I can put the transaction identifier into tables via triggers.
That's the million $ question. I don't know when transaction
begins or ends! And my original question was pointing to
a feature DRH had suggested way back in 2003. That would
have been the perfect way to do it (and probably the only
reliable way).
It would have been something along these lines:
CREATE TRIGGER ex1 AFTER BEGIN BEGIN
INSERT or UPDATE into TT values ('xid', lower(hex(randomblob(16)))
END;
CREATE TRIGGER ex2 BEFORE COMMIT BEGIN
DELETE from TT where key='xid'
END;
CREATE TRIGGER abc_ins AFTER INSERT on abc
UPDATE abc set xid = (select value from TT where key='xid')
END;
for this to work, we need a database intrinsic. please refer to
this link again:
http://osdir.com/ml/db.sqlite.general/2003-04/msg00137.html
----- Original Message ----
From: Pavel Ivanov <paiva...@gmail.com>
To: General Discussion of SQLite Database <sqlite-users@sqlite.org>
Sent: Wed, June 23, 2010 2:40:53 AM
Subject: Re: [sqlite] marking transaction boundaries
> The question following your suggestion is how do i maintain transaction
> start and stop times? I have no control of knowing when a transaction
> starts and ends and i said earlier, i cannot use transaction hooks.
And you didn't answer my question: how will you put this transaction
identifier into tables? How will you know which transaction each
INSERT or UPDATE belongs to?
_______________________________________________
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
