Sort Process Object

The Integrator Sort process object sorts the data flow based on a set of columns, changing the order in which the data appears in subsequent objects in the data flow. The Sort object will sort the data in memory if possible. Otherwise it will use temporary files to perform an external sort.

The sort preserves all input rows without collapsing or otherwise combining the data. Combining data after a sort can be accomplished by using the Squash process object.

Sort Attributes

Attribute	Type	Description
process_type (required)	String	Identifies the object as a Sort process object. The value of this string is "sort".
input (required)	String	Defines the object from which the data flow is arriving.
sort_columns (required)	Array of Strings	Defines the input columns that are used as keys in the sort, in the order they control the sort. The primary key appears first, the secondary key appears second, etc. If rows match on all sort column values, their order in the output will be arbitrary, that is, sort order for duplicates is undefined. If order stability is required, add more columns to the sort.
sort_size	Integer	Defines the approximate amount of memory assigned for the sort, in bytes. If the amount of data exceeds this memory, the sort object will switch to an external sort, using a set of temporary files for intermediate results. This attribute is optional; the default sort_size is 512,000,000 (512 megabytes) when using the 64-bit Integrator and 128,000,000 (128 megabytes) when using the 32-bit Integrator. Specify it without commas or quotes, for example, sort_size=250000000. The maximum sort sized is limited to 2,147,483,647 bytes (approximately 2GB). NOTE: This attribute is Sort Size in Visual Integrator.
numeric_columns	Boolean	Defines which sort columns are treated as numeric values. Values in these columns will be compared numerically, instead of alphabetically. NOTE: This attribute is Numeric in Visual Integrator.
temp_directory	String	Defines the directory which will be used to hold the temporary files for the sort. These files will be named: `s<name>.tmp` where <name> consists of the process ID of the Integrator followed by incrementing letters. Each execution of an Integ script has its own unique PID. Hence, when run simultaneously, each script has its own unique set of sort files. The temporary files are automatically deleted when the Integrator is successfully finished. This attribute is optional, but it is good practice to specify a temporary location known to have enough space for the sort. NOTE: This attribute is Temp Directory in Visual Integrator.
reverse	Boolean	Controls whether or not the sort is performed in reverse order. If the reverse attribute is present and not "false", then the sort will be performed in reverse order.
reverse_columns	Array of Strings	Defines which sort columns are sorted in reverse order. The values in these individual columns will be sorted in descending order unless modified by the row-level reverse attribute. The reverse_columns attribute and the reverse attribute are combined with an exclusive OR. If a column is listed in reverse_columns and the reverse attribute is "true", the reverse is applied twice and the column is sorted in ascending order. NOTE: This attribute is Reverse All in Visual Integrator.
locale	Boolean	Controls whether or not the sort comparisons are done using the locale collation sequence, to reflect international sort sequences. If set to "false" sort comparisons are not done using the locale collation sequence. If set to "true", comparisons are done using the current locale (as defined by the system). When set to "true", this attribute overrides the case_sensitive attribute. The default is "false". The sort order varies by platform. For example, Windows and UNIX use ASCII characters, while the OS/400 uses EBCDIC characters.
case_sensitive	Boolean	Controls whether or not the sort comparisons are case sensitive. If this attribute is "true", then the comparisons are done in a case sensitive fashion. This attribute is optional, but defaults to "true". (See locale attribute above.) NOTE: This attribute is Case Sensitive in Visual Integrator.
trace_after	Sub-object	Traces data flows leaving the specified object, which makes debugging scripts easier. This is equivalent to adding a Trace process object immediately after the current object. See Embedded Trace Object for more on using trace sub-objects.
trace_before	Sub-object	Traces data flows entering the specified object. This is equivalent to adding a Trace process object immediately before the current object. See Embedded Trace Object for more on using trace sub-objects.