Re: [sqlite] Bug in trigger: when comparing a value of an int column to a quoted value

[romtek]

Alexey, I am not clear about whether you reported my bug or yours. Do
you mind giving me a link to it so that I could see if I need to
report mine and let it go?



On Sun, Dec 25, 2011 at 11:16 AM, Alexey Pechnikov
 wrote:
> I think there are some problems with dynamic datatypes in sqlite.
> This may be redesigned in SQLite4 but not in SQLite3.
>
> P.S. Datatypes recognized differently after Shift-Insert SQLite3
> commands into Tcl shell. This is similar to problem with datatypes
> in triggers. I did report bug but for backward compability problem
> wil not be resolved by upstream. So I did write patch for myself.
>
> --
> 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-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users

Re: [sqlite] sqlite3_column_text() returning partial results

[Igor Tandetnik]

Matt Young  wrote:
> I used the strcpy_s function from Microsoft, the so called safe version
> that includes a char count. I used it under the Studio debugger.  Set a
> buffer of 200 chars to zero, set the char count to 20 in strcpy_s, and the
> debugger wrote in the top 180!!

As per documentation: "The debug versions of these functions first fill the 
buffer with 0xFD. To disable this behavior, use _CrtSetDebugFillThreshold." 
This is to make buffer overruns easier to spot under debugger. If you see a 
variable unexpectedly overwritten with 0xFD pattern, chances are high you are 
passing incorrect buffer length to strcpy_s or similar, and the buffer happens 
to be located in memory right before that varible.
-- 
Igor Tandetnik

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

Re: [sqlite] sqlite3_column_text() returning partial results

[Matt Young]

I used the strcpy_s function from Microsoft, the so called safe version
that includes a char count. I used it under the Studio debugger.  Set a
buffer of 200 chars to zero, set the char count to 20 in strcpy_s, and the
debugger wrote in the top 180!!

I freaked, didn't see that as safe at all!

On Sat, Dec 24, 2011 at 4:28 AM, Tony McC  wrote:

> On Wed, 21 Dec 2011 21:33:02 +
> "Jacob A. Camp"  wrote:
>
> > There it is, when I printed this value out to a txt file everything
> > was there.
> >
> > Apparently there is some sort of character limit when displaying
> > strings in debug mode in Visual Studio 2010, news to me...
> >
> > Thanks everyone!
>
> That's what the 2010 in the name of the product means: displayed strings
> are limited to that many characters (Studio is an acronym meaning Stop
> Displaying Under Debugger If Over [2010]). You were lucky to get a few
> more, but that is undocumented behaviour.  If you upgrade next year
> though you should be able to display a couple more characters.
>
> Tony
> ___
> 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] Bug in trigger: when comparing a value of an int column to a quoted value

[Alexey Pechnikov]

I think there are some problems with dynamic datatypes in sqlite.
This may be redesigned in SQLite4 but not in SQLite3.

P.S. Datatypes recognized differently after Shift-Insert SQLite3
commands into Tcl shell. This is similar to problem with datatypes
in triggers. I did report bug but for backward compability problem
wil not be resolved by upstream. So I did write patch for myself.

-- 
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