|previous : next|
Applies to ✅ Open Source Edition ✅ Express Edition ✅ Professional Edition ✅ Enterprise Edition
jOOQ supports the asterisk operator in projections both as a qualified asterisk (through Table.asterisk()) and as an unqualified asterisk (through DSL.asterisk()). It is also possible to omit the projection entirely, in case of which an asterisk may appear in generated SQL, if not all column names are known to jOOQ.
Whenever jOOQ generates an asterisk (explicitly, or because jOOQ doesn't know the exact projection), the column order, and the column set are defined by the database server, not jOOQ. If you're using generated code, this may lead to problems as there might be a different column order than expected, as well as too many or too few columns might be projected.
// Explicitly selects all columns available from BOOK - No asterisk create.select().from(BOOK).fetch(); // Explicitly selects all columns available from BOOK and AUTHOR - No asterisk create.select().from(BOOK, AUTHOR).fetch(); create.select().from(BOOK).crossJoin(AUTHOR).fetch(); // Renders a SELECT * statement, as columns are unknown to jOOQ - Implicit unqualified asterisk create.select().from(table(name("BOOK"))).fetch(); // Renders a SELECT * statement - Explicit unqualified asterisk create.select(asterisk()).from(BOOK).fetch(); // Renders a SELECT BOOK.* statement - Explicit qualified asterisk create.select(BOOK.asterisk()).from(BOOK).fetch(); create.select(BOOK.asterisk(), AUTHOR.asterisk()).from(BOOK, AUTHOR).fetch();
With all of the above syntaxes, the row type (as discussed below) is unknown to jOOQ and to the Java compiler.
It is worth mentioning that in many cases, using an asterisk is a sign of an inefficient query because if not all columns are needed, too much data is transferred between client and server, plus some joins that could be eliminated otherwise, cannot.