Home
Extending HCL OneDB™Beyond standard relational database objects, HCL OneDB™ can be extended to handle specialized data types, access methods, routines, and other objects. HCL OneDB™ includes many built-in extensions that are fully integrated in the database server. HCL OneDB™ also provides DataBlade® modules, which are packages of extended database objects for a particular purpose and that are installed separately from the database server. Alternatively, you can create your own user-defined objects for HCL OneDB™.
Creating extensionsYou can create user-defined data types, routines, access methods, and other database objects to suit your needs. You can use application programming interfaces to write user-defined routines and applications that access data in HCL OneDB™ databases.
DataBlade® API Function ReferenceThe HCL OneDB™ DataBlade® API Function Reference describes the DataBlade® API functions and the subset of functions that the DataBlade® API supports.
Function descriptionsThis section gives comprehensive reference information about the functions that are valid within DataBlade® API modules. It lists the function descriptions in alphabetical order, with syntax, usage information, and return values for the functions.
The mi_call() functionThe mi_call() function checks if there is sufficient stack space for the specified user-defined routine and extends the stack size if necessary.

The mi_call() function

The mi_call() function checks if there is sufficient stack space for the specified user-defined routine and extends the stack size if necessary.

Syntax

mi_integer mi_call(retval, routine, nargs, argument_list)
   mi_integer *retval;
   mi_integer (*routine)();
   mi_integer nargs;
   argument_list

retval: A pointer to the location of the user-defined-function return value.
routine: A pointer to the user-defined routine.
nargs: The number of the arguments (up to a maximum of 10) in the argument list.
argument_list: A comma-separated list of from 0 to 10 arguments to pass to the user-defined routine.


Valid in client LIBMI application?	Valid in user-defined routine?
No	Yes

Usage

The mi_call() function allocates stack space for the user-defined routine that routine references. The function determines the amount of memory to allocate based on the following information:

The return value, which retval references (for user-defined functions only)
The number of arguments, nargs (for all user-defined routines)

The mi_call() function creates a stack only if the current stack is not large enough to hold the arguments and possible return value of the UDR. If a new stack is required, mi_call() creates it and copies onto it the number of arguments that nargs specifies. Pass the actual argument values as a comma-separated list following the nargs value.

When mi_call() copies the arguments onto this new stack, the function uses the MI_DATUM size as the size of each argument. If a routine argument is larger than an MI_DATUM structure, mi_call() does not copy all the argument bytes. You must use the correct passing mechanism for routine arguments to ensure that they fit into the size of an MI_DATUM structure.

After arguments are copied to the stack, the mi_call() function calls the specified UDR and executes it. If the UDR is a user-defined function, mi_call() puts its return value at the address that retval references.

For more information, see the discussion of control modes for query data in the HCL OneDB™ DataBlade® API Programmer's Guide.

Return values

MI_NOMEM: The mi_call() function was unable to allocate virtual memory.
MI_TOOMANY: The nargs argument is greater than 10.
MI_CONTINUE: The current stack is large enough to hold the invocation of routine: no reallocation is required.
MI_DONE: The current stack was not large enough to hold the invocation of routine: mi_call() allocated a new thread stack, then called routine and put any return value in retval.