Hi,

I was just playing with the SchemaExport class to export the DDL creation script and I was wondering if the following formatting is intentional or should be improved:

create sequence DATASET_SEQUENCE start with 1 increment by 50;
create sequence MODEL_SEQUENCE start with 1 increment by 50;
create sequence USER_SEQUENCE start with 1 increment by 50;

    create table Dataset (
       ...
    );

    create table Model (
       ...
    );

    create table Model_User (
       model_id bigint not null,
        user_id bigint not null,
        primary key (model_id, user_id)
    );

    create table User (
       ...
    );
create unique index INDEX_ID on User (dataset_id, external_id, revision);

    alter table Model 
       add constraint FKj2cyptlmwgilu172586ym0vwk 
       foreign key (dataset_id) 
       references Dataset;

    ...

I've replaced any irrelevant part with "...", but to summarize:

1. create sequence statements are not separated by new lines like the other statements;
2. create table statements are 4-spaces indented by default;
3. alter table statements are 4-spaces indented by default;
4. in the create table statement, the row of the first column is 7-spaces indented while the other columns are 8-spaces indented (e.g. model_id and user_id of Model_User table);
5. between the last create table statement and the first create unique index statement there is no new line separation.

Unless there is a reason for this specific formatting, I would be happy to contribute a fix (the code of org.hibernate.engine.jdbc.internal.DDLFormatterImpl.java seems quite easy).