The implementation of the equals method is not Hibernate-safe

[jperezdelafuente]

Overview of the issue

Jhipster generates the equals methods like this:

@Override
    public boolean equals(Object o) {
        if (this == o) {
            return true;
        }
        if (!(o instanceof Foo)) {
            return false;
        }
        return id != null && id.equals(((Foo) o).id);
    }

Motivation for or Use Case

This implementation is problematic with entities wrapped in hibernate proxies:
https://blog.andrewbeacock.com/2008/08/how-to-implement-hibernate-safe-equals.html

In the hasCode() methods generated by Jhipster it appears

@Override
    public int hashCode() {
        // see https://vladmihalcea.com/how-to-implement-equals-and-hashcode-using-the-jpa-entity-identifier/
        return getClass().hashCode();
    }

but that implementation:
https://vladmihalcea.com/how-to-implement-equals-and-hashcode-using-the-jpa-entity-identifier/
is not respected with getId in the equals methods generated by jhipster

Suggest a Fix

A correct implementation for hibernate proxies would be (with "getId()" instead of "id"):

@Override
    public boolean equals(Object o) {
        if (this == o) {
            return true;
        }
        if (!(o instanceof Foo)) {
            return false;
        }
        return id != null && id.equals(((Foo) o).getId());
    }

JHipster Version(s)

Jhipster 8.0.0 - beta 3

• Checking this box is mandatory (this is just to show you read everything)

[mshima]

@jperezdelafuente can you contribute with a PR?

[vandenn3]

I also got a bug because of this implementation issue in generated equals. Thanks for creating this issue!

I see however that this change was already done in the past in #5604 , but then changed back here 1973fc6#diff-b2f9dbf1a993c204d3446d607494894a72f73dfb38228358c8e97b4d9d13da1c related to #8656 . I'm not sure that the rollback was intentional, as the goal of the fix for the issue 8656 was to allow comparison with subclasses (so getClass() was replaced by instanceof). I don't see any reason that would explain why the getters were removed in the fix of issue 8656.