Turbine 2.1 / Torque / Oracle / primarykey.vm

[Eric Schmidt <Er...@concur.com>]

I am working with Turbine 2.1 / Torque / Oracle and considering submittal of
a patch. I am new to all of this so I would appreciate it if someone lets me
know if I'm wasting my time; I haven't found any related discussion.

Using the org.apache.turbine.torque.TorqueSQLTask class and the release
version of torque\templates\sql\base\oracle\primarykey.vm results in a
schema.sql that does table creation that is dependent on a primary key
having been specified in the xml, despite the fact that the attribute can be
set to 'false'. The relevant vm segments:

table.vm:

  CREATE TABLE $table.Name
  (
   ...
  );

  #set ( $pk = $generator.parse("$basepath/primarykey.vm", "", "table",
$tbl) )
  #if($pk.length()>0)$pk#end

primarykey.vm:

  ALTER TABLE $table.Name
      ADD CONSTRAINT ${table.Name}_PK 
  PRIMARY KEY ($table.printPrimaryKey());

Because table.vm checks for the length of $pk which is always > 0 given the
format of primarykey.vm, all CREATE TABLE statements in the resulting sql
will be followed by the ALTER TABLE statement *with a potentially bogus
PRIMARY KEY output.

The patch I am testing would modify primarykey.vm as follows:

  #if($table.printPrimaryKey().length()>0)
    ALTER TABLE $table.Name
        ADD CONSTRAINT ${table.Name}_PK 
    PRIMARY KEY ($table.printPrimaryKey());
  #end

This results in $pk in table.vm having length 0 for tables having no primary
key, and thus the ALTER TABLE statement is not produced.

Any comments or suggestions would be appreciated.

Thanks,

Eric Schmidt


--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>

Re: Turbine 2.1 / Torque / Oracle / primarykey.vm

[John McNally <jm...@collab.net>]

I think you should be able to make the conditional based on whether any
pk's exist for the table as opposed to creating a string and checking
its length.

john mcnally

Eric Schmidt wrote:
> 
> I am working with Turbine 2.1 / Torque / Oracle and considering submittal of
> a patch. I am new to all of this so I would appreciate it if someone lets me
> know if I'm wasting my time; I haven't found any related discussion.
> 
> Using the org.apache.turbine.torque.TorqueSQLTask class and the release
> version of torque\templates\sql\base\oracle\primarykey.vm results in a
> schema.sql that does table creation that is dependent on a primary key
> having been specified in the xml, despite the fact that the attribute can be
> set to 'false'. The relevant vm segments:
> 
> table.vm:
> 
>   CREATE TABLE $table.Name
>   (
>    ...
>   );
> 
>   #set ( $pk = $generator.parse("$basepath/primarykey.vm", "", "table",
> $tbl) )
>   #if($pk.length()>0)$pk#end
> 
> primarykey.vm:
> 
>   ALTER TABLE $table.Name
>       ADD CONSTRAINT ${table.Name}_PK
>   PRIMARY KEY ($table.printPrimaryKey());
> 
> Because table.vm checks for the length of $pk which is always > 0 given the
> format of primarykey.vm, all CREATE TABLE statements in the resulting sql
> will be followed by the ALTER TABLE statement *with a potentially bogus
> PRIMARY KEY output.
> 
> The patch I am testing would modify primarykey.vm as follows:
> 
>   #if($table.printPrimaryKey().length()>0)
>     ALTER TABLE $table.Name
>         ADD CONSTRAINT ${table.Name}_PK
>     PRIMARY KEY ($table.printPrimaryKey());
>   #end
> 
> This results in $pk in table.vm having length 0 for tables having no primary
> key, and thus the ALTER TABLE statement is not produced.
> 
> Any comments or suggestions would be appreciated.
> 
> Thanks,
> 
> Eric Schmidt
> 
> --
> To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
> For additional commands, e-mail: <ma...@jakarta.apache.org>

--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>