Edit Question - KnowledgeHub

Title Body

We're using a JPA annotated type that looks like this (groovy code): @Entity @EqualsAndHashCode class TextNote extends Serializable { @Id Long id String text } When it was first written I was very new to JPA and wrote the SQL first, then made the annotated classes match the SQL. Reading up on PostgreSQL it seemed like the following was the table I wanted: CREATE TABLE textnote ( id bigint NOT NULL, text text ); This worked, and we had tables that looked like this: id | text -----+------------------------ 837 | really long text here What I want to do now is correct the JPA Entity to look like this: @Entity @EqualsAndHashCode class TextNote extends Serializable { @Id Long id @Lob String text } By adding the @Lob annotation the JPA provider (in my case, hibernate) could generate the DDL correctly for me in case we want to swap out databases. It also documents exactly what I want the text field to be. Now when a note gets created I see something like this: id | text -----+------------------------ 837 | 33427 Which is fine for new notes, as when I read it in code using String getText() it returns the really long text. Honestly I don't know how PostgreSQL implements the text type, nor should I need to in theory. However our production database already has many notes stored using old code without the @Lob annotation. Running the new code on an existing database generates issues like this: org.springframework.dao.DataIntegrityViolationException: Bad value for type long : not a number this time; SQL [n/a]; nested exception is org.hibernate.exception.DataException: Bad value for type long : not a number this time For existing notes, is there a way in SQL to migrate the old notes to use

Tags (comma-separated) Cancel