Emil Forslund
Emil Forslund
The connection pooling logic only makes sense for standalone applications. Enterprise applications often have their own connection pool already.
The pooling logic should be kept separate form the connection factory. In some cases, like with Java EE, you want to be able to supply a new Connection Factory without...
We currently use Tuples to store operands in field predicates, but this causes some problem since it requires all Speedment project to depend on the Tuples library. When we make...
The `Column` and `ColumnProperty` components hade an `OrderType` that can be either `ASC`, `DESC`, `NONE` or `null`. There is an ambiguity here that can cause problems.
If you have a table with hundreds of columns, it is very tiresome to set the "nullableImplementation" setting on every one if you want to generate regular Java beans instead...
Offseted time is useful to avoid loss of information when working between multiple databases with different local timezone. All times should then be converted into offset-time (relative to UTC/Greenwich in...
This will allow all open-source modules to be dependency managed in one place. The existing BOMs like `com.speedment.runtime:runtime-all` are good for managing one of the scopes, but we also need...
Right now every config option is in the same file, even generated configuration. It would be better if the files are hierarchial so that the tool can generate one file,...
Currently the custom streams can optimize predicate statements like this: ``` java Hare.stream() .filter(HareField.NAME.equal("Harry")) .findAny(); ``` And the resulting SQL would be: ``` sql SELECT * FROM `Hare` WHERE `name`...
In the current code translation system, every translator must create a file for every instance of the document they correspond to. Sometimes you want to generate a file for only...