Raphaël KINDT wrote:
>
> All events come sequentially (because it's a time line).
> I can have more than one event during the same time but I use 1 different
> bit to identify each event types.
>
> I think it's important to use this sequential constraint to do the delete
> request.
> But SQLite3
Alexey Pechnikov wrote:
>
> P.S. Is any method for multy-master replication of SQLite databases?
Nothing built in, but I believe several user have developed their own
systems to do this. You would probably be better off asking this
question in a new thread.
Dennis Cote
IGN KEY with SQLite3?
Thanks in advance for your answers.
Regards,
Raphaël
-Message d'origine-
De : [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] la part de Dennis Cote
Envoyé : mardi 17 juin 2008 18:59
À : General Discussion of SQLite Database
Objet : Re: [sqlite] Long delay for delete/se
В сообщении от Thursday 19 June 2008 02:37:57 Dennis Cote написал(а):
> I'm not sure if you even need unique id numbers for these records, or
> why you are concerned that there might be multiple records with the same
> save_date for that matter? Is this table linked to any others?
I'm using URL
Alexey Pechnikov wrote:
>
> I have same database on host A and host B. On _both_ hosts I can
> insert/update/delete/select rows and periodically synchronize databases.
>
Oh. That's a different issue that you didn't mention before. It
definitely makes things more complicated.
> CREATE TABLE
В сообщении от Thursday 19 June 2008 00:22:37 Dennis Cote написал(а):
> Alexey Pechnikov wrote:
> > I'm replicating my database using sqlite dump and load or sql queries.
> > I'm not sure that rowid is not different after that.
>
> If you assign each row an id (and even better make it an integer
Alexey Pechnikov wrote:
>
> I'm replicating my database using sqlite dump and load or sql queries. I'm
> not
> sure that rowid is not different after that.
>
If you assign each row an id (and even better make it an integer primary
key) and use those id columns to link related records then
В сообщении от Wednesday 18 June 2008 22:44:16 Dennis Cote написал(а):
> Alexey Pechnikov wrote:
> > I find get only _one_ row. I found correspond timestamp by other
> > questions. I don't want get more than one row.
>
> In that case you would be better off to get the rowid of that row using
> the
Alexey Pechnikov wrote:
>
> I find get only _one_ row. I found correspond timestamp by other questions. I
> don't want get more than one row.
>
In that case you would be better off to get the rowid of that row using
the other questions. Then you can get the row of interest directly using
the
В сообщении от Wednesday 18 June 2008 21:32:46 Dennis Cote написал(а):
> Alexey Pechnikov wrote:
> > There is problem for select one row by time - we must use query
> > select * from events where time || =julianday('2008-06-16 23:59:59.999');
> > and index is not used for this query.
> > Maybe
В сообщении от Tuesday 17 June 2008 20:59:19 Dennis Cote написал(а):
> I would recommend that you change your database schema an combine your
> timestamp string and subsecond integer field into a single floating
> point julian day number that provides both pieces of information in
> units of days
Raphaël KINDT wrote:
>
> I'm working on a video monitoring program that uses frame grabber (12
> cameras).
> This program records videos and detects some (input/output) events such as :
> motion detection, blue screens (cut camera cable), I/Os 1 to 24, and much
> more...
>
> I save all events in
Hi Raphaël,
Since you are dealing with intervals of numbers, perhaps recent
discussion of
rtree indexing will help.
Besides, separating instant into highly used level of unit may be also
useful.
For example, instant_date and instant_time.
Another rough method is applying PRAGMA synchronous =
13 matches
Mail list logo