[28.x] Classify restored E-Document table sensitivities

[Milica Đukić (djukicmilica)]

Why

The restored E-Document message infrastructure introduces tables whose runtime data sensitivity remains unclassified. This causes Data Classs Demo Data Tests.TestDataSensitivities to fail across localization test buckets.

This backports the classifications from #11170 and #11276 to releases/28.x.

Summary

• Classified E-Document Message fields as normal sensitivity data.
• Classified E-Doc. External Reference and E-Doc. Payment Occurrence fields as normal sensitivity data.

Fixes
AB#648705
AB#648955

[github-actions]

Could not find linked issues in the pull request description. Please make sure the pull request description contains a line that contains 'Fixes #' followed by the issue number being fixed. Use that pattern for every issue you want to link.

[alexei-dobriansky]

Good Sense Reviewer - Round 1

Recommendation: Accept

What this PR does

This change adds the three restored E-Document message tables to the data sensitivity initialization list in EDocumentSubscribers.Codeunit.al. The subscriber already classifies sensitive E-Document fields through DataClassificationEvalData.SetTableFieldsToNormal; adding tables 6432, 6433, and 6434 makes the restored CustomerContent fields use the same path. I checked the surrounding method and the analogous PEPPOL subscriber pattern; the added calls are in the right place and no follow-on code changes the classification result.

Problem-solution fit

Fit: Strong

The reported failure is an unclassified sensitive field on a restored E-Document table. The diff covers that table and the two related message/reference tables in the same message area, so the solution matches the validation issue without adding runtime behavior.

Suggestions

None.

Risk assessment and necessity

Risk: The regression surface is narrow: company data sensitivity setup for the E-Document app. The change does not alter posting, payment calculation, data writes, public APIs, or event signatures. The changed code only extends an existing classification list, and the current PR checks show the relevant W1 build and test jobs passing.

Necessity: The change is needed because restored E-Document message data left sensitive fields unclassified during validation. The scope is right: all three restored message-area tables are included, and no unrelated E-Document tables or behavior are changed.

---

[AI-PR-REVIEW] version=1 promptVersion=4 system=github pr=11278 round=1 by=alexei-dobriansky at=2026-09-09T22:21:24Z lastSha=899cf90f37af39fdb0011df84e6bf51af0e99abc reviewKey=4485d4211e3d719a1549bfb7563eb9f900936e6b7f9a393aa68405d23767a665 suggestions=none