Re: Breaking change in CLOB columns in PostgreSQL coming from DN 5.2.2
Define what you had before and what you have now.
Was the String field stored in a CLOB column (defined in the generated DDL as "CLOB") ? and presumably it was previously read/written using ClobColumnMapping?
And now (current code), it is read/written using LongVarcharColumnMapping or ClobColumnMapping?
Yes, it would be better if JDBC drivers actually provided proper support for all types, so when a user says CLOB they get a JDBC CLOB, not something else behind the scenes, which is the source of manu such "problems".
FWIW I have a field marked as jdbcType="CLOB" and using PostgreSQL v13 using JDBC v42.2. It creates a field of type "text" and uses LongVarcharColumnMapping. Read and write works fine.. Maybe you have something different to that ...
A project where there is but one person to maintain it and who uses a subset of what is available will cater for what they see best, and that won't necessarily include your projects requirements, but that is why this project has encouraged people to actually get involved and contribute tests etc so that their requirements are met.