[Bug 611936] Sqlite3: assertion `gnc_numeric_check(amt) == GNC_ERROR_OK' failed

[Jeff Kletsky]

In my opinion, since this is an OS-distribution-related problem rather 
than a core-to-the-library problem, spending a lot of valuable dev time 
on "fixing" it seems inappropriate.

I would suggest something along the lines of testing for the library 
bug, alerting the user if it exists (with a checkbox for "Don't tell me 
again"), and disabling the SQLite3 option if it exists.

There is test code distributed in 
libdbi-drivers-0.8.3-1/tests/test_dbi.c that may be useful, perhaps 
combined with tempfile. I don't know if there is a Windows equivalent of 
tempfile; another option would be to disable the test on Windows builds 
(or only enable it for Linux, as it seems to have only surfaced with 
Linux builds of the library).

Running test_dbi on an "Intel(R) Atom(TM) CPU  330   @ 1.60GHz"
user    0m0.028s
sys     0m0.008s

so it would seem that the "overhead" of performing the test for the 
"bad" data types:

the_longlong: in:-9223372036854775807 out:0<<
the_ulonglong: in:9223372036854775807 out:0<<

the_double: in:1.7976931348623157E+307 out:0.000000e+00<<

the_datetime: in:'2001-12-31 23:59:59' out:2001-12-31 0:0:0
the_datetime_tz: in:'2001-12-31 23:59:59 -10:00' out:2001-12-31 0:0:0

the_time: in:'23:59:59' out:0:0:0
the_time_tz: in:'23:59:59-10:00' out:0:0:0

on each start-up would not be a huge burden.

Jeff


On 01/04/2011 04:14 AM, GnuCash (bugzilla.gnome.org) wrote:
> https://bugzilla.gnome.org/show_bug.cgi?id=611936
>    GnuCash | SQL Backend | SVN
>
> Christian Stimming<stimming>  changed:
>
>             What    |Removed                     |Added
> ----------------------------------------------------------------------------
>               Status|RESOLVED                    |REOPENED
>                   CC|                            |stimming at tuhh.de
>           Resolution|NOTGNOME                    |
>
> --- Comment #32 from Christian Stimming<stimming at tuhh.de>  2011-01-04 12:14:30 UTC ---
> Reopening because it's a bug that seems to hit an increasing number of users,
> and we need to find a better way around it than just saying "try a different
> libdbi version".
>