Re: [sqlite] sqlite with Visual Basic

[Jalil Vaidya]

Seems like the only thing that can help you is in
yahoo group's file section ;-)

> I am not yahoo group member
Become a member, its free. Get the files and then
unsubscribe.

HTH,

Jalil Vaidya

--- Jorge <[EMAIL PROTECTED]> wrote:
> No, no...
> 
> there is not mentioned VBWrapper.zip
> 
> 
> 1) DLL for VB 6/5/4 like as ADO free!! by Javier
> Bermudez from Argentina.
>
http://free.hostdepartment.com/j/javierbermudez/archivos/sqlite/index.html
> 
> isn't active
> 
> Visual Basic 6
> http://www.ag-software.com/SQLite/default.asp
> it's OK, but it isn't VBWrapper.zip
> 
> SQLHelper:
> http://groups.yahoo.com/group/sqlite/files/ (Works
> with VB5 &
> VB6; Get both pssql.zip and psvbutls32.zip )
> I am not yahoo group member
> 
> SQLitePlus COM DLL wrapper for SQLite (coded with
> C++/ATL).
> http://www.sqliteplus.com ($$)
> it isn't my favourite price!!!
> 
> TIA
> 
> Jorge
> 
> 
> 
> - Original Message - 
> From: "Jalil Vaidya" <[EMAIL PROTECTED]>
> To: "Jorge" <[EMAIL PROTECTED]>; "Carlos
> Garces"
> <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>
> Sent: Thursday, June 17, 2004 5:50 PM
> Subject: Re: [sqlite] sqlite with Visual Basic
> 
> 
> Check out COM Wrappers / Visual Basic DLLs section
> in
> the wiki:
> http://www.sqlite.org/cvstrac/wiki?p=SqliteWrappers
> 
> HTH,
> 
> Jalil Vaidya
> 
> --- Jorge <[EMAIL PROTECTED]> wrote:
> > Hi,
> >
> > Are there any site whehe can I found the mentioned
> > VBWrapper.zip?
> >
> > I have tried in http://f4.grp.yahoofs.com/..., but
> > this URL isn't OK
> >
> > TIA
> >
> > Jorge Colaccini
> 
> 
> =
> 01001010
> 0111
> 01101100
> 01101001
> 01101100
> 
> 
> 
> 
> __
> Do you Yahoo!?
> New and Improved Yahoo! Mail - 100MB free storage!
> http://promotions.yahoo.com/new_mail
> 
>
-
> To unsubscribe, e-mail:
> [EMAIL PROTECTED]
> For additional commands, e-mail:
> [EMAIL PROTECTED]
> 
> 
> 
>
-
> To unsubscribe, e-mail:
> [EMAIL PROTECTED]
> For additional commands, e-mail:
> [EMAIL PROTECTED]
> 
> 


=
01001010
0111
01101100
01101001
01101100

__
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

RE: [sqlite] RE: SQLite performance with mid-size databases

[Richard Kuo]

Point certainly taken.  I was trying to avoid getting into too much
detail without first confirming if this was some well known issue.

I'll put together a standalone test app which will demonstrate the poor
loading/insert performance.  I'll get to the read performance after
that.

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] 
Sent: Thursday, June 17, 2004 3:15 AM
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]
Subject: Re: [sqlite] RE: SQLite performance with mid-size databases


Richard,

I notice you don't seem to be getting useful replies on this list, and I

think perhaps that it is in part a lack of useful input on your part
that 
is holding up the works. Here is my suggestion:
1) Take a copy of your working database
2) Try the exact sql query you believe to be scaling linearly
3) Prove to your satisfaction that this really is happening
4) Share with us the shortest example you can come up with that shows
the 
problem

If you can't manage to isolate it in this way, I suggest that you put
some 
debug code in to capture a log of all SQL statements you're issuing to 
sqlite while you're experiecing the problem. Try narrowing it down to a 
subset of those queries. On the other hand, perhaps is simply the number

of times you're making each kind of query. Keep an eye on exactly what
the 
most common operations are and find out if you're calling them in a way 
that scales linearly with your database size.

On a "stab in the dark" note, the COUNT operator is something often used

in scrolling tabular displays. It requires a linear scan of all table 
data, so if you're doing COUNTs don't. Try keeping your own records of
how 
many rows are in your tables using triggers (I belive you will find
sample 
code that does this in the list archives). LIMIT and OFFSET may also be 
used in such displays. Beware that OFFSET may be scanning through the 
results of your query and explicitly ignoring them in the same way as 
count would. Naturally, beware of any queries that don't fit naturally
to 
your indicies and use .explain to ensure that the indices are being
used.

Good luck.

Benjamin.





"Richard Kuo" <[EMAIL PROTECTED]>
17/06/2004 03:26 PM

 
To: "'Ionut Filip'" <[EMAIL PROTECTED]>
cc: <[EMAIL PROTECTED]>
Subject:[sqlite] RE: SQLite performance with mid-size
databases


Ionut,

Thank you for your input.  Do you mean precompilation and then
subsequent calls to the sqlite_bind API?

Am definitely using transactions, and the database is entirely read-only
with only the client in question accessing the db.

The poor and linearly scaling performance degradation with size is very
odd. I can make no sense of why this would happen unless SQLite has
terrible locality of reference issues.  I will post a schema later today
that may hopefully shed some light on the issue.

Richard

-Original Message-
From: Ionut Filip [mailto:[EMAIL PROTECTED] 
Sent: Tuesday, June 15, 2004 11:54 PM
To: [EMAIL PROTECTED]
Subject: SQLite performance with mid-size databases


Hi Richard,

Here is a tip to speed up the insert operations: use prepared statements
instead of plain INSERTs. If you have to insert more than one record
with the same format the performance increase is significant.

Transactions also speeds up db operations, there are more detailed
topics about this on the list.

> The speed results on the website [...]
Here I think you need to make sure the concurrency access to the DB is
not a bottleneck.

Ionut Filip

PS: You can also try to post a sample (partial) schema of your database.


-Original Message-
From: Richard Kuo [mailto:[EMAIL PROTECTED]
Sent: Wednesday, June 16, 2004 7:04 AM
To: [EMAIL PROTECTED]
Subject: [sqlite] SQLite performance with mid-size databases


Hi.  We are using SQLite to store and retrieve data rows where each
row is roughly 2K total in size and in a table of 15 columns.  The total
size of the database ranges from 100-300 MB.
 
The problem we are seeing is that query and insert performance is
unusually bad and scales up linearly with database size.  Compared to MS
Access, the query times are several times slower.  Frankly I was a bit
shocked at this considering that most people seem to think the
performance is good. However, I don't see anything that we are doing
wrong...we query the rows we want only by rowid.  I'm very puzzled that
this hasn't come up a lot in my searches of the mailing list, but
perhaps the slower query times aren't a concern for many of the
applications using SQLite.
 
Empirically speaking, we display our data in a scrolling 2
dimensional grid format.  With MS access, this grid responds
instantaneously when moving through the grid.  With SQLite, there is
very noticable stalling and lag and the disk i/o is higher than MS
Access by roughly a factor of 10.
 
I suppose I am looking to see if anyone is seeing the same results
that I am seeing, and wondering if this is 

Re: [sqlite] RE: SQLite performance with mid-size databases

[ben . carlyle]

"Ismail Kizir" <[EMAIL PROTECTED]>
17/06/2004 06:48 PM

 
To: <[EMAIL PROTECTED]>
cc: 
Subject:Re: [sqlite] RE: SQLite performance with mid-size databases


> I had the same problem as Richard's.
> Unfortunately, i have decided to use postgres and finally back to mysql.

If you're happy with your current solution, feel free to ignore the rest 
of this email. All one can hope for is to be happy with one's present 
solution :)

> I have a table of ~300,000 rows of newspaper articles, and its fulltext
> index tables.

Sqlite doesn't support fulltext indexes. I suggest explaining what you 
were trying to achieve, and how.

> Table opening time was very high.

There's no such thing as table opening time in sqlite :) What was being 
done to sqlite during "table open time"?

> Concurrency -even only for reading- was very very poor.

What was the nature of your concurrent access? Were any writers active? 
Did readers use transactions?

> Especially, i observed the linear write time just as Richard did.

I'm not sure what you're saying here. A single insert took O(n) time where 
n was the size of the database, or O(n) time where n is the size of the 
row data? Was it really O(n), or was it O(log(n)) as modern databases such 
as sqlite will normally guarantee? How much data was in the row you were 
inserting? How did you measure the linear growth (is this in the archives 
somewhere?)?

> I think SQLite is still a good solution for embedded system, but not 
even
> for mid-size databases.

I'm using sqlite in heavily-loaded system consisting of database files 
that are created over a two-day period. The average database file is about 
800 meg. After extensive testing early in the piece I concluded that only 
sqlite was suitable. All alternative technoligies I tested were far too 
slow for my (now aging) sun hardware. I tested sqlite, postgres, mysql, 
and sapdb. Anecdotally, I'd have to say that SQLite is ideal for targeted 
mid-size databases.

Benjamin.



-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [sqlite] SQLite performance with mid-size databases

[ben . carlyle]

Raymond,

So far on this list I have only seen a reports of scalability problems 
with sqlite when the real problems were specific queries being given to 
sqlite that were constructed in a way that sqlite does not process well. 
In every such case I can recall an alternative form of the query was able 
to be produced (usually by DRH personally) that did not exhibit the 
problem. Knowing the code, there is no reason for sqlite to be scaling 
linearly in queries unless it is being forced to do table scans.

It is true that sqlite doesn't optimise queries as well as major 
databases. It's not designed to. Users must take some care in constructing 
their queries and ensuring the queries are suited to sqlite's design if 
performance might be a problem. At the same time, such queries usually 
perform much better than those of major databases because of the vastly 
reduced optimisation and connection overhead that sqlite affords.

As with every database technology. If you care about performance you have 
to understand some things about the design of your underlying technology. 
For sqlite the design is simple and the experts are extremely responsive. 
If you're having problems and can provide a clear, specifc description of 
your problem you will get help. While queries such as "I think sqlite 
scales linearly, but I can't tell you want queries I'm issuing to make 
that happen" and "My scroll list seems slow, but I don't know what is 
happening between the GUI and the database or what queries are going on" 
are unlikely to solicit helpful response, "I am issuing this query on that 
database schema with about 100 thousand rows each carrying 2k of data" is 
likely to be something members of this list can help you solve.

I'd like to take this soapbox opportunity to again thank drh and his 
associates for the wonderful work they put into sqlite and their genine 
personal commitment to this project. Sqlite is a great product, and a 
well-targeted one.

Benjamin





Raymond Irving <[EMAIL PROTECTED]>
18/06/2004 12:09 AM

 
To: [EMAIL PROTECTED]
cc: 
Subject:Re: [sqlite] SQLite performance with mid-size databases



Well this does not sound good at all. I would think
that SQLite would at least do a better job at queries.

The issues with performance and scalability is of
great concern. I was planning on create some new apps
that use SQLite, but now I'm wondering is this little
database is capable of handle over 2 Gigabytes (GB) of
data even though the docs says it can handle up to 2
Terabytes (TB).

Does it really make sense to cut back on performance
in order to keep the library size small?

Will SQLite 3.0 fix these problems?

__
Raymond Irving

--- "D. Richard Hipp" <[EMAIL PROTECTED]> wrote:
> Richard Kuo wrote:
> > 
> > I suspect some unnecessary disk access has to be
> the problem...despite
> > the small amount of new guide information being
> queried out, disk bytes
> > read is several times higher than with MS access
> and scrolling back over
> > previously accessed areas of data is visibly
> faster...indicating that
> > the disk cache is very favorably impacting the
> speed of the queries.
> > 
> 
> If each of your rows contains 2K of data, that means
> each database entry
> is using about 2 overflow pages.  You can change
> this by increasing the
> page size.  Try recompiling SQLite after changing
> the SQLITE_PAGE_SIZE
> macro to 16384.
> 
> You might also trying switching to SQLite version
> 3.0.0 which will be
> released tomorrow.
> 
> Also tomorrow, I will be making available a database
> analysis tool
> for version 2.8 databases that will help us to
> better understand
> how information is stored on disk for your database,
> and possibly
> provide some clues about why you are having
> problems.





-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [sqlite] sqlite with Visual Basic

[Jorge]

No, no...

there is not mentioned VBWrapper.zip


1) DLL for VB 6/5/4 like as ADO free!! by Javier Bermudez from Argentina.
http://free.hostdepartment.com/j/javierbermudez/archivos/sqlite/index.html

isn't active

Visual Basic 6 http://www.ag-software.com/SQLite/default.asp
it's OK, but it isn't VBWrapper.zip

SQLHelper: http://groups.yahoo.com/group/sqlite/files/ (Works with VB5 &
VB6; Get both pssql.zip and psvbutls32.zip )
I am not yahoo group member

SQLitePlus COM DLL wrapper for SQLite (coded with C++/ATL).
http://www.sqliteplus.com ($$)
it isn't my favourite price!!!

TIA

Jorge



- Original Message - 
From: "Jalil Vaidya" <[EMAIL PROTECTED]>
To: "Jorge" <[EMAIL PROTECTED]>; "Carlos Garces"
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>
Sent: Thursday, June 17, 2004 5:50 PM
Subject: Re: [sqlite] sqlite with Visual Basic


Check out COM Wrappers / Visual Basic DLLs section in
the wiki:
http://www.sqlite.org/cvstrac/wiki?p=SqliteWrappers

HTH,

Jalil Vaidya

--- Jorge <[EMAIL PROTECTED]> wrote:
> Hi,
>
> Are there any site whehe can I found the mentioned
> VBWrapper.zip?
>
> I have tried in http://f4.grp.yahoofs.com/..., but
> this URL isn't OK
>
> TIA
>
> Jorge Colaccini


=
01001010
0111
01101100
01101001
01101100




__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

[sqlite] Assertion failure in pager.c

[James W. Walker]

How much should I worry about this?
Assertion (pPager->ckptJSize == 
pPager->nRec*JOURNAL_PG_SZ(journal_format)+JOURNAL_HDR_SZ(journal_format)) 
failed in "pager.c", line 2128

This is in version 2.8.13.  After resaving the file, the assertion no 
longer fails.
--
  James W. Walker, ScriptPerfection Enterprises, Inc.
  

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [sqlite] sqlite with Visual Basic

[Jalil Vaidya]

Check out COM Wrappers / Visual Basic DLLs section in
the wiki:
http://www.sqlite.org/cvstrac/wiki?p=SqliteWrappers

HTH,

Jalil Vaidya

--- Jorge <[EMAIL PROTECTED]> wrote:
> Hi,
> 
> Are there any site whehe can I found the mentioned
> VBWrapper.zip?
> 
> I have tried in http://f4.grp.yahoofs.com/..., but
> this URL isn't OK
> 
> TIA
> 
> Jorge Colaccini


=
01001010
0111
01101100
01101001
01101100




__
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail 

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

RE: [sqlite] Is there a demand for Client / Server Modules ? (It seems to be..)

[Hihn, Jason]

I'm joining this late, but I just updated the wiki.

Appearently SQLRelay already has this worked out?
http://sqlrelay.sourceforge.net

-Original Message-
From: Nicolas Girard [mailto:[EMAIL PROTECTED] 
Sent: Thursday, June 17, 2004 2:44 PM
To: [EMAIL PROTECTED]
Subject: Re: [sqlite] Is there a demand for Client / Server Modules ?
(It seems to be..)

On Thursday 17 June 2004 18:54, Alex K wrote:
> I didn't expect such a quick response...
>
> And unfortunatelly the link I gave was broken :-(
>
> So for people who don't want to wait until SourceForge release,
> I made a quick-and-dirty pre-release on my site:
> http://www.it77.de/sqlite/sqlite_sockets.htm
>
> Best regards
>
> Alex K.

Alex,
this is indeed a very nice contribution !

If you agree, I'll talk about this in the SQLite 3.0 announcement I
intend to 
post to linuxfr.org tomorrow, after Richard Hipp's released it. I'm
quite 
sure many people will be glad to hear about it, even if it is an early 
release.

In the meantime, let me ask you a few questions:

- do you intend to adapt your code to the forthcoming 3.0 version of
SQLite ?

- could you please detail your troubles under Linux, and perhaps add a
request 
for help from thread experts on your web page ? As your project is very 
useful, you may receive some help quickly (not from me, unfortunately
:-/ )

Many thanks anyway for your effort,

cheers
Nicolas

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [sqlite] Upping sizeof integer

[D. Richard Hipp]

Hihn, Jason wrote:
I'd like to know how hard it would be to make "integer primary key" be
64 bits wide. I have and application that will burn through 2^32 in
about 3 months in the main index table. I know I'll be well under the
size limitation of 2^41 though. 

64-bit rowids are support in SQLite version 3.0.0 which will be
released tomorrow morning.
--
D. Richard Hipp -- [EMAIL PROTECTED] -- 704.948.4565
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

[sqlite] Upping sizeof integer

[Hihn, Jason]

I searched the archives, and only found info relating to the proposal
for v3.

 

I'd like to know how hard it would be to make "integer primary key" be
64 bits wide. I have and application that will burn through 2^32 in
about 3 months in the main index table. I know I'll be well under the
size limitation of 2^41 though. 

 

I don't need to be compatible with any existing databases.

 

Thank you.

Re: [sqlite] SQLite performance with mid-size databases

[D. Richard Hipp]

Ulrik Petersen wrote:
this is also just a stab in the dark, and I subscribe to the digest 
version of the mailinglist, so I may not have the latest.

Perhaps you are doing something like
SELECT A.x, A.y
FROM A
WHERE   A.rowid = xxx
OR  A.rowid = yyy
OR  A.rowid = zzz
OR  A.rowid = ...
etc.etc. with may OR-conditions.
I have noticed that SQLite (and PostgreSQL, for that matter) slows down 
quite a bit when the number of WHERE-conditions reaches beyond a 
somewhat low number, say around 10.

In SQLite, any use of the OR operator in a WHERE clause more or
less shuts down the query optimizer, disables all indices, and
forces a full table scan.  If you want to write a query like the
one shown above, do it this way:
   SELECT * FROM A
   WHERE A.rowid IN (xxx,yyy,zzz,...);
The optimizer understands the IN operator just fine.
SQLite's query optimizer can handle up to 32 AND terms in the
WHERE expression before it begins to have problems.
--
D. Richard Hipp -- [EMAIL PROTECTED] -- 704.948.4565
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [sqlite] RE: SQLite performance with mid-size databases

[D. Richard Hipp]

Jacob Engstrand wrote:
>
> On 2004-06-17, at 16.39, [EMAIL PROTECTED] wrote:
>
>> If you have a large record size, then it also helps tremendously to put the
>>  indexed columns first in the row (I had a table with large blob data that
>> was terribly slow to query until I moved the small, non-blob columns
>> first). [That was found out by trial and error]
>>
>
>
> Really!? Let's see if I understand you correctly: Your SELECT statement is
> exactly the same, but you just changed the order of the columns in one table,
> and all of a sudden your SELECT query performed better?
>
SQLite stores data in column order.  The first column is stored first,
the second column second, and so forth.  The first 236 bytes of data
are stored directly on the B-Tree pages.  Any information beyond the
first 236 bytes is stored on separate overflow pages.
So if you are doing a query that requires checking the value
of some column whose data appears past the 236 byte mark
then overflow pages will need to be read from disk for each
row examined, even if the test turns out to be false and the
content of that row never makes it into the result set.
Reading all those extra overflow pages takes time.
The 236-byte figure is for SQLite version 2.8.  Version 3.0
will store up to 1000 bytes or maybe a little more before
resorting to using an overflow page - and even more than
that if you choose a page size larger than 1024 bytes.
--
D. Richard Hipp -- [EMAIL PROTECTED] -- 704.948.4565
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [sqlite] RE: SQLite performance with mid-size databases

[Jacob Engstrand]

On 2004-06-17, at 16.39, [EMAIL PROTECTED] wrote:
If you have a large record size, then it also helps tremendously to 
put the
indexed columns first in the row (I had a table with large blob data 
that
was terribly slow to query until I moved the small, non-blob columns 
first).
[That was found out by trial and error]


Really!? Let's see if I understand you correctly: Your SELECT statement 
is exactly the same, but you just changed the order of the columns in 
one table, and all of a sudden your SELECT query performed better?

I do believe you if you say so, but I'm just very curious as to why? If 
not too much trouble, would you please post the query and the table 
schema?

thanks
/jak

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

[sqlite] Re: Réf. : Re: [sqlite] Is there a demand for Client / Server Modules ?

[Alex K]

As you can see, there is old 
release.
It is correct. For most cases it is sufficient 

to replace sqlite.dll with given on 
homepage.
And run the server on the remote 
machine.
 
The SqliteBrowser is only for 
testing!!!
 
BUT:
 
Current State is not on the homepage. I am 
currently working on this.
In the new release the sqlite.dll will be 
sqlite_client.dll
You can still rename it to sqlite.dll and use as 
described above.
 
The Server will require original sqlite.dll (so 
you see, why i had to rename client lib)
Which you can get from www.sqlite.org. or compile with your favorite 
compiler.
 
The same procedure under Linux: 
Client uses sqlite_client.so for remote calls. 
Server accept the calls and uses 
original sqlite.so for local database 
access. 
 
Best regards
 
Alex K.
 
 

  - Original Message - 
  From: 
  Gregory Letellier 
  To: [EMAIL PROTECTED] 
  Sent: Thursday, June 17, 2004 6:01 
  PM
  Subject: Réf. : Re: [sqlite] Is there a 
  demand for Client / Server Modules ?
  
  

  
very good work !!
but if we want to use a remote sqlite base without sqlitebrowser, 
we must use the sqlite.dll you give in your site ? 
 
---Message 
original---
 

De : Alex K
Date : 06/17/04 
17:49:41
A : [EMAIL PROTECTED]
Sujet : Re: [sqlite] 
Is there a demand for Client / Server Modules ?
 
Sorry for broken link. It is http://www.it77.de/sqlite/sqlite.htm
 
 
- Original Message -
From: "Alex K" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, June 17, 2004 4:20 PM
Subject: [sqlite] Is there a demand for Client / Server Modules 
?
 
 
> I am about to register a project at SourceForge which
> provides the Client/Server facility to SQLite. Using
> SQLite over NFS is known as unsafe.
>
> Some description is in Wiki/ClientServer and at www.it77.de/sqlite
>
> The Windows version is currently working quite stable.
> The Linux version has sometimes strange problems (perhaps 
because of
> pthreads)
>
> Is there any demand for this ???
>
> -
> Alex K.
>
>
> 
-
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
 
 
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
 
.

  

  
  

Re: [sqlite] Is there a demand for Client / Server Modules ?

[Alex K]

Sorry for broken link. It is http://www.it77.de/sqlite/sqlite.htm


- Original Message - 
From: "Alex K" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, June 17, 2004 4:20 PM
Subject: [sqlite] Is there a demand for Client / Server Modules ?


> I am about to register a project at SourceForge which
> provides the Client/Server facility to SQLite. Using
> SQLite over NFS is known as unsafe.
> 
> Some description is in Wiki/ClientServer and at www.it77.de/sqlite
> 
> The Windows version is currently working quite stable.
> The Linux version has sometimes strange problems (perhaps because of
> pthreads)
> 
> Is there any demand for this ???
> 
> -
> Alex K.
> 
> 
> -
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
> 


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

[sqlite] Is there a demand for Client / Server Modules ?

[Alex K]

I am about to register a project at SourceForge which
provides the Client/Server facility to SQLite. Using
SQLite over NFS is known as unsafe.

Some description is in Wiki/ClientServer and at www.it77.de/sqlite

The Windows version is currently working quite stable.
The Linux version has sometimes strange problems (perhaps because of
pthreads)

Is there any demand for this ???

-
Alex K.


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [sqlite] SQLite performance with mid-size databases

[Raymond Irving]

See below:

--- Jacob Engstrand <[EMAIL PROTECTED]> wrote:
>  To disable the
> I1A index,
> rewrite the WHERE clause like this:
> 
>   SELECT * FROM t1 WHERE a+0=5 AND b='xyzzy';
> 
> To disable the I1B index you could write
> 
>   SELECT * FROM t1 WHERE a=5 AND b LIKE 'xyzzy';
> 

Why not add a feature to SQLite to allow the user to
choose the index to use? Maybe something like this:

SELECT * FROM t1 WHERE a=5 abd b='xyzzy' USE INDEX(a)

Wouldn't this be much more straiht forward and easier
to read?


__
Raymond Irving


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [sqlite] SQLite performance with mid-size databases

[Raymond Irving]

Well this does not sound good at all. I would think
that SQLite would at least do a better job at queries.

The issues with performance and scalability is of
great concern. I was planning on create some new apps
that use SQLite, but now I'm wondering is this little
database is capable of handle over 2 Gigabytes (GB) of
data even though the docs says it can handle up to 2
Terabytes (TB).

Does it really make sense to cut back on performance
in order to keep the library size small?

Will SQLite 3.0 fix these problems?

__
Raymond Irving

--- "D. Richard Hipp" <[EMAIL PROTECTED]> wrote:
> Richard Kuo wrote:
> > 
> > I suspect some unnecessary disk access has to be
> the problem...despite
> > the small amount of new guide information being
> queried out, disk bytes
> > read is several times higher than with MS access
> and scrolling back over
> > previously accessed areas of data is visibly
> faster...indicating that
> > the disk cache is very favorably impacting the
> speed of the queries.
> > 
> 
> If each of your rows contains 2K of data, that means
> each database entry
> is using about 2 overflow pages.  You can change
> this by increasing the
> page size.  Try recompiling SQLite after changing
> the SQLITE_PAGE_SIZE
> macro to 16384.
> 
> You might also trying switching to SQLite version
> 3.0.0 which will be
> released tomorrow.
> 
> Also tomorrow, I will be making available a database
> analysis tool
> for version 2.8 databases that will help us to
> better understand
> how information is stored on disk for your database,
> and possibly
> provide some clues about why you are having
> problems.
> 
> -- 
> D. Richard Hipp -- [EMAIL PROTECTED] -- 704.948.4565
> 
> 
>
-
> To unsubscribe, e-mail:
> [EMAIL PROTECTED]
> For additional commands, e-mail:
> [EMAIL PROTECTED]
> 
> 


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Re: [sqlite] ISO 8859

[Bertrand Mansion]

Greg Obleshchuk wrote:

>Hello,
>I had this guy email me about a problem about using ISO8859 character .
He
>has this in the DB 
>Handhilfsbetätigungssatz
> 
>but when queried using my wrapper it returns Handhilfsbet..tigungssatz
> 
>When using SQLITE.EXE the results are displayed fine.  When I debug the
code
>the char * returned in the call-back event for this column has this
>character as value -124 .  Which I assume is an overflowed 7 bit value.
Can
>anyone help me understand how to fix this in the source?

I am not an expert in encodings but just a guess, this might be related
to the fact that the "Handhilfsbetätigungssatz" word was typed on
another machine using another encoding ? For example, if your wrapper
does not enforce the use of ISO-8859-1, then if the user is on a Mac,
the characters inserted will use the Mac Roman encoding. Then, if you
display it back on Windows, the characters will show differently if you
don't convert them. 

Of course, I might be all wrong here as I am trying to figure that out
as well myself :)
In my application, I think I will let the user choose which encoding she
wants to use (through Preferences for example).

The safe way might be to use UTF-8.
But this might make the databases bigger ?
And there are apparently problems using the UTF-8 version of sqlite with
PHP (although I still don't understand what the problems can be ???).

Bertrand Mansion
Mamasam

-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

[sqlite] Need some solution for this problem.

[rajneesh . khanna]

HI all,
 I have / trying to port SQLite to VxWorks . But am having some
problems with the following functions
in file util.c
sqliteSetString() and sqliteSetNString() . When i try to create a database
for the first time i.e. i call sqlite_open() the following function gets
called

sqliteAddColumn() where the function sqliteSetNString() gets called but
memory allocation fails cos it asks for too much memory . In my case malloc
allows a maximum of 128mb . Still, am surprised how it manages to cross the
128mb limit.
Any help or suggestions would be really helpful .

Thanx
with regards
Rajneesh Khanna
DVD group,
Philips Innovation Campus
Ph: 5579000



-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

[sqlite] ISO 8859

[Greg Obleshchuk]

Hello,
I had this guy email me about a problem about using ISO8859 character .  He
has this in the DB 
Handhilfsbetätigungssatz
 
but when queried using my wrapper it returns Handhilfsbet..tigungssatz
 
When using SQLITE.EXE the results are displayed fine.  When I debug the code
the char * returned in the call-back event for this column has this
character as value -124 .  Which I assume is an overflowed 7 bit value.  Can
anyone help me understand how to fix this in the source?
 
thanks 
Greg