Skip to content

Widen production columns that are narrower than the Hibernate mapping #92

Open

Description

@fjlopez

Production Liquibase columns are shorter than the Hibernate mapping. Hibernate ddl-auto=validate still starts. These rows are INFO, not PROBLEM.

Measured 2026-10-19 with bash tools/scripts/compare-schema-drift.sh both --against jpa --fail-on none. Postgres: PROBLEM 0, INFO 60. Oracle: PROBLEM 0, INFO 63. STM_PAR_GS is not in the report. oracle_native_boolean is not in the report.

Character columns, both dialects (length_narrower_fixable):

  • STM_TOKEN_USER.CODE_OTP 150 < 255
  • STM_TREE.TRE_IMAGE_NAME 250 < 255
  • STM_TREE_NOD.TNO_STYLE 50 < 255
  • STM_TREE_NOD.TNO_VIEW_MODE 50 < 255

Oracle also reports STM_SERVICE_CHECK.SEC_OBSERVED 11 < 13 and STM_SERVICE_CHECK_SAMPLE.SCS_OBSERVED 11 < 13. Those are the timestamp type difference, tracked separately, not character columns to widen here.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    plannedActive backlog; exempt from stale bot

    Type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions