>On 12/13/2016 01:47 AM, Bob Friesenhahn wrote:
>> I am reworking code for a virtual table module and need to provide
>> proper error codes from the xColumn() and xRowid() callbacks for the
>> case where the row id does not exist, or the current row goes away.
>> This problem occurs because the
xRowid() is only called if the (hidden field) rowid is explicitly mentioned in
the select list, join condition or where clause(s) of a SELECT statement; or if
your VTAB supports write operations (see xUpdate documentation
http://www.sqlite.org/vtab.html#the_xupdate_method ).
-Ursprüngliche
Current documentation http://www.sqlite.org/rescode.html suggests you return
(267) SQLITE_CORRUPT_VTAB in this case. Other candidates would be SQLITE_IOERR
or SQLITE_NOTFOUND.
You may also call one or more of the sqlite_result_error() functions to set an
error string and/or an error code;
I'm not sure exactly what you're trying to do, but if sqlite3_close()
returns SQLITE_OK then it closed and if it returns SQLITE_BUSY then it's
not closed but possibly in a messy state (my selects have default handling
too but I think those are the only defined responses). If you know it
Hi, ALL,
I'm using following code to check for errors in debug mode:
int res = sqlite3_close( m_db );
if( res != SQLITE_OK )
{
// error handling
}
#ifdef DEBUG
sqlite3_stmt *statement = sqlite3_next_stmt( m_db, NULL );
if( statement )
const char *query =
On 12/14/2016 10:56 AM, Igor Korot wrote:
Keith,
On Tue, Dec 13, 2016 at 8:34 PM, Keith Medcalf wrote:
int res = sqlite3_close( m_db );
if( res == SQLITE_OK )
m_db = NULL;
else
{
// error handling
}
#ifdef DEBUG
sqlite3_stmt
Simon,
On Tue, Dec 13, 2016 at 8:33 PM, Simon Slavin wrote:
>
> On 14 Dec 2016, at 1:20am, Igor Korot wrote:
>
>> So are you saying that this code should be executed if sqlite3_close()
>> didn't return SQLITE_OK?
>
> Hold on. Closing the database is a
> On Dec 13, 2016, at 5:33 PM, Simon Slavin wrote:
>
> The only thing you should do if sqlite3_close() doesn’t work is to print an
> error message which includes the value returned. Because if you can’t close
> the database what are you going to do instead ?
Igor is
Keith,
On Tue, Dec 13, 2016 at 8:34 PM, Keith Medcalf wrote:
>
> int res = sqlite3_close( m_db );
> if( res == SQLITE_OK )
> m_db = NULL;
> else
> {
> // error handling
> }
> #ifdef DEBUG
> sqlite3_stmt *statement = sqlite3_next_stmt( m_db,
int res = sqlite3_close( m_db );
if( res == SQLITE_OK )
m_db = NULL;
else
{
// error handling
}
#ifdef DEBUG
sqlite3_stmt *statement = sqlite3_next_stmt( m_db, NULL );
if( statement )
const char *query = sqlite3_sql( statement );
#endif
Then you simply
On Tue, Dec 13, 2016 at 08:20:44PM -0500, Igor Korot wrote:
> > What more are you trying to find out?
>
> If I forget to finalize statement, I can use that sequence to find
> which query is dangling.
>
> So are you saying that this code should be executed if sqlite3_close()
> didn't return
Hi, Nathan,
On Tue, Dec 13, 2016 at 8:03 PM, Nathan Bossett wrote:
> I'm not sure exactly what you're trying to do, but if sqlite3_close()
> returns SQLITE_OK then it closed and if it returns SQLITE_BUSY then it's
> not closed but possibly in a messy state (my selects have
On 14 Dec 2016, at 1:20am, Igor Korot wrote:
> So are you saying that this code should be executed if sqlite3_close()
> didn't return SQLITE_OK?
Hold on. Closing the database is a special case.
The only thing you should do if sqlite3_close() doesn’t work is to print an
Nathan,
On Tue, Dec 13, 2016 at 8:36 PM, Nathan Bossett wrote:
> On Tue, Dec 13, 2016 at 08:20:44PM -0500, Igor Korot wrote:
>> > What more are you trying to find out?
>>
>> If I forget to finalize statement, I can use that sequence to find
>> which query is dangling.
>>
>>
Jens,
On Tue, Dec 13, 2016 at 8:38 PM, Jens Alfke wrote:
>
>> On Dec 13, 2016, at 5:33 PM, Simon Slavin wrote:
>>
>> The only thing you should do if sqlite3_close() doesn’t work is to print an
>> error message which includes the value returned.
Hi,
Could some one share the steps on how to build the sqlite-see with node js?
Thanks,
Rosina
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
On 12/12/16, Lipson, Ed wrote:
> Python 32bit, Windows 7 64bit
> How do I turn a real into a date?
That depends on how the real number represents a date. SQLite
supports two very popular encodings: Julian Day Number
(https://en.wikipedia.org/wiki/Julian_day) and Unix
Thanks, Jens. I'll keep an eye on that.
Take care,
Melody
-Original Message-
From: sqlite-users [mailto:sqlite-users-boun...@mailinglists.sqlite.org] On
Behalf Of Jens Alfke
Sent: Monday, December 12, 2016 12:32 PM
To: SQLite mailing list
Subject:
On 12/12/16, Rosina joy01 wrote:
> Hi,
>
> Could some one share the steps on how to build the sqlite-see with node js?
>
Sure. But first, can you please share with me (perhaps via reply to
d...@sqlite.org instead of back to this list) the email address
associated with
On Tue, Dec 13, 2016 at 9:41 AM, Simon Slavin wrote:
>
> On 13 Dec 2016, at 3:20pm, Paul Egli wrote:
>
> > Why must SQLite find rows where the value is NULL?
>
> Because the related fields in the offspring row might have NULL in them,
> and SQLite
On 13 Dec 2016, at 3:53pm, Paul Egli wrote:
> Well if i am missing something, i still cannot see it.
>
> Based on these quotes in the docs, i assume that a NULL in the child table
> means that it does not have a parent.
You are correct. I missed that.
So yes, the
On 13 Dec 2016, at 3:20pm, Paul Egli wrote:
> Why must SQLite find rows where the value is NULL?
Because the related fields in the offspring row might have NULL in them, and
SQLite needs to know how to find the parent row for that row.
Mind you, if the relating key
rather unintuitively excel uses the OLE automation timestamp that
records the number of days since 1899/12/30
http://sandersonforensics.com/forum/content.php?131-A-brief-history-of-time-stamps
https://msdn.microsoft.com/en-us/library/system.datetime.tooadate(v=vs.110).aspx
now no need for the
On Wed, Dec 7, 2016 at 3:11 PM, Simon Slavin wrote:
>
> On 7 Dec 2016, at 8:40pm, David Raymond wrote:
>
> > Question on making indexes for the child fields of foreign keys. I have
> a child table with a number of foreign keys on fields which the
On Tue, Dec 13, 2016 at 9:56 AM, Simon Slavin wrote:
>
> On 13 Dec 2016, at 3:53pm, Paul Egli wrote:
>
> > Well if i am missing something, i still cannot see it.
> >
> > Based on these quotes in the docs, i assume that a NULL in the child
> table
> >
> But that does not seems to be the whole story, since if I do:
>
> SELECT datetime(42713.1916667+julianday('1900-01-01'));
>
> I get 2016-12-11 04:36:00 instead of what you were expecting
> 2016-12-09 05:15. Something is a little off. I suspect that there
I think he is expecting
On 12/13/2016 11:02 PM, Paul Egli wrote:
On Tue, Dec 13, 2016 at 9:56 AM, Simon Slavin wrote:
On 13 Dec 2016, at 3:53pm, Paul Egli wrote:
Well if i am missing something, i still cannot see it.
Based on these quotes in the docs, i assume that a
Well, no. Microsoft intended to store the number of days since 1900/01/01 with
1900/01/01 being day 1. Unfortunately, Microsoft being Microsoft does not know
how to calculate leap years properly -- I think they still have problems with
the divisible by 100 and 400 bits, and will probably
On 13 Dec 2016, at 4:02pm, Paul Egli wrote:
> Great. Just to clarify, SQLite will already use "alternateChildIndex1" from
> the example? Or just that it would be possible as an enhancement request?
I don’t know the answer, but you can find out exactly what SQLite has
On 12/12/16, Rosina joy01 wrote:
> Hi,
>
> Could some one share the steps on how to build the sqlite-see with node js?
You are going to need to recompile the SQLite adaptor for Node.js. I
don't know how to do that, but as Node.js is open-source, surely it is
not too hard.
Hurray for intelligent logic! Also seems to work for composite foreign keys.
Thank you all for checking more into this. And thank you to the developers for
already having this in there. I think this would be worthy of having a couple
sentences written up for the "Foreign Key Support" page.
On Tue, Dec 13, 2016 at 8:07 PM, David Raymond
wrote:
> sqlite> create table parentTable
>...> (
>...> id integer primary key,
>...> a int not null,
>...> b int not null,
>...> otherStuff,
>...> unique (a, b)
>...> );
>
> sqlite>
For the specific case below I was just banging something out to test if it
worked, so didn't adhere to the best design practices. All your comments are
good ones for actual designs though.
-Original Message-
From: sqlite-users [mailto:sqlite-users-boun...@mailinglists.sqlite.org] On
33 matches
Mail list logo