On Sun, Dec 23, 2018 at 02:32:30PM +0100, Yann Ylavic wrote:
> Thanks Stefan, I didn't notice before in your proposed patch, but it
> looks like uint64_t casts should be apr_uint64_t too.
>
> Regards,
> Yann.
Right. I went ahead and fixed it in r1849630.
Thanks,
Stefan
On Sun, Dec 23, 2018 at 10:33 AM Stefan Sperling wrote:
>
> On Wed, Dec 19, 2018 at 07:03:39PM +0100, Stefan Sperling wrote:
> > On Wed, Dec 19, 2018 at 02:58:28PM +0100, Yann Ylavic wrote:
> > > On Wed, Dec 19, 2018 at 9:53 AM Stefan Sperling wrote:
> > > >
> > > > On Tue, Dec 18, 2018 at
On Wed, Dec 19, 2018 at 07:03:39PM +0100, Stefan Sperling wrote:
> On Wed, Dec 19, 2018 at 02:58:28PM +0100, Yann Ylavic wrote:
> > On Wed, Dec 19, 2018 at 9:53 AM Stefan Sperling wrote:
> > >
> > > On Tue, Dec 18, 2018 at 12:29:18AM +0100, Yann Ylavic wrote:
> > > > But yes, upcast is better,
On Wed, Dec 19, 2018 at 02:58:28PM +0100, Yann Ylavic wrote:
> On Wed, Dec 19, 2018 at 9:53 AM Stefan Sperling wrote:
> >
> > On Tue, Dec 18, 2018 at 12:29:18AM +0100, Yann Ylavic wrote:
> > > But yes, upcast is better, while at it I'd go for uint64_t...
> >
> > Like this?
>
> I think
On Wed, Dec 19, 2018 at 9:53 AM Stefan Sperling wrote:
>
> On Tue, Dec 18, 2018 at 12:29:18AM +0100, Yann Ylavic wrote:
> > But yes, upcast is better, while at it I'd go for uint64_t...
>
> Like this?
I think APR_UINT64_T_FMT/apr_uint64_t would be more portable ;)
Thanks for taking care of this
On Tue, Dec 18, 2018 at 12:29:18AM +0100, Yann Ylavic wrote:
> But yes, upcast is better, while at it I'd go for uint64_t...
Like this?
I've noticed that the same problem seems to exist in some other modules.
I'll send separate patches for those once this patch has settled.
Index:
On Mon, Dec 17, 2018 at 5:40 PM William A Rowe Jr wrote:
>
> On Sun, Dec 16, 2018 at 7:27 AM Yann Ylavic wrote:
>>
>>
>> Since it's logging only, it may be easier to cast to (long) each
>> total_in/out though.
>
> Downcast? Why not upcast to apr_off_t and use the _FMT macro as first
>
On Sun, Dec 16, 2018 at 7:27 AM Yann Ylavic wrote:
>
> Since it's logging only, it may be easier to cast to (long) each
> total_in/out though.
>
Downcast? Why not upcast to apr_off_t and use the _FMT macro as first
suggested?
On Sun, Dec 16, 2018 at 2:21 PM Yann Ylavic wrote:
>
> On Sun, Dec 16, 2018 at 2:14 PM Stefan Sperling wrote:
> >
> > The file /usr/include/zlib.h I have on OpenBSD -current has this:
> >
> > typedef struct z_stream_s {
> > [...]
> > z_off_t total_in; /* total nb of input bytes read so
On Sun, Dec 16, 2018 at 2:14 PM Stefan Sperling wrote:
>
> The file /usr/include/zlib.h I have on OpenBSD -current has this:
>
> typedef struct z_stream_s {
> [...]
> z_off_t total_in; /* total nb of input bytes read so far */
> [...]
> z_off_t total_out; /* total nb of bytes
On Sun, Dec 16, 2018 at 02:03:45PM +0100, Yann Ylavic wrote:
> On Sun, Dec 16, 2018 at 1:28 PM Stefan Sperling wrote:
> >
> > mod_deflates hard-codes some off_t format directives to "%ld".
> > It seems to me this code should use the macro provided by APR instead.
> >
> > Looking for another pair
On Sun, Dec 16, 2018 at 1:28 PM Stefan Sperling wrote:
>
> mod_deflates hard-codes some off_t format directives to "%ld".
> It seems to me this code should use the macro provided by APR instead.
>
> Looking for another pair of eyes. Does this patch look good to commit?
It seems that zlib defines
12 matches
Mail list logo
