Re: [sqlite] SQLite parsing of a .sql file

[Shaun Seckman (Firaxis)]

Checking for NULL on the statement fixed it :)

-Original Message-
From: sqlite-users-boun...@sqlite.org
[mailto:sqlite-users-boun...@sqlite.org] On Behalf Of Roger Binns
Sent: Friday, April 09, 2010 5:43 PM
To: General Discussion of SQLite Database
Subject: Re: [sqlite] SQLite parsing of a .sql file

On 04/09/2010 01:25 PM, Shaun Seckman (Firaxis) wrote:
> I'm running into a nasty problem towards the end of the file however.
> At the very end there are 3 characters left that are just whitespace
> characters.  When I pass this string into sqlite3_prepare_v2() I get a
> return value of 0 as if it's a valid statement.

http://sqlite.org/c3ref/prepare.html

   If the input text contains no SQL (if the input is an empty string
   or a comment) then *ppStmt is set to NULL

Similarly your file could contain this.  I suggest using it for testing.

    select 3 


 > However, when I attempt
> to call sqlite3_sql() I will get a crash in sqlite.

Don't do that then.  prepare already told you there was no SQL.  Note
that 
pzTail is still set correctly.

> If I call
> sqlite3_step() I receive the error code SQLITE_MISUSE.

Don't do that either :-)

> What would be very useful is if whitespace or comments were the only
> data passed into sqlite3_prepvare_v2() for it to return an error code
> that it was an empty statement or something similar.

It does, except an empty statement is not an error.  You get NULL
returned 
as documented.

> Otherwise there's
> no other way that I can tell how you could prevent the misuse code or
> crash to occur.

You are way overthinking all this.  Additionally you do not need to call

sqlite3_sql since zSql was the start of the statement and pzTail points
to 
the end.  This will be the case even when there are empty statements.

> I know there is sqlite_complete() but I would have expected prepare to
> perform something similar.  Also sqlite_complete assumes a NULL
> terminated string which is different from most other API functions
that
> allow for a string/length combination.

complete is unrelated to your problem unless you are trying to a read
the 
contents of a file in the smallest chunks possible.

Roger
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] Index and GLOB

[Mike Goins]

First thanks to all that have helped on previous occasions.

I'm a little confused about some results using an index and GLOB.  I'm
trying to optimize some queries to ensure they use an index after
reviewing the LIKE Optimization section at the sqlite website.

Using the latest binary, sqlite3-3.6.23.1.bin.gz

sqlite> CREATE TABLE tb_file (tb_file_key INTEGER NOT NULL PRIMARY KEY
AUTOINCREMENT , basename TEXT, extension TEXT, path TEXT, deleted
INTEGER default 0 );
sqlite> CREATE INDEX fullpath_idx on tb_file (basename, extension, path);
sqlite> INSERT INTO tb_file (basename, extension, path) VALUES ('a', 'b', 'c');
sqlite> select * from tb_file;
tb_f  basename   exte  path  dele
  -      
1 a  b c 0

sqlite> explain query plan SELECT tb_file_key, basename, extension,
path FROM tb_file WHERE basename GLOB 'a*' AND  extension GLOB 'b' AND
 path GLOB 'c';
orde  from   deta
  -  
0 0  TABLE tb_file WITH INDEX fullpath_idx
sqlite> explain query plan SELECT tb_file_key, basename, extension,
path FROM tb_file WHERE basename GLOB 'a' AND  extension GLOB 'b*' AND
 path GLOB 'c';
orde  from   deta
  -  
0 0  TABLE tb_file USING PRIMARY KEY ORDER BY


The first select uses the index since the the glob character is picked
up, while the second does not.   Is this expected?   Does it matter?

Thanks
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] SQLite parsing of a .sql file

[Shaun Seckman (Firaxis)]

So, using the tail pointer from sqlite3_prepare_v2() I'm now running
through a loop to process all statements in a file.  I'm presently using
pointer math in order to determine the length of the tail command since
no length value is returned as part of the function.

I'm running into a nasty problem towards the end of the file however.
At the very end there are 3 characters left that are just whitespace
characters.  When I pass this string into sqlite3_prepare_v2() I get a
return value of 0 as if it's a valid statement.  However, when I attempt
to call sqlite3_sql() I will get a crash in sqlite.  If I call
sqlite3_step() I receive the error code SQLITE_MISUSE.

What would be very useful is if whitespace or comments were the only
data passed into sqlite3_prepvare_v2() for it to return an error code
that it was an empty statement or something similar.  Otherwise there's
no other way that I can tell how you could prevent the misuse code or
crash to occur.

I know there is sqlite_complete() but I would have expected prepare to
perform something similar.  Also sqlite_complete assumes a NULL
terminated string which is different from most other API functions that
allow for a string/length combination.

Perhaps I missed something else in the documentation?

-Shaun

-Original Message-
From: sqlite-users-boun...@sqlite.org
[mailto:sqlite-users-boun...@sqlite.org] On Behalf Of Shaun Seckman
(Firaxis)
Sent: Friday, April 09, 2010 3:37 PM
To: General Discussion of SQLite Database
Subject: Re: [sqlite] SQLite parsing of a .sql file

Ah, I missed that in the documentation.  This is perfect!  Precisely
what I needed with much less work than copying the shell.c code.

Thanks a bunch!

-Original Message-
From: sqlite-users-boun...@sqlite.org
[mailto:sqlite-users-boun...@sqlite.org] On Behalf Of Dan Kennedy
Sent: Friday, April 09, 2010 3:24 PM
To: General Discussion of SQLite Database
Subject: Re: [sqlite] SQLite parsing of a .sql file


On Apr 10, 2010, at 1:51 AM, Adam DeVita wrote:

> Is this a 1 off import?  If so, perhaps the command line tool  
> can .read it.
>
> On Fri, Apr 9, 2010 at 2:13 PM, Shaun Seckman (Firaxis) <
> shaun.seck...@firaxis.com> wrote:
>
>> Greetings,
>>
>> I've got a .SQL file that contains multiple SQL insert statements for
>> various tables as well as comments embedded throughout.  When I
>> attempted to pass the file into sqlite3_exec, I found that only the
>> first SQL command was getting executed.

That should not be the case. sqlite3_exec() should execute the
whole script.

>> What is the best way to ensure
>> that all commands are executed?   Parsing the file line-by-line is
>> inaccurate as multiple statements may be on the same line and looking
>> for the next ';' character has parsing problems as well.
>>
>>
>>
>> The documents state that sqlite3_complete() only returns a 1 if the
>> statement is complete.  Were this method to return the index into the
>> character array to denote WHERE the SQL statement is complete, I  
>> could
>> use that to parse multiple statements.
>>
>>
>>
>> Does anyone have any suggestions?

See also the 5th parameter to sqlite3_prepare_v2().

This API allows you to compile the first statement in an SQL
script and returns a pointer to the start of the next statement
in the script at the same time.

   http://www.sqlite.org/c3ref/prepare.html



Dan.

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] SQLite parsing of a .sql file

[Shaun Seckman (Firaxis)]

Ah, I missed that in the documentation.  This is perfect!  Precisely
what I needed with much less work than copying the shell.c code.

Thanks a bunch!

-Original Message-
From: sqlite-users-boun...@sqlite.org
[mailto:sqlite-users-boun...@sqlite.org] On Behalf Of Dan Kennedy
Sent: Friday, April 09, 2010 3:24 PM
To: General Discussion of SQLite Database
Subject: Re: [sqlite] SQLite parsing of a .sql file


On Apr 10, 2010, at 1:51 AM, Adam DeVita wrote:

> Is this a 1 off import?  If so, perhaps the command line tool  
> can .read it.
>
> On Fri, Apr 9, 2010 at 2:13 PM, Shaun Seckman (Firaxis) <
> shaun.seck...@firaxis.com> wrote:
>
>> Greetings,
>>
>> I've got a .SQL file that contains multiple SQL insert statements for
>> various tables as well as comments embedded throughout.  When I
>> attempted to pass the file into sqlite3_exec, I found that only the
>> first SQL command was getting executed.

That should not be the case. sqlite3_exec() should execute the
whole script.

>> What is the best way to ensure
>> that all commands are executed?   Parsing the file line-by-line is
>> inaccurate as multiple statements may be on the same line and looking
>> for the next ';' character has parsing problems as well.
>>
>>
>>
>> The documents state that sqlite3_complete() only returns a 1 if the
>> statement is complete.  Were this method to return the index into the
>> character array to denote WHERE the SQL statement is complete, I  
>> could
>> use that to parse multiple statements.
>>
>>
>>
>> Does anyone have any suggestions?

See also the 5th parameter to sqlite3_prepare_v2().

This API allows you to compile the first statement in an SQL
script and returns a pointer to the start of the next statement
in the script at the same time.

   http://www.sqlite.org/c3ref/prepare.html



Dan.

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] SQLite parsing of a .sql file

[Roger Binns]

On 04/09/2010 11:13 AM, Shaun Seckman (Firaxis) wrote:
> What is the best way to ensure that all commands are executed?

Load the whole file into memory or memory map it.  Then call 
sqlite3_prepare/step/reset.  prepare will point you to the beginning of the 
next statement if there is one.

> The documents state that sqlite3_complete() only returns a 1 if the
> statement is complete.  Were this method to return the index into the
> character array to denote WHERE the SQL statement is complete, I could
> use that to parse multiple statements.

sqlite3_complete tells you if there are one (or more) complete statements. 
It would be useful if you were providing an interactive shell.  You could 
also use it with prepare/step/reset to read the file in statement sized chunks.

Roger
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] SQLite parsing of a .sql file

[Shaun Seckman (Firaxis)]

Ah, I believe it is something that the command line tool can .read.
Since I'm using just the static library, I'll have to copy the
functionality from shell.c into my own code.

Thanks!

-Original Message-
From: sqlite-users-boun...@sqlite.org
[mailto:sqlite-users-boun...@sqlite.org] On Behalf Of Adam DeVita
Sent: Friday, April 09, 2010 2:52 PM
To: General Discussion of SQLite Database
Subject: Re: [sqlite] SQLite parsing of a .sql file

Is this a 1 off import?  If so, perhaps the command line tool can .read
it.

On Fri, Apr 9, 2010 at 2:13 PM, Shaun Seckman (Firaxis) <
shaun.seck...@firaxis.com> wrote:

> Greetings,
>
> I've got a .SQL file that contains multiple SQL insert statements for
> various tables as well as comments embedded throughout.  When I
> attempted to pass the file into sqlite3_exec, I found that only the
> first SQL command was getting executed.  What is the best way to
ensure
> that all commands are executed?   Parsing the file line-by-line is
> inaccurate as multiple statements may be on the same line and looking
> for the next ';' character has parsing problems as well.
>
>
>
> The documents state that sqlite3_complete() only returns a 1 if the
> statement is complete.  Were this method to return the index into the
> character array to denote WHERE the SQL statement is complete, I could
> use that to parse multiple statements.
>
>
>
> Does anyone have any suggestions?
>
>
>
> -Shaun
>
>
>
>
>
> ___
> sqlite-users mailing list
> sqlite-users@sqlite.org
> http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
>



-- 
VerifEye Technologies Inc.
905-948-0015x245
7100 Warden Ave, Unit 3
Markham ON, L3R 8B5
Canada
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] SQLite parsing of a .sql file

[Dan Kennedy]

On Apr 10, 2010, at 1:51 AM, Adam DeVita wrote:

> Is this a 1 off import?  If so, perhaps the command line tool  
> can .read it.
>
> On Fri, Apr 9, 2010 at 2:13 PM, Shaun Seckman (Firaxis) <
> shaun.seck...@firaxis.com> wrote:
>
>> Greetings,
>>
>> I've got a .SQL file that contains multiple SQL insert statements for
>> various tables as well as comments embedded throughout.  When I
>> attempted to pass the file into sqlite3_exec, I found that only the
>> first SQL command was getting executed.

That should not be the case. sqlite3_exec() should execute the
whole script.

>> What is the best way to ensure
>> that all commands are executed?   Parsing the file line-by-line is
>> inaccurate as multiple statements may be on the same line and looking
>> for the next ';' character has parsing problems as well.
>>
>>
>>
>> The documents state that sqlite3_complete() only returns a 1 if the
>> statement is complete.  Were this method to return the index into the
>> character array to denote WHERE the SQL statement is complete, I  
>> could
>> use that to parse multiple statements.
>>
>>
>>
>> Does anyone have any suggestions?

See also the 5th parameter to sqlite3_prepare_v2().

This API allows you to compile the first statement in an SQL
script and returns a pointer to the start of the next statement
in the script at the same time.

   http://www.sqlite.org/c3ref/prepare.html



Dan.

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] SQLite parsing of a .sql file

[Adam DeVita]

Is this a 1 off import?  If so, perhaps the command line tool can .read it.

On Fri, Apr 9, 2010 at 2:13 PM, Shaun Seckman (Firaxis) <
shaun.seck...@firaxis.com> wrote:

> Greetings,
>
> I've got a .SQL file that contains multiple SQL insert statements for
> various tables as well as comments embedded throughout.  When I
> attempted to pass the file into sqlite3_exec, I found that only the
> first SQL command was getting executed.  What is the best way to ensure
> that all commands are executed?   Parsing the file line-by-line is
> inaccurate as multiple statements may be on the same line and looking
> for the next ';' character has parsing problems as well.
>
>
>
> The documents state that sqlite3_complete() only returns a 1 if the
> statement is complete.  Were this method to return the index into the
> character array to denote WHERE the SQL statement is complete, I could
> use that to parse multiple statements.
>
>
>
> Does anyone have any suggestions?
>
>
>
> -Shaun
>
>
>
>
>
> ___
> sqlite-users mailing list
> sqlite-users@sqlite.org
> http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
>



-- 
VerifEye Technologies Inc.
905-948-0015x245
7100 Warden Ave, Unit 3
Markham ON, L3R 8B5
Canada
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] SQLite parsing of a .sql file

[Shaun Seckman (Firaxis)]

Greetings,

I've got a .SQL file that contains multiple SQL insert statements for
various tables as well as comments embedded throughout.  When I
attempted to pass the file into sqlite3_exec, I found that only the
first SQL command was getting executed.  What is the best way to ensure
that all commands are executed?   Parsing the file line-by-line is
inaccurate as multiple statements may be on the same line and looking
for the next ';' character has parsing problems as well.

 

The documents state that sqlite3_complete() only returns a 1 if the
statement is complete.  Were this method to return the index into the
character array to denote WHERE the SQL statement is complete, I could
use that to parse multiple statements.

 

Does anyone have any suggestions?

 

-Shaun

 

 

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Poor query planning for virtual tables

[Alexey Pechnikov]

Hello!

On Friday 09 April 2010 18:27:52 Roger Binns wrote:
> > Virtual table _doesn't_ use any index but query planner show this:
> > 0|2|TABLE role_fts VIRTUAL TABLE INDEX 0:
> 
> It is showing the results of the BestIndex virtual table method.

Thanks. So the problem is in the FTS3 extension only.

Best regards, Alexey Pechnikov.
http://pechnikov.tel/
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Poor query planning for virtual tables

[Roger Binns]

On 04/09/2010 12:27 AM, Alexey Pechnikov wrote:
> Virtual table _doesn't_ use any index but query planner show this:
> 0|2|TABLE role_fts VIRTUAL TABLE INDEX 0:

It is showing the results of the BestIndex virtual table method.

Roger
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Export sqlite database to H2

[Vivien Malerba]

On 9 April 2010 13:37, Andreas Henningsson
 wrote:
> Hi
>
> I am about to change database from SQLite to H2 in a project of mine.
>
> Is there any tools I can use to do i it? What is the best way to export data
> from an SQLite database?
>

The easiest is probably that you export your data from SQLite as an
SQL dump, adapt it to the H2 dialect and use it to import your data
into your H2 database.

There is another solution however using the gda-sql tool from Libgda
(I'm the Libgda's maintainer) where you can have more control over
what get migrated from SQLite to H2 as the migration operation is done
by executing some "INSERT into  SELECT ... FROM " statements.

The idea is to use the gda-sql command to open 2 connections: one to
the SQLite database, and one to the H2 database and then "bind" them
together into a 3rd virtual connection in which you execute the
"INSERT into..." statements. If you want to explore this way, I can
give you more details.

Regards,

Vivien
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] sqlite3_step() transaction

[Simon Slavin]

On 9 Apr 2010, at 11:57am, Igor Tandetnik wrote:

> Christoph Walser, ETH Zurich wrote:
>> Is there a way to do all the 5000 rows in one transaction to minimize
>> database locking time? I know I can do a sqlite3_exec with "BEGIN
>> TRANSACTION", then do the inserts and then "COMMIT TRANSACTIONS" but how
>> does this work when using the bind, step and reset functions?
> 
> Yes it does.

Just to make it clear, Igor is recommending that you do explicitly call

BEGIN TRANSACTION
do the 5000 inserts
END TRANSACTION

as this will speed up your calls and reduce the amount of disk activity quite a 
lot.

Simon.
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Export sqlite database to H2

[Andreas Henningsson]

H2 is a java bases database. I have just started to use it and it works
quite well.
It have the datatypes you expect and a very natural SQL language.

Mips, I do not know.

It's open source I think.

The only reason I use it is because I want to have the possibility to store
files in the database. And the JDBC driver for H2 support this better then
SQLite.

/Andreas

On Fri, Apr 9, 2010 at 1:47 PM, Navaneeth Sen B
wrote:

> Hi Andreas,
>
> I have some questions.
> Can H2 completely replace SQLite?
> I would like to know more about H2.
> Does that have a port on Linux-MIPS platform?
> Does it have any licensing issues?(whether open source)
> Can it perform better than SQLite?
> Or what made you move to H2 from SQLite?
>
> Regards,
> Sen
>
> On 4/9/2010 5:07 PM, Andreas Henningsson wrote:
> > Hi
> >
> > I am about to change database from SQLite to H2 in a project of mine.
> >
> > Is there any tools I can use to do i it? What is the best way to export
> data
> > from an SQLite database?
> >
> > /Andreas
> >
> >
>
> ___
> sqlite-users mailing list
> sqlite-users@sqlite.org
> http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
>



-- 
Andreas Henningsson

"Vanligt sunt förnuft är inte särkilt vanligt." -- Voltaire
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Export sqlite database to H2

[Navaneeth Sen B]

Hi Andreas,

I have some questions.
Can H2 completely replace SQLite?
I would like to know more about H2.
Does that have a port on Linux-MIPS platform?
Does it have any licensing issues?(whether open source)
Can it perform better than SQLite?
Or what made you move to H2 from SQLite?

Regards,
Sen

On 4/9/2010 5:07 PM, Andreas Henningsson wrote:
> Hi
>
> I am about to change database from SQLite to H2 in a project of mine.
>
> Is there any tools I can use to do i it? What is the best way to export data
> from an SQLite database?
>
> /Andreas
>
>

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] Export sqlite database to H2

[Andreas Henningsson]

Hi

I am about to change database from SQLite to H2 in a project of mine.

Is there any tools I can use to do i it? What is the best way to export data
from an SQLite database?

/Andreas

-- 
Andreas Henningsson

"Vanligt sunt förnuft är inte särkilt vanligt." -- Voltaire
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] sqlite3_step() transaction

[Igor Tandetnik]

Christoph Walser, ETH Zurich wrote:
> I have a C application which generates around 1000 rows of data per
> second to insert in an SQLite database.
> What I now do is to collect 5000 rows in a buffer and then for each row,
> execute sqlite3_bind_*, sqlite3_step, sqlite3_reset.
> 
> I wonder now how sqlite does this internaly: with the above mentioned
> technique, is every row capsulated into a single transaction or are they
> one large transaction alltogether?

If you don't issue explicit BEGIN and COMMIT statements, then each row is a 
separate transaction.

> Is there a way to do all the 5000 rows in one transaction to minimize
> database locking time? I know I can do a sqlite3_exec with "BEGIN
> TRANSACTION", then do the inserts and then "COMMIT TRANSACTIONS" but how
> does this work when using the bind, step and reset functions?

Yes it does.
-- 
Igor Tandetnik

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] problem porting sqlite amalgamation 3.6.23.1 to vxworks 6.5 pentium3 DKM

[zhenhuan du]

I created a vxworks 6.5 Pentium3 downloadable kernel module project .
Sqlite3.c and sqlite3.h were imported into project.
But when I compiled the sqlite3.c  I got the following error output.
  I  wonder whether  this package can be directly put  into use without 
any problem with vxworks 6.5.
Hope someone can help me .Thanks in advance!


-
 



Build Started in Project 'sqlite3':   2010-04-09 15:31:58
Generation of makefiles started.
Generation of makefiles finished (Elapsed Time: 00:00).
Platform: Wind River VxWorks 6.5
Command: make --no-print-directory BUILD_SPEC=PENTIUMgnu DEBUG_MODE=1 
TRACE=1 PENTIUMgnu_DEBUG/sqlite3.o
Working Directory: 
F:/software_project/mpu_firmware/gmc96_mpu_firmware_vxworks_image_project_v1.3/gmc96_mpu_firmware_vxworks_image_project_v1.3/sqlite3
 

echo "building PENTIUMgnu_DEBUG/sqlite3.o"; ccpentium -g -mtune=pentium 
-march=pentium -ansi  -Wall  -MD -MP  
-ID:/WindRiver/vxworks-6.5/target/h 
-ID:/WindRiver/vxworks-6.5/target/h/wrn/coreip   -DCPU
=PENTIUM -DTOOL_FAMILY=gnu -DTOOL=gnu -D_WRS_KERNEL-o 
"PENTIUMgnu_DEBUG/sqlite3.o" -c "sqlite3.c"
building PENTIUMgnu_DEBUG/sqlite3.o
In file included from 
D:/WindRiver/vxworks-6.5/target/h/vxWorksCommon.h:296,
  from D:/WindRiver/vxworks-6.5/target/h/vxWorks.h:84,
  from 
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:21,
  from D:/WindRiver/vxworks-6.5/target/h/stdio.h:62,
  from sqlite3.c:6522:
D:/WindRiver/vxworks-6.5/target/h/version.h:82: error: syntax error 
before "char"
D:/WindRiver/vxworks-6.5/target/h/version.h:83: error: syntax error 
before "char"
D:/WindRiver/vxworks-6.5/target/h/version.h:84: error: syntax error 
before "char"
D:/WindRiver/vxworks-6.5/target/h/version.h:85: error: syntax error 
before "char"
D:/WindRiver/vxworks-6.5/target/h/version.h:87: error: syntax error 
before "const"
D:/WindRiver/vxworks-6.5/target/h/version.h:88: error: syntax error 
before "const"
D:/WindRiver/vxworks-6.5/target/h/version.h:89: error: syntax error 
before "const"
In file included from D:/WindRiver/vxworks-6.5/target/h/stdio.h:62,
  from sqlite3.c:6522:
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:39: error: syntax 
error before "UINT32"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:39: warning: no 
semicolon at end of struct or union
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:40: warning: type 
defaults to `int' in declaration of `attributes'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:40: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:41: error: syntax 
error before "type"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:41: warning: type 
defaults to `int' in declaration of `type'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:41: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:42: error: syntax 
error before "contextType"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:42: warning: type 
defaults to `int' in declaration of `contextType'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:42: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:43: warning: type 
defaults to `int' in declaration of `HANDLE'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:43: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:45: error: syntax 
error before '*' token
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:45: warning: type 
defaults to `int' in declaration of `HANDLE_ID'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:45: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:49: error: syntax 
error before "handleContextGet"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:49: error: syntax 
error before "handleId"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:50: warning: type 
defaults to `int' in declaration of `handleContextGet'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:50: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:51: error: syntax 
error before "handleId"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:53: error: syntax 
error before "handleContextSet"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:53: error: syntax 
error before "handleId"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:54: warning: type 
defaults to `int' in declaration of `handleContextSet'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:54: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:55: error: syntax 
error before 

[sqlite] problem porting sqlite amalgamation 3.6.23.1 to vxworks 6.5 DKM

[zhenhuan du]

I created a downloadable kernel module project .
Sqlite3.c and sqlite3.h were imported into project.
But when I compiled the sqlite3.c  I got the following error output.
  I  wonder whether  this package can be directly put  into use without 
any problem with vxworks 6.5.
Hope someone can help me .Thanks in advance!


-


Build Started in Project 'sqlite3':   2010-04-09 15:31:58
Generation of makefiles started.
Generation of makefiles finished (Elapsed Time: 00:00).
Platform: Wind River VxWorks 6.5
Command: make --no-print-directory BUILD_SPEC=PENTIUMgnu DEBUG_MODE=1 
TRACE=1 PENTIUMgnu_DEBUG/sqlite3.o
Working Directory: 
F:/software_project/mpu_firmware/gmc96_mpu_firmware_vxworks_image_project_v1.3/gmc96_mpu_firmware_vxworks_image_project_v1.3/sqlite3
echo "building PENTIUMgnu_DEBUG/sqlite3.o"; ccpentium -g -mtune=pentium 
-march=pentium -ansi  -Wall  -MD -MP  
-ID:/WindRiver/vxworks-6.5/target/h 
-ID:/WindRiver/vxworks-6.5/target/h/wrn/coreip   -DCPU
=PENTIUM -DTOOL_FAMILY=gnu -DTOOL=gnu -D_WRS_KERNEL-o 
"PENTIUMgnu_DEBUG/sqlite3.o" -c "sqlite3.c"
building PENTIUMgnu_DEBUG/sqlite3.o
In file included from D:/WindRiver/vxworks-6.5/target/h/vxWorksCommon.h:296,
  from D:/WindRiver/vxworks-6.5/target/h/vxWorks.h:84,
  from 
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:21,
  from D:/WindRiver/vxworks-6.5/target/h/stdio.h:62,
  from sqlite3.c:6522:
D:/WindRiver/vxworks-6.5/target/h/version.h:82: error: syntax error 
before "char"
D:/WindRiver/vxworks-6.5/target/h/version.h:83: error: syntax error 
before "char"
D:/WindRiver/vxworks-6.5/target/h/version.h:84: error: syntax error 
before "char"
D:/WindRiver/vxworks-6.5/target/h/version.h:85: error: syntax error 
before "char"
D:/WindRiver/vxworks-6.5/target/h/version.h:87: error: syntax error 
before "const"
D:/WindRiver/vxworks-6.5/target/h/version.h:88: error: syntax error 
before "const"
D:/WindRiver/vxworks-6.5/target/h/version.h:89: error: syntax error 
before "const"
In file included from D:/WindRiver/vxworks-6.5/target/h/stdio.h:62,
  from sqlite3.c:6522:
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:39: error: syntax 
error before "UINT32"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:39: warning: no 
semicolon at end of struct or union
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:40: warning: type 
defaults to `int' in declaration of `attributes'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:40: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:41: error: syntax 
error before "type"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:41: warning: type 
defaults to `int' in declaration of `type'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:41: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:42: error: syntax 
error before "contextType"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:42: warning: type 
defaults to `int' in declaration of `contextType'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:42: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:43: warning: type 
defaults to `int' in declaration of `HANDLE'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:43: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:45: error: syntax 
error before '*' token
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:45: warning: type 
defaults to `int' in declaration of `HANDLE_ID'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:45: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:49: error: syntax 
error before "handleContextGet"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:49: error: syntax 
error before "handleId"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:50: warning: type 
defaults to `int' in declaration of `handleContextGet'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:50: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:51: error: syntax 
error before "handleId"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:53: error: syntax 
error before "handleContextSet"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:53: error: syntax 
error before "handleId"
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:54: warning: type 
defaults to `int' in declaration of `handleContextSet'
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:54: warning: data 
definition has no type or storage class
D:/WindRiver/vxworks-6.5/target/h/private/handleLibP.h:55: error: syntax 
error before "handleInit"

Re: [sqlite] Poor query planning for virtual tables

[Alexey Pechnikov]

Hello!

On Friday 09 April 2010 01:28:18 Jay A. Kreibich wrote:
>   Exactly like every other query the database makes?  Why would
>   this be different?

Virtual table _doesn't_ use any index but query planner show this:
0|2|TABLE role_fts VIRTUAL TABLE INDEX 0:

Why planner don't show full-teable scan as "TABLE role_fts VIRTUAL TABLE"?

Best regards, Alexey Pechnikov.
http://pechnikov.tel/
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] sqlite3_step() transaction

[Christoph Walser, ETH Zurich]

Hi everybody,

I have a C application which generates around 1000 rows of data per
second to insert in an SQLite database.
What I now do is to collect 5000 rows in a buffer and then for each row,
execute sqlite3_bind_*, sqlite3_step, sqlite3_reset.

I wonder now how sqlite does this internaly: with the above mentioned
technique, is every row capsulated into a single transaction or are they
one large transaction alltogether?
Is there a way to do all the 5000 rows in one transaction to minimize
database locking time? I know I can do a sqlite3_exec with "BEGIN
TRANSACTION", then do the inserts and then "COMMIT TRANSACTIONS" but how
does this work when using the bind, step and reset functions?

Thanks for any help,
Christoph

___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users