Re: About Databases...
From: Dennis Lee Bieber (wlfraed_at_ix.netcom.com)
Date: 03/12/05
- Next message: steve: "Tkinter WEIRDNESS or Python WEIRDNESS?"
- Previous message: Skip Montanaro: "Re: csv module and unicode, when or workaround?"
- In reply to: andrea_gavana_at_tin.it: "About Databases..."
- Next in thread: Robert Kern: "Re: About Databases..."
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Sat, 12 Mar 2005 05:01:24 GMT
On Fri, 11 Mar 2005 23:32:48 +0100, andrea_gavana@tin.it declaimed the
following in comp.lang.python:
> oil field). As you may have understood, I know almost NOTHING about databases
Then your first approach should be a book on relational database
design, normalization (and when to violate normalization), and SQL.
Preferably a book that is NOT specific to ANY Relational DBMS. Though
I'd probably avoid any book that has chapters on Hierarchical DBMS,
CODASYL/DBTG Network DBMS, or Relational Algebra and/or Relational
Calculus -- Unless you're running on 1970s "big iron", but in that case,
you may not have Python by which to even attempt to access them... <G>
After you've gotten the generics down, then you can consider the
specifics of any available RDBMS...
Python modules likely exist for: M$ JET (I'm presuming you're on
a Windows box, including the server), M$ SQL Server/MSDE, MySQL, MaxDB
by MySQL (their name for the repackaged SAP DB), PostgreSQL, Firebird
(the descendent of Interbase). I'm pretty sure all of the above run on
WinNT, and most are even available for W9x.
> In general, my data will be numeric (floats, integers). Will a binary storage
> (if it is possible) reduce the size of the DB? And what about speed in storing/retrieving
> data?
>
I don't know of any RDBMS that /doesn't/ support binary
numerics, though they may do so with 32-bit-only integer and 64-bit-only
floats (many though have short and long ints, and single or double
floats). Many also have either a decimal or currency data type. However,
IN the SQL query, your data will be formatted as text during the
insertion, and converted back to binary by the RDBMS server.
If you need to store what are essentially large arrays, which
will never by accessed via the RDBMS piecemeal (IOW: you will retrieve
or store the entire array at once) you'll want to check on the
capabilities of BLOBs (Binary Large OBjects, as I recall) in each
candidate RDBMS.
-- > ============================================================== < > wlfraed@ix.netcom.com | Wulfraed Dennis Lee Bieber KD6MOG < > wulfraed@dm.net | Bestiaria Support Staff < > ============================================================== < > Home Page: <http://www.dm.net/~wulfraed/> < > Overflow Page: <http://wlfraed.home.netcom.com/> <
- Next message: steve: "Tkinter WEIRDNESS or Python WEIRDNESS?"
- Previous message: Skip Montanaro: "Re: csv module and unicode, when or workaround?"
- In reply to: andrea_gavana_at_tin.it: "About Databases..."
- Next in thread: Robert Kern: "Re: About Databases..."
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
