Saturday, October 15, 2011

Problem - GWT logging not working

When you inherit certain modules in .gwt.xml file, it disables GWT logging. This happened to me when I tried inheriting Requestfactory. Some modules inherit com.google.gwt.logging.LoggingDisabled  internally and thus affect your logging.

To fix this you should explicitly enable logging after you have inherited all your modules. At the end of your list of inherits add the following to your .gwt.xml file-

<set-property name="gwt.logging.enabled" value="TRUE"/>

Executing Postgres crosstab query as a prepared statement


I was executing a crosstab query as a prepared statement in Java(in a GWT app) and getting the following error -

PSQLException - Can't use query methods that take a query string on a PreparedStatement.

With some helpful folks from stackflow, I was able to resolve the error with the following code -

String query = "SELECT * FROM crosstab(
                      'SELECT rowid, a_name, value
                       FROM test WHERE a_name = ''att2''
                                    OR a_name = ''att3''
                      ORDER BY 1,2'
) AS ct(row_name text, cat_1 text, cat_2 text, cat_3 text);";

PreparedStatement stat = conn.prepareStatement(query);
ResultSet rs = stat.getResultSet();

//Note that, it is executeQuery() and not executeQuery(query)
stat.executeQuery();
rs = stat.getResultSet();
while (rs.next()) {
    //TODO
}





Thanks!

PostgreSQL crosstab query - Rotate a table about a pivot

An interesting feature of relational databases(postgres in this case) is the ability to rotate the table about a pivot. So if you have data like this-
 id | rowid | key | value
---+------+----+-------
  1 | test1 | key1      | val1
  2 | test1 | key2      | val2
  3 | test1 | key3      | val3
  4 | test1 | key4      | val4
  5 | test2 | key1      | val5
  6 | test2 | key2      | val6
  7 | test2 | key3      | val7
  8 | test2 | key4      | val8

And want to have a result set like this -

rowid | key1 | key2 | key3 | key4
------+------+-----+-----+------
 test1  | val1  | val2  | val3  | val4
 test2  | val5  | val6  | val7  | val8


It can be achieved by a "crosstab" query in a postgres database -



Update: 02/14/2013
The following can be achieved by a crosstab and also by an interesting SQL which I came across here - http://stackoverflow.com/questions/14863985/postgres-crosstab-maybe