# Re: [PATCH 1/1] Bypass expensive content comparsion during rename detection 2 messages from 2006-12-14 to 2006-12-14. Participants: Junio C Hamano, Shawn Pearce. Thread: https://gitlist.dev/t/43573 ## Junio C Hamano, 2006-12-14 11:26 Subject: Re: [PATCH 1/1] Bypass expensive content comparsion during rename detection Message-ID: <200612141126.kBEBQme7021704@hera.kernel.org> URL: https://gitlist.dev/e/200612141126.kBEBQme7021704%40hera.kernel.org ``` estimate_similarity() is called after it has determined that one and two do not match exactly, and it relies on populate_filespec with at least size-only has been called on them so that it can reject filepair with vastly different sizes, still without looking at (thus without loading) the contents. ``` ## Shawn Pearce, 2006-12-14 12:13 Subject: Re: [PATCH 1/1] Bypass expensive content comparsion during rename detection Message-ID: <20061214121336.GN1747@spearce.org> URL: https://gitlist.dev/e/20061214121336.GN1747%40spearce.org In-Reply-To: <200612141126.kBEBQme7021704@hera.kernel.org> ``` Junio C Hamano wrote: > estimate_similarity() is called after it has determined that one > and two do not match exactly, and it relies on populate_filespec > with at least size-only has been called on them so that it can > reject filepair with vastly different sizes, still without > looking at (thus without loading) the contents. *ouch* That hurts. OK, it makes a lot of sense, but its crazy. About as crazy as the comment I put in my revised patch... ;-) -- ```