From: René Scharfe <l.s.r@web.de>
Date: Thu, 12 May 2016 22:12:12 GMT
Subject: Re: [Bug?] log -p -W showing the whole file for a patch that adds to the end?
Message-ID: <5734FFBC.5070505@web.de>
In-Reply-To: <xmqq4ma5msrd.fsf@gitster.mtv.corp.google.com>

Am 11.05.2016 um 00:51 schrieb Junio C Hamano:
> The helper function get_func_line() however gets confused when a
> hunk adds a new function at the very end, and returns -1 to signal
> that it did not find a suitable "function header line", i.e. the
> beginning of previous function.  The caller then takes this signal
> and shows from the very beginning of the file.  We end up showing
> the entire file, starting from the very beginning to the end of the
> newly added lines.
>
> We can avoid this by using the original hunk boundary when the
> helper gives up.

It's a more conservative fallback in this case, but it regresses e.g. if 
you have a long list of includes at the top of a C file -- it won't show 
the full list anymore with -W.

The problem is that we are trying to access lines in the preimage that 
are only actually in the postimage.  Simply switching to useing the 
postimage unconditionally is not enough -- we'd have the same problem 
with the complementary case of lines at the end being removed.

I wonder if using the postimage as a fallback suffices.  Need to look at 
this more closely.

René