hii
thank you for suggestion.
but when reading the value from sqlite header it printing same value no
change i can see i pasting the c++ code below
/****************************************************CODE****************************************************/
#include <stdio.h>
#include <iostream>
using namespace std;
int main ()
{
FILE * pFile;
int c;
int n = 0;
char fcc[35] = {0};
int i = 0;
int j =0;
while(j < 4)
{
int n = 0;
sleep(10); // sleep is used so that i can make any change in
database so that value of counter will change
pFile=fopen ("demo.sqlite","r"); //demo.sqlite is my sqlite file
if (pFile==NULL)
{
perror ("Error opening file");
exit(-1);
}
else
{
do
{
c = fgetc (pFile);
n++;
if(n >23 && n < 28)
{
printf("%x",c); // printing the value of header whic in
hex format
i++;
}
}while (c != EOF);
fclose (pFile);
cout<<endl;
}
j++;
}
On Mon, Sep 12, 2011 at 6:45 PM, Simon Slavin <slav...@bigfraud.org> wrote:
>
> On 12 Sep 2011, at 1:35pm, Richard Hipp wrote:
>
> > On Mon, Sep 12, 2011 at 8:02 AM, Igor Tandetnik <itandet...@mvps.org>
> wrote:
> >
> >> Simon Slavin <slav...@bigfraud.org> wrote:
> >>> <http://www.sqlite.org/c3ref/total_changes.html>
> >>>
> >>> int sqlite3_total_changes(sqlite3*);
> >>>
> >>> My understanding (which might be wrong) is that this count includes all
> >>> changes made by all connections to that database: not
> >>> only changes made using your connection but also chances made by
> another
> >>> computer, process, or thread.
> >>
> >> I'm 99% sure your understanding is wrong [snip]
> >
> > Igor is right. The sqlite3_total_changes() function only reports the
> number
> > of rows that have been changed by the same database connection that
> issued
> > the sqlite3_total_changes() call.
>
> Okay, I find that page in the documentation ambiguous then. Could that
> sentence, or something like it, be added for clarity ?
>
> > There is no API for accessing the database change counter. But you can
> read
> > it yourself by looking at bytes 24-27 of the database file.
>
> How hard would it be to implement this safely using sqlite3's own
> filehandle to the database file ? I know we're meant to treat sqlite3* as a
> black box but could it be done relatively safely, with a warning that it
> might fail under certain weird conditions ? Would one use
> sqlite3_file_control() ?
>
> Alternatively, suppose one was doing a lot of in-memory caching for a
> SQLite database but didn't want to block other apps from accessing it. Is
> there a clean way to say "Has anyone been messing with this but me ?" ?
>
> Simon.
> _______________________________________________
> sqlite-users mailing list
> sqlite-users@sqlite.org
> http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
>
--
Akash Agrawal
_______________________________________________
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
