At 03:43 PM 12/22/2004, you wrote:
See section 6.0 in http://www.sqlite.org/lockingv3.html.
That article is on SQLite version 3, but the methods for
corrupting a database apply equally well to version 2.
Thanks. Unfortunately, none of these seem terribly likely. The user
reported that nothing
Are you using your Python binding? If yes, maybe you can make the "commit"
action invoke a callback which updates said private table. Every code path
that writes to tables does so within a transaction, no?
Of course, if your code commits by issuing an SQL statement in place
instead of calling a
On Wed, Dec 22, 2004 at 06:53:15AM -0800, Roger Binns wrote:
> >If concurrent access is by instances of your program only, they can
> >rendezvous via a private table.
>
> Yes, but that involves modifying every single code path that could
> update any of the many tables I use to update the
Dan Keeley wrote:
Hi,
I've finally got round to building sqlite. Now i'm after the odbc
driver for it.
Only thing is I can't seem to find the page on the sqlite website that
lists other programs, such as an odbc driver for use with sqlite?
Is there a recommended odbc driver to use?
Dan,
The
Michael Hunley wrote:
What else could cause this "corruption?"
See section 6.0 in http://www.sqlite.org/lockingv3.html.
That article is on SQLite version 3, but the methods for
corrupting a database apply equally well to version 2.
Is there any way to repair the db so the user does not lose his
Dan Keeley wrote:
Hi,
I've finally got round to building sqlite. Now i'm after the odbc
driver for it.
Only thing is I can't seem to find the page on the sqlite website that
lists other programs, such as an odbc driver for use with sqlite?
Is there a recommended odbc driver to use?
You do not
> >
> >
> >You're altering tables you don't understand the structure of?
> >
> >
> >
> >
> Think about a php forum... someone could choose to make a mod and add
> a
> column for the birthday in the users table... you can't know if
> someone did it or not...
Arg! Just invites people to crash
Jay wrote:
--- Paolo Vernazza <[EMAIL PROTECTED]> wrote:
Tell your friends that they need to read the FAQ.
http://www.sqlite.org/faq.html#q13
Sorry, Richard, but I think that it isn't so simple you must also
store triggers and indexes... and maybe you don't know what's the
table
On Tue, 21 Dec 2004, Paolo Vernazza wrote:
>D. Richard Hipp wrote:
>
>>
>>
>> Tell your friends that they need to read the FAQ.
>> http://www.sqlite.org/faq.html#q13
>
>Sorry, Richard, but I think that it isn't so simple you must also
>store triggers and indexes... and maybe you don't know
CARIOTOGLOU MIKE wrote:
> I can verify this :
>
> Select "columnName" from someTable gives columns with quoted names !
> ("ColumnName") instead of ColumnName
>
> This *feels* like a bug
>
I believe this is a bug as well.
The following script demonstrates the problems using the sqlite3 shell. I
D. Richard Hipp wrote:
> Before doing the "DELETE TABLE t1" you can run this query:
>
> SELECT sql FROM sqlite_master WHERE tbl_name!=name AND
> tbl_name='t1';
>
> That query will give you the complete text of all CREATE INDEX
> and CREATE TRIGGER statements associated with table t1. Save
>
> > You missed his point. When you use the steps in the FAQ to alter
> > a table, one of the steps involves deleting it. When you do that
> > SQLite also deletes all existing triggers and indices on the
> > table.
> >
> > Consequently one piece of code that is simple (alter table)
> suddenly
Roger Binns wrote:
You're altering tables you don't understand the structure of?
You missed his point. When you use the steps in the FAQ to alter
a table, one of the steps involves deleting it. When you do that
SQLite also deletes all existing triggers and indices on the
table.
Consequently
If concurrent access is by instances of your program only, they can
rendezvous via a private table.
Yes, but that involves modifying every single code path that could
update any of the many tables I use to update the private table.
The counter Richard mentioned is matches exactly what my
You're altering tables you don't understand the structure of?
You missed his point. When you use the steps in the FAQ to alter
a table, one of the steps involves deleting it. When you do that
SQLite also deletes all existing triggers and indices on the
table.
Consequently one piece of code
--- Paolo Vernazza <[EMAIL PROTECTED]> wrote:
> > Tell your friends that they need to read the FAQ.
> > http://www.sqlite.org/faq.html#q13
>
> Sorry, Richard, but I think that it isn't so simple you must also
>
> store triggers and indexes... and maybe you don't know what's the
> table
>
This bug is a beauty: text column comparison in joins may follow NUMERIC
rules, even if
the column(s) have text affinity!
consider this case. Tables A,B are a typical one-to-many relation, where B
has N records
for each A, and one field (key) relates them.
CREATE TABLE a(
key VARCHAR(2),
I can verify this :
Select "columnName" from someTable gives columns with quoted names !
("ColumnName") instead of ColumnName
This *feels* like a bug
> -Original Message-
> From: red forks [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, December 22, 2004 6:07 AM
> To: [EMAIL PROTECTED]
Hi,
I've finally got round to building sqlite. Now i'm after the odbc driver
for it.
Only thing is I can't seem to find the page on the sqlite website that lists
other programs, such as an odbc driver for use with sqlite?
Is there a recommended odbc driver to use?
Rgds,
Dan
On Tue, Dec 21, 2004 at 07:58:46PM -0800, Roger Binns wrote:
> Richard already answered. There is no mystery about who else is writing to
> the db - it will be another instance of my program.
If concurrent access is by instances of your program only, they can
rendezvous via a private table.
20 matches
Mail list logo
