5.7.1 column-name statement

Organization of this subsection
(1) Format
(2) Operands
(3) Data storage in the database when the data type specified in the column structure information file does not match the data type in the table
(4) Examples of type operand specifications in the column-name statement for an input data file in pdrorg-output binary format

(1) Format

 column-name[{[,type=data-type

             [,null={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value)}]]

             [,element=elements-count][,elmtype=arrayed-data-format]

             [,nullset=null-value-option][,mode={text|bin}]

             [,enclose_del={yes|no}][,filldata=filler-data]

             [,sequence=([authorization-identifier.]sequence-generator-identifier[,replace=
               {'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')|null|force}])]

         |[,func=([authorization-identifier.]function-name

             ,param=argument-type[,type=data-type

             [,null={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')}]]

             [,mode={text|bin}][,enclose_del={yes|no}]

             [,filldata=filler-data]

             [,param=argument-type[,type=data-type

               [,null={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')}]]

               [,mode={text|bin}][,enclose_del={yes|no}]

               [,filldata=filler-data]]...)]

           }]
           [,filedir=path-name]

Some of the column-name statement's operands are not specifiable depending on the format of input data file. The following table shows whether each operand is specifiable for each format.

Type of specified columnOperandInput data format
DAT formatFixed-size data formatpdrorg-output binary format
Predefined type (column specification)column-nameRRR
typeNRO#3, #7
nullNOO
elementOON
elmtypeOON
nullsetOON
modeNO#4N
enclose_delNO#5N
filldataNON
sequenceOOO
Abstract data type (function specification)column-nameRRR
funcO#1O#1O#1
paramO#2O#2O#2
typeNO#2O#3
nullNOO
modeNO#4N
enclose_delNO#5N
filldataNON
filedirO#6O#6N
Legend:
R: Required
O: Optional
N: Not specifiable
#1
Required if a function is specified.
#2
Required if the func operand is specified with the function.
#3
Specifiable if the data type of the column or the data type specified in param with function specification is CHAR, MCHAR, NCHAR, BLOB, or BINARY. If any one of the following conditions is applicable, make sure that the operand is specified:
  • Loading unload data from a BLOB column to a column of an abstract data type for which the BINARY or BLOB parameter has been specified
  • Loading unload data from the BLOB parameter of an abstract data type to a column of the BLOB or BINARY type
  • Loading unload data from a BLOB column to a BINARY column
  • Loading unload data from the BLOB parameter of an abstract data type to a column of the abstract data type for which the BINARY parameter has been specified
#4
Specifiable only when the type operand value is char and the table definition is CHAR or MCHAR.
#5
Specifiable only when the type operand value is char and the table definition is VARCHAR or MVARCHAR.
#6
Specifiable if the data type of the column is XML and the -K f option is specified.
#7
Specifiable if the data type of the column specified in param is CHAR, MCHAR, NCHAR, BLOB, or BINARY. Specification is mandatory if you expand a character string data type whose column is in the fixed-size data format to the length defined at the destination during data loading.

(2) Operands

(a) column-name

Specifies the name of the table column in which input data is to be stored. Note that a reserved column name cannot be specified. If a reserved column name is specified, pdload issues the KFPL25103-E message and terminates abnormally.

If the column-name is enclosed in double quotation marks ("), it is treated as being case sensitive; otherwise, it is treated as all uppercase letters. If the column-name contains a space, enclose the entire column-name in double quotation marks.

(b) type=data-type

Specifies the data type of the column in the input data. The table below describes the specification of data types and argument types and whether the specification is permitted.

Table 5-42 Specification of data types and argument types and whether the specification is permitted

SpecificationWhether data type can be specifiedCoding rules
Fixed-size data formatpdload unload file
integerYNNone
smallintYN
char(n)#1, #2YYn indicates the number of bytes in the input data, and m indicates the number of characters in the input data.#6
The maximum value of n is the CHAR or MCHAR length in bytes [Figure] 2 (if the data type of the target column is BINARY, the maximum length of the BINARY type in bytes). The maximum value of m is the length of NCHAR in digits. These values need not match the column length defined for the target table. If data is to be converted to the numeric type, do not include any null character (\0).
nchar(m)#1, #2YY
mchar(n)#1, #2YY
varchar(n)YNn indicates the number of bytes in the input data, and m indicates the number of characters in the input data.#6
The maximum value of n is the maximum length of VARCHAR or MVARCHAR in bytes [Figure] 2. The maximum value of m is the maximum length of NVARCHAR in digits. These values need not match the column length defined for the target table.
nvarchar(m)YN
mvarchar(n)YN
floatYNNone
smallfltYN
dec(precision[,decimal-places])YNNone
dateYNNone
interval-year-to-dayYN
timeYN
interval-hour-to-secondYN
yy-mm-dd#3YNYou can use this format only when storing character data indicating the last two digits of the year in a DATE-type column. Any one-byte character can be specified as the separator between the year, month, and date (for example, yy-mm-dd or yy/mm/dd).
yymmdd#3YNYou can use this format only when storing unsigned packed-format data indicating the last two digits of the year in a DATE-type column.
timestamp(p)YNp indicates the length of the seconds fraction part (0, 2, 4, or 6).
binary(n)YY#5If you are specifying this data type in the type operand in the case of pdrorg-output binary format, omit (n).
n indicates the maximum length of the BINARY type in bytes.
binaryprmNYNone
blob(n)YY#4
  • When specifying in the type operand for the fixed-size data format
    n indicates the maximum length of the LOB input file name and may not match the length defined for the LOB column. Use this format for binary-format data that contains its length in the first four bytes.
  • When specifying in the type operand for the pdrorg-output binary format
    Omit (n).
blobprmNYNone
adec(precision[,decimal-places])YNThis input data format specifies the data to be stored in a DECIMAL column as a decimal number in character string format. Because the data does not contain a decimal point, the precision and decimal places must be specified in the type operand to determine the decimal point. For details about the adec data format, see 5.5.3(3)(a) ADEC type (ASCII CODE DECIMAL).
bdec(n)YNThis input data format specifies the data to be stored in a DECIMAL column as binary data. n specifies the length of the binary data. Because the data does not contain a decimal point, the location of decimal point is determined based on the data type of the column. For details about the bdec data format, see 5.5.3(3)(b) BDEC type (BINARY DECIMAL).
Legend:
Y: Can be specified
N: Cannot be specified
#1
Specify a left-justified column value.
#2
If the characters are to be converted to a numeric value, there is no need to left-justify them.
#3
The last two digits of the year are expanded to the format 19yy and then stored.
#4
The data type cannot be specified if the data in the corresponding column has the BLOB type. Do not specify the type statement.
#5
The data type cannot be specified if the data in the corresponding column has the BINARY type. Do not specify the type statement.
#6
The following shows specification examples.
When the character encoding of the input data is UTF-8:
[Figure]
  • Contents of column structure information file:

    C1,type=char(3)

When the character encoding of the input data is UTF-16:
[Figure]
  • Contents of column structure information file:

    C1,type=char(6)

To determine whether or not the input data is stored in the database when the data type specified in the column structure information file does not match the data type of the table column, see (3) Data storage in the database when the data type specified in the column structure information file does not match the data type in the table.

(c) null={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')}

Specifies that the null value is to be stored in specified columns of the table.

Because the column data is compared with the input data, specify a comparison value that corresponds to the data type specified in the type operand.

The available comparison method depends on the type of input data. The following table shows the comparison method for each type of input data:

Input data typeComparison method
Numeric comparisonCharacter comparisonHexadecimal comparison
DECIMALYNY
DATEYNY
INTERVAL YEAR TO DAYYNY
TIMEYNY
INTERVAL HOUR TO SECONDYNY
TIMESTAMPYNY
CHARNYY
VARCHARNYY
NCHARNYY
NVARCHARNYY
MCHARNYY
MVARCHARNYY
BINARYNYY
BLOBNYY
INTEGERYNY
SMALLINTYNY
FLOATYNY
SMALLFLTYNY
yymmdd formatYNY
yy-mm-dd formatNYY
ADECNYY
BDECNNY
Legend:
Y: Can be specified
N: Cannot be specified
Rules
  1. The null value is stored in the columns containing the value specified in this operand.
  2. This operand is not applicable to a column with the NOT NULL constraint, a cluster key component column, or a primary key component column.
  3. For an input data file in pdrorg-output binary format, the null value is specified by setting the offset to 0 for LOB and BINARY column data and the LOB and BINARY parameters of an abstract data type; therefore, this operand cannot be specified for these columns. When these data types have real data, to treat them as the null value by setting the offset to 0, you must first have deleted the real data part from the corresponding input data (simply editing the corresponding offset to 0 does not result in correct processing).
  4. For an input data file in fixed-size data format, the system performs comparison after dividing the data into columns; thus, all trailing spaces are deleted before data comparison.
numeric-comparison-value ~((up to 255 characters))
For numeric input data, this operand specifies the numeric value to be compared.
Rules
  1. The specification method is the same as for numeric data in the DAT format, but do not include spaces.
  2. The numeric attribute applies to the INTEGER, SMALLINT, FLOAT, SMALLFLT, DEC, DATE, TIME, INTERVAL YEAR TO DAY, INTERVAL HOUR TO SECOND and TIMESTAMP data types, and to the yymmdd format.
  3. If a numeric value of TIMESTAMP type is specified and the fraction part is shorter than the defined value, the system assumes 0.
Example
  • DATE: yyyy-mm-dd
    Example: January 1, 1995 [Figure] 1995-01-01
  • TIME: hh:mm:ss
    Example: 12:01:01 [Figure] 12:01:01
  • INTERVAL YEAR TO DAY: yyyymmdd.
    Example: 1,900 years, 1 month, and 1 day [Figure] 19000101.
  • INTERVAL HOUR TO SECOND: hhmmss.
    Example: 12 hours, 1 minute, and 1 second [Figure] 120101.
  • TIMESTAMP: yyyy-mm-dd hh:mm:ss.nnnn
    Example: 1900-01-01 at 01:10:30:45
    [Figure]1900-01-01 01:10:30.4500 or 1900-01-01 01:10:30.45
(begin-position,{c|x}'comparison-value')
For a character comparison value, specify this information when the input data has the character attribute or is binary or LOB data. The character attribute applies to the CHAR, NCHAR, MCHAR, VARCHAR, NVARCHAR, and MVARCHAR data types and to the yy-mm-dd format.
For a hexadecimal comparison value, this information can be specified even if the input data has the numeric attribute (comparison is available with all data types).
begin-position ~<unsigned integer> ((1-32000))
Specifies in bytes the beginning position of the storage data for the target column that is to be compared with the comparison value in order to determine whether or not the input data is treated as the null value.
If the specified comparison start position satisfies the following condition, a control statement error results:
  • In DAT format
    Comparison start position specified for begin-position > 32,000
  • In fixed-size or binary format
    Comparison start position specified for begin-position > definition length of the corresponding column (for a national character string, definition length[Figure] 2)
The system compares the portion of the input data beginning at the specified position with the specified comparison value. If they match, the system treats the input data as the null value. The following shows examples:
Input data of CHAR(5)Result of comparison with null=(2,'XYZ')Handling of input data
*XYZ*Perfectly matches the comparison value.Treated as the null value.
XYZ**Matches the comparison value, but the beginning position does not match.Treated as the input data.
*XY**Does not match the comparison value.
If the charset operand is specified in the option statement, take into account the length of the characters for the comparison start position that is specified for begin-position.
If the comparison value 'XYZ' in UTF-8 character encoding is to be compared with input data in UTF-16 character encoding beginning in character position 2, 3 must be specified as the comparison start position (such as null=(3,c'XYZ')).
c:
Specifies that the comparison value is a character string. Specify a column in HiRDB's default character encoding, even if the input data file has been created using some other character encoding (pdload internally converts to the character encoding used in the input data file).
If the specified comparison value satisfies the following condition, a control statement error results:
  • Length of comparison value in bytes > 255
    If the character encoding used in the input data file is not HiRDB's default character encoding, the length of the comparison value in bytes is the value obtained after conversion.
  • Comparison start position specified for begin-position + length of comparison value in bytes - 1 > definition length of the corresponding column (for a national character string, definition length[Figure] 2)
x:
Specifies that the comparison value is a hexadecimal value. Specify a column in HiRDB's default character encoding, even if the input data file has been created using some other character encoding.
Two hexadecimal characters (0 to f) are treated as 1 byte.
The specified characters are not case sensitive.
If the specified character string does not consist of hexadecimal characters or if the specified comparison value satisfies the following condition, a control statement error results:
  • Length of comparison value in bytes > 254
  • Comparison start position specified for begin-position + length of comparison value in bytes - 1 > definition length of the corresponding column (for a national character string, definition length[Figure] 2)
For a numeric-type column, the comparison value may be the same as the input data. In such a case, specify the comparison value as hexadecimal characters. The following shows examples:
Data type of columnValue treated as the null valueSpecification following the part null=
DECIMALCharacter string NULL(1,x'4e554c4c')
DATE0000-00-00(1,x'00000000​')
INTERVAL YEAR TO DAYByte 1 is 0xff.(1,x'ff')
TIME99:99:99(1,x'999999')
INTERVAL HOUR TO SECONDByte 2 is 0xff.(2,x'ff')
TIMESTAMPBytes 2 and 3 are 0xff.(2,x'ffff')
INTEGER#Character string NULL(1,x'4e554c4c')
SMALLINT#Character string NL(1,x'4e4c')
FLOAT#Character string **NULL**(1,x'2a2a4e554c4c2a2a')
SMALLFLT#Character string NULL(1,x'4e554c4c')
#: A specified hexadecimal comparison value may match some value (in the example of SMALLINT, this comparison value matches the big-endian 20044). To treat the value as the null value, you must specify a value that is not stored.
comparison-value ~((up to 255 characters))
Specifies the characters to be compared with the input data. If you are specifying x, be sure to specify an even number of characters (with a maximum of 254 characters).
(d) element=elements-count

~<1 to number of elements specified for the corresponding column in table definition>

For a repetition column, specify the number of elements to be specified in the input data file.

Specify the element operand if the number of elements specified in the input data file is less than the number of elements specified for the corresponding column during table definition.

If the element operand is omitted, the system assumes the maximum number of elements specified for the corresponding column during table definition.

Rules
  1. If the arrayed data format is the VV format, the system always uses the number of elements specified in the input data file. Therefore, this operand is ignored whether or not it is specified.
  2. An error results in the case of an input data file in pdrorg-output binary format.
(e) elmtype=arrayed-data-format

For a repetition column, specify its array data format.

arrayed-data-format
Specify one of the following formats:
  • ff: FF format
  • fv: FV format
  • vv: VV format
For details about the arrayed data format, see 5.5.4 Arrayed data format for tables containing repetition columns.
Rules
  1. If you omit the elmtype operand and also omit the array statement in the control information file, the system assumes the following format for the arrayed data format:
    • DAT format file: FF format
    • Binary format file: VV format
    • Fixed-size data format file: FF format
  2. An error results in the case of input data file in pdrorg-output binary format.
(f) nullset=null-value-option

This specifies how null value storage is to be handled if a null value is specified from an element through the end of specified elements for an FF arrayed data format.

null-value-option
c: Nothing is stored in specified element.
e: Null value is stored in specified element.
Rules
  1. When the nullset operand is omitted, the system assumes the following value:
    • If an array statement is not specified in the control information file: c
    • If an array statement is specified in the control information file: Value specified in the nullset operand of the array statement
  2. An error results in the case of an input data file in pdrorg-output binary format.
(g) mode={text|bin}

Specifies the input data storage method when the input data is a fixed-length character string (type=char(n)) and the column data type of the storage target is CHAR or MCHAR. This operand is applicable only to the fixed-size data format.

Criteria
To store the following characters as part of the input data, specify bin:
  • Trailing single-byte spaces (0x20), tab characters (\t), and null characters (\0) in the input data
  • Quotation marks (") at both ends of the input data
To delete the above characters from the input data, specify text (default value).
text:
Specifies that the input data is to be edited as follows and then stored:
  • Delete the trailing single-byte spaces (0x20), tab characters (\t), and null characters (\0) from the input data.
  • Delete the quotation marks (") at both ends of the input data.
  • If the input data is shorter than the defined length of the storage column, pad the column with single-byte spaces.
bin:
Specifies that the input data is to be edited as follows and then stored:
  • If the input data is shorter than the defined length of the storage column, pad the column with single-byte spaces.
When bin is specified, space conversion cannot be performed (the spacelvl operand cannot be specified in the option statement).
(h) enclose_del={yes|no}

Specifies the input data storage method when the input data is a fixed-length character string (type=char(n)) and the column data type of the storage target is VARCHAR or MVARCHAR. This operand is applicable only to the fixed-size data format.

Criteria
To store the quotation marks (") at both ends of the input data as part of the input data, specify no.
To delete the quotation marks at the ends of the input data, specify yes (default value).
yes:
Specifies that the input data is to be edited as follows and then stored:
  • Delete the trailing single-byte spaces (0x20), tab characters (\t), and null characters (\0) from the input data.
  • Delete the quotation marks (") at both ends of the input data.
no:
Specifies that the input data is to be edited as follows and then stored:
  • Delete the trailing single-byte spaces (0x20), tab characters (\t), and null characters (\0) from the input data.
To store single-byte spaces, tab characters, and null characters as part of the input data, enclose them in quotation marks.
(i) filldata=filler-data

Specifies the trailing filler data that is to be used when BINARY data is specified as CHAR type in the fixed-size data format. The filler data is 1-byte data in the range 0 to 255 expressed in hexadecimal (X'00" to X'ff'). When this operand is omitted, 0 is assumed.

Rules
  1. When converting data from CHAR to BINARY, the system treats the data without the consecutive trailing fillers as the data section of the BINARY data.
  2. A control statement error results if this operand is specified for an input data file that is not in the fixed-size data format or for a column that is not the BINARY type.
  3. Specifying this operand for a parameter other than a BINARY-type parameter results in a control statement error.
  4. If a file containing BINARY data is specified for the BINARY type parameter, you can use as the fill character filldata, single-byte space, tab, or \0. However, if the charset operand is specified in the option statement, filldata cannot be used (it will be ignored if specified).
(j) sequence=([authorization-identifier.]sequence-generator-identifier[,replace={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')|null|force}])

Specifies that sequence numbers acquired from the sequence generator are to be stored as input data.

[authorization-identifier.]sequence-generator-identifier
Specifies the identifier of the sequence generator from which sequence numbers are to be acquired.
If the authorization identifier is omitted, the user executing pdload (authorization identifier specified in the -u option or the PDUSER environment variable) is assumed.
replace={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')|null|force}
Specifies a condition for replacing input data with a sequence number acquired from the sequence generator.
If input data matches this operand's value, data in the corresponding column is replaced with a sequence number acquired from the sequence generator. You must specify the correct comparison value for the data attribute specified in the type operand because the value is compared based on the data type of the input data.
'numeric-comparison-value'
Specifies a numeric value to be compared when the input data has the numeric attribute.
For the specification method, see the description of the null option in (c) null={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')}.
(begin-position,{c|x}'comparison-value')
Specifies a character string to be compared by using a start position and a comparison value when the input data has the character string attribute.
For the specification method, see the description of the null option in (c) null={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')}.
null
Specifies that the input data is to be replaced with the sequence number if the input data value is the null. This operand is applicable when the input data file is in DAT or extended DAT format.
force
Specifies that the input data is to be replaced with a sequence number unconditionally.
Rules
  1. If you use pdparaload or if you load data in units of RDAREAs, do not specify this operand for partitioning key columns. If you do specify this operand in such a case, a sequence number obtained from the sequence generator will be used as the partitioning key value. As a result, data might not be stored because the corresponding partitioning key value will no longer be subject to storage in the specified RDAREA.
(k) filedir=path-name

Specifies the directory containing a parameter attribute data file when the -K f option is specified to read parameter attribute data in a XML-type column based on one file per data.

Rules
  1. The path name specified here is combined with the file path name in the input data file to obtain the path name of the file to be read as parameter attribute data. The length of the combined file path name must not exceed 1,023 bytes.
  2. Specifying this option when there is no XML-type column results in an error.
  3. If you omit this operand, you must specify the absolute path name of the file to be read as the parameter attribute data in the input data file.
  4. If you specify the absolute path to the storage directory for each column in this operand, you must specify the file path relative to the directory specified in this operand in the input data file.
Example
In this example, the XML type is defined for column 2. The following shows an example of coding in the input data file in DAT format.
  • Data in the input data file

    column-1  column-2     column-3
    00001,xml001.txt,2007-10-01
    00002,xml002.txt,2007-10-02

  • Coding in the column structure information file

    column-1
    column-2,filedir=/user/xml_data
    column-3

  • File referenced by pdload

    /user/xml_data/xml001.txt
    /user/xml_data/xml002.txt

(l) func=([authorization-identifier.]function-name,param=argument-type[,type=data-type[,null={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')}]][,null={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')}]][,mode={text|bin}][,enclose_del={yes|no}][,filldata=filler-data][,...])

If a specified column name has an abstract data type, this operand specifies information about the constructor function that generates the values to be stored.

For data storage in the database when the input parameter specified in the column structure information file has a different data type than the argument of the constructor function to be called, see (3) Data storage in the database when the data type specified in the column structure information file does not match the data type in the table.

[authorization-identifier.]function-name
When generating the data to be stored by calling a constructor function supplied by a plug-in, specify the name of the function.
You cannot specify a function that produces a return value different from the abstract data type of the column to be stored, or that has an abstract data type argument.
If the authorization identifier is omitted, the system assumes the user who defined the abstract data type.
param=argument-type
Specify the data type of the parameter that is to be passed to the constructor function.
If there are multiple parameters, specify them in the order based on the function's input format.
The table below shows the data type of the parameter that is passed to the constructor function and whether the data type can be specified.

Table 5-43 Data type of the parameter that is passed to the constructor function and whether the data type can be specified

Data type of parameterData type specification methodWhether data type can be specified
ClassificationData typeDAT formatFixed-size data formatpdload unload file
Numeric dataINTEGERintegerYYY
SMALLINTsmallintYYY
DECIMALdec()YYY
adec()YYY
bdec()YYY
FLOATfloatYYY
SMALLFLTsmallfltYYY
Character string dataCHARACTERcharYYY
VARCHARvarcharYYY
National character dataNCHARncharYYY
NVARCHARnvarcharYYY
Mixed character string dataMCHARmcharYYY
MVARCHARmvarcharYYY
Date dataDATEdateYYY
yy-mm-ddNNN
yymmddNNN
Time dataTIMEtimeYYY
Timestamp dataTIMESTAMPtimestampYYY
Date interval dataINTERVAL YEAR TO DAYinterval-year-to-dayYYY
Time interval dataINTERVAL HOUR TO SECONDinterval-hour-to-secondYYY
Large object dataBLOBblobYYY
blobprmNNN
Binary dataBINARYbinaryYYY
binaryprmNNN
Legend:
Y: Can be specified.
N: Cannot be specified.
type=data-type
If the input data file is a fixed-size data format file, specify the data type of the parameter to be input. If there are multiple parameters, specify them in the order based on the function's input format.
The table below shows the data type specification method and whether the data type can be specified.

Table 5-44 Data type specification method and whether the data type can be specified

ClassificationData typeData type specification methodWhether data type can be specified
Fixed-size data formatpdload unload file
Numeric dataINTEGERintegerYN
SMALLINTsmallintYN
DECIMALdec(precision[,decimal-places])YN
adec(precision[,decimal-places])YN
bdec(n)YN
FLOATfloatYN
SMALLFLTsmallfltYN
Character string dataCHARACTERchar(n)YN
VARCHARvarchar(n)YN
National character dataNCHARnchar(m)YN
NVARCHARnvarchar(m)YN
Mixed character string dataMCHARmchar(n)YN
MVARCHARmvarchar(n)YN
Date dataDATEdateYN
yy-mm-ddNN
yymmddNN
Time dataTIMEtimeYN
Timestamp dataTIMESTAMP[(p)]timestamp(p)YN
Date interval dataINTERVAL YEAR TO DAYinterval-year-to-dayYN
Time interval dataINTERVAL HOUR TO SECONDinterval-hour-to-secondYN
Large object dataBLOBblob(n)#YY
blobprmNY
Binary dataBINARYbinary(n)#YY
binaryprmNY
Legend:
Y: Can be specified.
N: Cannot be specified.
#: For a pdload unload file, (n)can be omitted.
null={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')}
For an input data file in the fixed-size data format, specify this operand to handle the input parameter as the null value.
The input parameter's data type is subject to comparison. For the specification method, see (c) null={'numeric-comparison-value'|(begin-position,{c|x}'comparison-value')}, previously.
mode={text|bin}
For details about this specification, see (g) mode={text|bin}.
enclose_del={yes|no}
For details about this specification, see (h) enclose_del={yes|no}.
filldata=filler-data
For details about this specification, see (i) filldata=filler-data.
Specification order in func operand:
The following shows the order in which param, type, null, filldata, mode, and enclose_del are specified in the func operand:
  • In the fixed-size data format

    [Figure]

  • In the binary format

    [Figure]

  • In the DAT format

    [Figure]

(3) Data storage in the database when the data type specified in the column structure information file does not match the data type in the table

Table 5-45 Data storage in the database when the data type specified in the column structure information file does not match the data type in the table (1/2) and Table 5-46 Data storage in the database when the data type specified in the column structure information file does not match the data type in the table (2/2) show whether input data is stored in the database when a data type specified in the column structure information file does not match the data type in the table.

Table 5-45 Data storage in the database when the data type specified in the column structure information file does not match the data type in the table (1/2)

Data type specified in column structure information fileData type in table
INSICHNCMCVCNVMVFLSM
integerY------------------
smallint--Y----------------
charYYY#1YYYYYYY
nchar------Y#1------------
mchar--------Y#1----------
varchar----------Y#2--------
nvarchar------------Y------
mvarchar--------------Y#2----
float----------------Y--
smallflt------------------Y
dec--------------------
date--------------------
interval-year- to-day--------------------
time--------------------
interval-hour- to-second--------------------
timestamp--------------------
binary--------------------
yy-mm-dd--------------------
yymmdd--------------------
blob--------------------
adec--------------------
bdec--------------------
Legend:
IN: INTEGER
SI: SMALLINT
CH: CHAR
NC: NCHAR
MC: MCHAR
VC: VARCHAR
NV: NVARCHAR
MV: MVARCHAR
FL: FLOAT
SM: SMALLFLT
Y: Can be stored.
--: Cannot be stored.
#1
Data is stored according to the length of input data of the fixed-length character string data type and the definition length for the storage table as follows:
No.Specification of character set for target columnRelationship between n and mData storage method
1Omittedn = mThe input data is stored as is in the database.
2n > mAn error results.
3n < mThe data following (n + 1) is padded with space characters.
4Specifiedn' = mThe input data is stored as is in the database.
5n' > mThe input data obtained after character encoding conversion is truncated to fit the definition length of the target, and then is stored in the database. For details, see Table 5-17 Relationships among input data length, defined column length, and lengover operand.
6n' < mThe data following (n + 1) is padded with space characters.
#2
Data is stored according to the length of input data of the variable-length character string data type and the definition length for the storage table as follows:
No.Specification of character set for target columnRelationship between n and mData storage method
1Omittedn[Figure] mThe input data is stored as is in the database.
2n > mThe input data is truncated to fit the definition length of the target, and then is stored in the database. For details, see Table 5-17 Relationships among input data length, defined column length, and lengover operand.
3Specifiedn'[Figure] mThe input data is stored as is in the database.
4n' > mThe input data obtained after character encoding conversion is truncated to fit the definition length of the target, and then is stored in the database. For details, see Table 5-17 Relationships among input data length, defined column length, and lengover operand.
Legend:
n: Input data length
m: Definition length of the target table
n': Data length after character encoding conversion

Table 5-46 Data storage in the database when the data type specified in the column structure information file does not match the data type in the table (2/2)

Data type specified in column structure information fileData type in table
DECDATEIYTDTIMEIHTSTMSTBINBLOB
integer----------------
smallint----------------
charYYYYYYYY
nchar----------------
mchar----------------
varchar----------------
nvarchar----------------
mvarchar----------------
float----------------
smallflt----------------
decY--------------
date--Y------------
interval-year-to-day----Y----------
time------Y--------
interval-hour-to-second--------Y------
timestamp----------Y----
binary------------Y--
yy-mm-dd--Y------------
yymmdd--Y------------
blob------------Y#Y
blobprm------------Y#Y#
adecY--------------
bdecY--------------
Legend:
DEC: DECIMAL
DATE: DATE
IYTD: INTERVAL YEAR TO DAY
TIME: TIME
IHTS: INTERVAL HOUR TO SECOND
TMST: TIMESTAMP
BIN: BINARY
BLOB: BLOB
Y: Can be stored.
--: Cannot be stored.

#: Data can be stored for an input data file in pdrorg-output binary format.

(4) Examples of type operand specifications in the column-name statement for an input data file in pdrorg-output binary format

The following table provides examples of type operand specifications in the column-name statement for an input data file in pdrorg-output binary format:

Input data typeData type after conversion
BLOB typeBLOB parameterBINARY typeBINARY parameter
BLOB type--func=(F_BLOB,
param=blob)
type=blobfunc=(F_BIN,
param=binary,
type=blob)
BLOB parametertype=blobprmfunc=(F_BLOB,
param=blob)
type=blobprmfunc=(F_BIN,
param=binary,
type=blobprm)
BINARY typeNN--N
BINARY parameterNNNfunc=(F_BIN,
param=binary)
Legend:
--: Specification is not necessary.
N: The corresponding input data type cannot be converted to the indicated data type.
F_BIN: This is a plug-in-provided constructor function of an abstract data type that takes BINARY type as a parameter.
F_BLOB: This is a plug-in-provided constructor function that takes BLOB type as a parameter.