[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Bug report: Stata crashes when inserting dataset with missing data via ODBC.
Thanks to Kevin Turner at StataCorp my problem seems to
bee solved with the next official update (next week). In
addition, Kevin has offered me a special tailored fix of the
problem until then!
I have to say that StataCorp is still able to surprise me. This
time with what they call "outstanding technical support". I
couldn't agree more.
> This problem was fixed and will be in the next update.
> The issue is only present with -insert- and has to do with
> missing values were were being inserted as large numerical
> instead of
> just NULL. The initial reason for leaving the large numerical
> values was to
> preserve the multiple missing values that Stata supports (.a
> However,as time went by, it became apparent that many
> simply could not
> handle the larger values in the context that they were being
> inserted. Many
> times the result was a crash. So the fix was made to convert
> Stata missing
> values to NULL.
> Michael Ingre (Michael.Ingre@ipm.ki.se) writes:
> > I have a problem when trying to insert data via an ODBC
> >. odbc insert _all, t(table) dsn(myodbcconnection) create
> >When the above command is executed the first 4 rows of
> inserted into
> >the table. Then Stata crashes (exits to the OS) and has to
> >The first 4 rows contains complete data and the fifth row
> cell with
> >missing data. If I enter a value in the cell with missing
> and try
> >again, 10 more rows will be inserted into the database,
> crashing on
> >the first row with missing data.
> >The problem seems to be related to missing data. Other
> functions seems
> >to be working fine (as far as I know): list, query, load
> >My system config is:
> >Stata ver 8.2 31 Oct 2003
> >Mac OS X ver 10.2.8 (6R73)
> >MySQL ver 4.15
> >MyODBC ver 3.51.06
* For searches and help try: