SCLM provides support for Build translators that operate on more than one source member in a single invocation. This type of translator is known as an input list translator. SCLM users can use existing translators that support this feature or write new user-defined translators to take advantage of the feature. The IBM® Ada/370 Compiler is the only SCLM-supported translator that can use input lists.

The SCLM Input List feature can increase the performance of an SCLM Build. Instead of SCLM calling a translator once for each member to be built, SCLM calls the translator passing a list of members to be built. SCLM attempts to place as many members as possible on each input list, thereby limiting the number of translator invocations. The project manager specifies the maximum number of members passed to a translator on an invocation in the language definition that includes the translator. This feature is most useful when using translators that have a high startup overhead to run. Fewer invocations mean increased speed for the SCLM Build process.

An input list translator receives a file that contains a list of data sets that a Build action is performed against. It returns a file that contains a return code for each data set in the input list and, optionally, a set of unique outputs for each data set in the input list.

Two translators, FLMTPRE and FLMTPST, serve as the interfaces between SCLM and the input list translator.

• The FLMTPRE translator generates a list of data sets that an input list translator can use as input.
• The FLMTPST translator passes the return code information that an input list translator provides for every data set on the input list back to SCLM.

For more information, refer to FLMTPRE and FLMTPST.