Informix tests fixes

[VladoKuruc]

Fixes for some of the Informix tests

---

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license
and can be relicensed under the terms of the LGPL v2.1 license in the future at the maintainers' discretion.
For more information on licensing, please check here.

---

[hibernate-github-bot]

Thanks for your pull request!

This pull request does not follow the contribution rules. Could you have a look?

❌ All commit messages should start with a JIRA issue key matching pattern HHH-\d+
    ↳ Offending commits: [449fa85]

› This message was automatically generated.

[mbladel]

I'm not sure we want to spread special cases across the main test suite for a community-supported dialect.

[VladoKuruc]

I'm not sure we want to spread special cases across the main test suite for a community-supported dialect.

I believe it is legitimate for tests to also consider community dialects. Especially if the database is also available in Docker and it is documented in the README. What do you think about that @beikov ?"

[VladoKuruc]

I suggest including community dialect tests in CI build, but configured to allow the build to continue even if these tests fail. This would enable to monitor the evolution of dialect support and detect regressions without jeopardizing the overall CI process.

[mbladel]

@VladoKuruc as I already said, your PR looks fine but I'm reluctant to special-case community dialects in the core test suite. Also, testing community-supported dialects is up to the user / vendor that maintains them and using CI would consume resources. However, I can bring this up with the team and see what they think of it.

Note: you can just skip the failing tests locally while they are not yet implemented / supported by the dialect you're testing

[beikov]

Actually, I believe that using the default length for LONGVARCHAR which should be Length.LONG is on purpose. Why do you have to change the size here? If varchar has a max capacity, it should be handled by implementing Dialect#getMaxVarcharLength()

[VladoKuruc]

I understand that the maximum length for the LONGVARCHAR type is set in the getMaxVarcharLength() method, which is correctly set to 32739 for Informix. The problem is that Informix also has a row length limit. The Forest entity is only used in the test org.hibernate.orm.test.annotations.entity.BasicHibernateAnnotationsTest, where I don't think the column length is critical.

create table Forest (
        id integer not null,
        length int8 not null,
        longDescription lvarchar(32600),
        bigText varchar(255),
        name varchar(255),
        smallText varchar(255),
        country blob,
        primary key (id)
    )" via JDBC [The operation causes a rowsize to exceed the allowable limit (32767).]

Of course, I could ignore these tests for InformixDialect, but I think more coverage is beneficial.

[beikov]

Other than what I commented, this looks ok to me.