
The Java client accepts control files containing multiple jobs.
jobs
{
job-definition [{job-definition ] ...
}
job-definition:
jobname { load | extract }[options job-options]
{ job-entry-list }
job-options:
(
option_name = option_value [, option_name = option_value
)
job-entry-list:
job-entry [, job-entry ]...
job-entry:
source sourcename target { targetname | ( targetname-list )}
map mapname [ options job-entry-options ]
targetname-list:
targetname [, targetname ]...
job-entry-options
(
option_name = option_value [, option_name = option_value ] ...
)
job-definition
one or more occurrences of job-definition, separated by a comma (,). The last
job-definition should not use the comma separator.
jobname
The name you choose for this job. jobname must be unique within the jobs namespace.
jobname is case-insensitive and cannot exceed 64 bytes.
load | extract
“load” or “extract” means this job is a load job or an extract job, respectively. A load job
is used to load data into SQL tables, while an extract job is used to extract data from SQL
tables.
options
one or more occurrences of job-options, separated by a comma (,). The last job-options
should not use the comma separator.
The options you define at the job level override global options defined in the options section.
The following job options are supported:
• “baddatafile” (page 47)
• “commitsize” (page 47)
• “deleteonerror” (page 48)
• “discards” (page 48)
• “errors” (page 49)
• “faileddatafile” (page 49)
• “operation” (page 52)
• “parallel” (page 52)
• “rowsetsize” (page 53)
• “truncate” (page 54)
Control File Organization and Syntax 43
Commentaires sur ces manuels