[PATCH:setxkbmap] Use snprintf instead of trying to pre-measure sprintf result sizes.
Van de Bugger
van.de.bugger at gmail.com
Tue Feb 15 12:48:37 PST 2011
On Tue, 2011-02-15 at 10:48 +1000, Peter Hutterer wrote:
> On Thu, Feb 10, 2011 at 12:00:11AM -0800, Alan Coopersmith wrote:
> > Signed-off-by: Alan Coopersmith <alan.coopersmith at oracle.com>
> > ---
> > setxkbmap.c | 7 +++----
> > 1 files changed, 3 insertions(+), 4 deletions(-)
> >
> > diff --git a/setxkbmap.c b/setxkbmap.c
> > index 868b727..9593ba4 100644
> > --- a/setxkbmap.c
> > +++ b/setxkbmap.c
> > @@ -626,14 +626,13 @@ findFileInPath(char *name, char *subdir)
> > }
> > for (i = 0; (i < numInclPath); i++)
> > {
> > - if ((strlen(inclPath[i]) + strlen(subdir) + strlen(name) + 2) >
> > + if (snprintf(buf, PATH_MAX, "%s/%s%s", inclPath[i], subdir, name) >=
> > PATH_MAX)
> > {
> > VMSG3(0, "Path too long (%s/%s%s). Ignored.\n", inclPath[i],
> > subdir, name);
> > continue;
> > }
> > - sprintf(buf, "%s/%s%s", inclPath[i], subdir, name);
> > fp = fopen(name, "r");
##### I guess it is a bug. We do printing to `buf', checking lentgh, but
open file named by `name'. See for my next patch.
> > if ((verbose > 7) || ((!fp) && (verbose > 5)))
> > MSG2("%s file %s\n", (fp ? "Found" : "Didn't find"), buf);
> > @@ -845,13 +844,13 @@ applyRules(void)
> > * we succeed with */
> > for (i = 0; (i < numInclPath) && (!rules); i++)
> > {
> > - if ((strlen(inclPath[i]) + strlen(rfName) + 8) > PATH_MAX)
> > + if (snprintf(buf, PATH_MAX, "%s/rules/%s", inclPath[i],
> > + svValue[RULES_NDX]) >= PATH_MAX)
> > {
> > VMSG2(0, "Path too long (%s/rules/%s). Ignored.\n",
> > inclPath[i], rfName);
> > continue;
> > }
> > - sprintf(buf, "%s/rules/%s", inclPath[i], svValue[RULES_NDX]);
> > rules = XkbRF_Load(buf, svValue[LOCALE_NDX], True, True);
> > }
> > }
> > --
> > 1.7.3.2
>
> I can't seem to find the rev this patch was based on, it gives me conflicts
> with and without Van's patches (just pushed those to master).
>
> Reviewed-by: Peter Hutterer <peter.hutterer at who-t.net> but a follow-up for
> strlen(buf) would be nice.
>
> Cheers,
> Peter
> _______________________________________________
> xorg-devel at lists.x.org: X.Org development
> Archives: http://lists.x.org/archives/xorg-devel
> Info: http://lists.x.org/mailman/listinfo/xorg-devel
More information about the xorg-devel
mailing list