Comment # 8
on bug 8339
from Evan Huus
Thanks for your work. I've committed it in r48049 with two changes:
- Minor update to use the new WS_DLL_PUBLIC define (instead of 'extern') that
Balint added a few days ago.
- Removed the 0-length case from proto_item_add_bitmask_tree. It is not valid
to pass 0 length to proto_item_add_bitmask, so if you want to be able to
support 0 length in proto_item_add_bitmask_len (which I'm not sure you do?)
then you'll have to do extra work in that function.
Regarding get_uint_value and get_int_value, just so I'm clear I understand the
issue: in the 'default' case when length > 4 and encoding is big-endian, it's
taking the four bytes from the wrong end of the set? I don't think that case is
hit much (if at all) so if you want to provide a patch fixing it then I'd
accept that.
Also, if I understand your previous comment correctly, bug #7933 is no longer
relevant and can be closed?
You are receiving this mail because:
- You are watching all bug changes.