Different database servers use different ways to quote and escape identifiers.
E.g. \"foo bar\" vs `foo bar` vs [foo bar], or \"10\"\"\" vs \"10\\\"\", or identifier
Have a look at
DatabaseMetaData.getIdentifierQuoteString()
I never used it but it sounds good :-)
getExtraNameCharacters()
could also be of some help
I think the answer to your question is that if you are writing a database neutral application using JDBC, then you need to use database neutral names, and not things that require special escaping per database vendor.
There is nothing I know of in the JDBC which supports that. A ORM product will deal with such things.
Edit: If you are writing an ORM, then I would think need a seperate SQL generation class for each supported database, just to handle the various syntax involved, so you would have to write that. You can certainly look at the source code of the various open source ORM's out there and see how they handle it.
Since Java 9, the Statement interface provides various methods for engine-specific quoting:
enquoteIdentifier
for SQL identifiers (e.g. schema, table, column names)enquoteLiteral
for string literals (e.g. char, varchar, text literals)enquoteNCharLiteral
for National Character Set literalsStatement stmt = connection.createStatement();
String query = String.format(
"SELECT id FROM %s WHERE name = %s",
stmt.enquoteIdentifier("table", false),
stmt.enquoteLiteral("it's"));
ResultSet resultSet = stmt.executeQuery(query);
However, whenever possible (i.e. for values in CRUD queries), use prepared statements instead.
Statement stmtFormat = connection.createStatement();
String query = String.format(
"SELECT id FROM %s WHERE name = ?",
stmtFormat.enquoteIdentifier("table", false);
PreparedStatement stmt = connection.createPreparedStatement(query);
stmt.setString(1, "it's");
ResultSet resultSet = stmt.executeQuery();