[sword-devel] ListKey issue (bug?)
greg.hellings at gmail.com
Sun May 17 19:26:08 MST 2009
To me, none of that answers the question why everyone in 64-bit was
able to replicate the problem, while people in 32-bit machines weren't
able to replicate the bug. That just seems like an inconsistency.
Are we willing to just say it's something that the versions of the
compilers handled differently, or might there be a bug somewhere near
this in our own code related 64-bit systems?
On Sun, May 17, 2009 at 7:40 PM, Troy A. Griffitts <scribe at crosswire.org> wrote:
> Again, the details with this, if interested are:
> results++ mandates increment then return of a temporary which represented
> the object before the increment.
> You are assigning your retval to the internal text value of this temporary,
> which goes out of scope immediately after the assignment (calling the d-tor
> and thus freeing the memory).
> Hope this helps.
> Matthew Talbert wrote:
>> It turns out that if we use ++results instead of results++ we don't
>> have a problem. So it would seem that there is still something funny
>> going on.
>> Nevertheless we will not count on the result persisting.
>> sword-devel mailing list: sword-devel at crosswire.org
>> Instructions to unsubscribe/change your settings at above page
> sword-devel mailing list: sword-devel at crosswire.org
> Instructions to unsubscribe/change your settings at above page
More information about the sword-devel