RE: [PEAR-DEV] Package Proposal : DB_Dump
| From: | Lukas Smith | Date: | Sun, 01 Jun 2003 11:22:56 +0000 |
| Subject: | RE: [PEAR-DEV] Package Proposal : DB_Dump | ||
| References: | 1 | Groups: | php.pear.dev |
| Request: | Send a blank email to pear-dev+get-16876@lists.php.net to get a copy of this message | ||
> From: Alexey Borzov [mailto:borz_off@cs.msu.su]
> Sent: Sunday, June 01, 2003 12:39 PM
>
> Hi!
>
> indyone :o) wrote:
> > Hi,
> > My name is Ioannis and i have developed a new class which i'm
> > calling it DB_Dump and as the name says, it has to do with the
dumping
> > the data of a database.
> > This class needs PEAR::DB to work. The main uses of this class is to
> > create a dump file of the tables of a database, and to execute a
file
> > with dumped data.
>
> What is the purpose of using an abstraction layer if the only
supported
> database
> is MySQL?
It didnt sound like only supporting MySQl was his the entire scope of
his goals.
> > To make this class possible i have used some functions from
phpMyAdmin.
> > The dumped data is also compatible with phpMyAdmin's dump files.
>
> Isn't it simpler just to use phpMyAdmin then?
See above
> > I have the source here
> > http://cassiopea.ee.teiath.gr/~indyone/dbdump/
> >
> > If you have any questions, suggestions or comments, please do
contact
> me.
>
> I don't think you'll ever be able to support all databases that
PEAR::DB
> supports. You are able to support MySQL just because it is *very*
limited.
>
> Consider the following: to properly dump PostgreSQL database, you will
> need to dump:
> 1) Tables
> 2) Sequences
> 3) Indexes
> 4) Custom functions
> 5) Triggers
> 6) Custom types
> 7) Rules
> 8) Views
> ...
> N) Data
True to that.
However if more than 1 Database is supported and atleast a significant
subset of those features can be supported I would say the project is
worthwhile.
MDB already has a portion of these things available and hopefully with
the next version of MDB things will become more flexible, as currently
things are very focused on maximum portability. So maybe MDB might
become a worthwhile alternative for this project.
Anyways tableinfo can certainly be improved to make it possible to
support enough features/data structures to make it worthwhile for a
large number of people. It doesn't have to solve every possible
situation where you might want a dump, it just needs to solve enough so
that there is an added value over phpMyAdmin. Maybe even the same
functionality as phpMyAdmin with PEAR CS would be a value added because
iirc the phpMyAdmin code is very nasty(?)
Regards,
Lukas