On 17.02.2015 17:56, Philip Martin wrote: > Philip Martin <philip.mar...@wandisco.com> writes: > >> Essentially every time we compare start and end we need to make sure >> that we handle SVN_INVALID_REVNUM properly. I think the functions >> listed in the issue are a few locations I identified that need to be >> checked. > Some of the APIs in svn_ra.h document that SVN_INVALID_REVNUM can be > passed to mean HEAD, other functions in the file are silent on the > matter. Is passing SVN_INVALID_REVNUM to mean HEAD supported by > svn_ra_get_file_revs2, svn_ra_replay_range, etc.
Sadly, the RA layer is not consistent there ... I wouldn't be surprised if it was inconsistent amongst different RA implementations, too. Needs code review, IMO. -- Brane
