If a column ofA has elements with equal real parts, then use imag(A) to break ties.
'abs' — Sort rows ofA by abs(A) whenA is real or complex. If a column ofA has elements with equal magnitude, then use angle(A) in the interval (-π,π] to break ties.

Output Arguments

collapse all

Sorted array, returned as a column vector or matrix. B is the same size as A.

Sorted table, returned as a table or timetable with the same variables astblA.

Sort index, returned as a column vector. The sort index describes the rearrangement of the rows in the input such that B = A(index,:).

The sortrows function uses a stable sorting algorithm. So, when the input contains repeated values, the sort index preserves the original order from the input, regardless of sorting direction. For example, if A = [1 1; 2 2; 1 2; 2 2], then [Ba,Ia] = sortrows(A,'ascend') returns the sort index Ia = [1; 3; 2; 4] and [Bd,Id] = sortrows(A,'descend') returns the sort index Id = [2; 4; 3; 1].

Data Types: double

Extended Capabilities

expand all

Thesortrows function supports tall arrays with the following usage notes and limitations:

Sorting by row names is not supported.

For more information, see Tall Arrays.

Usage notes and limitations:

Cell or string input array is not supported.
If A is complex with all zero imaginary parts, then MATLAB® might convert A toreal(A) before callingsortrows(A). In this case, MATLAB sorts the rows of A byreal(A), but the generated code sorts the rows ofA by abs(A). To make the generated code match MATLAB, use sortrows(real(A)) orsortrows(A,'ComparisonMethod','real'). See Code Generation for Complex Data with Zero-Valued Imaginary Parts (MATLAB Coder).
If tblA is a table or timetable, then the input argument vars must be constant.
If tblA is a table or timetable, and it has a variable that is a cell array of character vectors with multiple columns, then you cannot sort tblA using the values in that variable.
The vars input argument does not support pattern expressions.

Usage notes and limitations:

The first input argument must not be a cell array.
If A is complex with all zero imaginary parts, then MATLAB might convert A toreal(A) before callingsortrows(A). In this case, MATLAB sorts the rows of A byreal(A), but the generated code sorts the rows ofA by abs(A). To make the generated code match MATLAB, use sortrows(real(A)) orsortrows(A,'ComparisonMethod','real').

The sortrows function supports GPU array input with these usage notes and limitations:

Sorting cell arrays is not supported.
Sparse inputs are not supported.

For more information, see Run MATLAB Functions on a GPU (Parallel Computing Toolbox).

Usage notes and limitations:

Sorting cell arrays is not supported.
Table, timetable, datetime and duration inputs are not supported.

For more information, see Run MATLAB Functions with Distributed Arrays (Parallel Computing Toolbox).

Version History

Introduced before R2006a