Warren Young a écrit :
> Mathieu Pagé wrote:
>>
>> <code>
>> mysqlpp::Row row;
>> mysqlpp::Row::size_type i;
>> for (i = 0; row = res.at(i); ++i)
>> cout <<row.at(0) <<endl;
>> </code>
>>
>> however I get an length_error exception on "row = res.at(i)".
>
> I think you're expecting Row::at() to return false (or, rather, an
> object that can be converted to false) when you walk off the end of
> the result set. MySQL++ doesn't do that sort of thing, by design. It
> signals errors with exceptions, just as you saw.
>
> That is the fall-back behavior when you turn off exceptions, however.
> Read the user manual for details.
>
> You might also look at examples/simple3.cpp. It operates in a way
> very similar to what you're doing, without needing special treatment.
> It does that by using a different type of query than you are using.
>
Hi M. Young
Thanks you for your answer.
I looked at simple.cpp and tried to do the same thing in my code (using
"res.fetch_row()") and I got the same error. Actually I just realised,
looking at the stack trace when the error occurs, that the exception is
trown during the execution of the operator= function that copy the row.
I also tried to execute example/resetdb tonight and it raise an
exception too. Maybe there is a problem with my compiled version of MySQL++.
