Re: [PATCH] comctl32/listview: Fix printf format (PVS-Studio)

[Alexandre Julliard]

Nikolay Sivov  writes:
> @@ -587,7 +587,7 @@ static const char* debugscrollinfo(const SCROLLINFO *pScrollInfo)
>      int len, size = DEBUG_BUFFER_SIZE;
>  if (pScrollInfo == NULL) return "(null)";
> - len = snprintf(buf, size, "{cbSize=%d, ", pScrollInfo->cbSize);
> - len = snprintf(buf, size, "{cbSize=%u, ", pScrollInfo->cbSize);

That doesn't seem very necessary. I don't mind a few fixes like that,
but I'm concerned that there would be a million of them. How many do you
expect?

[PATCH] comctl32/listview: Fix printf format (PVS-Studio)

[Michael Stefaniuc]

On 10/28/2015 01:35 PM, Alexandre Julliard wrote:
> Nikolay Sivov  writes:
> > @@ -587,7 +587,7 @@ static const char* debugscrollinfo(const SCROLLINFO *pScrollInfo)
> >      int len, size = DEBUG_BUFFER_SIZE;
> >  if (pScrollInfo == NULL) return "(null)";
> > - len = snprintf(buf, size, "{cbSize=%d, ", pScrollInfo->cbSize);
> > - len = snprintf(buf, size, "{cbSize=%u, ", pScrollInfo->cbSize);
>
> That doesn't seem very necessary. I don't mind a few fixes like that,
> but I'm concerned that there would be a million of them. How many do you
> expect?

PVS-Studio has 266 defects for this. 3 of those are a double <==> long
double mismatch. The rest are all signed/unsigned stuff.

bye
    michael

[PATCH] comctl32/listview: Fix printf format (PVS-Studio)

[Nikolay Sivov]

On 28.10.2015 16:11, Michael Stefaniuc wrote:
> On 10/28/2015 01:35 PM, Alexandre Julliard wrote:
> > Nikolay Sivov  writes:
> > > @@ -587,7 +587,7 @@ static const char* debugscrollinfo(const SCROLLINFO *pScrollInfo)
> > >       int len, size = DEBUG_BUFFER_SIZE;
> > >   if (pScrollInfo == NULL) return "(null)";
> > > - len = snprintf(buf, size, "{cbSize=%d, ", pScrollInfo->cbSize);
> > > - len = snprintf(buf, size, "{cbSize=%u, ", pScrollInfo->cbSize);
> >
> > That doesn't seem very necessary. I don't mind a few fixes like that,
> > but I'm concerned that there would be a million of them. How many do you
> > expect?
>
> PVS-Studio has 266 defects for this. 3 of those are a double <==> long
> double mismatch. The rest are all signed/unsigned stuff.

Right, many could be combined to a single patch, and some parts warnings 
from programs/ we can ignore probably,
so it would be less. So not a million, but maybe about 50 total or so. 
I'm okay skipping this if you think it's useless.
> bye
>    michael

[PATCH] comctl32/listview: Fix printf format (PVS-Studio)

[Alexandre Julliard]

Nikolay Sivov  writes:
> On 28.10.2015 16:11, Michael Stefaniuc wrote:
> > On 10/28/2015 01:35 PM, Alexandre Julliard wrote:
> > > Nikolay Sivov  writes:
> > > > @@ -587,7 +587,7 @@ static const char* debugscrollinfo(const SCROLLINFO *pScrollInfo)
> > > >       int len, size = DEBUG_BUFFER_SIZE;
> > > >         if (pScrollInfo == NULL) return "(null)";
> > > > - len = snprintf(buf, size, "{cbSize=%d, ", pScrollInfo->cbSize);
> > > > - len = snprintf(buf, size, "{cbSize=%u, ", pScrollInfo->cbSize);
> > >
> > > That doesn't seem very necessary. I don't mind a few fixes like that,
> > > but I'm concerned that there would be a million of them. How many do you
> > > expect?
> >
> > PVS-Studio has 266 defects for this. 3 of those are a double <==> long
> > double mismatch. The rest are all signed/unsigned stuff.
>
> Right, many could be combined to a single patch, and some parts
> warnings from programs/ we can ignore probably,
> so it would be less. So not a million, but maybe about 50 total or
> so. I'm okay skipping this if you think it's useless.

If it's around 50 patches, that's OK.