Regards,
Olivier Mascia, integral.be/om
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 842 bytes
Desc: Message signed with OpenPGP using GPGMail
URL:
<http://mailinglists.sqlite.org/cgi-bin/mailman/p
Greetings All,
I use the Full Text Search facilities extensively. The feature is intuitive
and works beautifully and super fast even with 100GB+ databases. I must
applaud the SQLITE dev team on such amazing work.
I have two questions with regards to FTS.
1) I'm going out on a limb here:
As some
--
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 842 bytes
Desc: Message signed with OpenPGP using GPGMail
URL:
<http://mailinglists.sqlite.org/cgi-bin/mailman/private/sqlite-users/attachments/20160221/4d9028de/attachment.pgp>
select f1,
(select max(f2)
from mytable
where f1 = o.f1
and (f3 & 8)) as f2
from mytable as o
where (f3 & 8)
group by f1;
> -Original Message-
> From: sqlite-users-bounces at mailinglists.sqlite.org [mailto:sqlite-users-
> bounces
On 2/21/16, Olivier Mascia wrote:
>
> The way I had read that documentation is: if your function is meant to be
> deterministic, returning the same result for the very same input, at least
> for the duration of the SQL statement (or of course always), then you should
> tag it as SQLITE_DETERMINIST
On Sun, Feb 21, 2016 at 6:09 PM, Olivier Mascia wrote:
> Dear all,
>
> Is it possible to implement a SQL function (
> https://www.sqlite.org/c3ref/create_function.html), which implementation
> would be able to return the same value for the duration of the current
> transaction?
>
This _might_ su
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 842 bytes
Desc: Message signed with OpenPGP using GPGMail
URL:
<http://mailinglists.sqlite.org/cgi-bin/mailman/private/sqlite-users/attachments/20160221/bc1a4da2/attachment.pgp>
Hi,
Oracle is distributing its BerkeleyDB including an interesting fusion
with SQLite
where they take all from SQLite except the final storage engine b-tree
and pager included.
As a product it is possible to obtain a dbsql binary that is a clone of
sqlite3 command line
but working with BDB data
Of course it is.
The function merely needs to store a value associated with a connection if the
value does not exist (for that connection) when it is called and use a
commit_hook and rollback_hook so that it knows when it needs to clear the value
for the connection. Then for every call on a c
On 2/21/2016 12:09 PM, Olivier Mascia wrote:
> Is it possible to implement a SQL function
> (https://www.sqlite.org/c3ref/create_function.html), which implementation
> would be able to return the same value for the duration of the current
> transaction?
>
> In other words, assume I'd like to imp
On 2016/02/21 1:49 AM, Richard Hipp wrote:
> On 2/20/16, Dave Baggett wrote:
>> Question: can I force SQLite to keep an index purely in memory,
> No. Because if you did, other processes updating the table would have
> no way of also updating the index.
>
Of course - but how about being able to
My "dbsql" is reporting version 3.8.3.1:
./dbsql --version
3.8.3.1 2014-02-11 14:52:19 ea3317a4803d71d88183b29f1d3086f46d68a00e
I grabbed the berkeleydb code from Oracle's website, then compiled it with
SQL compatibility.
On Sun, Feb 21, 2016 at 9:37 AM, javaj1811 at elxala.com
wrote:
> Hi,
>
Both Igor's and Simon's solutions work on my test system.
The only additional comment is that Simon's works in Sqlite as expected.
However running the same on PostgreSQL bombs with an error complaining
about an sub-query with no name! Interesting? Perhaps watch out for
portability with that one. I
Is anyone aware of a design doc for any of the FTS implementations? Looking
for something a bit more technical than the docs. If not, where in the
source would you recommend starting? Thanks!
14 matches
Mail list logo
