From: John Tapsell Date: Wed, 01 May 2013 07:23:32 GMT Subject: Re: git log -p unexpected behaviour Message-ID: In-Reply-To: <7vvc73bvp9.fsf@alter.siamese.dyndns.org> On 30 April 2013 21:38, Junio C Hamano wrote: > John Tapsell writes: > >> On 30 April 2013 20:44, Junio C Hamano wrote: >>> John Tapsell writes: >>> >>>> Is there no way to fix --cc to work even in the edge cases? >>> >>> Can you clarify what you mean by "fix" and "edge cases"? >> >> My understanding is that even with -cc there will be changes that >> won't be seen - and hence why --cc could be even more of a "security >> risk", no? > > Combined diff is a way to show a tricky conflict resolved in a > tricky way, so that the tricky-ness of the resolution can be > examined. A trivial resolution that takes one side is not shown > because it is not usually interesting. I don't really understand your point sorry. In this trivial resolution case, you would still just see the commit that added the code in a later commit. No? There couldn't be a case where you add or change a line of code, but not see it with --cc ? > This design choice of course > have to trust people involved in the project do not pull from > untrustworthy sources. > > You would need "log -p -m" (without any pathspec) for the kind of > "security" you are talking about. Note that "-p -m --first-parent" > is not necessarily enough. This results in seeing the same change more than once though, right?