db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Daniel John Debrunner <...@apache.org>
Subject Re: Changes to comparable method in TypeCompiler
Date Thu, 19 Apr 2007 14:43:34 GMT
Mamta Satoor wrote:

> I think what you are suggesting is to move comparable method out from 
> the TypeCompiler and into DataTypeDescriptor altogether. So, the 
> existing code, where we use TypeCompiler to decide if 2 types can be 
> compared or not should now call a method on DTD to determine 
> comparability. This might be cleaner than stuffing collation information 
> in CharTypeCompiler but I am just wondering why was comparable not 
> defined on DTD at the very start. Why do we go through TypeCompiler and 
> what functionality does TypeCompiler provide that DTD does not? In other 
> words, I don't understand the connection between TypeCompiler and DTD 
> and how they fit together.

It's not that TypeCompiler provides functionality that DTD does not, but 
instead DTD has functionality/information that TypeCompiler does not. 
Ignoring the "compiler" aspect for the moment there are two components 
to a DataTypeDescriptor, the underlying SQL type (INTEGER, CHAR, 
VARCHAR, XML etc.) represented as TypeId and attributes of the 
descriptor (nullablity, length, precision, scale and now collation).

Thus

  DTD = TypeId + {attributes}

Some functionality is applicable to a type regardless of a specific 
DTD's attributes, thus methods for that functionality can be declared on 
TypeId instead of DTD.

Some functionality on the other hand needs the attribute information as 
well, say the display length of a type is a function of its 
length/precision&scale and its underlying SQL type.

The collation changes have moved the comparable check from being only 
reliant on the SQL type (TypeId) to being dependent on the type's 
attributes (collation type and implicit/explicit). Thus the original 
location for the comparable method made sense, but now does not.

The TypeCompiler/TypeId split was due to an early plan to have a 
execute-only version of the technology, this never happened as there was 
no demand for it. One of the benefits of a SQL engine is the ability to 
execute arbitrary queries, which would not be available in an execute 
only version. Code cleanup could be done here which probably would 
decrease the footrprint of derby.

HTH,
Dan.



Mime
View raw message