Ensure Job Compilation Parity Across DataStage Interfaces

Problem:
There should be functional parity in BASIC code generation across all DataStage job compilation methods, including:

• Thick client (legacy Designer)
• DataStage Flow Designer (DFD)
• Server-side / REST API compilation

Currently, jobs can compile successfully using the legacy Windows-based Designer but fail when compiled through DFD or server-side compilation methods.

Example:

We are using the built-in Transformer function DIGITS inside a Sequence job.

Compilation results:
• Legacy Designer (Thick Client): Compiles successfully
• DataStage Flow Designer (DFD): Compilation fails

Compilation error:

Array 'DIGITS' never dimensioned.

We opened a support case and IBM responded:

"Sequencers are not Parallel but Server jobs. When Server jobs are compiled, BASIC code is generated. The thick client and DFD use different mechanisms to generate the BASIC code. DFD compilation does not support all BASIC functions like the thick client, so the reported compilation error is expected behavior."

This suggests Sequence job functionality is not fully supported in DFD, however the official documentation states Sequence jobs are supported:
https://www.ibm.com/docs/en/iis/11.7.0?topic=designer-sequence-jobs

This inconsistency creates confusion for customers attempting to modernize development workflows.

Similar issues appear in other enhancement requests including:
ISINT-I-839
ISINT-I-743

The common response has been to continue using the thick client. This is not a viable long-term solution.

Impact:

Organizations attempting to standardize on web-based development tooling in Cloud Pak for Data struggle to do so if jobs compile differently between the platforms.

As long as these inconsistencies exist, customers must maintain the legacy Windows-based Designer even when adopting web-based tooling and REST APIs.

While thick client compilation can technically be automated, it requires unnecessary platform dependencies, increases operational complexity, and prevents CI/CD pipelines from being fully implemented using supported server-side or API-based tooling.

Requested Enhancement:

Compilation behavior should be consistent across all DataStage interfaces. Jobs should compile the same way regardless of whether compilation is invoked through:

• Legacy Designer (thick client)
• DataStage Flow Designer
• Next-Gen UI
• REST/API invocation

If architectural limitations prevent a unified compilation mechanism, IBM should provide documentation on differences between compilation paths, including:

• A list of unsupported Server BASIC functions when compiling outside of the thick client
• Supported alternative approaches for affected functionality
• Compilation warnings when jobs use functions that are unsupported or incompatible across the thick client, DataStage Flow Designer, Next-Gen UI, or REST/API compilation paths

This information should be included in the official product documentation so customers can design jobs that are compatible with supported compilation paths.