Welcome
Extending Informix®
Beyond standard relational database objects, HCL Informix® can be extended to handle specialized data types, access methods, routines, and other objects. Informix includes many built-in extensions that are fully integrated in the database server. Informix also provides 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 Informix.
Creating extensions
You 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 Informix® databases.
Function Reference
The IBM® Informix® DataBlade® API Function Reference describes the functions and the subset of functions that the supports.
Function descriptions
This section gives comprehensive reference information about the functions that are valid within modules. It lists the function descriptions in alphabetical order, with syntax, usage information, and return values for the functions.
The mi_save_set_delete() function
The mi_save_set_delete() function removes a row from a save set.

HCL Informix® V15.0.0 documentation
Welcome to the documentation for HCL Informix® 15.0.0 and related client tools and products.
Product overview
HCL Informix® is a fast and scalable database server that manages traditional relational, object-relational, and dimensional databases. Its small footprint and self-managing capabilities are suited to embedded data-management solutions.
Installing
These topics describe how to install HCL Informix® database servers, client products, and modules.
Administering
In addition to administering the database server, you can tune performance, replicate data, and archive data.
Migrating and upgrading
You can upgrade to the 15.0.0 release of HCL Informix® or migrate from other database servers to Informix. Upgrading is an in-place migration method that uses your existing hardware and operating system software. Some changes to the Informix database server can affect upgrading from a previous release.
Client APIs and tools
You can use the HCL Informix® implementation of client APIs to develop applications for Informix database servers.
Embedding Informix®
When you embed HCL Informix®, you can use enterprise-class high-availability and high performance with embeddability features such as easy programmability, a small disk and memory footprint, and silent deployment.
Extending Informix®
Beyond standard relational database objects, HCL Informix® can be extended to handle specialized data types, access methods, routines, and other objects. Informix includes many built-in extensions that are fully integrated in the database server. Informix also provides 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 Informix.
- Informix® extensions and modules
  These topics describe how to use built-in database extensions and separately installed modules.
- Creating extensions
  You 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 Informix® databases.
  - Programmer's Guide
    The HCL® Informix® DataBlade® API Programmer's Guide describes information about the , the C-language application programming interface (API) provided with HCL Informix.
  - Function Reference
    The IBM® Informix® DataBlade® API Function Reference describes the functions and the subset of functions that the supports.
    - Categories of functions
      This section summarizes by categories the functions that the provides. It divides functions that are valid within modules into several functional categories and lists the functions within each category.
    - Function descriptions
      This section gives comprehensive reference information about the functions that are valid within modules. It lists the function descriptions in alphabetical order, with syntax, usage information, and return values for the functions.
      - The ax_reg() function
        The ax_reg() function allows modules or applications using user-defined routines (UDRs) to register XA-compliant, external data sources (also called resource managers) with the HCL Informix® transaction manager. The registration is dynamic and is applicable for the current transaction only. The must register participating data sources into each transaction.
      - The ax_unreg() function
        The ax_unreg() function allows modules or applications using user-defined routines (UDRs) to unregister previously registered XA-compliant, external data sources (also called resource managers) from transactions.
      - The biginttoint2() function
        The biginttoint2() function converts a BIGINT type number to an int2 type number.
      - The bigintcvint2() function
        The bigintcvint2() function converts an int2 type number to a BIGINT type number.
      - The biginttoint4() function
        The biginttoint4() function converts a BIGINT type number to an int4 type number.
      - The bigintcvint4() function
        The bigintcvint4() function converts an int4 type number to a BIGINT type number.
      - The biginttoasc() function
        The biginttoasc() function converts a BIGINT type value to a C char type value.
      - The bigintcvasc() function
        The bigintcvasc() function converts a C char type value to a BIGINT type number.
      - The bigintcvdec() function
        The bigintcvdec() function converts a decimal type number to a BIGINT type number.
      - The biginttodec() function
        The biginttodec() function converts a BIGINT type number to a decimal type number.
      - The biginttodbl() function
        The biginttodbl() function converts a BIGINT type number to a double type number.
      - The bigintcvdbl() function
        The bigintcvdbl() function converts a double type number to a BIGINT type number.
      - The biginttoflt() function
        The biginttoflt() function converts a BIGINT type number to a float type number.
      - The bigintcvflt() function
        The bigintcvflt() function converts a float type number to a BIGINT type number.
      - The bigintcvifx_int8() function
        The bigintcvifx_int8() function converts and int8 type number to a BIGINT type number.
      - The biginttoifx_int8() function
        The biginttoifx_int8() function converts a BIGINT type number to an int8 type number.
      - The bycmpr() function
        The bycmpr() function compares two groups of contiguous bytes for a given length. This function returns the result of the comparison.
      - The bycopy() function
        The bycopy() function copies a given number of bytes from one location to another.
      - The byfill() function
        The byfill() function fills a specified area with one character.
      - The byleng() function
        The byleng() function returns the number of significant characters in a string, not counting trailing blanks.
      - The decadd() function
        The decadd() function adds two decimal type values.
      - The deccmp() function
        The deccmp() function compares two decimal type numbers.
      - The deccopy() function
        The deccopy() function copies a value from one decimal structure to another.
      - The deccvasc() function
        The deccvasc() function converts a value held as printable characters in a C char type into a decimal type number.
      - The deccvdbl() function
        The deccvdbl() function converts a C double type number into a decimal type number.
      - The deccvint() function
        The deccvint() function converts a C int type number into a decimal type number.
      - The deccvlong() function
        The deccvlong() function converts an int4 type value into a decimal type value.
      - The decdiv() function
        The decdiv() function divides two decimal type values.
      - The dececvt() and decfcvt() functions
        The dececvt() and decfcvt() functions are analogous to the subroutines under ECVT(3) in the UNIX™ Programmer's Manual.
      - The decmul() function
        The decmul() function multiplies two decimal type values.
      - The decround() function
        The decround() function rounds a decimal type number to fractional digits.
      - The decsub() function
        The decsub() function subtracts two decimal type values.
      - The dectoasc() function
        The dectoasc() function converts a decimal type number to a C char type value.
      - The dectodbl() function
        The dectodbl() function converts a decimal type number into a C double type number.
      - The dectoint() function
        The dectoint() function converts a decimal type number into a C int type number.
      - The dectolong() function
        The dectolong() function converts a decimal type number into an int4 type number.
      - The dectrunc() function
        The dectrunc() function truncates a rounded decimal type number to fractional digits.
      - The dtaddinv() function
        The dtaddinv() function adds an interval value to a datetime value. The result is a datetime value.
      - The dtcurrent() function
        The dtcurrent() function assigns the current date and time to a datetime variable.
      - The dtcvasc() function
        The dtcvasc() function converts a string that conforms to ANSI SQL standard for a DATETIME value to a datetime value.
      - The dtcvfmtasc() function
        The dtcvfmtasc() function uses a formatting mask to convert a character string to a datetime value.
      - The dtextend() function
        The dtextend() function extends a datetime value to a different qualifier. Extending is the operation of adding or dropping fields of a DATETIME value to make it match a given qualifier.
      - The dtsub() function
        The dtsub() function subtracts one datetime value from another. The result is an interval value.
      - The dtsubinv() function
        The dtsubinv() function subtracts an interval value from a datetime value. The result is a datetime value.
      - The dttoasc() function
        The dttoasc() function converts the field values of a datetime variable to an ASCII string that conforms to ANSI SQL standards.
      - The dttofmtasc() function
        The dttofmtasc() function uses a formatting mask to convert a datetime variable to a character string.
      - The ifx_int8add() function
        The ifx_int8add() function adds two int8 type values.
      - The ifx_int8cmp() function
        The ifx_int8cmp() function compares two int8 type numbers.
      - The ifx_int8copy() function
        The ifx_int8copy() function copies one int8 structure to another.
      - The ifx_int8cvasc() function
        The ifx_int8cvasc() function converts a value held as printable characters in a C char type into an int8 type number.
      - The ifx_int8cvdbl() function
        The ifx_int8cvdbl() function converts a C double type number into an int8 type number.
      - The ifx_int8cvdec() function
        The ifx_int8cvdec() function converts a decimal type value into an int8 type value.
      - The ifx_int8cvflt() function
        The ifx_int8cvflt() function converts a C float type number into an int8 type number.
      - The ifx_int8cvint() function
        The ifx_int8cvint() function converts a C int type number into an int8 type number.
      - The ifx_int8cvlong() function
        The ifx_int8cvlong() function converts a C long type value into an int8 type value.
      - The ifx_int8div() function
        The ifx_int8div() function divides two int8 type values.
      - The ifx_int8mul() function
        The ifx_int8mul() function multiplies two int8 type values.
      - The ifx_int8sub() function
        The ifx_int8sub() function subtracts two int8 type values.
      - The ifx_int8toasc() function
        The ifx_int8toasc() function converts an int8 type number to a C char type value.
      - The ifx_int8todbl() function
        The ifx_int8todbl() function converts an int8 type number into a C double type number.
      - The ifx_int8todec() function
        The ifx_int8todec() function converts an int8 type number into a decimal type number.
      - The ifx_int8toflt() function
        The ifx_int8toflt() function converts an int8 type number into a C float type number.
      - The ifx_int8toint() function
        The ifx_int8toint() function converts an int8 type number into a C int type number.
      - The ifx_int8tolong() function
        The ifx_int8tolong() function converts an int8 type number into a C long type number.
      - The incvasc() function
        The incvasc() function converts a string that conforms to the ANSI SQL standard for an INTERVAL value to an interval value.
      - The incvfmtasc() function
        The incvfmtasc() function uses a formatting mask to convert a character string to an interval value.
      - The intoasc() function
        The intoasc() function converts the field values of an interval variable to an ASCII string that conforms to the ANSI SQL standard.
      - The intofmtasc() function
        The intofmtasc() function uses a formatting mask to convert an interval variable to a character string.
      - The invdivdbl() function
        The invdivdbl() function divides an interval value by a numeric value.
      - The invdivinv() function
        The invdivinv() function divides an interval value by another interval value.
      - The invextend() function
        The invextend() function copies an interval value under a different qualifier.
      - The invmuldbl() function
        The invmuldbl() function multiplies an interval value by a numeric value.
      - The ldchar() function
        The ldchar() function copies a fixed-length string into a null-terminated string and removes any trailing blanks.
      - The mi_alloc() function
        The mi_alloc() function allocates a block of user memory of a specified size and returns a pointer to that block.
      - The mi_binary_to_date() function
        The mi_binary_to_date() function creates a text (string) representation of a date from the internal (binary) DATE representation.
      - The mi_binary_to_datetime() function
        The mi_binary_to_datetime() function creates a text (string) representation of a date, time, or date and time value from the binary DATETIME representation.
      - The mi_binary_to_decimal() function
        The mi_binary_to_decimal() function creates a text (string) representation of a decimal value from the internal (binary) DECIMAL representation.
      - The mi_binary_to_money() function
        The mi_binary_to_money() function creates a text (string) representation of a monetary value from the internal (binary) MONEY representation.
      - The mi_binary_query() function
        The mi_binary_query() function reports whether the last SQL statement sent on a particular connection returns results in a binary representation.
      - 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.
      - The mi_call_on_vp() function
        The mi_call_on_vp() function enables you to switch execution to a specified virtual processor (VP) to execute a specified C function.
      - The mi_cast_get() function
        The mi_cast_get() function looks up a registered cast function by the type identifiers of its source and target data types and creates its function descriptor.
      - The mi_class_id() function
        The mi_class_id() function obtains the VP-class identifier for a specified virtual-processor (VP) class.
      - The mi_class_maxvps() function
        The mi_class_maxvps() function obtains the maximum number of virtual processors (VPs) in a VP class.
      - The mi_class_name() function
        The mi_class_name() function obtains the name of a virtual-processor (VP) class.
      - The mi_class_numvp() function
        The mi_class_numvp() function obtains the number of active virtual processors (VPs) in a VP class.
      - The mi_client() function
        The mi_client() function dynamically determines whether a module is running in the database server or as a client application.
      - The mi_client_locale() function
        The mi_client_locale() function returns the name of the client locale.
      - The mi_close() function
        The mi_close() function closes a connection.
      - The mi_close_statement() function
        The mi_close_statement() function closes an open cursor.
      - The mi_collection_card() function
        The mi_collection_card() function returns a count of the elements in a collection (that is, its cardinality).
      - The mi_collection_close() function
        The mi_collection_close() function closes a collection and frees the collection descriptor.
      - The mi_collection_copy() function
        The mi_collection_copy() function copies a collection to a new collection variable.
      - The mi_collection_create() function
        The mi_collection_create() function creates a collection.
      - The mi_collection_delete() function
        The mi_collection_delete() function deletes a single element of a collection.
      - The mi_collection_fetch() function
        The mi_collection_fetch() function fetches a single element from a collection.
      - The mi_collection_free() function
        The mi_collection_free() function frees a collection.
      - The mi_collection_insert() function
        The mi_collection_insert() function inserts a single element into a collection.
      - The mi_collection_open() function
        The mi_collection_open() function opens a collection.
      - The mi_collection_open_with_options() function
        The mi_collection_open_with_options() function opens a collection in a specified open mode.
      - The mi_collection_update() function
        The mi_collection_update() function updates a collection.
      - The mi_column_count() function
        The mi_column_count() function obtains the number of columns in a row descriptor.
      - The mi_column_default() function
        The mi_column_default() function retrieves the default value of the specified column.
      - The mi_column_default_string() function
        The mi_column_default_string() function retrieves the default value of the specified column in string format.
      - The mi_column_id() function
        The mi_column_id() function obtains the column identifier of a specified column from a row descriptor.
      - The mi_column_name() function
        The mi_column_name() function obtains the name of a specified column from a row descriptor.
      - The mi_column_nullable() function
        The mi_column_nullable() function indicates whether a specified column in a row descriptor can contain SQL NULL values.
      - The mi_column_precision() function
        The mi_column_precision() function obtains the precision of the specified column from a row descriptor.
      - The mi_column_scale() function
        The mi_column_scale() function obtains the scale of the specified column from a row descriptor.
      - The mi_column_type_id() function
        The mi_column_type_id() function obtains the type identifier of the specified column from a row descriptor.
      - The mi_column_typedesc() function
        The mi_column_typedesc() function obtains the type descriptor of the specified column from a row descriptor.
      - The mi_command_is_finished() function
        The mi_command_is_finished() function reports whether the current statement has finished executing.
      - The mi_current_command_name() function
        The mi_current_command_name() function returns the name of the SQL statement that invoked the C UDR.
      - The mi_dalloc() function
        The mi_dalloc() function allocates the specified amount of memory for the specified memory duration and returns a pointer to the allocated block.
      - The mi_date_to_binary() function
        The mi_date_to_binary() function converts a text (string) representation of a date value to its binary (internal) DATE representation.
      - The mi_date_to_string() function
        The mi_date_to_string() function creates a text (string) representation of a date value from the binary (internal) DATE representation.
      - The mi_datetime_compare() function
        The mi_datetime_compare() function compares two binary (internal) DATETIME values and returns an integer value that indicates whether the first value is before, equal to, or after the second value.
      - The mi_datetime_to_binary() function
        The mi_datetime_to_binary() function converts a text (string) representation of a date, time, or date and time value to its binary (internal) DATETIME representation.
      - The mi_datetime_to_string() function
        The mi_datetime_to_string() function creates an ANSI SQL standard text (string) representation of a date, time, or date and time value from the binary (internal) DATETIME representation.
      - The mi_db_error_raise() function
        The mi_db_error_raise() function raises an error or warning and sends the message to the calling program.
      - The mi_dbcreate() function
        The mi_dbcreate() function creates a new database on the given connection.
      - The mi_dbdrop() function
        The mi_dbdrop() function drops a database from the given connection.
      - The mi_decimal_to_binary() function
        The mi_decimal_to_binary() function converts a text (string) representation of a decimal value to its binary (internal) DECIMAL representation.
      - The mi_decimal_to_string() function
        The mi_decimal_to_string() function creates a text (string) representation of a decimal value from the binary (internal) DECIMAL representation.
      - The mi_default_callback() function
        The mi_default_callback() function is the system-default callback for all callbacks in a client LIBMI application.
      - The mi_disable_callback() function
        The mi_disable_callback() function disables a callback for a single event or for all events.
      - The mi_drop_prepared_statement() function
        The mi_drop_prepared_statement() function drops a previously prepared statement.
      - The mi_enable_callback() function
        The mi_enable_callback() function enables a callback for a specified event type.
      - The mi_errmsg() function
        The mi_errmsg() function retrieves an error or warning message from an error descriptor into a user-allocated buffer.
      - The mi_error_desc_copy() function
        The mi_error_desc_copy() function returns a copy of a specified error descriptor.
      - The mi_error_desc_destroy() function
        The mi_error_desc_destroy() function frees an error descriptor that the mi_error_desc_copy() function allocated.
      - The mi_error_desc_finish() function
        The mi_error_desc_finish() function completes processing of the current exception list.
      - The mi_error_desc_is_copy() function
        The mi_error_desc_is_copy() function determines whether the specified error descriptor is a user copy.
      - The mi_error_desc_next() function
        The mi_error_desc_next() function gets the next error descriptor from the list of current exceptions that are associated with the current SQL statement.
      - The mi_error_level() function
        The mi_error_level() function retrieves from an error descriptor the exception level associated with an exception or an error level associated with a client LIBMI error.
      - The mi_error_sql_state() function
        The mi_error_sql_state() function retrieves the value of the SQLSTATE status variable from an error descriptor.
      - The mi_error_sqlcode() function
        The mi_error_sqlcode() function retrieves the value of the HCL Informix® SQLCODE status variable from an error descriptor.
      - The mi_exec() function
        The mi_exec() function sends an SQL statement to the database server for parsing, optimization, and execution.
      - The mi_exec_prepared_statement() function
        The mi_exec_prepared_statement() function sends a prepared statement to the database server for execution.
      - The mi_fetch_statement() function
        The mi_fetch_statement() function fetches specified rows from the database server into a cursor that is associated with an opened prepared statement.
      - The mi_file_allocate() function
        The mi_file_allocate() function ensures that a specified number of files are available to be opened.
      - The mi_file_close() function
        The mi_file_close() function closes an operating-system file.
      - The mi_file_errno() function
        The mi_file_errno() function returns the value of the system errno variable after a file input/output (I/O) operation. This value is the last errno value generated during an mi_file* function call and comes from the computer where the file is located. This function does not translate the value from one platform to another.
      - The mi_file_open() function
        The mi_file_open() function opens an operating-system file.
      - The mi_file_read() function
        The mi_file_read() function reads a specified number of bytes from an open operating-system file into a buffer.
      - The mi_file_seek() function
        The mi_file_seek() function sets the file seek position for the next read or write operation on the open file.
      - The mi_file_seek8() function
        The mi_file_seek8() function sets the file seek position for the next read or write operation on an open file of length greater that 2 GB.
      - The mi_file_sync() function
        The mi_file_sync() function forces a write to disk of all pages in an operating-system file.
      - The mi_file_tell() function
        The mi_file_tell() function returns the current file seek position for an operating-system file, relative to the beginning of the file.
      - The mi_file_tell8() function
        The mi_file_tell8() function returns the current file seek position, relative to the beginning of the file, for an operating-system file of length greater than 2 GB.
      - The mi_file_to_file() function
        The mi_file_to_file() function copies files between the database server and a client computer.
      - The mi_file_unlink() function
        The mi_file_unlink() function unlinks (removes) a file that mi_file_open() previously opened.
      - The mi_file_write() function
        The mi_file_write() function writes a specified number of bytes to an open operating-system file.
      - The mi_fix_integer() function
        The mi_fix_integer() function converts the specified 4-byte integer to or from the byte order of the client computer.
      - The mi_fix_smallint() function
        The mi_fix_smallint() function converts the specified 2-byte integer to or from the byte order of the client computer.
      - The mi_fp_argisnull() function
        The mi_fp_argisnull() accessor function determines whether the argument of a user-defined routine is an SQL NULL value from its associated MI_FPARAM structure.
      - The mi_fp_arglen() function
        The mi_fp_arglen() accessor function obtains the length of an argument of a user-defined routine from its associated MI_FPARAM structure.
      - The mi_fp_argprec() function
        The mi_fp_argprec() accessor function obtains the precision of an argument of a user-defined routine from its associated MI_FPARAM structure.
      - The mi_fp_argscale() function
        The mi_fp_argscale() accessor function obtains the scale of an argument of a user-defined routine from its associated MI_FPARAM structure.
      - The mi_fp_argtype() function
        The mi_fp_argtype() accessor function obtains the type identifier for the data type of an argument of a user-defined routine from the argument's associated MI_FPARAM structure.
      - The mi_fp_funcname() function
        The mi_fp_funcname() function obtains the name of a user-defined routine (UDR) using its associated MI_FPARAM structure.
      - The mi_fp_funcstate() function
        The mi_fp_funcstate() accessor function obtains user-state information for the user-defined routine from its associated MI_FPARAM structure.
      - The mi_fp_getcolid() function
        The mi_fp_getcolid() accessor function obtains the column identifier of the column that is associated with the user-defined routine from its MI_FPARAM structure.
      - The mi_fp_getfuncid() function
        The mi_fp_getfuncid() accessor function obtains the routine identifier for a user-defined routine (UDR) in its associated MI_FPARAM structure.
      - The mi_fp_getrow() function
        The mi_fp_getrow() accessor function obtains the row structure that is associated with the user-defined routine from its MI_FPARAM structure.
      - The mi_fp_nargs() function
        The mi_fp_nargs() accessor function obtains the number of arguments for the UDR routine from its associated MI_FPARAM structure.
      - The mi_fp_nrets() function
        The mi_fp_nrets() accessor function obtains the number of return values for the UDR from its associated MI_FPARAM structure.
      - The mi_fp_request() function
        The mi_fp_request() accessor function obtains the iterator status for an iterator function from an associated MI_FPARAM structure.
      - The mi_fp_retlen() function
        The mi_fp_retlen() accessor function obtains the length of a return value of a user-defined function from its associated MI_FPARAM structure.
      - The mi_fp_retprec() function
        The mi_fp_retprec() accessor function obtains the precision of a return value of a user-defined function from its associated MI_FPARAM structure.
      - The mi_fp_retscale() function
        The mi_fp_retscale() accessor function obtains the scale of a return value of a user-defined function from its associated MI_FPARAM structure.
      - The mi_fp_rettype() function
        The mi_fp_rettype() accessor function obtains the type identifier for the data type of a return value of a user-defined function from its associated MI_FPARAM structure.
      - The mi_fp_returnisnull() function
        The mi_fp_returnisnull() accessor function determines whether the return value of a user-defined function is an SQL NULL from its associated MI_FPARAM structure.
      - The mi_fp_setargisnull() function
        The mi_fp_setargisnull() accessor function sets the value of an argument of a user-defined routine to an SQL NULL in its associated MI_FPARAM structure.
      - The mi_fp_setarglen() function
        The mi_fp_setarglen() accessor function sets the length of an argument of a user-defined routine in its associated MI_FPARAM structure.
      - The mi_fp_setargprec() function
        The mi_fp_setargprec() accessor function sets the precision of a fixed-point or floating-point argument of a user-defined routine in its associated MI_FPARAM structure.
      - The mi_fp_setargscale() function
        The mi_fp_setargscale() accessor function sets the scale of an argument of a user-defined routine in its associated MI_FPARAM structure.
      - The mi_fp_setargtype() function
        The mi_fp_setargtype() accessor routine sets the type identifier for the data type of an argument of a user-defined routine in its associated MI_FPARAM structure.
      - The mi_fp_setcolid() function
        The mi_fp_setcolid() accessor function sets the column identifier of the column that is associated with the user-defined routine from its MI_FPARAM structure.
      - The mi_fp_setfuncid() function
        The mi_fp_setfuncid() accessor function sets the routine identifier for a user-defined routine in its associated MI_FPARAM structure.
      - The mi_fp_setfuncstate() function
        The mi_fp_setfuncstate() accessor function sets the user-state pointer for the user-defined routine in its associated MI_FPARAM structure.
      - The mi_fp_setisdone() function
        The mi_fp_setisdone() accessor function sets the iterator-completion flag for an iterator function in its associated MI_FPARAM structure.
      - The mi_fp_setnargs() function
        The mi_fp_setnargs() accessor function sets the number of arguments for the user-defined routine in its associated MI_FPARAM structure.
      - The mi_fp_setnrets() function
        The mi_fp_setnrets() accessor function sets the number of return values for the user-defined function in its associated MI_FPARAM structure.
      - The mi_fp_setretlen() function
        The mi_fp_setretlen() accessor function sets the length of a return value of a user-defined function from its associated MI_FPARAM structure.
      - The mi_fp_setretprec() function
        The mi_fp_setretprec() accessor function sets the precision of a return value of a user-defined function in its associated MI_FPARAM structure.
      - The mi_fp_setretscale() function
        The mi_fp_setretscale() accessor function sets the scale of a return value of a user-defined function in its associated MI_FPARAM structure.
      - The mi_fp_setrettype() function
        The mi_fp_setrettype() accessor function sets the type identifier for the data type of a return value of a user-defined function in its associated MI_FPARAM structure.
      - The mi_fp_setreturnisnull() function
        The mi_fp_setreturnisnull() accessor function sets the value of a return value of a user-defined function to an SQL NULL in its associated MI_FPARAM structure.
      - The mi_fp_setrow() function
        The mi_fp_setrow() accessor function sets the row structure that is associated with the user-defined routine in its associated MI_FPARAM structure.
      - The mi_fp_usr_fparam() function
        The mi_fp_usr_fparam() accessor function determines whether the database server or the developer has allocated the specified MI_FPARAM structure.
      - The mi_fparam_allocate() function
        The mi_fparam_allocate() function allocates an MI_FPARAM structure and returns a pointer to this structure.
      - The mi_fparam_copy() function
        The mi_fparam_copy() function copies an existing MI_FPARAM structure to a newly allocated MI_FPARAM structure field by field.
      - The mi_fparam_free() function
        The mi_fparam_free() function deallocates resources used by an MI_FPARAM structure allocated on behalf of a user-defined routine.
      - The mi_fparam_get() function
        The mi_fparam_get() function retrieves a pointer to the MI_FPARAM structure that is associated with the function descriptor of a user-defined routine.
      - The mi_fparam_get_current() function
        The mi_fparam_get_current() function obtains the MI_FPARAM structure for the currently running user-defined routine (UDR).
      - The mi_free() function
        The mi_free() routine frees the user memory that was previously allocated with the mi_alloc(), mi_dalloc(), or mi_zalloc() function.
      - The mi_func_commutator() function
        The mi_func_commutator() function obtains the name of a commutator function for a user-defined function.
      - The mi_func_desc_by_typeid() function
        The mi_func_desc_by_typeid() function looks up a registered user-defined routine by its routine identifier and creates its function descriptor.
      - The mi_func_handlesnulls() function
        The mi_func_handlesnulls() function determines whether a user-defined routine (UDR) can handle SQL NULL values.
      - The mi_func_isvariant() function
        The mi_func_isvariant() function determines whether a user-defined function is variant.
      - The mi_func_negator() function
        The mi_func_negator() function obtains the name of a negator function for a user-defined function.
      - The mi_funcarg_get_argtype() function
        The mi_funcarg_get_argtype() function returns the argument type for the companion-UDR argument of a cost or selectivity function.
      - The mi_funcarg_get_colno() function
        The mi_funcarg_get_colno() function returns the column number for the column associated with the companion-UDR argument of a cost or selectivity function.
      - The mi_funcarg_get_constant() function
        The mi_funcarg_get_constant() function returns the constant value of a companion-UDR argument of a cost or selectivity function.
      - The mi_funcarg_get_datalen() function
        The mi_funcarg_get_datalen() function returns the data length of an argument for the companion UDR of a cost or selectivity function.
      - The mi_funcarg_get_datatype() function
        The mi_funcarg_get_datatype() function returns the type identifier for the data type of an argument for the companion UDR of a cost or selectivity function.
      - The mi_funcarg_get_distrib() function
        The mi_funcarg_get_distrib() function returns the data-distribution information for the column associated with the companion-UDR argument of a cost or selectivity function.
      - The mi_funcarg_get_routine_id() function
        The mi_funcarg_get_routine_id() function returns the routine identifier for the companion UDR of a cost or selectivity function.
      - The mi_funcarg_get_routine_name() function
        The mi_funcarg_get_routine_name() function returns the routine name for the companion UDR of a cost or selectivity function.
      - The mi_funcarg_get_tabid() function
        The mi_funcarg_get_tabid() function returns the table identifier for the column associated with the companion-UDR argument of a cost or selectivity function.
      - The mi_funcarg_isnull() function
        The mi_funcarg_isnull() function determines whether the companion-UDR argument of a cost or selectivity function contains the SQL NULL value.
      - The mi_get_bigint() function
        The mi_get_bigint() function copies an mi_bigint (BIGINT) value, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_bytes() function
        The mi_get_bytes() function copies the given number of bytes, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_connection_info() function
        The mi_get_connection_info() function populates a connection-information descriptor with current connection parameters for an open connection.
      - The mi_get_connection_option() function
        The mi_get_connection_option() function returns information about the database of the current connection.
      - The mi_get_connection_user_data() function
        The mi_get_connection_user_data() function obtains the address of user data associated with an open connection.
      - The mi_get_cursor_table() function
        The mi_get_cursor_table() function obtains the name of the database table that is associated with a specified cursor.
      - The mi_get_database_info() function
        The mi_get_database_info() function populates a database-information descriptor with current database parameters for an open connection.
      - The mi_get_date() function
        The mi_get_date() function copies an mi_date (DATE) value, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_datetime() function
        The mi_get_datetime() function copies an mi_datetime (DATETIME) value, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_db_locale() function
        The mi_get_db_locale() function returns the value of the current database locale.
      - The mi_get_dbnames() function
        The mi_get_dbnames() function retrieves the names of all databases available on the database server, corresponding to the logged-in connection.
      - The mi_get_decimal() function
        The mi_get_decimal() function copies an mi_decimal (DECIMAL) value, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_default_connection_info() function
        The mi_get_default_connection_info() function populates a connection-information descriptor with the default connection parameters.
      - The mi_get_default_database_info() function
        The mi_get_default_database_info() function populates a database-information descriptor with the default database parameters.
      - The mi_get_double_precision() function
        The mi_get_double_precision() function copies an mi_double_precision (FLOAT) value, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_duration_size() function
        The mi_get_duration_size() function returns the TOTAL pool size of the specified memory duration.
      - The mi_get_id() function
        The mi_get_id() function enables you to obtain statement identifiers or session identifiers.
      - The mi_get_int8() function
        The mi_get_int8() function copies an mi_int8 (INT8) value, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_integer() function
        The mi_get_integer() function copies an mi_integer (INTEGER) value, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_interval() function
        The mi_get_interval() function copies an mi_interval (INTERVAL) value, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_lo_handle() function
        The mi_get_lo_handle() function copies an LO handle, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_memptr_duration() function
        The mi_get_memptr_duration() function returns the memory duration of a specified pointer.
      - The mi_get_money() function
        The mi_get_money() function copies an mi_money (MONEY) value, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_next_sysname() function
        The mi_get_next_sysname() function retrieves the next database server name.
      - The mi_get_parameter_info() function
        The mi_get_parameter_info() function populates a parameter-information descriptor with the current session parameters.
      - The mi_get_real() function
        The mi_get_real() function copies an mi_real (SMALLFLOAT) value, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_result() function
        The mi_get_result() function returns the status of the current statement.
      - The mi_get_row_desc() function
        The mi_get_row_desc() function obtains a pointer to the row descriptor for the specified row.
      - The mi_get_row_desc_from_type_desc() function
        The mi_get_row_desc_from_type_desc() function obtains a row descriptor that is associated with the specified type descriptor.
      - The mi_get_row_desc_without_row() function
        The mi_get_row_desc_without_row() function obtains the row descriptor for the current statement.
      - The mi_get_serverenv() function
        The mi_get_serverenv() function obtains information from the server environment.
      - The mi_get_session_connection() function
        The mi_get_session_connection() function obtains a session-duration connection descriptor.
      - The mi_get_smallint() function
        The mi_get_smallint() function copies an mi_smallint (SMALLINT) value, converting any difference in alignment or byte order on the client computer to that of the server computer.
      - The mi_get_statement_row_desc() function
        The mi_get_statement_row_desc() function obtains the row descriptor for a prepared statement.
      - The mi_get_string() function
        The mi_get_string() function copies an mi_string (CHAR(x)) value from a buffer.
      - The mi_get_type_source_type() function
        The mi_get_type_source_type() function obtains a type descriptor for the source of a distinct type.
      - The mi_get_transaction_id() function
        The mi_get_transaction_id() function obtains the current internal transaction ID.
      - The mi_get_vardata() function
        The mi_get_vardata() accessor function returns a pointer to the data contained in a varying-length structure (such as mi_lvarchar).
      - The mi_get_vardata_align() function
        The mi_get_vardata_align() accessor function obtains a pointer to the data in a varying-length structure (such as mi_lvarchar) and adjusts for any initial padding required to align the data.
      - The mi_get_varlen() function
        The mi_get_varlen() accessor function returns the length of the data stored in a varying-length structure (such as mi_lvarchar).
      - The mi_hdr_status() function
        The mi_hdr_status() function returns the high-availability cluster replication status of the current server.
      - The mi_init_library() function
        The mi_init_library() function initializes the library.
      - The mi_interrupt_check() function
        The mi_interrupt_check() function checks for a user interrupt.
      - The mi_interval_compare() function
        The mi_interval_compare() function compares two binary (internal) INTERVAL values and returns an integer value that indicates whether the first value is before, equal to, or after the second value.
      - The mi_interval_to_string() function
        The mi_interval_to_string() function creates an ANSI SQL standards text (string) representation of an interval value from its binary (internal) INTERVAL representation.
      - The mi_issmall_data() function
        The mi_issmall_data() macro determines whether storage of data is or is not in a smart large object.
      - The mi_last_serial() function
        The mi_last_serial() function returns a SERIAL value that the database server has generated for the most recent INSERT statement on a SERIAL column.
      - The mi_last_serial8() function
        The mi_last_serial8() function obtains the SERIAL8 value that the database server generated for the most recent INSERT statement on a SERIAL8 column.
      - The mi_library_version() function
        The mi_library_version() function returns the version and version date of the currently being used.
      - The mi_lo_alter() function
        The mi_lo_alter() function alters the storage characteristics of an existing smart large object.
      - The mi_lo_close() function
        The mi_lo_close() function closes an open smart large object.
      - The mi_lo_colinfo_by_ids() function
        The mi_lo_colinfo_by_ids() function sets the fields of an LO-specification structure to the column-level storage characteristics for specified row descriptor and column identifier.
      - The mi_lo_colinfo_by_name() function
        The mi_lo_colinfo_by_name() function sets the fields of an LO-specification structure to the column-level storage characteristics for a specified database column.
      - The mi_lo_copy() function
        The mi_lo_copy() function creates a copy of a smart large object and opens the copy.
      - The mi_lo_create() function
        The mi_lo_create() function creates a new smart large object and opens it for access within a module.
      - The mi_lo_decrefcount() function
        The mi_lo_decrefcount() function decrements the reference count of a smart large object.
      - The mi_lo_delete_immediate() function
        The mi_lo_delete_immediate() function deletes a smart large object.
      - The mi_lo_expand() function
        The mi_lo_expand() function copies multirepresentational data to a new smart large object.
      - The mi_lo_filename() function
        The mi_lo_filename() function constructs a file name for smart-large-object data based on an LO handle and a file name specification.
      - The mi_lo_from_buffer() function
        The mi_lo_from_buffer() function copies a specified number of bytes from a user-defined buffer to an existing smart large object.
      - The mi_lo_from_file() function
        The mi_lo_from_file() function copies the contents of an operating-system file on the server or client computer to a new smart large object.
      - The mi_lo_from_file_by_lofd() function
        The mi_lo_from_file_by_lofd() function copies the contents of an operating-system file on the server or client computer to an open smart large object.
      - The mi_lo_from_string() function
        The mi_lo_from_string() function converts an LO handle in its text representation to its binary representation.
      - The mi_lo_increfcount() function
        The mi_lo_increfcount() function increments the reference count of a smart large object.
      - The mi_lo_invalidate() function
        The mi_lo_invalidate() function marks the LO handle as invalid.
      - The mi_lo_lock() function
        The mi_lo_lock() function obtains a byte-range lock on the specified number of bytes in a smart large object.
      - The mi_lo_lolist_create() function
        The mi_lo_lolist_create() function converts an array of LO handles into an MI_LO_LIST structure.
      - The mi_lo_open() function
        The mi_lo_open() function opens an existing smart large object for access.
      - The mi_lo_ptr_cmp() function
        The mi_lo_ptr_cmp() function compares two LO handles to determine if they reference the same smart large object.
      - The mi_lo_read() function
        The mi_lo_read() function reads a specified number of bytes of data from an open smart large object into a buffer.
      - The mi_lo_readwithseek() function
        The mi_lo_readwithseek() function performs a seek operation and then reads a specified number of bytes of data from an open smart large object.
      - The mi_lo_release() function
        The mi_lo_release() function tells the database server that the resources that are associated with a temporary smart large object can be released.
      - The mi_lo_seek() function
        The mi_lo_seek() function sets the LO seek position for the next read or write operation on an open smart large object.
      - The mi_lo_spec_free() function
        The mi_lo_spec_free() function frees an LO-specification structure.
      - The mi_lo_spec_init() function
        The mi_lo_spec_init() function allocates and initializes the default system storage characteristics that are used to create a smart large object.
      - The mi_lo_specget_def_open_flags() function
        The mi_lo_specget_def_open_flags() function obtains from an LO-specification structure the default open-mode flag for a smart large object.
      - The mi_lo_specget_estbytes() function
        The mi_lo_specget_estbytes() function obtains from an LO-specification structure the estimated size, in bytes, of a smart large object.
      - The mi_lo_specget_extsz() function
        The mi_lo_specget_extsz() function obtains from an LO-specification structure the allocation extent size, in kilobytes, of a smart large object.
      - The mi_lo_specget_flags() function
        The mi_lo_specget_flags() function obtains from an LO-specification structure the attributes flag for a smart large object.
      - The mi_lo_specget_maxbytes() function
        The mi_lo_specget_maxbytes() accessor function obtains the maximum size of a smart large object (in bytes) from an LO-specification structure.
      - The mi_lo_specget_sbspace() function
        The mi_lo_specget_sbspace() function obtains from an LO-specification structure the name of an sbspace where a smart large object is stored.
      - The mi_lo_specset_def_open_flags() function
        The mi_lo_specset_def_open_flags() function sets the default open-mode flag for a smart large object.
      - The mi_lo_specset_estbytes() function
        The mi_lo_specset_estbytes() function sets the estimated size of a smart large object.
      - The mi_lo_specset_extsz() function
        The mi_lo_specset_extsz() function sets the allocation extent size for a smart large object.
      - The mi_lo_specset_flags() function
        The mi_lo_specset_flags() function sets the attributes flag for a smart large object.
      - The mi_lo_specset_maxbytes() function
        The mi_lo_specset_maxbytes() function sets the maximum number of bytes allowed for a smart large object.
      - The mi_lo_specset_sbspace() function
        The mi_lo_specset_sbspace() function sets the sbspace name of a smart large object in an LO-specification structure.
      - The mi_lo_stat() function
        The mi_lo_stat() function puts information about the current status of an open smart large object into an LO-status structure.
      - The mi_lo_stat_atime() function
        The mi_lo_stat_atime() function returns from an LO-status structure the last-access time for a smart large object.
      - The mi_lo_stat_cspec() function
        The mi_lo_stat_cspec() function returns from an LO-status structure the storage characteristics for a smart large object.
      - The mi_lo_stat_ctime() function
        The mi_lo_stat_ctime() function returns from an LO-status structure the last-change time for a smart large object.
      - The mi_lo_stat_free() function
        The mi_lo_stat_free() function frees an LO-status structure.
      - The mi_lo_stat_mtime_sec() function
        The mi_lo_stat_mtime_sec() function returns from an LO-status structure the last-modification time, in seconds, of a smart large object.
      - The mi_lo_stat_mtime_usec() function
        The mi_lo_stat_mtime_usec() function returns from an LO-status structure the microsecond component of the last-modification time for a smart large object.
      - The mi_lo_stat_refcnt() function
        The mi_lo_stat_refcnt() function returns from an LO-status structure the reference count of a smart large object.
      - The mi_lo_stat_size() function
        The mi_lo_stat_size() function obtains from an LO-status structure the size, in bytes, of a smart large object.
      - The mi_lo_stat_uid() function
        The mi_lo_stat_uid() function returns from an LO-status structure the user identifier of the owner of a smart large object.
      - The mi_lo_tell() function
        The mi_lo_tell() function returns the current LO seek position for an open smart large object, relative to the beginning of the smart large object.
      - The mi_lo_to_buffer() function
        The mi_lo_to_buffer() function copies a specified number of bytes from a smart large object to a user-defined buffer.
      - The mi_lo_to_file() function
        The mi_lo_to_file() function copies a smart large object to an operating-system file on the client or server computer.
      - The mi_lo_to_string() function
        The mi_lo_to_string() function converts an LO handle in binary representation to its text representation.
      - The mi_lo_truncate() function
        The mi_lo_truncate() function truncates a smart large object at a specified byte position.
      - The mi_lo_unlock() function
        The mi_lo_unlock() releases a byte-range lock on a smart large object.
      - The mi_lo_utimes() function
        The mi_lo_utimes() function enables you to set the last-access and last-modification time of a smart large object.
      - The mi_lo_validate() function
        The mi_lo_validate() function checks whether a given LO handle is valid.
      - The mi_lo_write() function
        The mi_lo_write() function writes a specified number of bytes to an open smart large object.
      - The mi_lo_writewithseek() function
        The mi_lo_writewithseek() function performs a seek operation and then writes a specified number of bytes of data to an open smart large object.
      - The mi_lock_memory() function
        The mi_lock_memory() function locks and waits for a named-memory mutex specified by name and duration.
      - The mi_lvarchar_to_string() function
        The mi_lvarchar_to_string() function returns the data in a varying-length structure as a null-terminated string.
      - The mi_module_lock() function
        The mi_module_lock() function prevents the shared-object file of the current UDR from being unloaded.
      - The mi_money_to_binary() function
        The mi_money_to_binary() function converts a text (string) representation of a monetary value to its binary (internal) MONEY representation.
      - The mi_money_to_string() function
        The mi_money_to_string() function creates a text (string) representation of a monetary value from the binary (internal) MONEY representation.
      - The mi_named_alloc() function
        The mi_named_alloc() function allocates a named block of memory of the specified size in the specified memory duration.
      - The mi_named_free() function
        The mi_named_free() function frees a block of named memory.
      - The mi_named_get() function
        The mi_named_get() function retrieves the address of a named-memory block.
      - The mi_named_zalloc() function
        The mi_named_zalloc() function allocates and initializes a named-memory block in the specified memory duration.
      - The mi_new_var() function
        The mi_new_var() function creates a new varying-length structure for text data.
      - The mi_next_row() function
        The mi_next_row() function retrieves the next row from the cursor of a query.
      - The mi_open() function
        The mi_open() function establishes a connection to a database server.
      - The mi_open_prepared_statement() function
        The mi_open_prepared_statement() function sends a prepared statement that is a query to the database server for execution and opens a cursor for the retrieved rows of the query.
      - The mi_parameter_count() function
        The mi_parameter_count() function returns the number of input parameters in a prepared statement.
      - The mi_parameter_nullable() function
        The mi_parameter_nullable() function indicates whether the column associated with a specified input parameter in a prepared statement can contain an SQL NULL value.
      - The mi_parameter_precision() function
        The mi_parameter_precision() function obtains the precision of the column associated with the specified input parameter in a prepared statement.
      - The mi_parameter_scale() function
        The mi_parameter_scale() function returns the scale of the column that is associated with a specified input parameter in a prepared statement.
      - The mi_parameter_type_id() function
        The mi_parameter_type_id() function returns the type identifier of the column that is associated with the specified input parameter in a prepared statement.
      - The mi_parameter_type_name() function
        The mi_parameter_type_name() function returns the data type name of the column that is associated with a specified input parameter in a prepared statement.
      - The mi_prepare() function
        The mi_prepare() function sends an SQL statement to the database server to be prepared and returns a statement descriptor for the prepared statement.
      - The mi_process_exec() function
        The mi_process_exec() function forks and executes a new process, returning before this new process completes.
      - The mi_put_bigint() function
        The mi_put_bigint() function copies an mi_bigint (BIGINT) value, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_bytes() function
        The mi_put_bytes() function copies the given number of bytes, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_date() function
        The mi_put_date() function copies an mi_date (DATE) value, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_datetime() function
        The mi_put_datetime() function copies an mi_datetime (DATETIME) value, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_decimal() function
        The mi_put_decimal() function copies an mi_decimal (DECIMAL) value, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_double_precision() function
        The mi_put_double_precision() function copies an mi_double_precision (FLOAT) value, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_int8() function
        The mi_put_int8() function copies an mi_int8 (INT8) value, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_integer() function
        The mi_put_integer() function copies an mi_integer (INTEGER) value, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_interval() function
        The mi_put_interval() function copies an mi_interval (INTERVAL) value, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_lo_handle() function
        The mi_put_lo_handle() function copies an LO handle, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_money() function
        The mi_put_money() function copies an mi_money (MONEY) value, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_real() function
        The mi_put_real() function copies an mi_real (SMALLFLOAT) value, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_smallint() function
        The mi_put_smallint() function copies an mi_smallint (SMALLINT) value, converting any difference in alignment or byte order on the server computer to that of the client computer.
      - The mi_put_string() function
        The mi_put_string() function copies an mi_string (CHAR(x)) value to a buffer.
      - The mi_query_finish() function
        The mi_query_finish() function finishes execution of the current statement.
      - The mi_query_interrupt() function
        The mi_query_interrupt() function interrupts the current statement.
      - The mi_realloc() function
        The mi_realloc() function reallocates a block of user memory to a specified size and returns a pointer to that block.
      - The mi_register_callback() function
        The mi_register_callback() function registers a callback function for a single event type or for all event types.
      - The mi_result_command_name() function
        The mi_result_command_name() function returns the name of the SQL statement that is the current statement.
      - The mi_result_reference() function
        The mi_result_reference() function returns the reference of the last object inserted on the connection.
      - The mi_result_row_count() function
        The mi_result_row_count() function returns the number of rows affected by the current statement.
      - The mi_retrieve_callback() function
        The mi_retrieve_callback() function retrieves the handle of a registered callback.
      - The mi_routine_end() function
        The mi_routine_end() function releases resources associated with a function descriptor.
      - The mi_routine_exec() function
        The mi_routine_exec() function executes the registered user-defined routine or cast function associated with a specified function descriptor.
      - The mi_routine_get() function
        The mi_routine_get() function looks up a registered user-defined routine by a routine signature that is a character string and creates its function descriptor.
      - The mi_routine_get_by_typeid() function
        The mi_routine_get_by_typeid() function looks up a registered user-defined routine (UDR) on the local database server by a routine signature that the function builds from a list of arguments. This function also creates a function descriptor for the UDR.
      - The mi_routine_id_get() function
        The mi_routine_id_get() accessor function returns the routine identifier for the user-defined routine that the specified function descriptor describes.
      - The mi_row_create() function
        The mi_row_create() function creates a row, based on a row descriptor and column data.
      - The mi_row_desc_create() function
        The mi_row_desc_create() function creates a row descriptor from a specified type identifier.
      - The mi_row_desc_free() function
        The mi_row_desc_free() routine frees a row descriptor.
      - The mi_row_free() function
        The mi_row_free() function frees a row structure.
      - The mi_save_set_count() function
        The mi_save_set_count() function returns the number of rows in a save set.
      - The mi_save_set_create() function
        The mi_save_set_create() function creates a save set on the current connection.
      - The mi_save_set_delete() function
        The mi_save_set_delete() function removes a row from a save set.
      - The mi_save_set_destroy() function
        The mi_save_set_destroy() function destroys a save set and frees its resources.
      - The mi_save_set_get_first() function
        The mi_save_set_get_first() function retrieves the first row from a save set.
      - The mi_save_set_get_last() function
        The mi_save_set_get_last() function retrieves the last row from a save set.
      - The mi_save_set_get_next() function
        The mi_save_set_get_next() function traverses a save set in the forward direction.
      - The mi_save_set_get_previous() function
        The mi_save_set_get_previous() function traverses a save set in the backward direction.
      - The mi_save_set_insert() function
        The mi_save_set_insert() function appends a copy of a row to the end of a save set.
      - The mi_save_set_member() function
        The mi_save_set_member() function determines whether a row is a member of any open save set.
      - The mi_server_connect() function
        The mi_server_connect() function establishes a connection between a client LIBMI application and a database server.
      - The mi_server_library_version() function
        The mi_server_library_version() function returns the name of the database server with its major and minor version numbers.
      - The mi_server_reconnect() function
        The mi_server_reconnect() function re-establishes a dropped connection.
      - The mi_set_connection_user_data() function
        The mi_set_connection_user_data() function associates the address of user data with an open connection.
      - The mi_set_default_connection_info() function
        The mi_set_default_connection_info() function sets the default connection parameters with values from a connection-information descriptor that the user provides.
      - The mi_set_default_database_info() function
        The mi_set_default_database_info() function sets the default database parameters with values from a database-information descriptor that the user provides.
      - The mi_set_large() function
        The mi_set_large() macro sets the threshold-tracking field of a multirepresentational opaque type to indicate that the multirepresentational data is stored in a smart large object.
      - The mi_set_parameter_info() function
        The mi_set_parameter_info() function sets the session parameters with values from a parameter-information descriptor that the user provides.
      - The mi_set_vardata() function
        The mi_set_vardata() accessor routine stores data in the data portion of the data in a varying-length structure (such as mi_lvarchar).
      - The mi_set_vardata_align() function
        The mi_set_vardata_align() accessor routine stores data in the data portion of a varying-length structure (such as mi_lvarchar) and adjusts for any initial padding required to align the data.
      - The mi_set_varlen() function
        The mi_set_varlen() accessor routine sets the length of the data in a varying-length structure (such as mi_lvarchar).
      - The mi_set_varptr() function
        The mi_set_varptr() function sets the data pointer in a varying-length structure (such as mi_lvarchar).
      - The mi_stack_limit() function
        The mi_stack_limit() function checks if the specified amount of space is available on the user stack.
      - The mi_statement_command_name() function
        The mi_statement_command_name() function returns the name of an SQL prepared statement.
      - The mi_stream_clear_eof() function
        The mi_stream_clear_eof() function clears the end-of-file marker for a stream.
      - The mi_stream_close() function
        The mi_stream_close() function closes a stream.
      - The mi_stream_eof() function
        The mi_stream_eof() function determines whether the current stream seek position is at the end of the stream.
      - The mi_stream_get_error() function
        The mi_stream_get_error() function returns the last error status for an open stream and clears the error condition.
      - The mi_stream_getpos() function
        The mi_stream_getpos() function returns the current seek position of an open stream.
      - The mi_stream_init() function
        The mi_stream_init() function initializes a user-defined stream.
      - The mi_stream_length() function
        The mi_stream_length() function obtains the length of an open stream.
      - The mi_stream_open_fio() function
        The mi_stream_open_fio() function opens a new stream on an operating-system file.
      - The mi_stream_open_mi_lvarchar() function
        The mi_stream_open_mi_lvarchar() function opens a new stream on varying-length data.
      - The mi_stream_open_str() function
        The mi_stream_open_str() function opens a new stream on a character string.
      - The mi_stream_read() function
        The mi_stream_read() function reads a specified number of bytes from a stream into a buffer.
      - The mi_stream_seek() function
        The mi_stream_seek() function sets the stream seek position for the next read or write operation on an open stream.
      - The mi_stream_set_eof() function
        The mi_stream_set_eof() function sets the end-of-file marker for a stream.
      - The mi_stream_set_error() function
        The mi_stream_set_error() function sets the last error status for an open stream.
      - The mi_stream_setpos() function
        The mi_stream_setpos() function sets the stream seek position for the next read or write operation on an open stream.
      - The mi_stream_tell() function
        The mi_stream_tell() function returns the current seek position of an open stream, relative to the beginning of the stream.
      - The mi_stream_write() function
        The mi_stream_write() function writes a specified number of bytes to an open stream.
      - The mi_streamread_boolean() function
        The mi_streamread_boolean() function reads an mi_boolean (BOOLEAN) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_collection() function
        The mi_streamread_collection() function reads a collection structure (LIST, SET, MULTISET) from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_date() function
        The mi_streamread_date() function reads an mi_date (DATE) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_datetime() function
        The mi_streamread_datetime() function reads an mi_datetime (DATETIME) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_decimal() function
        The mi_streamread_decimal() function reads an mi_decimal (DECIMAL) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_double() function
        The mi_streamread_double() function reads an mi_double_precision (FLOAT) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_int8() function
        The mi_streamread_int8() function reads an mi_int8 (INT8) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_integer() function
        The mi_streamread_integer() function reads an mi_integer (INTEGER) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_interval() function
        The mi_streamread_interval() function reads an mi_interval (INTERVAL) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_lo() function
        The mi_streamread_lo() function reads smart-large-object data from a stream and copies the data to a smart large object in the default sbspace.
      - The mi_streamread_lo_by_lofd() function
        The mi_streamread_lo_by_lofd() function reads smart-large-object data from a stream and copies the data to the open smart large object that a specified LO file descriptor identifies.
      - The mi_streamread_lvarchar() function
        The mi_streamread_lvarchar() function reads a varying-length structure (mi_lvarchar) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_money() function
        The mi_streamread_money() function reads an mi_money (MONEY) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_real() function
        The mi_streamread_real() function reads an mi_real (SMALLFLOAT) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_row() function
        The mi_streamread_row() function reads a row structure (row type) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_smallint() function
        The mi_streamread_smallint() function reads an mi_smallint (SMALLINT) value from a stream, converting any difference in the stream representation to that of the internal representation.
      - The mi_streamread_string() function
        The mi_streamread_string() function reads an mi_string (CHAR(x)) value from a stream.
      - The mi_streamwrite_boolean() function
        The mi_streamwrite_boolean() function writes an mi_boolean (BOOLEAN) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_collection() function
        The mi_streamwrite_collection() function writes a collection structure (LIST, SET, MULTISET) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_date() function
        The mi_streamwrite_date() function writes an mi_date (DATE) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_datetime() function
        The mi_streamwrite_datetime() function writes an mi_datetime (DATETIME) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_decimal() function
        The mi_streamwrite_decimal() function writes an mi_decimal (DECIMAL) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_double() function
        The mi_streamwrite_double() function writes an mi_double_precision (FLOAT) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_int8() function
        The mi_streamwrite_int8() function writes an mi_int8 (INT8) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_integer() function
        The mi_streamwrite_integer() function writes an mi_integer (INTEGER) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_interval() function
        The mi_streamwrite_interval() function writes an mi_interval (INTERVAL) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_lo() function
        The mi_streamwrite_lo() function writes a smart large object to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_lvarchar() function
        The mi_streamwrite_lvarchar() function writes a varying-length structure (mi_lvarchar) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_money() function
        The mi_streamwrite_money() function writes an mi_money (MONEY) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_real() function
        The mi_streamwrite_real() function writes an mi_real (SMALLFLOAT) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_row() function
        The mi_streamwrite_row() function writes a row structure (row type) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_smallint() function
        The mi_streamwrite_smallint() function writes an mi_smallint (SMALLINT) value to a stream, converting any difference in the internal representation to that of the stream representation.
      - The mi_streamwrite_string() function
        The mi_streamwrite_string() function writes an mi_string (CHAR(x)) value to a stream.
      - The mi_string_to_date() function
        The mi_string_to_date() function converts a text (string) representation of a date value to its binary (internal) DATE representation.
      - The mi_string_to_datetime() function
        The mi_string_to_datetime() function converts a text (string) representation of a date, time, or date and time value to its internal (binary) DATETIME representation.
      - The mi_string_to_decimal() function
        The mi_string_to_decimal() function converts a text (string) representation of a decimal value to its binary (internal) DECIMAL representation.
      - The mi_string_to_interval() function
        The mi_string_to_interval() function converts a text (string) representation of an interval value to its binary (internal) INTERVAL representation.
      - The mi_string_to_lvarchar() function
        The mi_string_to_lvarchar() function converts a null-terminated string to a varying-length structure.
      - The mi_string_to_money() function
        The mi_string_to_money() function converts a text (string) representation of a monetary value to its binary (internal) MONEY representation.
      - The mi_switch_mem_duration() function
        The mi_switch_mem_duration() function switches the current memory duration.
      - The mi_sysname() function
        The mi_sysname() function obtains the name of the default database server.
      - The mi_system() function
        The mi_system() function allows you to execute operating system commands from within a module or C UDR.
      - The mi_td_cast_get() function
        The mi_td_cast_get() function looks up a registered cast function by the type descriptors of its source and target data types and creates its function descriptor.
      - The mi_tracefile_set() function
        The mi_tracefile_set() function sets the trace-output file. The database server writes trace messages to this file.
      - The mi_tracelevel_set() function
        The mi_tracelevel_set() function sets the trace level for specified trace classes.
      - The mi_transaction_state() function (Server)
        The mi_transaction_state() function returns the current transaction state (none, implicit, or explicit).
      - The mi_transition_type() function
        The mi_transition_type() accessor function obtains the type of state transition from a transition descriptor.
      - The mi_trigger_event() function
        The mi_trigger_event() function returns the trigger event information for the current trigger event.
      - The mi_trigger_get_new_row() function
        The mi_trigger_get_new_row() function returns the name of the new row value if the MI_TRIGGER_FOREACH_EVENT bit is set, which implies that neither the MI_TRIGGER_BEFORE_EVENT nor the MI_TRIGGER_AFTER_EVENT bits are set.
      - The mi_trigger_get_old_row() function
        The mi_trigger_get_old_row() function returns the name of the old row value if the MI_TRIGGER_FOREACH_EVENT bit is set, which implies that the MI_TRIGGER_BEFORE_EVENT or the MI_TRIGGER_AFTER_EVENT bits are not set. In addition, mi_trigger_get_old_row() returns all the columns in the requested row, not just those columns into which data was explicitly inserted. Columns with default data are also returned.
      - The mi_trigger_level() function
        The mi_trigger_level() function returns the nesting level value of the current trigger.
      - The mi_trigger_name() function
        The mi_trigger_name() function returns the name of the currently executing trigger.
      - The mi_trigger_tabname() function
        The mi_trigger_tabname() function returns the triggering table or view name on which the current trigger action statement was started.
      - The mi_try_lock_memory() function
        The mi_try_lock_memory() function requests a lock on a named-memory block specified by name and memory duration.
      - The mi_type_align() function
        The mi_type_align() function obtains the alignment for a data type from its type descriptor.
      - The mi_type_byvalue() function
        The mi_type_byvalue() function checks a type descriptor to determine whether a data type is passed by value or by reference.
      - The mi_type_constructor_typedesc() function
        The mi_type_constructor_typedesc() function obtains a type descriptor for a constructed data type.
      - The mi_type_element_typedesc() function
        The mi_type_element_typedesc() function obtains the type descriptor for the elements of a collection data type from its type descriptor.
      - The mi_type_full_name() function
        The mi_type_full_name() function obtains the full name (owner.typename) of a data type from its type descriptor.
      - The mi_type_length() function
        The mi_type_length() function obtains the length of a data type from its type descriptor.
      - The mi_type_maxlength() function
        The mi_type_maxlength() function obtains the maximum length of a data type from its type descriptor.
      - The mi_type_owner() function
        The mi_type_owner() function obtains the owner of a data type from its type descriptor.
      - The mi_type_precision() function
        The mi_type_precision() function obtains the precision of a data type from its type descriptor.
      - The mi_type_qualifier() function
        The mi_type_qualifier() function obtains the qualifier of a data type from its type descriptor.
      - The mi_type_scale() function
        The mi_type_scale() function obtains the scale of the data type from its type descriptor.
      - The mi_type_typedesc() function
        The mi_type_typedesc() function creates a type descriptor, given a type identifier.
      - The mi_type_typename() function
        The mi_type_typename() function obtains the name of a data type from its type descriptor.
      - The mi_typedesc_typeid() function
        The mi_typedesc_typeid() function obtains the type identifier of a data type from its type descriptor.
      - The mi_typeid_equals() function
        The mi_typeid_equals() function determines whether two type identifiers represent the same data type.
      - The mi_typeid_is_builtin() function
        The mi_typeid_is_builtin() function determines whether a type identifier is for a built-in data type.
      - The mi_typeid_is_collection() function
        The mi_typeid_is_collection() function determines whether a type identifier is for a collection.
      - The mi_typeid_is_complex() function
        The mi_typeid_is_complex() function determines whether a type identifier is for a complex data type.
      - The mi_typeid_is_distinct() function
        The mi_typeid_is_distinct() function determines whether a type identifier is for a distinct data type.
      - The mi_typeid_is_list() function
        The mi_typeid_is_list() function determines whether a type identifier is for a LIST collection data type.
      - The mi_typeid_is_multiset() function
        The mi_typeid_is_multiset() function determines whether a type identifier is for a MULTISET collection data type.
      - The mi_typeid_is_row() function
        The mi_typeid_is_row() function determines whether a type identifier is for an SQL row data type.
      - The mi_typeid_is_set() function
        The mi_typeid_is_set() function determines whether a type identifier is for a SET collection data type.
      - The mi_typename_to_id() function
        The mi_typename_to_id() function creates a type identifier for a data type, given the type name in LVARCHAR format.
      - The mi_typename_to_typedesc() function
        The mi_typename_to_typedesc() function creates a type descriptor for a data type, given the type name in LVARCHAR format.
      - The mi_typestring_to_id() function
        The mi_typestring_to_id() function creates a type identifier for a data type, given the type name as a null-terminated string.
      - The mi_typestring_to_typedesc() function
        The mi_typestring_to_typedesc() function creates a type descriptor for a data type, given the type name as a null-terminated string.
      - The mi_udr_lock() function
        The mi_udr_lock() function locks an instance of a UDR onto the virtual processor (VP) on which it begins execution.
      - The mi_unlock_memory() function
        The mi_unlock_memory() function unlocks a named-memory block specified by name and memory duration.
      - The mi_unregister_callback() function
        The mi_unregister_callback() function unregisters a callback for an event type or for all event types.
      - The mi_value() function
        The mi_value() function retrieves a column value from a row, given the column identifier.
      - The mi_value_by_name() function
        The mi_value_by_name() function retrieves a column value from a row, given the column name.
      - The mi_var_copy() function
        The mi_var_copy() function creates a copy of a varying-length structure.
      - The mi_var_free() function
        The mi_var_free() function frees the specified varying-length structure.
      - The mi_var_to_buffer() function
        The mi_var_to_buffer() function copies data from a varying-length structure to a buffer.
      - The mi_version_comparison() function
        The mi_version_comparison() function compares the version of two instances of HCL Informix®.
      - The mi_vpinfo_classid() function
        The mi_vpinfo_classid() function obtains the VP-class identifier for the VP class in which the current UDR is running.
      - The mi_vpinfo_isnoyield() function
        The mi_vpinfo_isnoyield() function determines whether the virtual processor (VP) on which the current UDR is executing is part of a nonyielding user-defined VP class.
      - The mi_vpinfo_vpid() function
        The mi_vpinfo_vpid() function obtains the VP identifier of the virtual processor (VP) on which the user-defined routine (UDR) is executing.
      - The mi_xa_get_current_xid() function
        The mi_xa_get_current_xid() function returns the pointer to the current XID structure for an XA-compliant, external data source. The XID structure is defined in the $INFORMIXDIR/incl/public/xa.h file.
      - The mi_xa_get_xadatasource_rmid() function
        The mi_xa_get_xadatasource_rmid() function gets the resource manager ID that was previously created in the database for an XA-compliant, external data source.
      - The mi_xa_register_xadatasource() function
        The mi_xa_register_xadatasource() function registers an XA data source with the current transaction. The registration is dynamic and is applicable for the current transaction only. The must register participating data sources into each transaction.
      - The mi_xa_unregister_xadatasource() function
        The mi_xa_unregister_xadatasource() function unregisters the previously registered XA data source from the transaction.
      - The mi_yield() function
        The mi_yield() routine yields processing to other database server threads.
      - The mi_zalloc() function
        The mi_zalloc() function allocates and zero-fills a block of memory of the given size and returns a pointer to a block of user memory of the specified size.
      - The rdatestr() function
        The rdatestr() function converts an internal DATE value to a character string.
      - The rdayofweek() function
        The rdayofweek() function returns the day of the week as an integer value for an internal DATE.
      - The rdefmtdate() function
        The rdefmtdate() function uses a formatting mask to convert a character string to an internal DATE format.
      - The rdownshift() function
        The rdownshift() function changes all the uppercase characters within a null-terminated string to lowercase characters.
      - The rfmtdate() function
        The rfmtdate() function uses a formatting mask to convert an internal DATE value to a character string.
      - The rfmtdec() function
        The rfmtdec() function uses a formatting mask to convert a decimal value to a character string.
      - The rfmtdouble() function
        The rfmtdouble() function uses a formatting mask to convert a C double value to a character string.
      - The rfmtlong() function
        The rfmtlong() function uses a formatting mask to convert a C long value to a character string.
      - The rjulmdy() function
        The rjulmdy() function creates an array of three short integer values that represent the month, day, and year from an internal DATE value.
      - The rleapyear() function
        The rleapyear() function returns 1 (TRUE) when the argument that is passed to it is a leap year and 0 (FALSE) when it is not.
      - The rmdyjul() function
        The rmdyjul() function creates an internal DATE from an array of three short integer values that represent month, day, and year.
      - The rstod() function
        The rstod() function converts a null-terminated string into a C double value.
      - The rstoi() function
        The rstoi() function converts a null-terminated string into a short integer value.
      - The rstol() function
        The rstol() function converts a null-terminated string into a long integer value.
      - The rstrdate() function
        The rstrdate() function converts a character string to an internal DATE.
      - The rtoday() function
        The rtoday() function returns the system date as a long integer value.
      - The rupshift() function
        The rupshift() function changes all the characters within a null-terminated string to uppercase characters.
      - The stcat() function
        The stcat() function concatenates one null-terminated string to the end of another.
      - The stchar() function
        The stchar() function stores a null-terminated string in a fixed-length string, padding the end with blanks, if necessary.
      - The stcmpr() function
        The stcmpr() function compares two null-terminated strings.
      - The stcopy() function
        The stcopy() function copies a null-terminated string from one location in memory to another location.
      - The stleng() function
        The stleng() function returns the length, in bytes, of a null-terminated string that you specify.
  - Developers Kit
    The Developers Kit (DBDK) guides contain information about the tools you can use to develop and package modules, which extend the functionality of HCL Informix® databases.
  - J/Foundation Developer's Guide
    The HCL® J/Foundation Developer's Guide describes how to write user-defined routines (UDRs) in the Java™ programming language for Informix®.
  - R-Tree Index User's Guide
    The HCL® Informix® R-Tree Index User's Guide describes the HCL Informix R-tree secondary access method and how to access and use its components.
  - User-Defined Routines and Data Types Developer's Guide
    The HCL® Informix® User-Defined Routines and Data Types Developer's Guide describes how to define new data types and enable user-defined routines (UDRs) to extend HCL Informix.
  - Virtual-Table Interface Guide
    The IBM® Informix® Virtual-Table Interface Programmer's Guide explains how to create a primary access method with the Virtual-Table Interface (VTI) so that users have a single SQL interface to HCL Informix tables and to data that does not conform to the storage scheme of HCL Informix.
  - Virtual-Index Interface Guide
    The IBM® Informix® Virtual-Index Interface Programmer's Guide explains how to create a secondary access method with the Virtual-Index Interface (VII) to extend the built-in indexing schemes of HCL Informix typically with a module.
Data warehousing
In addition to designing and implementing Informix® dimensional databases, you can use tools to create data warehouse applications and optimize your data warehouse queries.
Designing databases
The first step in creating a relational database is to construct a data model, which is a precise, complete definition of the data you want to store. After you prepare your data model, you must implement it as a database and tables. To implement your data model, you first select a data type for each column and then you create a database and tables and populate the tables with data. You can also implement fragmentation strategies and control access to your data.
JSON compatibility
You can use the popular JSON-oriented query language created by MongoDB to interact with data stored in HCL Informix®.
Security
You can secure your Informix® database server and the data that is stored in your Informix databases. You can encrypt data, secure connections, control user privileges and access, and audit data security.
SQL programming
You can use the HCL Informix® implementation of the SQL language to develop applications for Informix database servers.
Troubleshooting HCL Informix®
Several troubleshooting techniques, tools, and resources are available for resolving problems that you encounter in your HCL Informix® database server environment.

The mi_save_set_delete() function

The mi_save_set_delete() function removes a row from a save set.

Syntax

mi_integer mi_save_set_delete(row)
   MI_ROW *row;

row: A pointer to a row in a save set.


Valid in client LIBMI application?	Valid in user-defined routine?
Yes	Yes

Usage

The mi_save_set_delete() function is a destructor function for a save set.

Return values

MI_OK: The row was deleted from the save set.
MI_ERROR: The function was not successful.