db2Load API - Load data into a table

Posted on 2010-07-23 12:26 Prayer 阅读(1412) 评论(0) 编辑收藏引用所属分类: DB2 、DB2CLI

http://publib.boulder.ibm.com/infocenter/db2luw/v9/index.jsp?topic=/com.ibm.db2.udb.admin.doc/doc/r0008704.htm

DB2 Version 9 for Linux, UNIX, and Windows

db2Load API - Load data into a table

Loads data into a DB2 table. Data residing on the server may be in the form of a file, cursor, tape, or named pipe. Data residing on a remotely connected client may be in the form of a fully qualified file, a cursor, or named pipe. Although faster than the import utility, the load utility does not support loading data at the hierarchy level or loading into a nickname.

Authorization

One of the following:

sysadm
dbadm
load authority on the database and:
- INSERT privilege on the table when the load utility is invoked in INSERT mode, TERMINATE mode (to terminate a previous load insert operation), or RESTART mode (to restart a previous load insert operation)
- INSERT and DELETE privilege on the table when the load utility is invoked in REPLACE mode, TERMINATE mode (to terminate a previous load replace operation), or RESTART mode (to restart a previous load replace operation)
- INSERT privilege on the exception table, if such a table is used as part of the load operation.

Note: In general, all load processes and all DB2 server processes are owned by the instance owner. All of these processes use the identification of the instance owner to access needed files. Therefore, the instance owner must have read access to the input files, regardless of who invokes the command.

Required connection

Database. If implicit connect is enabled, a connection to the default database is established. Utility access to Linux, UNIX, or Windows database servers from Linux, UNIX, or Windows clients must be a direct connection through the engine and not through a DB2 Connect gateway or loop back environment.

Instance. An explicit attachment is not required. If a connection to the database has been established, an implicit attachment to the local instance is attempted.

API include file

db2ApiDf.h

API and data structure syntax

SQL_API_RC SQL_API_FN
db2Load (
db2Uint32 versionNumber,
void * pParmStruct,
struct sqlca * pSqlca);
typedef SQL_STRUCTURE db2LoadStruct
{
struct sqlu_media_list *piSourceList;
struct sqlu_media_list *piLobPathList;
struct sqldcol *piDataDescriptor;
struct sqlchar *piActionString;
char *piFileType;
struct sqlchar *piFileTypeMod;
char *piLocalMsgFileName;
char *piTempFilesPath;
struct sqlu_media_list *piVendorSortWorkPaths;
struct sqlu_media_list *piCopyTargetList;
db2int32 *piNullIndicators;
struct db2LoadIn *piLoadInfoIn;
struct db2LoadOut *poLoadInfoOut;
struct db2PartLoadIn *piPartLoadInfoIn;
struct db2PartLoadOut *poPartLoadInfoOut;
db2int16 iCallerAction;
} db2LoadStruct;
typedef SQL_STRUCTURE db2LoadUserExit
{
db2Char iSourceUserExitCmd;
struct db2Char *piInputStream;
struct db2Char *piInputFileName;
struct db2Char *piOutputFileName;
db2Uint16 *piEnableParallelism;
} db2LoadUserExit;
typedef SQL_STRUCTURE db2LoadIn
{
db2Uint64 iRowcount;
db2Uint64 iRestartcount;
char *piUseTablespace;
db2Uint32 iSavecount;
db2Uint32 iDataBufferSize;
db2Uint32 iSortBufferSize;
db2Uint32 iWarningcount;
db2Uint16 iHoldQuiesce;
db2Uint16 iCpuParallelism;
db2Uint16 iDiskParallelism;
db2Uint16 iNonrecoverable;
db2Uint16 iIndexingMode;
db2Uint16 iAccessLevel;
db2Uint16 iLockWithForce;
db2Uint16 iCheckPending;
char iRestartphase;
char iStatsOpt;
db2Uint16 iSetIntegrityPending;
struct db2LoadUserExit *piSourceUserExit;
} db2LoadIn;
typedef SQL_STRUCTURE db2LoadOut
{
db2Uint64 oRowsRead;
db2Uint64 oRowsSkipped;
db2Uint64 oRowsLoaded;
db2Uint64 oRowsRejected;
db2Uint64 oRowsDeleted;
db2Uint64 oRowsCommitted;
} db2LoadOut;
typedef SQL_STRUCTURE db2PartLoadIn
{
char *piHostname;
char *piFileTransferCmd;
char *piPartFileLocation;
struct db2LoadNodeList *piOutputNodes;
struct db2LoadNodeList *piPartitioningNodes;
db2Uint16 *piMode;
db2Uint16 *piMaxNumPartAgents;
db2Uint16 *piIsolatePartErrs;
db2Uint16 *piStatusInterval;
struct db2LoadPortRange *piPortRange;
db2Uint16 *piCheckTruncation;
char *piMapFileInput;
char *piMapFileOutput;
db2Uint16 *piTrace;
db2Uint16 *piNewline;
char *piDistfile;
db2Uint16 *piOmitHeader;
SQL_PDB_NODE_TYPE *piRunStatDBPartNum;
} db2PartLoadIn;
typedef SQL_STRUCTURE db2LoadNodeList
{
SQL_PDB_NODE_TYPE *piNodeList;
db2Uint16 iNumNodes;
} db2LoadNodeList;
typedef SQL_STRUCTURE db2LoadPortRange
{
db2Uint16 iPortMin;
db2Uint16 iPortMax;
} db2LoadPortRange;
typedef SQL_STRUCTURE db2PartLoadOut
{
db2Uint64 oRowsRdPartAgents;
db2Uint64 oRowsRejPartAgents;
db2Uint64 oRowsPartitioned;
struct db2LoadAgentInfo *poAgentInfoList;
db2Uint32 iMaxAgentInfoEntries;
db2Uint32 oNumAgentInfoEntries;
} db2PartLoadOut;
typedef SQL_STRUCTURE db2LoadAgentInfo
{
db2int32 oSqlcode;
db2Uint32 oTableState;
SQL_PDB_NODE_TYPE oNodeNum;
db2Uint16 oAgentType;
} db2LoadAgentInfo;
SQL_API_RC SQL_API_FN
db2gLoad (
db2Uint32 versionNumber,
void * pParmStruct,
struct sqlca * pSqlca);
typedef SQL_STRUCTURE db2gLoadStruct
{
struct sqlu_media_list *piSourceList;
struct sqlu_media_list *piLobPathList;
struct sqldcol *piDataDescriptor;
struct sqlchar *piActionString;
char *piFileType;
struct sqlchar *piFileTypeMod;
char *piLocalMsgFileName;
char *piTempFilesPath;
struct sqlu_media_list *piVendorSortWorkPaths;
struct sqlu_media_list *piCopyTargetList;
db2int32 *piNullIndicators;
struct db2gLoadIn *piLoadInfoIn;
struct db2LoadOut *poLoadInfoOut;
struct db2gPartLoadIn *piPartLoadInfoIn;
struct db2PartLoadOut *poPartLoadInfoOut;
db2int16 iCallerAction;
db2Uint16 iFileTypeLen;
db2Uint16 iLocalMsgFileLen;
db2Uint16 iTempFilesPathLen;
} db2gLoadStruct;
typedef SQL_STRUCTURE db2gLoadIn
{
db2Uint64 iRowcount;
db2Uint64 iRestartcount;
char *piUseTablespace;
db2Uint32 iSavecount;
db2Uint32 iDataBufferSize;
db2Uint32 iSortBufferSize;
db2Uint32 iWarningcount;
db2Uint16 iHoldQuiesce;
db2Uint16 iCpuParallelism;
db2Uint16 iDiskParallelism;
db2Uint16 iNonrecoverable;
db2Uint16 iIndexingMode;
db2Uint16 iAccessLevel;
db2Uint16 iLockWithForce;
db2Uint16 iCheckPending;
char iRestartphase;
char iStatsOpt;
db2Uint16 iUseTablespaceLen;
db2Uint16 iSetIntegrityPending;
struct db2LoadUserExit *piSourceUserExit;
} db2gLoadIn;
typedef SQL_STRUCTURE db2gPartLoadIn
{
char *piHostname;
char *piFileTransferCmd;
char *piPartFileLocation;
struct db2LoadNodeList *piOutputNodes;
struct db2LoadNodeList *piPartitioningNodes;
db2Uint16 *piMode;
db2Uint16 *piMaxNumPartAgents;
db2Uint16 *piIsolatePartErrs;
db2Uint16 *piStatusInterval;
struct db2LoadPortRange *piPortRange;
db2Uint16 *piCheckTruncation;
char *piMapFileInput;
char *piMapFileOutput;
db2Uint16 *piTrace;
db2Uint16 *piNewline;
char *piDistfile;
db2Uint16 *piOmitHeader;
void *piReserved1;
db2Uint16 iHostnameLen;
db2Uint16 iFileTransferLen;
db2Uint16 iPartFileLocLen;
db2Uint16 iMapFileInputLen;
db2Uint16 iMapFileOutputLen;
db2Uint16 iDistfileLen;
} db2gPartLoadIn;

db2Load API parameters

versionNumber: Input. Specifies the version and release level of the structure passed as the second parameter pParmStruct.
pParmStruct: Input. A pointer to the db2LoadStruct structure.
pSqlca: Output. A pointer to the sqlca structure.

db2LoadStruct data structure parameters

piSourceList

Input. A pointer to an sqlu_media_list structure used to provide a list of source files, devices, vendors, pipes, or SQL statements.

The information provided in this structure depends on the value of the media_type field. Valid values (defined in sqlutil header file, located in the include directory) are:

SQLU_SQL_STMT: If the media_type field is set to this value, the caller provides an SQL query through the pStatement field of the target field. The pStatement field is of type sqlu_statement_entry. The sessions field must be set to the value of 1, since the load utility only accepts a single SQL query per load.
SQLU_SERVER_LOCATION: If the media_type field is set to this value, the caller provides information through sqlu_location_entry structures. The sessions field indicates the number of sqlu_location_entry structures provided. This is used for files, devices, and named pipes.
SQLU_CLIENT_LOCATION: If the media_type field is set to this value, the caller provides information through sqlu_location_entry structures. The sessions field indicates the number of sqlu_location_entry structures provided. This is used for fully qualified files and named pipes. Note that this media_type is only valid if the API is being called via a remotely connected client.
SQLU_TSM_MEDIA: If the media_type field is set to this value, the sqlu_vendor structure is used, where filename is the unique identifier for the data to be loaded. There should only be one sqlu_vendor entry, regardless of the value of sessions. The sessions field indicates the number of TSM sessions to initiate. The load utility will start the sessions with different sequence numbers, but with the same data in the one sqlu_vendor entry.
SQLU_OTHER_MEDIA: If the media_type field is set to this value, the sqlu_vendor structure is used, where shr_lib is the shared library name, and filename is the unique identifier for the data to be loaded. There should only be one sqlu_vendor entry, regardless of the value of sessions. The sessions field indicates the number of other vendor sessions to initiate. The load utility will start the sessions with different sequence numbers, but with the same data in the one sqlu_vendor entry.
SQLU_REMOTEFETCH: If the media_type field is set to this value, the caller provides information through an sqlu_remotefetch_entry structure. The sessions field must be set to the value of 1.

piLobPathList

Input. A pointer to an sqlu_media_list structure. For IXF, ASC, and DEL file types, a list of fully qualified paths or devices to identify the location of the individual LOB files to be loaded. The file names are found in the IXF, ASC, or DEL files, and are appended to the paths provided.

The information provided in this structure depends on the value of the media_type field. Valid values (defined in sqlutil header file, located in the include directory) are:

SQLU_LOCAL_MEDIA: If set to this value, the caller provides information through sqlu_media_entry structures. The sessions field indicates the number of sqlu_media_entry structures provided.
SQLU_TSM_MEDIA: If set to this value, the sqlu_vendor structure is used, where filename is the unique identifier for the data to be loaded. There should only be one sqlu_vendor entry, regardless of the value of sessions. The sessions field indicates the number of TSM sessions to initiate. The load utility will start the sessions with different sequence numbers, but with the same data in the one sqlu_vendor entry.
SQLU_OTHER_MEDIA: If set to this value, the sqlu_vendor structure is used, where shr_lib is the shared library name, and filename is the unique identifier for the data to be loaded. There should only be one sqlu_vendor entry, regardless of the value of sessions. The sessions field indicates the number of other vendor sessions to initiate. The load utility will start the sessions with different sequence numbers, but with the same data in the one sqlu_vendor entry.

piDataDescriptor

Input. Pointer to an sqldcol structure containing information about the columns being selected for loading from the external file.

If the pFileType parameter is set to SQL_ASC, the dcolmeth field of this structure must be set to SQL_METH_L. The user specifies the start and end locations for each column to be loaded.

If the file type is SQL_DEL, dcolmeth can be either SQL_METH_P or SQL_METH_D. If it is SQL_METH_P, the user must provide the source column position. If it is SQL_METH_D, the first column in the file is loaded into the first column of the table, and so on.

If the file type is SQL_IXF, dcolmeth can be one of SQL_METH_P, SQL_METH_D, or SQL_METH_N. The rules for DEL files apply here, except that SQL_METH_N indicates that file column names are to be provided in the sqldcol structure.

piActionString

Input. Pointer to an sqlchar structure, followed by an array of characters specifying an action that affects the table.

The character array is of the form:

"INSERT|REPLACE|RESTART|TERMINATE
INTO tbname [(column_list)]
[FOR EXCEPTION e_tbname[NOUNIQUEEXC][NORANGEEXC]]"

INSERT: Adds the loaded data to the table without changing the existing table data.
REPLACE: Deletes all existing data from the table, and inserts the loaded data. The table definition and the index definitions are not changed.
RESTART: Restarts a previously interrupted load operation. The load operation will automatically continue from the last consistency point in the load, build, or delete phase.
TERMINATE: Terminates a previously interrupted load operation, and rolls back the operation to the point in time at which it started, even if consistency points were passed. The states of any table spaces involved in the operation return to normal, and all table objects are made consistent (index objects may be marked as invalid, in which case index rebuild will automatically take place at next access). If the table spaces in which the table resides are not in load pending state, this option does not affect the state of the table spaces.
The load terminate option will not remove a backup pending state from table spaces.
tbname: The name of the table into which the data is to be loaded. The table cannot be a system table or a declared temporary table. An alias, or the fully qualified or unqualified table name can be specified. A qualified table name is in the form schema.tablename. If an unqualified table name is specified, the table will be qualified with the CURRENT SCHEMA.
(column_list): A list of table column names into which the data is to be inserted. The column names must be separated by commas. If a name contains spaces or lowercase characters, it must be enclosed by quotation marks.
FOR EXCEPTION e_tbname: Specifies the exception table into which rows in error will be copied. The exception table is used to store copies of rows that violate unique index rules, range constraints and security policies.
NORANGEEXC: Indicates that if a row is rejected because of a range violation it will not be inserted into the exception table.
NOUNIQUEEXC: Indicates that if a row is rejected because it violates a unique constraint it will not be inserted into the exception table.

piFileType

Input. A string that indicates the format of the input data source. Supported external formats (defined in sqlutil) are:

SQL_ASC: Non-delimited ASCII.
SQL_DEL: Delimited ASCII, for exchange with dBase, BASIC, and the IBM Personal Decision Series programs, and many other database managers and file managers.
SQL_IXF: PC version of the Integration Exchange Format, the preferred method for exporting data from a table so that it can be loaded later into the same table or into another database manager table.
SQL_CURSOR: An SQL query. The sqlu_media_list structure passed in through the piSourceList parameter is either of type SQLU_SQL_STMT or SQLU_REMOTEFETCH, and refers to an SQL query or a table name.

piFileTypeMod

Input. A pointer to the sqlchar structure, followed by an array of characters that specify one or more processing options. If this pointer is NULL, or the structure pointed to has zero characters, this action is interpreted as selection of a default specification.

Not all options can be used with all of the supported file types. See related link "File type modifiers for the load utility."

piLocalMsgFileName

Input. A string containing the name of a local file to which output messages are to be written.

piTempFilesPath

Input. A string containing the path name to be used on the server for temporary files. Temporary files are created to store messages, consistency points, and delete phase information.

piVendorSortWorkPaths

Input. A pointer to the sqlu_media_list structure which specifies the Vendor Sort work directories.

piCopyTargetList

Input. A pointer to an sqlu_media_list structure used (if a copy image is to be created) to provide a list of target paths, devices, or a shared library to which the copy image is to be written.

The values provided in this structure depend on the value of the media_type field. Valid values for this parameter (defined in sqlutil header file, located in the include directory) are:

SQLU_LOCAL_MEDIA: If the copy is to be written to local media, set the media_type to this value and provide information about the targets in sqlu_media_entry structures. The sessions field specifies the number of sqlu_media_entry structures provided.
SQLU_TSM_MEDIA: If the copy is to be written to TSM, use this value. No further information is required.
SQLU_OTHER_MEDIA: If a vendor product is to be used, use this value and provide further information via an sqlu_vendor structure. Set the shr_lib field of this structure to the shared library name of the vendor product. Provide only one sqlu_vendor entry, regardless of the value of sessions. The sessions field specifies the number of sqlu_media_entry structures provided. The load utility will start the sessions with different sequence numbers, but with the same data provided in the one sqlu_vendor entry.

piNullIndicators

Input. For ASC files only. An array of integers that indicate whether or not the column data is nullable. There is a one-to-one ordered correspondence between the elements of this array and the columns being loaded from the data file. That is, the number of elements must equal the dcolnum field of the pDataDescriptor parameter. Each element of the array contains a number identifying a location in the data file that is to be used as a NULL indicator field, or a zero indicating that the table column is not nullable. If the element is not zero, the identified location in the data file must contain a Y or an N. A Y indicates that the table column data is NULL, and N indicates that the table column data is not NULL.

piLoadInfoIn

Input. A pointer to the db2LoadIn structure.

poLoadInfoOut

Input. A pointer to the db2LoadOut structure.

piPartLoadInfoIn

Input. A pointer to the db2PartLoadIn structure.

poPartLoadInfoOut

Output. A pointer to the db2PartLoadOut structure.

iCallerAction

Input. An action requested by the caller. Valid values (defined in sqlutil header file, located in the include directory) are:

SQLU_INITIAL: Initial call. This value (or SQLU_NOINTERRUPT) must be used on the first call to the API.
SQLU_NOINTERRUPT: Initial call. Do not suspend processing. This value (or SQLU_INITIAL) must be used on the first call to the API.
If the initial call or any subsequent call returns and requires the calling application to perform some action prior to completing the requested load operation, the caller action must be set to one of the following:
SQLU_CONTINUE: Continue processing. This value can only be used on subsequent calls to the API, after the initial call has returned with the utility requesting user input (for example, to respond to an end of tape condition). It specifies that the user action requested by the utility has completed, and the utility can continue processing the initial request.
SQLU_TERMINATE: Terminate processing. Causes the load utility to exit prematurely, leaving the table spaces being loaded in LOAD_PENDING state. This option should be specified if further processing of the data is not to be done.
SQLU_ABORT: Terminate processing. Causes the load utility to exit prematurely, leaving the table spaces being loaded in LOAD_PENDING state. This option should be specified if further processing of the data is not to be done.
SQLU_RESTART: Restart processing.
SQLU_DEVICE_TERMINATE: Terminate a single device. This option should be specified if the utility is to stop reading data from the device, but further processing of the data is to be done.

db2LoadUserExit data structure parameters

iSourceUserExitCmd: Input. The fully qualified name of an executable that will be used to feed data to the utility. For security reasons, the executable must be placed within the sqllib/bin directory on the server. This parameter is mandatory if the piSourceUserExit structure is not NULL.
The piInputStream, piInputFileName, piOutputFileName and piEnableParallelism fields are optional.
piInputStream: Input. A generic byte-stream that will be passed directly to the user-exit application via STDIN. You have complete control over what data is contained in this byte-stream and in what format. The load utility will simply carry this byte-stream over to the server and pass it into the user-exit application by feeding the process' STDIN (it will not codepage convert or modify the byte-stream). Your user-exit application would read the arguments from STDIN and use the data however intended.
One important attribute of this feature is the ability to hide sensitive information (such as userid/passwords).
piInputFileName: Input. Contains the name of a fully qualified client-side file, whose contents will be passed into the user-exit application by feeding the process' STDIN.
piOutputFileName: Input. The fully qualified name of a server-side file. The STDOUT and STDERR streams of the process which is executing the user-exit application will be streamed into this file. When piEnableParallelism is TRUE, multiple files will be created (one per user-exit instance), and each file name will be appended with a 3 digit numeric node-number value, such as <filename>.000).
piEnableParallelism: Input. A flag indicating that the utility should attempt to parallelize the invocation of the user-exit application.

db2LoadIn data structure parameters

iRowcount

Input. The number of physical records to be loaded. Allows a user to load only the first rowcnt rows in a file.

iRestartcount

Input. Reserved for future use.

piUseTablespace

Input. If the indexes are being rebuilt, a shadow copy of the index is built in tablespace iUseTablespaceName and copied over to the original tablespace at the end of the load. Only system temporary table spaces can be used with this option. If not specified then the shadow index will be created in the same tablespace as the index object.

If the shadow copy is created in the same tablespace as the index object, the copy of the shadow index object over the old index object is instantaneous. If the shadow copy is in a different tablespace from the index object a physical copy is performed. This could involve considerable I/O and time. The copy happens while the table is offline at the end of a load.

This field is ignored if iAccessLevel is SQLU_ALLOW_NO_ACCESS.

This option is ignored if the user does not specify INDEXING MODE REBUILD or INDEXING MODE AUTOSELECT. This option will also be ignored if INDEXING MODE AUTOSELECT is chosen and load chooses to incrementally update the index.

iSavecount

The number of records to load before establishing a consistency point. This value is converted to a page count, and rounded up to intervals of the extent size. Since a message is issued at each consistency point, this option should be selected if the load operation will be monitored using db2LoadQuery - Load Query. If the value of savecnt is not sufficiently high, the synchronization of activities performed at each consistency point will impact performance.

The default value is 0, meaning that no consistency points will be established, unless necessary.

iDataBufferSize

The number of 4KB pages (regardless of the degree of parallelism) to use as buffered space for transferring data within the utility. If the value specified is less than the algorithmic minimum, the required minimum is used, and no warning is returned.

This memory is allocated directly from the utility heap, whose size can be modified through the util_heap_sz database configuration parameter.

If a value is not specified, an intelligent default is calculated by the utility at run time. The default is based on a percentage of the free space available in the utility heap at the instantiation time of the loader, as well as some characteristics of the table.

iSortBufferSize

Input. This option specifies a value that overrides the SORTHEAP database configuration parameter during a load operation. It is relevant only when loading tables with indexes and only when the iIndexingMode parameter is not specified as SQLU_INX_DEFERRED. The value that is specified cannot exceed the value of SORTHEAP. This parameter is useful for throttling the sort memory used by LOAD without changing the value of SORTHEAP, which would also affect general query processing.

iWarningcount

Input. Stops the load operation after warningcnt warnings. Set this parameter if no warnings are expected, but verification that the correct file and table are being used is desired. If the load file or the target table is specified incorrectly, the load utility will generate a warning for each row that it attempts to load, which will cause the load to fail. If warningcnt is 0, or this option is not specified, the load operation will continue regardless of the number of warnings issued.

If the load operation is stopped because the threshold of warnings was exceeded, another load operation can be started in RESTART mode. The load operation will automatically continue from the last consistency point. Alternatively, another load operation can be initiated in REPLACE mode, starting at the beginning of the input file.

iHoldQuiesce

Input. A flag whose value is set to TRUE if the utility is to leave the table in quiesced exclusive state after the load, and to FALSE if it is not.

iCpuParallelism

Input. The number of processes or threads that the load utility will spawn for parsing, converting and formatting records when building table objects. This parameter is designed to exploit intra-partition parallelism. It is particularly useful when loading presorted data, because record order in the source data is preserved. If the value of this parameter is zero, the load utility uses an intelligent default value at run time. Note: If this parameter is used with tables containing either LOB or LONG VARCHAR fields, its value becomes one, regardless of the number of system CPUs, or the value specified by the user.

iDiskParallelism

Input. The number of processes or threads that the load utility will spawn for writing data to the table space containers. If a value is not specified, the utility selects an intelligent default based on the number of table space containers and the characteristics of the table.

iNonrecoverable

Input. Set to SQLU_NON_RECOVERABLE_LOAD if the load transaction is to be marked as non-recoverable, and it will not be possible to recover it by a subsequent roll forward action. The rollforward utility will skip the transaction, and will mark the table into which data was being loaded as "invalid". The utility will also ignore any subsequent transactions against that table. After the roll forward is completed, such a table can only be dropped. With this option, table spaces are not put in backup pending state following the load operation, and a copy of the loaded data does not have to be made during the load operation. Set to SQLU_RECOVERABLE_LOAD if the load transaction is to be marked as recoverable.

iIndexingMode

Input. Specifies the indexing mode. Valid values (defined in sqlutil header file, located in the include directory) are:

SQLU_INX_AUTOSELECT: LOAD chooses between REBUILD and INCREMENTAL indexing modes.
SQLU_INX_REBUILD: Rebuild table indexes.
SQLU_INX_INCREMENTAL: Extend existing indexes.
SQLU_INX_DEFERRED: Do not update table indexes.

iAccessLevel

Input. Specifies the access level. Valid values are:

SQLU_ALLOW_NO_ACCESS: Specifies that the load locks the table exclusively.
SQLU_ALLOW_READ_ACCESS: Specifies that the original data in the table (the non-delta portion) should still be visible to readers while the load is in progress. This option is only valid for load appends, such as a load insert, and will be ignored for load replace.

iLockWithForce

Input. A boolean flag. If set to TRUE load will force other applications as necessary to ensure that it obtains table locks immediately. This option requires the same authority as the FORCE APPLICATIONS command (SYSADM or SYSCTRL).

SQLU_ALLOW_NO_ACCESS loads may force conflicting applications at the start of the load operation. At the start of the load the utility may force applications that are attempting to either query or modify the table.

SQLU_ALLOW_READ_ACCESS loads may force conflicting applications at the start or end of the load operation. At the start of the load the load utility may force applications that are attempting to modify the table. At the end of the load the load utility may force applications that are attempting to either query or modify the table.

iCheckPending

This parameter is being deprecated as of Version 9.1. Use the iSetIntegrityPending parameter instead.

iRestartphase

Input. Reserved. Valid value is a single space character ' '.

iStatsOpt

Input. Granularity of statistics to collect. Valid values are:

SQLU_STATS_NONE: No statistics to be gathered.
SQLU_STATS_USE_PROFILE: Statistics are collected based on the profile defined for the current table. This profile must be created using the RUNSTATS command. If no profile exists for the current table, a warning is returned and no statistics are collected.

iSetIntegrityPending

Input. Specifies to put the table into set integrity pending state. If the value SQLU_SI_PENDING_CASCADE_IMMEDIATE is specified, set integrity pending state will be immediately cascaded to all dependent and descendent tables. If the value SQLU_SI_PENDING_CASCADE_DEFERRED is specified, the cascade of set integrity pending state to dependent tables will be deferred until the target table is checked for integrity violations. SQLU_SI_PENDING_CASCADE_DEFERRED is the default value if the option is not specified.

piSourceUserExit

Input. A pointer to the db2LoadUserExit structure.

db2LoadOut data structure parameters

oRowsRead: Output. Number of records read during the load operation.
oRowsSkipped: Output. Number of records skipped before the load operation begins.
oRowsLoaded: Output. Number of rows loaded into the target table.
oRowsRejected: Output. Number of records that could not be loaded.
oRowsDeleted: Output. Number of duplicate rows deleted.
oRowsCommitted: Output. The total number of processed records: the number of records loaded successfully and committed to the database, plus the number of skipped and rejected records.

db2PartLoadIn data structure parameters

piHostname

Input. The hostname for the iFileTransferCmd parameter. If NULL, the hostname will default to "nohost". This parameter is deprecated.

piFileTransferCmd

Input. File transfer command parameter. If not required, it must be set to NULL. This parameter is deprecated. Use the piSourceUserExit parameter instead.

piPartFileLocation

Input. In PARTITION_ONLY, LOAD_ONLY, and LOAD_ONLY_VERIFY_PART modes, this parameter can be used to specify the location of the partitioned files. This location must exist on each database partition specified by the piOutputNodes option.

For the SQL_CURSOR file type, this parameter cannot be NULL and the location does not refer to a path, but to a fully qualified file name. This will be the fully qualified base file name of the partitioned files that are created on each output database partition for PARTITION_ONLY mode, or the location of the files to be read from each database partition for LOAD_ONLY mode. For PARTITION_ONLY mode, multiple files may be created with the specified base name if there are LOB columns in the target table. For file types other than SQL_CURSOR, if the value of this parameter is NULL, it will default to the current directory.

piOutputNodes

Input. The list of Load output database partitions. A NULL indicates that all nodes on which the target table is defined.

piPartitioningNodes

Input. The list of partitioning nodes. A NULL indicates the default.

piMode

Input. Specifies the load mode for partitioned databases. Valid values (defined in db2ApiDf header file, located in the include directory) are:

- DB2LOAD_PARTITION_AND_LOAD: Data is distributed (perhaps in parallel) and loaded simultaneously on the corresponding database partitions.
- DB2LOAD_PARTITION_ONLY: Data is distributed (perhaps in parallel) and the output is written to files in a specified location on each loading database partition. For file types other than SQL_CURSOR, the name of the output file on each database partition will have the form filename.xxx, where filename is the name of the first input file specified by piSourceList and xxx is the database partition number.For the SQL_CURSOR file type, the name of the output file on each database partition will be determined by the piPartFileLocation parameter. Refer to the piPartFileLocation parameter for information about how to specify the location of the database partition file on each database partition.
Note: This mode cannot be used for a CLI LOAD.
DB2LOAD_LOAD_ONLY: Data is assumed to be already distributed; the distribution process is skipped, and the data is loaded simultaneously on the corresponding database partitions. For file types other than SQL_CURSOR, the input file name on each database partition is expected to be of the form filename.xxx, where filename is the name of the first file specified by piSourceList and xxx is the 13-digit database partition number. For the SQL_CURSOR file type, the name of the input file on each database partition will be determined by the piPartFileLocation parameter. Refer to the piPartFileLocation parameter for information about how to specify the location of the database partition file on each database partition.
Note: This mode cannot be used when loading a data file located on a remote client, nor can it be used for a CLI LOAD.
DB2LOAD_LOAD_ONLY_VERIFY_PART: Data is assumed to be already distributed, but the data file does not contain a database partition header. The distribution process is skipped, and the data is loaded simultaneously on the corresponding database partitions. During the load operation, each row is checked to verify that it is on the correct database partition. Rows containing database partition violations are placed in a dumpfile if the dumpfile file type modifier is specified. Otherwise, the rows are discarded. If database partition violations exist on a particular loading database partition, a single warning will be written to the load message file for that database partition. The input file name on each database partition is expected to be of the form filename.xxx, where filename is the name of the first file specified by piSourceList and xxx is the 13-digit database partition number.
Note: This mode cannot be used when loading a data file located on a remote client, nor can it be used for a CLI LOAD.
DB2LOAD_ANALYZE: An optimal distribution map with even distribution across all database partitions is generated.

piMaxNumPartAgents

Input. The maximum number of partitioning agents. A NULL value indicates the default, which is 25.

piIsolatePartErrs

Input. Indicates how the load operation will react to errors that occur on individual database partitions. Valid values (defined in db2ApiDf header file, located in the include directory) are:

DB2LOAD_SETUP_ERRS_ONLY: In this mode, errors that occur on a database partition during setup, such as problems accessing a database partition or problems accessing a table space or table on a database partition, will cause the load operation to stop on the failing database partitions but to continue on the remaining database partitions. Errors that occur on a database partition while data is being loaded will cause the entire operation to fail and rollback to the last point of consistency on each database partition.
DB2LOAD_LOAD_ERRS_ONLY: In this mode, errors that occur on a database partition during setup will cause the entire load operation to fail. When an error occurs while data is being loaded, the database partitions with errors will be rolled back to their last point of consistency. The load operation will continue on the remaining database partitions until a failure occurs or until all the data is loaded. On the database partitions where all of the data was loaded, the data will not be visible following the load operation. Because of the errors in the other database partitions the transaction will be aborted. Data on all of the database partitions will remain invisible until a load restart operation is performed. This will make the newly loaded data visible on the database partitions where the load operation completed and resume the load operation on database partitions that experienced an error.
Note: This mode cannot be used when iAccessLevel is set to SQLU_ALLOW_READ_ACCESS and a copy target is also specified.
DB2LOAD_SETUP_AND_LOAD_ERRS: In this mode, database partition-level errors during setup or loading data cause processing to stop only on the affected database partitions. As with the DB2LOAD_LOAD_ERRS_ONLY mode, when database partition errors do occur while data is being loaded, the data on all database partitions will remain invisible until a load restart operation is performed.
Note: This mode cannot 1be used when iAccessLevel is set to SQLU_ALLOW_READ_ACCESS and a copy target is also specified.
DB2LOAD_NO_ISOLATION: Any error during the Load operation causes the transaction to be aborted. If this parameter is NULL, it will default to DB2LOAD_LOAD_ERRS_ONLY, unless iAccessLevel is set to SQLU_ALLOW_READ_ACCESS and a copy target is also specified, in which case the default is DB2LOAD_NO_ISOLATION.

piStatusInterval

Input. Specifies the number of megabytes (MB) of data to load before generating a progress message. Valid values are whole numbers in the range of 1 to 4000. If NULL is specified, a default value of 100 will be used.

piPortRange

Input. The TCP port range for internal communication. If NULL, the port range used will be 6000-6063.

piCheckTruncation

Input. Causes Load to check for record truncation at Input/Output. Valid values are TRUE and FALSE. If NULL, the default is FALSE.

piMapFileInput

Input. Distribution map input filename. If the mode is not ANALYZE, this parameter should be set to NULL. If the mode is ANALYZE, this parameter must be specified.

piMapFileOutput

Input. Distribution map output filename. The rules for piMapFileInput apply here as well.

piTrace

Input. Specifies the number of records to trace when you need to review a dump of all the data conversion process and the output of hashing values. If NULL, the number of records defaults to 0.

piNewline

Input. Forces Load to check for newline characters at end of ASC data records if RECLEN file type modifier is also specified. Possible values are TRUE and FALSE. If NULL, the value defaults to FALSE.

piDistfile

Input. Name of the database partition distribution file. If a NULL is specified, the value defaults to "DISTFILE".

piOmitHeader

Input. Indicates that a distribution map header should not be included in the database partition file when using DB2LOAD_PARTITION_ONLY mode. Possible values are TRUE and FALSE. If NULL, the default is FALSE.

piRunStatDBPartNum

Specifies the database partition on which to collect statistics. The default value is the first database partition in the output database partition list.

db2LoadNodeList data structure parameters

piNodeList: Input. An array of node numbers.
iNumNodes: Input. The number of nodes in the piNodeList array. A 0 indicates the default, which is all nodes on which the target table is defined.

db2LoadPortRange data structure parameters

iPortMin: Input. Lower port number.
iPortMax: Input. Higher port number.

db2PartLoadOut data structure parameters

oRowsRdPartAgents

Output. Total number of rows read by all partitioning agents.

oRowsRejPartAgents

Output. Total number of rows rejected by all partitioning agents.

oRowsPartitioned

Output. Total number of rows partitioned by all partitioning agents.

poAgentInfoList

Output. During a load operation into a partitioned database, the following load processing entities may be involved: load agents, partitioning agents, pre-partitioning agents, file transfer command agents and load-to-file agents (these are described in the Data Movement Guide). The purpose of the poAgentInfoList output parameter is to return to the caller information about each load agent that participated in a load operation. Each entry in the list contains the following information:

oAgentType: A tag indicating what kind of load agent the entry describes.
oNodeNum: The number of the database partition on which the agent executed.
oSqlcode: The final sqlcode resulting from the agent's processing.
oTableState: The final status of the table on the database partition on which the agent executed (relevant only for load agents).

It is up to the caller of the API to allocate memory for this list prior to calling the API. The caller should also indicate the number of entries for which they allocated memory in the iMaxAgentInfoEntries parameter. If the caller sets poAgentInfoList to NULL or sets iMaxAgentInfoEntries to 0, then no information will be returned about the load agents.

iMaxAgentInfoEntries

Input. The maximum number of agent information entries allocated by the user for poAgentInfoList. In general, setting this parameter to 3 times the number of database partitions involved in the load operation should be sufficient.

oNumAgentInfoEntries

Output. The actual number of agent information entries produced by the load operation. This number of entries will be returned to the user in the poAgentInfoList parameter as long as iMaxAgentInfoEntries is greater than or equal to oNumAgentInfoEntries. If iMaxAgentInfoEntries is less than oNumAgentInfoEntries, then the number of entries returned in poAgentInfoList is equal to iMaxAgentInfoEntries.

db2LoadAgentInfo data structure parameters

oSqlcode

Output. The final sqlcode resulting from the agent's processing.

oTableState

Output. The purpose of this output parameter is not to report every possible state of the table after the load operation. Rather, its purpose is to report only a small subset of possible tablestates in order to give the caller a general idea of what happened to the table during load processing. This value is relevant for load agents only. The possible values are:

DB2LOADQUERY_NORMAL: Indicates that the load completed successfully on the database partition and the table was taken out of the LOAD IN PROGRESS (or LOAD PENDING) state. In this case, the table still could be in SET INTEGRITY PENDING state due to the need for further constraints processing, but this will not reported as this is normal.
DB2LOADQUERY_UNCHANGED: Indicates that the load job aborted processing due to an error but did not yet change the state of the table on the database partition from whatever state it was in prior to calling db2Load. It is not necessary to perform a load restart or terminate operation on such database partitions.
DB2LOADQUERY_LOADPENDING: Indicates that the load job aborted during processing but left the table on the database partition in the LOAD PENDING state, indicating that the load job on that database partition must be either terminated or restarted.

oNodeNum

Output. The number of the database partition on which the agent executed.

oAgentType

Output. The agent type. Valid values (defined in db2ApiDf header file, located in the include directory) are :

DB2LOAD_LOAD_AGENT
DB2LOAD_PARTITIONING_AGENT
DB2LOAD_PRE_PARTITIONING_AGENT
DB2LOAD_FILE_TRANSFER_AGENT
DB2LOAD_LOAD_TO_FILE_AGENT

db2gLoadStruct data structure specific parameters

iFileTypeLen: Input. Specifies the length in bytes of iFileType parameter.
iLocalMsgFileLen: Input. Specifies the length in bytes of iLocalMsgFileName parameter.
iTempFilesPathLen: Input. Specifies the length in bytes of iTempFilesPath parameter.

db2gLoadIn data structure specific parameters

iUseTablespaceLen: Input. The length in bytes of piUseTablespace parameter.

db2gPartLoadIn data structure specific parameters

piReserved1: Reserved for future use.
iHostnameLen: Input. The length in bytes of piHostname parameter.
iFileTransferLen: Input. The length in bytes of piFileTransferCmd parameter.
iPartFileLocLen: Input. The length in bytes of piPartFileLocation parameter.
iMapFileInputLen: Input. The length in bytes of piMapFileInput parameter.
iMapFileOutputLen: Input. The length in bytes of piMapFileOutput parameter.
iDistfileLen: Input. The length in bytes of piDistfile parameter.

Usage notes

Data is loaded in the sequence that appears in the input file. If a particular sequence is desired, the data should be sorted before a load is attempted.

The load utility builds indexes based on existing definitions. The exception tables are used to handle duplicates on unique keys. The utility does not enforce referential integrity, perform constraints checking, or update summary tables that are dependent on the tables being loaded. Tables that include referential or check constraints are placed in set integrity pending state. Summary tables that are defined with REFRESH IMMEDIATE, and that are dependent on tables being loaded, are also placed in set integrity pending state. Issue the SET INTEGRITY statement to take the tables out of set integrity pending state. Load operations cannot be carried out on replicated summary tables.

For clustering indexes, the data should be sorted on the clustering index prior to loading. The data need not be sorted when loading into an multi-dimensionally clustered (MDC) table.

Related information

db2Export API - Export data from a database

db2Import API - Import data into a table, hierarchy, nickname or view

db2LoadQuery API - Get the status of a load operation

LOAD command

sqldcol data structure

sqlu_media_list data structure

dtformat.sqc -- Load and import data format extensions (C)

tbload.sqc -- How to load into a partitioned database (C)

tbmove.sqc -- How to move table data (C)

tbmove.sqC -- How to move table data (C++)

Loading data

参考主题

( 反馈 )

最近一次更新时间 2007-11-12

http://publib.boulder.ibm.com/infocenter/db2luw/v9/topic/com.ibm.db2.udb.admin.doc/doc/r0008704.htm

只有注册用户登录后才能发表评论。


相关文章: DB2 缓冲池调优Bufferpools db2 查看修改端口号 not in/not exists 的 null 陷阱 sql 在not in 子查询有null值情况下经常出现的陷阱 DB2 对于null和''的区别用where in遇到null时的解决方法1 db2之in和exist的性能区别 Unix环境下嵌入式C程序编译 DB2的sqC文件名前八位不能重复的解决办法分布式 DBA: 创建和使用分区表

网站导航: 博客园 IT新闻 BlogJava 知识库博问管理

Prayer