On Wed, Jul 1, 2015 at 6:15 AM, Pan Xinhui <xinhuix.pan@xxxxxxxxx> wrote:your advice sounds good, I will have a try. and welcome for review. :)
In __bitmap_parselist we can accept whitespaces on head or tail
during every parsing procedure.
If input has valid ranges, there is no reason to reject the user.
fixes are:
1) if input ends with ',', bit 0 might be set unexpectedly.
now we check if any digit is available after every loop.
2) if input has '0-', bit 0 might be set unexpectedly,
now we return -EINVAL as this kind of input is definitely wrong.
3) minor code style fix in __bitmap_parse.
and avoid in-loop incrementation of ndigits.
Why not three patches, so it becomes easier to see which is which?
sorry for my lack knowledge of comment rules. thanks for pointing out my mistakes.
commit 2528a8b also add some check, but it's still not enough.
it only correct the result in fix 1 above.
I believe the convention is to have at least 12 characters of the
sha1, with the title behind it: 2528a8b8f457 (__bitmap_parselist: fix
bug in empty string handling). Using only seven characters still risks
collisions.
Thanks,--
Frans