Re: [sqlite] Performance with journal_mode = off

[Kees Nuyt]

On Wed, 27 Mar 2013 17:55:00 -0400, Jeff Archer
 wrote:

>On Wed, Mar 27, 2013 at 5:46 PM, David King  wrote:
>>
>> > I am populating a database with 5764 records using the exact same data set
>> > each time into a newly created file.
>> > When I use no explicit transactions (default atomic commit) it takes 17.7
>> > seconds.
>> > When I set journal_mode = off, same operation takes 5.5 seconds.
>> > If I do all 5764 inserts within a single transaction only 2.5 seconds.
>>
>>
>> That sounds about right, yeah. With journalling,
>> most disk writes have to be done twice (once to the
>> journal and once to the data file).
>
> Which is why I expected journal_mode = off to make it faster.  But it
> is 3 seconds faster when I leave journaling enabled and do all writes
> within a single transaction.

I miss one test case: both journal_mode = off, and all 5764 inserts
within a single transaction.

Give that one a thought (and a try) and you'll understand the
difference.

-- 
Groet, Cordialement, Pozdrawiam, Regards,

Kees Nuyt

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

Re: [sqlite] Timeline for full ALTER TABLE support?

[Roger Binns]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 26/03/13 14:40, Tim Gustafson wrote:
> Is there a timeline or road map anywhere that could inform us as to
> when we might see that feature added?

In addition to what others have said, if SQLite did implement full ALTER
TABLE support it is unlikely to be implemented any different than what is
already in the FAQ. Note that there are pragmas to temporarily disable
various things like foreign keys and constraint checks.

  http://www.sqlite.org/faq.html#q11

The storage format for SQLite has each row written out as the values for
each column sequentially.  If you want to delete a column or re-order them
then every row has to be rewritten.

Appending a column was implemented by requiring a column default for the
new column, and if the row read had too few columns then obtain the
missing ones from the defaults.  This didn't require a rewrite of every row.

You can achieve what you want via various means, most recommended first:

- - update your code so that it only needs to append columns

- - change the underlying tables as in FAQ q11 above

- - create a view that presents the schema you want composed of the
underlying tables, with triggers for insert to redirect columns as appropriate

- - convince/pay someone to make a patch that implements what you need and
maintain your fork forwards

- - convince/pay the SQLite team to add it to the core, keeping in mind that
they care about library size and the 'Lite' bit

Roger
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.11 (GNU/Linux)

iEYEARECAAYFAlFTj5sACgkQmOOfHg372QQmEACfUehTgbHqtIHVvhcFOY36bUKn
g88AoNrlsPZX2vGDp6hqlanmNDM8yNFo
=iUmI
-END PGP SIGNATURE-
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Timeline for full ALTER TABLE support?

[Larry Brasfield]

Referring to a page at 
http://web.archive.org/web/20020926232103/http://www.sqlite.org/omitted.html, 
clearly dated mid-2002, Tim Gustafson effluviated:


It's not that it's surprising.  The page currently says:

Those features near the top of the list are likely to be added in the
near future.

That's a lie.  There are no plans to implement anything on this list,
and the page should say so.


The word "lie" is ill-chosen.  Of the top 3 items in that 10-item list, 
two have been implemented.  The "likely to be added" statement is 
arguable true, even in retrospect.  To suggest that any deceit was 
intended or that the statement was known to be false when made is 
unsupported by the evidence and indicates either ignorance of what the 
word "lie" means or bad faith.


--
Larry Brasfield

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

Re: [sqlite] Performance with journal_mode = off

[Simon Slavin]

On 27 Mar 2013, at 9:55pm, Jeff Archer  wrote:

> Which is why I expected journal_mode = off to make it faster.  But it
> is 3 seconds faster when I leave journaling enabled and do all writes
> within a single transaction.

>> When I set journal_mode = off, same operation takes 5.5 seconds.
>> If I do all 5764 inserts within a single transaction only 2.5 seconds.

Reasonable figures.  With 5764 writes to the disk in separates transactions you 
have quite a lot of reading of data plus 5764 attempts to update the database 
file.  The updates have to be done in the right order, and each update has to 
wait for the disk to be rotated into the right position, though each update 
changes only a small amount of data (probably two sectors).

With 5764 writes to the disk in the same transaction, you have the same amount 
of reading needed, but only 1 attempt to write to the database file.  Although 
all the same sectors need writing, each one needs writing once only and the 
sectors can be written in any order (your drivers will probably optimize the 
order or writes they happen in whatever order will be fastest).

So yes, you save more time by doing your updates in one transaction than doing 
them separately.  In fact I'd thought that your figures would be further apart 
than 55%.

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

Re: [sqlite] Performance with journal_mode = off

[Jeff Archer]

On Wed, Mar 27, 2013 at 5:46 PM, David King  wrote:
>
> > I am populating a database with 5764 records using the exact same data set
> > each time into a newly created file.
> > When I use no explicit transactions (default atomic commit) it takes 17.7
> > seconds.
> > When I set journal_mode = off, same operation takes 5.5 seconds.
> > If I do all 5764 inserts within a single transaction only 2.5 seconds.
>
>
> That sounds about right, yeah. With journalling, most disk writes have to be 
> done twice (once to the journal and once to the data file).

Which is why I expected journal_mode = off to make it faster.  But it
is 3 seconds faster when I leave journaling enabled and do all writes
within a single transaction.
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Performance with journal_mode = off

[David King]

> I am populating a database with 5764 records using the exact same data set
> each time into a newly created file.
> When I use no explicit transactions (default atomic commit) it takes 17.7
> seconds.
> When I set journal_mode = off, same operation takes 5.5 seconds.
> If I do all 5764 inserts within a single transaction only 2.5 seconds.


That sounds about right, yeah. With journalling, most disk writes have to be 
done twice (once to the journal and once to the data file).
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] Performance with journal_mode = off

[Jeff Archer]

Could someone please confirm if this makes sense.  It is not what I
expected.  I have repeated several times so I believe these are the correct
numbers.
I am populating a database with 5764 records using the exact same data set
each time into a newly created file.
When I use no explicit transactions (default atomic commit) it takes 17.7
seconds.
When I set journal_mode = off, same operation takes 5.5 seconds.
If I do all 5764 inserts within a single transaction only 2.5 seconds.


Jeff Archer
Vice President of Software Engineering
Nanotronics Imaging
jsarc...@nanotronicsimaging.com
<330>819.4615
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] Version 3.7.16.1 bug fix release in about a week

[Richard Hipp]

The query optimizer enhancements that were added as part of version 3.7.15
were a little overzealous in optimizing out certain ORDER BY clauses, which
means that in some cases the result rows were not being sorted and were
coming out in an incorrect order.  See
http://www.sqlite.org/src/info/a179fe7465 for details.

Because of this issue, we have decided to do a bug-fix patch release,
version 3.7.16.1, in about a week.  Several other fixes will be included in
the same release.  Visit
http://www.sqlite.org/draft/releaselog/3_7_16_1.html for a complete list of
change.

Builds of the amalgamation source code files for the release candidate can
be obtained from http://www.sqlite.org/draft/download.html

Please post to this mailing list, or email directly to me, if you know of
any other problems that this release needs to address or if you know of any
reason why we should delay the release.

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

Re: [sqlite] Ubuntu Linking

[brian_f_john...@yahoo.com]

I found that reference to CC in a January posting in this mailing list.

I am obviously not a C programmer but I have to figure it out anyway.

The problem is resolved. I ended up finding a different example that had the 
correct syntax.

Thanks,

Brian
  


 From: James K. Lowden 
To: sqlite-users@sqlite.org 
Cc: "brian_f_john...@yahoo.com" ; General Discussion 
of SQLite Database  
Sent: Wednesday, March 27, 2013 11:42 AM
Subject: Re: [sqlite] Ubuntu Linking
  
On Wed, 27 Mar 2013 09:33:39 -0700 (PDT)
"brian_f_john...@yahoo.com"  wrote:

> I ran: cc -O -c sqlite3.c shell.c  That created .o files after 1
> warning about function exprDup having a memset with constant zero
> length.
> 
> I compiled my program:  cc -I$HOME/lstp -O -c lstp_sql.c and have
> a .o for it as well
> 
> Now I am trying to figure out how to link them all together into an
> executable. I have tried many different combinations and none seem to
> work.

So you don't care about shell.c, right?  That's just the command shell;
the DBMS is implemented in sqlite3.c.  

I would think you want something like

    cc -o appname -I$HOME/lstp lstp_sql.c sqlite3.o

except that "cc" is actually obsolete!  ;-)  Without using the -std
option, you're telling the compiler to accept code predating even the
C89 standard, which is two standards ago.  

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

Re: [sqlite] Ubuntu Linking

[James K. Lowden]

On Wed, 27 Mar 2013 09:33:39 -0700 (PDT)
"brian_f_john...@yahoo.com"  wrote:

> I ran: cc -O -c sqlite3.c shell.c  That created .o files after 1
> warning about function exprDup having a memset with constant zero
> length.
> 
> I compiled my program:  cc -I$HOME/lstp -O -c lstp_sql.c and have
> a .o for it as well
> 
> Now I am trying to figure out how to link them all together into an
> executable. I have tried many different combinations and none seem to
> work.

So you don't care about shell.c, right?  That's just the command shell;
the DBMS is implemented in sqlite3.c.  

I would think you want something like

cc -o appname -I$HOME/lstp lstp_sql.c sqlite3.o

except that "cc" is actually obsolete!  ;-)  Without using the -std
option, you're telling the compiler to accept code predating even the
C89 standard, which is two standards ago.  

--jkl


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

[sqlite] NSA internal amusement magazine

[Simon Slavin]

Newly declassified, issues of the internal amusement magazine (i.e. this 
mailing list) from the NSA in the 1970s to 1990s:



Largely redacted, but contains everything from a description of the speech 
developed in Russian prison camps to trivia contests (where 'trivia' means 'the 
sort of thing that would interest an NSA employee').

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

Re: [sqlite] Timeline for full ALTER TABLE support?

[Tim Gustafson]

> It really shouldn't be that surprising, this is pretty standard for open
> source software. Somebody has to do it, which means either writing code
> yourself, waiting for somebody else to get around to writing code, or
> putting up $$$ so somebody else spends their time writing code for you.

It's not that it's surprising.  The page currently says:

Those features near the top of the list are likely to be added in the
near future.

That's a lie.  There are no plans to implement anything on this list,
and the page should say so.

-- 

Tim Gustafson
t...@ucsc.edu
831-459-5354
Baskin Engineering, Room 313A
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Timeline for full ALTER TABLE support?

[Ryan Johnson]

On 27/03/2013 12:14 PM, Tim Gustafson wrote:

Clemens' analysis of the likelihood of seeing ALTER TABLE anytime soon is
correct.

Might I suggest that the "omitted" page then be updated to
unambiguously state that there is no plan to even implement the
missing features, so that people aren't left wondering?  Thanks!
It really shouldn't be that surprising, this is pretty standard for open 
source software. Somebody has to do it, which means either writing code 
yourself, waiting for somebody else to get around to writing code, or 
putting up $$$ so somebody else spends their time writing code for you.


Ryan

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

[sqlite] Ubuntu Linking

[brian_f_john...@yahoo.com]

Hi,
 
I ran: cc -O -c sqlite3.c shell.c  That created .o files after 1 warning about 
function exprDup having a memset with constant zero length.

I compiled my program:  cc -I$HOME/lstp -O -c lstp_sql.c and have a .o for it 
as well

Now I am trying to figure out how to link them all together into an executable. 
I have tried many different combinations and none seem to work.

What I am trying to accomplish is to have an interface api sitting on top of 
sqlite that I can call from a different program that will read and write data.

Thanks,

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

Re: [sqlite] Timeline for full ALTER TABLE support?

[Tim Gustafson]

> Clemens' analysis of the likelihood of seeing ALTER TABLE anytime soon is
> correct.

Might I suggest that the "omitted" page then be updated to
unambiguously state that there is no plan to even implement the
missing features, so that people aren't left wondering?  Thanks!

-- 

Tim Gustafson
t...@ucsc.edu
831-459-5354
Baskin Engineering, Room 313A
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Timeline for full ALTER TABLE support?

[Richard Hipp]

On Wed, Mar 27, 2013 at 12:06 PM, Dominique Devienne wrote:

> On Wed, Mar 27, 2013 at 2:41 PM, Clemens Ladisch 
> wrote:
>
> > Tim Gustafson wrote:
> > > That page also says that things are listed there in the order they're
> > > likely to be implemented in SQLite.  Is there a timeline or road map
> > > anywhere that could inform us as to when we might see that feature
> > > added?
> >
> > No.  (There isn't even a timeline for 'regular' development.)
> >
>
> I don't mean to be disrespectful in any way, but may I ask whether you
> officially work on SQLite to make that statement?
>
> I'm afraid I'm fairly new to this list, and beside obviously D. Richard
> Hipp, I don't know who's working for http://www.hwaci.com and neither
> http://www.hwaci.com/sw/sqlite/prosupport.html nor
> http://www.hwaci.com/staff.html help in this regard, so I'm just curious
> on
> who can speak authoritatively on such answers, and in general on this list
> about SQLite. That's all :)  Thanks, --DD
>

Clemens' analysis of the likelihood of seeing ALTER TABLE anytime soon is
correct.


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

Re: [sqlite] Timeline for full ALTER TABLE support?

[Dominique Devienne]

On Wed, Mar 27, 2013 at 2:41 PM, Clemens Ladisch  wrote:

> Tim Gustafson wrote:
> > That page also says that things are listed there in the order they're
> > likely to be implemented in SQLite.  Is there a timeline or road map
> > anywhere that could inform us as to when we might see that feature
> > added?
>
> No.  (There isn't even a timeline for 'regular' development.)
>

I don't mean to be disrespectful in any way, but may I ask whether you
officially work on SQLite to make that statement?

I'm afraid I'm fairly new to this list, and beside obviously D. Richard
Hipp, I don't know who's working for http://www.hwaci.com and neither
http://www.hwaci.com/sw/sqlite/prosupport.html nor
http://www.hwaci.com/staff.html help in this regard, so I'm just curious on
who can speak authoritatively on such answers, and in general on this list
about SQLite. That's all :)  Thanks, --DD
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] How do I get older versions of the sources?

[Wayne Bradney]

I was failing at the clone command with a "not authorized to clone" error.
I was going against http://www2.sqlite.org/cgi/src , since the download page 
indicated that as the closest one to me.
When I go against http://www.sqlite.org/cgi/src , I can clone.

Thanks



 From: Stephan Beal 
To: General Discussion of SQLite Database  
Sent: Wednesday, March 27, 2013 10:18 AM
Subject: Re: [sqlite] How do I get older versions of the sources?
 
On Wed, Mar 27, 2013 at 3:15 PM, Wayne Bradney wrote:

> Might seem like a dumb questions, but I can figure it out.fossil clone
> command gives me: Not authorized to cloneHitting the URL in the browser
> gives me much info about cryptic tag ids but no apparent way to get to full
> sources for an older version.
> Help?
>

With the clone command you can only fetch ALL versions. You then need to
"open" the repo and "checkout" ("co") the version (tag name) you want.

Try:

fossil clone http://www.sqlite.org/cgi/src sqlite3.fsl
mkdir sqlite3
cd sqlite3
fossil open ../sqlite3.fsl
fossil co VERSION_TAG_NAME_YOU_WANT

(untested, but i think that'll do it)

-- 
- stephan beal
http://wanderinghorse.net/home/stephan/
http://gplus.to/sgbeal
___
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] How do I get older versions of the sources?

[Stephan Beal]

On Wed, Mar 27, 2013 at 3:15 PM, Wayne Bradney wrote:

> Might seem like a dumb questions, but I can figure it out.fossil clone
> command gives me: Not authorized to cloneHitting the URL in the browser
> gives me much info about cryptic tag ids but no apparent way to get to full
> sources for an older version.
> Help?
>

With the clone command you can only fetch ALL versions. You then need to
"open" the repo and "checkout" ("co") the version (tag name) you want.

Try:

fossil clone http://www.sqlite.org/cgi/src sqlite3.fsl
mkdir sqlite3
cd sqlite3
fossil open ../sqlite3.fsl
fossil co VERSION_TAG_NAME_YOU_WANT

(untested, but i think that'll do it)

-- 
- stephan beal
http://wanderinghorse.net/home/stephan/
http://gplus.to/sgbeal
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Wanted - simple DATA editor for sqlite tables

[Stephen Chrzanowski]

The black and white of it, you're looking at something home brew then.

No one is going to have anything that is plain and simple enough that does
exactly what you want it to do without a lot of extra fluff, and with the
additional "kind-of-a-restriction" of being under Linux, I can't offer
anything up off the top of my head.  Even under Windows I can't think of a
bare bones table data editor that does JUST that.  (I'm a Windows desktop
developer but I know my way around a Linux machine rather well as it is my
primary job).  The tools that do already exist, and are already out there
for public consumption (Anything between commercial and public domain) will
end up having more of the fluff to reach out to a wider range of people.

Considering that if Product A has exactly what you're looking for right now
with absolutely no fluff, and Product B which has exactly what you're
looking for PLUS extra fluff, more people are going to Product B because of
the fluff that they COULD use later on down the road.  Look at Win 98 vs
Win 7.  Both let you open up Calc.exe, they run Office, OpenOffice, Java,
FireFox, etc, etc...  But, Win7 supports multi-threaded CPUs, 64-bit
instructions, wider range support of games, better support for graphics and
hardware.  If I were to offer you either operating system for free, which
would you take?

Unfortunately, I feel that you're going to have to build something to suit
your needs, even though for basic tools, I still scream out to not reinvent
the wheel.
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] How do I get older versions of the sources?

[Wayne Bradney]

Might seem like a dumb questions, but I can figure it out.fossil clone command 
gives me: Not authorized to cloneHitting the URL in the browser gives me much 
info about cryptic tag ids but no apparent way to get to full sources for an 
older version.
Help? 
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] Timeline for full ALTER TABLE support?

[Clemens Ladisch]

Tim Gustafson wrote:
> I see that full ALTER TABLE support is in position number two on the
> list of things that SQLite doesn't support:
>
> http://www.sqlite.org/omitted.html

Once Upon A Time™, that list had more entries:


By now, only the bottom of the list is left.

> That page also says that things are listed there in the order they're
> likely to be implemented in SQLite.  Is there a timeline or road map
> anywhere that could inform us as to when we might see that feature
> added?

No.  (There isn't even a timeline for 'regular' development.)


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

Re: [sqlite] ANN: user-defined functions

[Jean-Christophe Deschamps]

... so I think you must be using a really old copy.


That's correct: I downloaded this source years ago.

And yes, argc for SQLite functions has a different semantics from main().

Thanks for the heads up.

--
j...@antichoc.net  


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

Re: [sqlite] Question about behavior when journal_mode = off

[Richard Hipp]

On Wed, Mar 27, 2013 at 8:05 AM, Jeff Archer <
jsarc...@nanotronicsimaging.com> wrote:

> If all connections (1 per thread) of all processes (multiple simultaneous)
> issue command "PRAGMA journal_mode = off", is access to a common database
> file located on the local machine still synchronized correctly between the
> random accesses (reads and writes) occurring across all?
>

Yes.  The journal handles rollback and atomic commit (both of which you
lose when you go to journal_mode=off) but isolation is handled with file
locking on the original database file.

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

[sqlite] Question about behavior when journal_mode = off

[Jeff Archer]

If all connections (1 per thread) of all processes (multiple simultaneous)
issue command "PRAGMA journal_mode = off", is access to a common database
file located on the local machine still synchronized correctly between the
random accesses (reads and writes) occurring across all?
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

[sqlite] Timeline for full ALTER TABLE support?

[Tim Gustafson]

Hi,

I see that full ALTER TABLE support is in position number two on the
list of things that SQLite doesn't support:

http://www.sqlite.org/omitted.html

That page also says that things are listed there in the order they're
likely to be implemented in SQLite.  Is there a timeline or road map
anywhere that could inform us as to when we might see that feature
added?

-- 

Tim Gustafson
t...@ucsc.edu
831-459-5354
Baskin Engineering, Room 313A
___
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] ANN: user-defined functions

[Chris]

> At 15:46 25/03/2013, you wrote:
> 
> >   The sqrt() function takes only one argument, at least.
> 
> It checks   assert( argc==2 ); at line 503 AFAIK.

Under some conventions, argc is one more than you might initially
expect, so it's worth looking a little deeper. Think of main() and also
C functions registered as Tcl commands - in both cases argv[0] tells you
what name you were called by. However, that is not the case here! The
most recent entry in the change log says:

> 2010-01-06 Correct check for argc in squareFunc

... so I think you must be using a really old copy.

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

[sqlite] Install sqlite-netFx40-setup-bundle silently

[Jeff Williams]

Is it possible and if so how to install sqlite-netFx40-setup-bundle in a 
silent mode.
I would like to do the install as part of a software package but would 
prefer the users are not presented with the install screens.

Also what would be the easiest way to determine if 
sqlite-netFx40-setup-bundle is already installed.

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