I did try using batch of transactions with synchronous=OFF PRAGMA set. The
performance improved slightly. But as db file started to grow larger and larger
in size, the performance degraded considerably. Is it expected? Or Is there a
work around for this?
Thanks
Raghu
-Original
Max,
Thanks.. Looks like I will have to optimize using transactions. I will try
that
-Original Message-
From: sqlite-users-boun...@sqlite.org [mailto:sqlite-users-boun...@sqlite.org]
On Behalf Of Max Vlasov
Sent: Tuesday, December 15, 2009 3:56 PM
To: General Discussion of SQLite
Hello!
On Tuesday 15 December 2009 19:05:31 D. Richard Hipp wrote:
> Maintaining a fork of SQLite using Fossil is not difficult.
The are some problems of usage:
1. Then I use reverse proxy-server Fossil does not check
the existence of X-Forwarded-For header and does not ask
login/password. Yes,
Hello!
On Tuesday 15 December 2009 19:05:31 D. Richard Hipp wrote:
> Maintaining a fork of SQLite using Fossil is not difficult. A sketch
> of one solution can be found at http://www.sqlite.org/
> privatebranch.html and there is an updated version of that document
> at
Cariotoglou Mike wrote:
> Just an idea :
>
> Parameters bound via sql_bind... *should* have affinity, since they
> are manifestly typed.
Affinity is determined and necessary conversions are encoded into the virtual
machine by sqlite_prepare, before any parameters are bound.
Just an idea :
Parameters bound via sql_bind... *should* have affinity, since they are
manifestly typed.
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
-Original Message-
From: sqlite-users-boun...@sqlite.org [mailto:sqlite-users-boun...@sqlite.org]
On Behalf Of P Kishor
Sent: Tuesday, December 15, 2009 2:48 PM
To: General Discussion of SQLite Database
Subject: Re: [sqlite] BACK API Questions
On Tue, Dec 15, 2009 at 2:44 PM,
On Tue, Dec 15, 2009 at 2:44 PM, Raghavendra Thodime
wrote:
> Hi Martin,
> Thanks for your reply.
>
> You mean to say that I need to modify sqlite3 code to choose which pages
> that need to be copied and copy them myself?
You don't need to modify sqlite3 code, but you
Hi Martin,
Thanks for your reply.
You mean to say that I need to modify sqlite3 code to choose which pages
that need to be copied and copy them myself? Or is there some other API that I
can use for this purpose? My main problem is I will have huge influx of real
time data which I
Hello!
On Monday 14 December 2009 11:33:32 Jens Miltner wrote:
> ...
You did not understand me. The manually named columns
may not be renamed automatically. But in the test the
column "id" was renamed to "id:1":
sqlite> create table test (id INTEGER PRIMARY KEY);
sqlite> insert into test
Hello!
On Tuesday 15 December 2009 18:35:18 wcl...@gfs-hofheim.de wrote:
> I have made you a patch file for just the READONLY and ENFORCE table
> constraints.
Please send me the original patch. All features are interesting and I want
to test them.
> I think if I post the patch here it will get
Hello!
On Tuesday 15 December 2009 19:17:31 wcl...@gfs-hofheim.de wrote:
> Trying to attach a non-existent database as "readonly" will return the
> error "unable to open database file".
This message is not helpful becouse is used in other situations. Is it
possible to change it?
Best
Hello Raghu,
it seems to mee that you misunderstand the backup api. This api does not
"know" or care for your schema and tables, but backups the database
block for block. In this sense, the api does not know "old" contents and
can not add "new" content.
The backup api can not help you in your
Any idea
-Original Message-
From: sqlite-users-boun...@sqlite.org [mailto:sqlite-users-boun...@sqlite.org]
On Behalf Of Raghavendra Thodime
Sent: Tuesday, December 15, 2009 10:15 AM
To: sqlite-users@sqlite.org
Subject: [sqlite] BACK API Questions
Hi,
I am using in-memory db
On Tue, Dec 15, 2009 at 06:58:03PM +0100, Angelo scratched on the wall:
> Hi to everybody,
>
> I'm using sqlite 3.6.16 both shell and C api.
> I need to make automatic check (from C code) of the opened database files
> (not
> connections) . With sqlite3_column_database_name I get just main,
pragma database_list returns record set as any select statement. So
you can capture this record set from C code as you do with any other
select statement and based on that make your checking as you need.
Pavel
On Tue, Dec 15, 2009 at 12:58 PM, Angelo wrote:
> Hi to
D. Richard Hipp wrote on 15/12/2009 17:05:31:
> (1) Have your patches been fully documented and have you generated
> automated test cases that provide 100% MC/DC and branch test
> coverage? I'm guessing not and yet those are requirements for new
> features in the core.
I believe they are
Alexey Pechnikov wrote on 15/12/2009 15:15:42:
>> 3. ENFORCE constraint for table columns, for example: CREATE TABLE t(i
>> enforce integer, j enforce text). This optional constraint enforces
type
>> checking so that an entry must match the column type (i.e. integer,
real,
>> numeric, text,
Hi,
I am using in-memory db where I will execute only inserts. Periodically, I
am trying to backup my in-memory db to file based db. But as soon as I backup,
I want to clear up the in-memory db so that I will have enough memory for
subsequent inserts. So next time when I back up I want to
On Dec 15, 2009, at 10:53 AM, Cariotoglou Mike wrote:
> Quoting from the documentation on Type Affinity:
>
By coincidence, I was working on rewriting and clarifying that
documentation just last night. The latest draft can be seen at
http://www.sqlite.org/draft/datatype3.html
My belief
On Dec 15, 2009, at 8:26 AM, wcl...@gfs-hofheim.de wrote:
>
> This is a quick email just to gauge interest in a number of patches
> I've
> created for Sqlite over the last year or so.
Your best bet would probably be to clone the SQLite Fossil repository
and publish your own private branch as
I was just wondering whether it might be possible in some future version
of SQLite to expose the pragma and compile time options settings in the
form of a table:
select * from sqlite_pragma ;
select * from sqlite_options ;
This might also allow update to set the value of a
> AH, GOT IT. values do NOT have affinity. so I would
> either need to cast the expression OR the value.
>
> is this correct ?
Yes, that's correct, you have to cast either left or right part to
assign affinity to it and to force conversion of another part.
Pavel
On Tue, Dec 15, 2009 at 10:53
> Why not just make epc column "unique on conflict ignore"?
> i.e.
> CREATE TABLE readmode( epc text unique on conflict ignore, col2 text... );
>
>> one EPC ( say:'e2003411b802010994095761' i.e. 24char ) out of 5000 records.
>> and select query take 2-3 sec. to give result output.
>>
>> By any
Quoting from the documentation on Type Affinity:
"
SQLite may attempt to convert values between the numeric storage classes
(INTEGER and REAL) and TEXT before performing a comparison. Whether or
not any conversions are attempted before the comparison takes place
depends on the nominal affinity
John Brooks wrote on 15/12/2009 15:13:05:
> What license do you put these under?
Sorry, I forgot to say in my last email: I'm happy for these to end up in
the public domain like Sqlite -- I don't think realistically I could do
otherwise given the blessing-style licence that the main Sqlite
Apparently result of count() and probably all other aggregate
functions (as well as result of any function at all) has no affinity.
So when you compare it to anything having no affinity too you have no
type transformation during comparison. And thus values have to be
exactly the same including
John Brooks wrote on 15/12/2009 15:13:05:
> I love the idea of READONLY and ENFORCE. I would certainly make use of
> those in amalgamation form. What license do you put these under?
>
> You should really publish these all somewhere, there are some great
> features. Good work.
Thank you.
I have
2009/12/15 D. Richard Hipp :
>
>>
>
> Because string '2' is not the same thing as integer 2.
>
> sqlite3> select 2='2';
> 0
> sqlite3>
>
Why
sqlite> select cast( 2 as integer ) = '2';
1
>
> D. Richard Hipp
> d...@hwaci.com
>
Simon
___
>
> essence of problem:
>
> why is there a difference in output between the following selects?
>
> sqlite> select * from t1 where (select count(*) from t2 where
> t2.id=t1.id) = '2';
> sqlite> select * from t1 where cast((select count(*) from t2 where
> t2.id=t1.id)
> as integer)= '2';
> 2
>
2009/12/15 Cariotoglou Mike :
> Simon, thanks for the answer. your code helped track down the issue,
> which I *still* believe to be a bug:
>
.
.
.
>
> select * from
> (select *,"
> cast((select count(*) from ITEM_ARTIST where
> ARTIST_id=artists.artist_id) as int) CNT
> from
On Tue, Dec 15, 2009 at 6:16 AM, Florian Schricker wrote:
> Hello Simon,
>
>
>> From: Simon Slavin
>> On 14 Dec 2009, at 8:52am, Florian Schricker wrote:
>>
> Primary keys are Oper, Product, Category, Name and CreateTS
There is only one
Original Message
Subject: Re: [sqlite] SQLite version 3.6.21
From: Andreas Schwab
To: sqlite-users@sqlite.org
Date: 2009-12-08 02:41
> $ ./sqlite3 :memory: 'create table test(integer)'
> Segmentation fault
We have the same problem and with this posted
Hello!
On Tuesday 15 December 2009 16:26:00 wcl...@gfs-hofheim.de wrote:
> 3. ENFORCE constraint for table columns, for example: CREATE TABLE t(i
> enforce integer, j enforce text). This optional constraint enforces type
> checking so that an entry must match the column type (i.e. integer,
Florian Schricker wrote:
> Hello Simon,
>
SNIP
> Please excuse me for being so blunt:
> You have no idea what I'm supposed to do here nor do you have any idea
> for whatever historic reason the database or table is designed "as is"
> here nor (and finally) do you have no idea about my education
Simon, thanks for the answer. your code helped track down the issue,
which I *still* believe to be a bug:
The issue is not with parameters, after all, but with comparison of
values, and something "magic" about count(), I suspect.
It has to do with the fact that the layer I am using (PHP PDO)
On 12/15/09 13:16 , "Florian Schricker" wrote:
> Please excuse me for being so
> blunt:
You have no idea what I'm supposed to do here nor do you have any
> idea
for whatever historic reason the database or table is designed "as
> is"
here nor (and finally) do you have no
Hi all,
This is a quick email just to gauge interest in a number of patches I've
created for Sqlite over the last year or so. What I want to know is
whether anyone would like me to post one or more of these patches here? I
have thought about posting to http://www.sqlite.org/contrib but since
Hello Simon,
> From: Simon Slavin
> On 14 Dec 2009, at 8:52am, Florian Schricker wrote:
>
Primary keys are Oper, Product, Category, Name and CreateTS
>>> There is only one primary key per table.
>>
>> So to say the primary key is "Oper, Product, Category, Name,
Hi,
I am using in-memory db where I will execute only inserts. Periodically, I
am trying to backup my in-memory db to file based db. But as soon as I backup,
I want to clear up the in-memory db so that I will have enough memory for
subsequent inserts. So next time when I back up I want to
2009/12/15 Cariotoglou Mike :
> I checked in the bug database, and this does not seem to have been
> reported, and IMHO it is definitely a bug.
> workarounds exist,as pointed out by me and others. still, I would like
> to hear from the core team whether this is
> recognized as a
Hello
F Aha! now I am starting to understand the utility of that data.sqlite file!
F Thanks again, Simon, and I will look for a PHP forum. Do you know a good
F one?
I've find that the PHP manual:
http://www.php.net/manual/en/
answers almost all my questions.
Swithun.
Aha! now I am starting to understand the utility of that data.sqlite file!
Thanks again, Simon, and I will look for a PHP forum. Do you know a good
one?
Frank L.
Simon Slavin-3 wrote:
>
>
> On 15 Dec 2009, at 8:24am, FrankLane wrote:
>
>> In page 1 I have defined:
>>
>> $dbHandle = new
I checked in the bug database, and this does not seem to have been
reported, and IMHO it is definitely a bug.
workarounds exist,as pointed out by me and others. still, I would like
to hear from the core team whether this is
recognized as a bug, and will be dealt with at some point in time.
btw, I
2009/12/15 Ram Mandavkar :
>
> Hello All,
>
> I am New to SQL-Lite DB,
> I had one ASSET table having EPC as varchar(50) field and 5000 records in
> it.
>
> If i want to search
> (
> CREATE TRIGGER "main"."TR_MULTIREAD_BEFORE" BEFORE INSERT ON READMODE
> BEGIN
On 15 Dec 2009, at 8:24am, FrankLane wrote:
> In page 1 I have defined:
>
> $dbHandle = new PDO('sqlite:data.sqlite');
>
> $var='hello';
>
> I can pass a $var variable to page 2 using using the URL as in
>
> echo " page2.php?var=$var GO TO PAGE 2 ";
Instead do this:
echo "
Hello All,
I am New to SQL-Lite DB,
I had one ASSET table having EPC as varchar(50) field and 5000 records in
it.
If i want to search
(
CREATE TRIGGER "main"."TR_MULTIREAD_BEFORE" BEFORE INSERT ON READMODE
BEGIN
SELECT CASE
WHEN M.a = NEW.EPC THEN
Hello - I have two pages. In page 1 I have defined:
$dbHandle = new PDO('sqlite:data.sqlite');
$var='hello';
I can pass a $var variable to page 2 using using the URL as in
echo " page2.php?var=$var GO TO PAGE 2 ";
and I want to execute an sqlite/PHP command in page 2 like:
48 matches
Mail list logo