FAQ
Hi!

Though I am not yet registered as an author on pause,
I would like to get comments on a name for a module.
I do not stick with the current name SQL::Schema::*.
The opposite is true. Currently the module only supports
DDL statements within Oracle's SQL dialect and is only
capable to extract a subset of an Oracle database schema.

Some of the names I have in mind are:

Oracle::DDL::*
Oracle::Schema::*
SQL::DDL::Oracle::*
SQL::Oracle::*

Currently I do like the second most, I think.

Torsten

P.S.: I attached some of the docs and a tarball.

--
Torsten Hentschel private
todd@bayleys.ping.de

Search Discussions

  • Tim Bunce at Apr 11, 2000 at 9:14 pm
    It would be nice to allow the same interface to work for multiple
    database types. So,

    $schema = SQL::Schema->new($dbh);

    seems appealing. In other words a thin SQL::Schema 'wrapper' module
    that then loads a "SQL::Schema::$driver_name" module to do the work.

    Tim.
    On Tue, Apr 11, 2000 at 06:58:10PM +0200, Torsten Hentschel wrote:
    Hi!

    Though I am not yet registered as an author on pause,
    I would like to get comments on a name for a module.
    I do not stick with the current name SQL::Schema::*.
    The opposite is true. Currently the module only supports
    DDL statements within Oracle's SQL dialect and is only
    capable to extract a subset of an Oracle database schema.

    Some of the names I have in mind are:

    Oracle::DDL::*
    Oracle::Schema::*
    SQL::DDL::Oracle::*
    SQL::Oracle::*

    Currently I do like the second most, I think.

    Torsten

    P.S.: I attached some of the docs and a tarball.

    --
    Torsten Hentschel private
    todd@bayleys.ping.de
  • Torsten Hentschel at Apr 13, 2000 at 9:02 am

    On Tue, 11 Apr 2000, Tim Bunce wrote:
    It would be nice to allow the same interface to work for multiple
    database types. So,

    $schema = SQL::Schema->new($dbh);

    seems appealing. In other words a thin SQL::Schema 'wrapper' module
    that then loads a "SQL::Schema::$driver_name" module to do the work.

    Tim.
    Sounds reasonable to me. And I agree that this is
    a "nice to have".
    Just some points against:

    - Other databases (like DB/2, Informix) come with tools
    to extract the database schema as plain SQL;
    so, as there is no need to write such tools, I do not
    expect that someone will volunteer to reimplement
    them in perl (except as a study).

    - The databases do vary widely when it comes to DDL
    statements. There will be only a few elements they have
    in common. So a general interface will be either hard
    to implement (complex mechanisms to distinguish between
    proprietary and common SQL elements / objects)
    or it will be this general that it might become
    nearly useless as the provided methods limit you
    to a very basic use.

    Nevertheless, it might make sense to start with a very
    general and simple interface and see what happens.
    If there is no interest we can still make the occupied
    name space available again.

    Torsten

    --
    Torsten Hentschel privat
    todd@bayleys.ping.de

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupmodules @
categoriesperl
postedApr 11, '00 at 5:14p
activeApr 13, '00 at 9:02a
posts3
users2
websitecpan.org...

2 users in discussion

Torsten Hentschel: 2 posts Tim Bunce: 1 post

People

Translate

site design / logo © 2021 Grokbase