Comment # 8
on bug 10081
from Michael Mann
(In reply to comment #7)
> Does it need to be emem memory? If it's in the PDML export code it should
> probably just be g_malloced and manually freed since there are no exceptions
> that would cause it to be lost?
That thought had crossed my mind, but the issue is the byte string is being
created in proto_construct_match_selected_string (not in print.c where PDML is
outputted). However in print.c where proto_construct_match_selected_string is
called, the comment above it is:
/* XXX - this is a hack until we can just call
* fvalue_to_string_repr() for *all* FT_* types. */
So is the "right" fix to address it there?
You are receiving this mail because:
- You are watching all bug changes.
