Reads a row or a column from an array into a vector of numbers or strings
[C++]
template< class T >
BOOL ReadVector(
std::vector<T>& vecResult,
const char* pszArgName,
CString& strError,
long* plUBound = 0,
long lRowOrCol = XLA_UNIQUE_VECTOR,
long lFlags = XLA_ARRAY_FLAGS_STD,
long lLBound = 0,
long lIgnoreItems = 0
) const;
template< class T >
BOOL ReadVector(
std::vector<T>& vecResult,
const char* pszArgName,
CXlOper& xloError,
long* plUBound = 0,
long lRowOrCol = XLA_UNIQUE_VECTOR,
long lFlags = XLA_ARRAY_FLAGS_STD,
long lLBound = 0,
long lIgnoreItems = 0
) const;
A vector containing items of type T, whose contents will be replaced by the contents of the CXlOper.
The following types are directly supported for T:
The name of the argument that is being read. This will be used in any error strings that are generated.
Reference to an error string buffer. If the function fails for any reason, a useful error string will be placed here, e.g. "MyArgName: expected number in cell 3".
Pointer to upper bound. If it is null, then it is ignored. If it is not null then it is used to check that the upper bounds of various arrays are consistent.
See Bounded input arrays for more information on the use of upper bounds.
Zero-based index of row or column to be extracted. If it is out of range, the function will fail.
If lRowOrCol is set to XLA_UNIQUE_VECTOR, then the only row of a single-row array (or the only column of a one-column array) will be returned: if the array has more than one row and more than one column, the function will report an error and return FALSE.
Flags controlling the conversion. See matrix & vector flags for a full list of flag values.
This will be used as a lower bound in the output vector. lLBound cells containing the default value will be inserted at the start of the output vector.
The first lIgnoreItems of input will be ignored. This option can be useful for excluding labels.
If any error occurs, a description will be put into xloError. It can then conveniently be returned as the result of the add-in function.
The function returns TRUE if the conversion was successful,
FALSE if it failed for any reason. If it fails, an explanatory error string will be written
to strError or xloError.
This should be returned to the user.
This function extracts a single row of numbers or strings from an array argument.
Using the default flag values, it is reasonably 'smart':
by default it truncates any empty cells at the end (right or bottom) of the input range. However, a wide variety of other options are available, if you use other flag values
and optional arguments.
See matrix & vector flags for a full list of flag values. See Bounded input arrays
for more information on the use of upper bounds.
If the CXlOper contains a reference to a range of cells, rather than
the values of the cells, then this function will fail, and will return a
message stating that the argument must be coerced before use. You should write code to Coerce the input to a value, e.g.:
extern "C" __declspec( dllexport )
LPXLOPER ReadXlOper(const CXlOper* Arg1)
{
CXlOper xloResult;
//}}XLP_SRC
CXlOper valArg1;
if (valArg1.Coerce(*Arg1) != 0)
return CXlOper::RetError(xlerrValue);
std::vector<bool> vecB;
if (!valArg1.ReadVector(vecB, "Arg1", strError))
return CXlOper::RetString(strError);
...
}
It is generally simpler and faster to use a COper type to handle an array input. You should only use CXlOper to handle an array input if you need a CXlOper feature, such as a heterogeneous input that can contain either a value or a reference (COper does not support references).
For more information on the use of CXlOper for inputs, see Comparison of COper, CXlOper and CXlArray types.
Header: xllplus.h
CXlOper Class
| CXlOper Methods
| Bounded input arrays
| CXlOper::ReadMatrix()