On Thu, 2011-01-06 at 14:24:15 -0800, Alan Coopersmith wrote: > On 01/ 5/11 05:32 PM, Guillem Jover wrote: > > This will make the return value be 1 less than written. And len should > > be checked here too, in case this call failed, which could happen due to > > memory exhaustion for example (depending on the implementation). > > Though in the specific case of xrdb, the calls only check if the return > value is -1 or not, so being off by one doesn't matter here, I've gone > ahead and fixed this, and the other issues you raised, so that it's > correct if this gets copied into more apps or the callers get modified > in the future.
If the second vsnprintf call fails then the returned value will be < -1, which will not match against the error checks from the callers. > Thanks for the review, revised patch follows. The revised patch does not seem to differ from the original? regards, guillem _______________________________________________ [email protected]: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
