file size parsed/passed incorrectly/incompletely in cli_qpathinfo*()

[Jeremy Allison]

On Wed, Mar 16, 2005 at 01:16:31PM -0500, Derrell.Lipman at UnwiredUniverse.com wrote:
> Jeremy Allison <jra at samba.org> writes:
> 
> > On Wed, Mar 16, 2005 at 02:07:26PM +0100, Henrik Nordstrom wrote:
> >> 
> >> Not even sure size_t ever was the traditional. For as long as I can see in 
> >> the history of UNIX off_t as been "the" type for file sizes and positions, 
> >> and size_t the size of I/O buffers etc. Before that long resp int appears 
> >> to have been used for the purpose.
> >
> > Yeah you're right. I think it was just a bug introduced when we
> > created that library. I vaguely remember it being my fault.... :-).
> 
> No need to kick yourself. :-) I've already fixed it, for all of the attribute
> functions.  (There are still numerous uses of size_t which I didn't change.
> They may or may not be appropriate.)  The patch will be provided after I do
> some testing, probably later today or tomorrow.
> 
> I had to make some changes to torture.c for this patch, but I have no idea how
> that is used.  I've heard numerous references to a torture test, but have never
> run it.  smbtorture (which I presume to be the executable that runs it) does
> not seem to built by a simple "make".  Is this something I should verify still
> works, and if so, how do I use it?

You do make bin/smbtorture. Don't worry too much about it - if you break
it we'll fix it (as the build farm depends on it) but the real torture
tester these days is in Samba4.

Jeremy.