The MWFlags class contains a set of array formatting
and data conversion flags (See Rules for Data Conversion Between .NET and MATLAB for
more information on conversion between MATLAB® and COM Automation
types.) All MATLAB
Compiler SDK™ COM components contain a reference
to an MWFlags object that can modify data conversion
rules at the object level. This class contains these properties and
method:
The ArrayFormatFlags property controls array
formatting (as a matrix or a cell array) and the application of these
rules to nested arrays. The MWArrayFormatFlags class
is a noncreatable class accessed through an MWFlags class
instance. This class contains six properties:
This property of type mwArrayFormat controls
the formatting of arrays passed as input parameters to MATLAB
Compiler SDK class
methods. The default value is mwArrayFormatMatrix.
The behaviors indicated by this flag are listed in the next table.
Array Formatting Rules for Input Arrays
| Value | Behavior |
|---|---|
| Converts arrays according to the default conversion rules listed in Rules for Data Conversion Between .NET and MATLAB. |
| Coerces all arrays into cell arrays. Input scalar or numeric array arguments are converted to cell arrays with each cell containing a scalar value for the respective index. |
| Coerces all arrays into matrices. When an input argument
is encountered that is an array of |
This property governs the level at which to apply the rule set
by the InputArrayFormat property for nested arrays
(an array of Variants is passed and each element
of the array is an array itself). It is not necessary to modify this
flag for varargin parameters. The data conversion
code automatically increments the value of this flag by 1 for varargin cells,
thus applying the InputArrayFormat flag to each
cell of a varargin parameter. The default value
is 0.
This property of type mwArrayFormat controls
the formatting of arrays passed as output parameters to class methods.
The default value is mwArrayFormatAsIs. The behaviors
indicated by this flag are listed in the next table.
Array Formatting Rules for Output Arrays
| Value | Behavior |
|---|---|
| Converts arrays according to the default conversion rules listed in Rules for Data Conversion Between .NET and MATLAB. |
| Coerces all arrays into matrices. When an output cell
array argument is encountered (the default behavior converts it to
an array of |
| Coerces all output arrays into arrays of |
This property is similar to the InputArrayIndFalg property,
as it governs the level at which to apply the rule set by the OutputArrayFormat property
for nested arrays. As with the input case, this flag is automatically
incremented by 1 for a varargout parameter. The
default value of this flag is 0.
This flag applies to Excel® ranges only. When the target
output from a method call is a range of cells in an Excel worksheet,
and the output array size and shape is not known at the time of the
call, setting this flag to True instructs the data
conversion code to resize each Excel range to fit the output
array. Resizing is applied relative to the upper left corner of each
supplied range. The default value for this flag is False.
Setting this flag to True transposes the
output arguments. This flag is useful when processing an output parameter
from a method call on a COM component, where the MATLAB function
returns outputs as row vectors, and you desire to place the data into
columns. The default value for this flag is False.
The DataConversionFlags property controls
how input variables are processed when type coercion is needed. The MWDataConversionFlags class
is a noncreatable class accessed through an MWFlags class
instance. This class contains these properties:
This property converts all numeric input arguments to one specific MATLAB type.
This flag is useful is when variables maintained within the Visual Basic® code
are different types, e.g., Long, Integer,
etc., and all variables passed to the compiled MATLAB code must
be doubles. The default value for this property is mwTypeDefault,
which uses the default rules in Rules for Data Conversion Between .NET and MATLAB.
This property sets the date bias for performing COM to MATLAB numeric
date conversions. The default value of this property is 693960, representing
the difference between the COM Date type and MATLAB numeric
dates. This flag allows existing MATLAB code that already performs
the increment of numeric dates by 693960 to be used unchanged with
COM components. To process dates with such code, set this property
to 0.
This example uses data conversion flags to reshape the output from a method compiled from a MATLAB function that produces an output vector of unknown length.
function p = myprimes(n)
if length(n)~=1, error('N must be a scalar'); end
if n < 2, p = zeros(1,0); return, end
p = 1:2:n;
q = length(p);
p(1) = 2;
for k = 3:2:sqrt(n)
if p((k+1)/2)
p(((k*k+1)/2):k:q) = 0;
end
end
p = (p(p>0));
This function produces a row vector of all the prime numbers
between 0 and n. Assume that this function is included
in a class named myclass that is included in a
component named mycomponent with a version of 1.0.
The subroutine takes an Excel range and a Double as
inputs, and places the generated prime numbers into the supplied range.
The MATLAB function produces a row vector, although you want
the output in column format. It also produces an unknown number of
outputs, and you do not want to truncate any output. To handle these
issues, set the TransposeOutput flag and the AutoResizeOutput flag
to True. In previous examples, the Visual Basic CreateObject function
creates the necessary classes. This example uses an explicit type
declaration for the aClass variable. As with previous
examples, this function assumes that MWInitApplication has
been previously called.
Sub GenPrimes(R As Range, n As Double)
Dim aClass As mycomponent.myclass
On Error GoTo Handle_Error
Set aClass = New mycomponent.myclass
aClass.MWFlags.ArrayFormatFlags.AutoResizeOutput = True
aClass.MWFlags.ArrayFormatFlags.TransposeOutput = True
Call aClass.myprimes(1, R, n)
Exit Sub
Handle_Error:
MsgBox (Err.Description)
End Sub
This property converts dates passed as input parameters to method
calls on MATLAB
Compiler SDK classes. The default value is mwDateFormatNumeric.
The behaviors indicated by this flag are shown in the following table.
Conversion Rules for Input Dates
| Value | Behavior |
|---|---|
| Convert dates to numeric values as indicated by the rule listed in Rules for Data Conversion Between .NET and MATLAB. |
| Convert input dates to strings. |
This property processes an output argument as a date. By default,
numeric dates that are output parameters from compiled MATLAB functions
are passed as Doubles that need to be decremented
by the COM date bias (693960) as well as coerced to COM dates. Set
this flag to True to convert all output values
of type Double.
This property is an enumeration and can have two possible values: mwReplaceNaN and mwReplaceZero.
To treat empty cells referenced by input parameters as zeros,
set the value to mwReplaceZero. To treat empty
cells referenced by input parameters as NaNs (Not
a Number), set the value to mwReplaceNaN.
By default, the value is mwReplaceZero.
Creates a copy of an MWFlags object.
| Argument | Type | Description |
|---|---|---|
|
| Reference to an uninitialized |
None
Clone allocates a new MWFlags object and
creates a deep copy of the object's contents. Call this function when
a separate object is required instead of a shared copy of an existing
object reference.