On 09/09/2016 12:06 PM, Keith Medcalf wrote:
Richard,
Can a single sqlite3_create_function call define a function which is both a
scalar function and an aggregate function, or are two calls to create_function
required, one defining the scalar and the other the aggregate version? (using
the s
Richard,
Can a single sqlite3_create_function call define a function which is both a
scalar function and an aggregate function, or are two calls to create_function
required, one defining the scalar and the other the aggregate version? (using
the same function name)
On Thursday, 8 September,
On 9/8/2016 1:15 PM, Stephan Mueller wrote:
" > I'd prefer to commit after each update
" You can't commit on a single connection either, while there's an
" unfinalized SELECT statement traversal going on. So you aren't gaining
" anything by trying to interleave SELECT and updates on the same conn
Thanks Keith, for the detailed response. My plan now is to update SQLite
to a WAL-capable version, enable WAL mode, use two connections instead
of one and wrap the SELECT in a transaction. With luck, I improve overall
performance (because of WAL mode) as well as gaining correctness.
Many thanks,
Thanks Kevin. I had considered the temporary table approach, but with
potentially millions of updates, I am concerned about the time spent
writing the 100MB+ of temp data, spoiling the interactive user
experience. However, I may try it (as you say, it should be simple) to see
the actual performan
Igor wrote:
" On 9/7/2016 6:11 PM, Stephan Mueller wrote:
" > I understand that a way to ensure "SELECT is unperturbed" semantics is to
" > use separate connections for SELECT and updates.
" If you go down that route, make sure you are using WAL journaling mode;
" it won't work otherwise.
I see t
I sent this long-winded question.
thanks,
stephan();
-Original Message-
From: sqlite-users [mailto:sqlite-users-boun...@mailinglists.sqlite.org] On
Behalf Of Bob McFarlane
Sent: Wednesday, September 07, 2016 5:10 PM
To: 'SQLite mailing list'
Subject: Re: [sqlite] Clarification on "No Iso
It seems to me that the simplest, most portable approach for this sort of
thing would to be having the SELECT create a temporary table of the desired
actions, and not apply them until after the select has concluded. This
would work in any database -- it does not depend on precise semantics of
WAL,
Darn. Oh well.
On Sep 8, 2016 9:34 AM, "Stephen Chrzanowski" wrote:
> Apparently, no. Error is "cannot create trigger on system table".
>
> On Thu, Sep 8, 2016 at 11:19 AM, Stephen Chrzanowski
> wrote:
>
> > Interesting idea. I'll try that on a scratch DB when I get the
> > chance. (Spen
Apparently, no. Error is "cannot create trigger on system table".
On Thu, Sep 8, 2016 at 11:19 AM, Stephen Chrzanowski
wrote:
> Interesting idea. I'll try that on a scratch DB when I get the
> chance. (Spent too much time in this thread, rather than doing work I'm
> being paid to do. heh)
On Wed, Sep 7, 2016 at 7:08 PM, Dan Kennedy wrote:
> On 09/07/2016 11:08 PM, Dominique Devienne wrote:
>
>> Initial design was to copy the DB file (app is "shutdown", so no
>> connection
>> to that DB file).
>> Drop the 5 big tables. Vaccum it (the new DB). Add the new tables.
>> But the 25GB cop
Interesting idea. I'll try that on a scratch DB when I get the
chance. (Spent too much time in this thread, rather than doing work I'm
being paid to do. heh)
On Thu, Sep 8, 2016 at 11:07 AM, Scott Robison
wrote:
> On Sep 8, 2016 8:16 AM, "Richard Hipp" wrote:
> >
> > On 9/8/16, Stephen Ch
Roger. Since I don't deal with pre-release versions much I guess my brain
automatically skipped over the timestamp and other number, as it's used to the
(3.15.0) part being unique on its own. (Which it obviously can't be while in
development)
Don't mind me folks, nothing to see here, nothing t
On Sep 8, 2016 8:16 AM, "Richard Hipp" wrote:
>
> On 9/8/16, Stephen Chrzanowski wrote:
> >
> > However, the rabbit I was hoping to pull out of the hat was that the
change
> > in version numbers be done automatically when I make a change in the 3rd
> > party DB management tool.
I don't know if t
That is kind of the line I was thinking of going. I don't want to rely on
MD5 as an ID, as by chance I might get the same numbers back, especially if
I revert back to an already bit-exact existing schema. (For instance, I do
a schema change, apply the update, but then find I have to revert to the
On 9/8/16, David Raymond wrote:
> Random question from when I just loaded up that URL: Is it supposed to have
> a futuristic SQLite version?
>
> SQLite Version: 2016-08-22 20:10:01 [7839519349] (3.15.0)
Yes. That's called "dogfooding"
(https://en.wikipedia.org/wiki/Eating_your_own_dog_food
Random question from when I just loaded up that URL: Is it supposed to have a
futuristic SQLite version?
SQLite Version: 2016-08-22 20:10:01 [7839519349] (3.15.0)
-Original Message-
From: sqlite-users [mailto:sqlite-users-boun...@mailinglists.sqlite.org] On
Behalf Of Richard Hipp
Sen
On 2016-09-08 12:17, Paul Sanderson wrote:
> How does this work for you to get all the even rows
>
> SELECT ROWID
> FROM table
> WHERE ROWID & 0x01 = 0x00
I think you should be able to create an expression index as well:
https://www.sqlite.org/lang_createtable.html#rowid
- Matthias-Christian
_
On 9/8/16, Stephen Chrzanowski wrote:
>
> However, the rabbit I was hoping to pull out of the hat was that the change
> in version numbers be done automatically when I make a change in the 3rd
> party DB management tool.
Perhaps make your application schema-version number a hash of the SQL
for th
Suggestion noted, and accepted, but, implementation of database table vs
user_schema, I'm not sure yet. I know your point isn't about where the
data is, but reference to another mechanism that is available to me.
Thanks.
However, the rabbit I was hoping to pull out of the hat was that the change
On 9/8/16, Stephen Chrzanowski wrote:
> That said, since I know that the backup API will make the schema version
> change, is it safe to read the schema_version at the beginning of the
> application, retain it for the life time of the application, then after the
> backup is written, rewrite the sc
I build my queries usually via binding within the function I'm calling,
which I don't think changes the schema version, since I'm only seeing a
bump of +1 at application close during either a vacuum or the backup API
call.
That said, since I know that the backup API will make the schema version
ch
Laura,
On Thu, Sep 8, 2016 at 5:22 AM, Laura BERGOENS
wrote:
> Hi everyone,
>
> So I've put indexes myself on the most used tables in my program, then
> ANALYZE the db, and now it flies.
> The automatic creation of index probably took a lot of time, and it affects
> even more performance when it'
> On 6 Sep 2016, at 10:19, Stephan Beal wrote:
>
> A counter-opinion, though apparently in the small minority: i _absolutely
> despise_ fixed-width web site layouts.
+1, including layouts with fixed maximum width.
___
sqlite-users mailing list
sqlit
Simon Slavin wrote:
> ... if your table has lots of rows and/or you do lots of queries like
> this, you should consider keeping a copy of the lower 16 bits as
> another integer column.
In these case, you should consider using an expression index:
sqlite> create table t(x,y);
sqlite> create index
How does this work for you to get all the even rows
SELECT ROWID
FROM table
WHERE ROWID & 0x01 = 0x00
Paul
www.sandersonforensics.com
skype: r3scue193
twitter: @sandersonforens
Tel +44 (0)1326 572786
http://sandersonforensics.com/forum/content.php?195-SQLite-Forensic-Toolkit
-Forensic Toolkit fo
On 8 Sep 2016, at 11:07am, Dave Blake wrote:
> Looking for the best way to query a table with an integer column by value
> of the lower 16 bits of the data in that column. Does SQLite support
> bitwise logic?
Yes. You can use the following
& | ~ << >>
AND OR NOT SHIFTLEFT SHIFTRIGHT
However
Looking for the best way to query a table with an integer column by value
of the lower 16 bits of the data in that column. Does SQLite support
bitwise logic?
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org
On 9/8/16, Simon Slavin wrote:
>
> On 8 Sep 2016, at 10:43am, Richard Hipp wrote:
>
>> On 9/7/16, David Empson wrote:
>>>
>>> every ATTACH DATABASE or
>>> DETACH DATABASE triggers a schema change...
>>
>> That's because the set of tables and indexes available to the query
>> planner changes, and
On 8 Sep 2016, at 10:43am, Richard Hipp wrote:
> On 9/7/16, David Empson wrote:
>>
>> every ATTACH DATABASE or
>> DETACH DATABASE triggers a schema change...
>
> That's because the set of tables and indexes available to the query
> planner changes, and so all of the SQL statements need to be
On 9/7/16, David Empson wrote:
>
> every ATTACH DATABASE or
> DETACH DATABASE triggers a schema change...
That's because the set of tables and indexes available to the query
planner changes, and so all of the SQL statements need to be reparsed
and replanned, to take into account those changes.
--
Hi everyone,
So I've put indexes myself on the most used tables in my program, then
ANALYZE the db, and now it flies.
The automatic creation of index probably took a lot of time, and it affects
even more performance when it's always same queries on same tables that are
repeated over and over again
On 8 Sep 2016, at 7:27am, Laura BERGOENS wrote:
> I took notes of everything you guys said, and I'll spend my morning doing
> this :
>
> Running ANALYZE once per DB I'mUsing, Create indexes myself and contact DB
> Browser to share them the thing.
Do those two steps the other way around. ANALY
33 matches
Mail list logo
