Databases: Which one's right for me?

Aaron Watters aaron at reportlab.com
Sun Jan 11 12:58:16 EST 2004


Rene Pijlman <reply.in.the.newsgroup at my.address.is.invalid> wrote in message news:<qu7000dot866j3qd9hi6hgkh4np5nuq7fa at 4ax.com>...
> Marc:
> >Basically I need a db that will travel with my executable script and
> >faithfully store and edit the data needed in that script.
> 
> Have a look at ZODB.
> 
> Introduction:
> http://www.python.org/workshops/2000-01/proceedings/papers/fulton/zodb3.html

BTW, I would like to see somewhere an explanation of how
ZODB concurrency control and recovery work.  Please inform if
there is some information.  I'm not encouraged by the fact that
in what documentation I can find the definitions are not the
standard ones.  For example

zodb.pdf:
"""
Isolation means that two programs or threads running in two different 
transactions cannot see each other's changes until
they commit their transactions.
"""
Database System Concepts, 4th edition, Silberchatz, Korth, Sudarshan p.566
"""
Isolation: Even though multiple transactions may execute concurrently
the system guarantees that for every pair of transactions Ti and Tj
it appears to Ti that either Tj finished executing before Tj started
or Tj started executing after Ti finished.
"""
The zodb definition is far weaker than the standard one (at least in
this document).  Maybe they really meant something different?

For an example of what I'm looking for please see the xsdb documentation
  http://xsdb.sourceforge.net/concurrency.html
or the gadfly documentation
  http://gadfly.sourceforge.net/recover.html

  -- Aaron Watters
===
...and somewhere in the eprom
that hacker he switched context
but in is cryptic code I found
a statement I could parse:
you got to know when to kludge up
know when to core dump
know when to single step
know when to "run"
you never count your source lines
while your sittin' at the keyboard
there'll be time enough for metrics
when the hackings done.
  The Hacker (with apologies to The Gambler)



More information about the Python-list mailing list