From: Junio C Hamano Date: Mon, 03 Mar 2014 18:40:08 GMT Subject: Re: [PATCH] commit.c:record_author_date() use skip_prefix() instead of starts_with() Message-ID: In-Reply-To: <53144450.3080006@alum.mit.edu> Michael Haggerty writes: >> - if (!starts_with(buf, "author ")) { >> + if (!skip_prefix(buf, "author ")) { > > If this is the only change, there is not much point, is there? How does > this help? Perhaps there is some way to take advantage of the > difference between starts_with() and skip_prefix() to simplify the rest > of the function? I admit I lost track, but wasn't there a discussion to use starts_with/ends_with when appropriate (namely, the caller is absolutely not interested in what the remainder of the string is after skipping the prefix), moving away from skip_prefix()? Isn't this change going in the wrong direction?