SQL/MX 3.2 Management Manual (H06.25+, J06.14+)

• import automatically performs constraint validations while loading table rows. Performance
considerations constraint validation are described in “Managing Constraints to Improve import
Performance” (page 197).
• import automatically performs triggered actions while loading table rows. Performance
considerations for performing triggered actions are described in “Managing Triggers to
Improve import Performance” (page 197).
Summary of import Options
Table 14 (page 192) describes the most important import command options.
Table 14 import Command Options
Descriptionimport Option
Disables all triggers before the data is inserted, which prevents trigger actions from occurring.-D
Specifies the first row and the number of rows to import from the input file. The first row is
designated 0 (zero). import terminates when the number of rows specified has been inserted
or when it reaches the end of the input file.
-F and -C
Specifies the name of the input file that contains the data to be imported. The input file must be
an OSS or Guardian text file. Guardian file names must be specified in OSS format (for example,
/G/DATA01/subvolume/file-name).
-I
Specifies the name of the DataLoader/MX process that can provide ASCII data to import.-IP
This option allows you to ignore some parsing and execution errors and to automatically generate
the error log filenames and error message log filenames.
-LES
Specifies the pathname parsing-errormsg-filename of an OSS output file to which
import logs the error messages that correspond with the logged data rows that have parsing
-PM
errors. The name of the error message file must be specified in OSS format, for example
/usr/jdoe/errfile, may be the same as the one specified with the -E option, but must
not be the same as any other output file.
Specifies the name summary-filename of a non-audited, OSS unstructured file (or Guardian file
of type 180 if the filename is under /G) to which import writes the number of input rows read
-SF
so far, the number of those input rows that were skipped (either due to the use of the -F
<first-row> option or because they were comment lines), the elapsed time since starting to read
the first input row, and the elapsed time since skipping the initial rows specified via the -F
<first-row> option.
Specifies the number of rows (1 to 2147483647) that import must read from the input file
between updates of the information written to the summary file. If this option is not specified,
-SI
the default shall be to write the results summary information only at the completion of the import
operation.
Specifies the number of records to process before a commit. If import returns an error before
the record count reaches the number specified with this option, all changes to the database
-T
within that transaction are rolled back. If the error occurs after the record count has surpassed
the number of records specified, all rows up to failure, minus the -T option value, are rolled
back in the table.
Specifies the name of the file that contains format specifications for the input file. The format file
must be an OSS or Guardian text file. The Guardian file name must be specified in OSS format.
The format file cannot exceed 60,000 bytes.
nl
-U
The format file specified with -U cannot contain a circumflex (^) character or a comma (,) and
cannot contain a minus sign (-) as the first character of the name.
Specifies the input file type. The possible values for the input file type are DELIM and FIXED. If
you do not specify this parameter, import assumes the default DELIM file type (a delimited
-W
input file). If you do not specify this parameter and there is no format file, import assumes the
default delimiters for a delimited input file.
Specifies the pathname exec-error-filename of an OSS output file to which import logs
data rows that have execution errors. After the rows in the error file are edited to resolve the
-XE
192 Reorganizing SQL/MX Tables and Maintaining Data