As you might know, SQLite3 does not define the order in which
aggregate functions are applied to the values to be aggregated. For
traditional aggregation functions (min(), max(), ...) this makes
perfect sense. For group_concat(), however, this undefined order is
obnoxious.
Apparently MySQL has
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 04/24/2011 02:03 PM, David L wrote:
> This can't be the desired behaviour, can it?
It isn't. The team have fixed the problem:
http://www.sqlite.org/src/timeline
Roger
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.10 (GNU/Linux)
Comment:
On Sun, 24 Apr 2011 23:03:03 +0200, "David L"
wrote:
>Hello Sqlite-users!
>
>I have stumbled upon the fact that foreign keys are not always enforced.
>
>Specifically, that happened when I used the INSERT statement with a SELECT
>clause, which apparently caused invalid values
Hello Sqlite-users!
I have stumbled upon the fact that foreign keys are not always enforced.
Specifically, that happened when I used the INSERT statement with a SELECT
clause, which apparently caused invalid values to be inserted.
How else would you explain the following?:
PRAGMA
On 24/04/2011, at 10:49 AM, Mickey Mestel wrote:
> we are using sqlite on the iPhone, in conjunction with SQLCipher, so the
> sqlite version is compiled in with SQLCipher. the version of sqlite is
> 3.7.2, and 4.3 of iOS.
>
> i have an issue that suddenly started appearing all of a sudden.
5 matches
Mail list logo