Rollback is a problem with simpler software.
You are basically specifying Sqlite in your requirements, and it does
all that wiith a small footprint based on the capability.
You might contact Dr Hipp and enquire about a commercial version of
Sqlite which strips out the SQL compiler and uses precompiled SQL.
Raymond Hurst wrote:
I am an extreme newbie at this and I am open to suggestions:
My requirements are:
Smallest footprint possible
ACID requirements (transactional rollback a must)
Be able to create tables with fields that have name/value pairs.
Not sure what else!
If a dBase derivative meets these requirements I'm all for it and just
need to be pointed in the right direction.
Ray
John Stanton wrote:
Why not implement a simple index file system like one of the DBase
derivatives? If you have no OS you will find porting something like
Sqlite or Berkely DB a challenge. If you are not looking for SQL then
you can achieve your goal in maybe 20K of memory.
[EMAIL PROTECTED] wrote:
---- P Kishor <[EMAIL PROTECTED]> wrote:
On 3/29/07, Ludvig Strigeus <[EMAIL PROTECTED]> wrote:
Hi,
I'm looking at using Sqlite as a storage backend for a program.
Using SQL is
a little bit overkill and much more than we need. How complicated
would it
be to interface to the btree subsystem directly? Sqlite seems very
modular
from the looks of it, but has anyone attempted anything like this
before?
The functionality I need is key->value maps with support for lookup
by an
exact key, insertion, iteration of all keys in the database, removal.
Transactions would also preferably be needed.
why bother with SQLite then? Use the right tool for the job -- use
BerkeleyDB.
--
Puneet Kishor http://punkish.eidesis.org/
Nelson Inst. for Env. Studies, UW-Madison http://www.nelson.wisc.edu/
Open Source Geospatial Foundation http://www.osgeo.org/education/
---------------------------------------------------------------------
collaborate, communicate, compete
=====================================================================
-----------------------------------------------------------------------------
To unsubscribe, send email to [EMAIL PROTECTED]
-----------------------------------------------------------------------------
Size is a constraint for me.
I see that SQLite can be around 170KB where BerkeleyDB is around 500K.
I also see that the SQL statements can be converted to byte code.
Is this byte code more efficient that the SQL statement in code size?
I'm looking at embedding a DB of some type into a Single Board
Computer with no OS.
-----------------------------------------------------------------------------
To unsubscribe, send email to [EMAIL PROTECTED]
-----------------------------------------------------------------------------
-----------------------------------------------------------------------------
To unsubscribe, send email to [EMAIL PROTECTED]
-----------------------------------------------------------------------------
-----------------------------------------------------------------------------
To unsubscribe, send email to [EMAIL PROTECTED]
-----------------------------------------------------------------------------
-----------------------------------------------------------------------------
To unsubscribe, send email to [EMAIL PROTECTED]
-----------------------------------------------------------------------------
