Use force-quoting in `R2dbcMappingContext` by default

[ShenFeng312]

I have a table where the field is desc. Since it's a reserved keyword in PostgreSQL, I had to use the annotation @Column("\"desc\"") to allow it to be written properly. However, when reading the data, I found that it cannot be read correctly because in org.springframework.data.relational.core.conversion.MappingRelationalConverter#read, the org.springframework.data.relational.core.conversion.RowDocumentAccessor#hasValue method doesn't recognize the field. I believe some logic should be added here to solve this issue.

[ShenFeng312]

can we add a annotation like @ResultSetKey("desc") or other way to support it @mp911de @schauder

[schauder]

With @Column("\"desc\"") you make the double quotes part of the column name. Is that intended? SQL identifier in @Column, @Table and similar annotations are used exactly as they are given, and properly quoted to avoid interpretation as key words, or automatic changes to upper or lower case.

Therefore I guess you really want to use @Column("desc").

If that still doesn't work, please provide a full reproducer, preferably base on https://github.com/schauder/issue-jdbc-1993

[ShenFeng312]

@schauder Thank you for your help, and sorry I forgot to mention that this bug occurs under r2dbc. I tried using the demo provided in your repository for testing, but I’m not sure how to use TestcontainersConfiguration. I believe when using r2dbc, saving data causes an error. The logs I printed indicate that it is not treating field and table names as keywords.

[ShenFeng312]

The cause of the bug seems to be that the default values provided by org.springframework.data.r2dbc.mapping.R2dbcMappingContext are inconsistent with JDBC. Should I submit a PR to fix this issue?

	public R2dbcMappingContext(NamingStrategy namingStrategy) {
		super(namingStrategy);
// use setForceQuote(true) to support keyword fields
		setForceQuote(false);
	}

[mp911de]

We initially didn't want to enable forceQuote for R2DBC to retain compatibility. It would make sense to align for the next major revision, so feel free to submit a pull request. I assume that updating all tests will be the major effort here.