{"thread":{"id":"43573","subject":"Re: [PATCH 1/1] Bypass expensive content comparsion during rename detection","startedAt":"2006-12-14T11:26:48Z","lastAt":"2006-12-14T12:13:36Z","messageCount":2,"participants":["Junio C Hamano","Shawn Pearce"],"isPatch":true,"patchVersion":1,"patchTotal":1},"messages":[{"id":"295619","messageId":"200612141126.kBEBQme7021704@hera.kernel.org","threadId":"43573","inReplyTo":null,"subject":"Re: [PATCH 1/1] Bypass expensive content comparsion during rename detection","fromName":"Junio C Hamano","fromEmail":"junio@hera.kernel.org","sentAt":"2006-12-14T11:26:48Z","receivedAt":"2006-12-14T11:26:48Z","isPatch":true,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"estimate_similarity() is called after it has determined that one\nand two do not match exactly, and it relies on populate_filespec\nwith at least size-only has been called on them so that it can\nreject filepair with vastly different sizes, still without\nlooking at (thus without loading) the contents.\n"},{"id":"296732","messageId":"20061214121336.GN1747@spearce.org","threadId":"43573","inReplyTo":"200612141126.kBEBQme7021704@hera.kernel.org","subject":"Re: [PATCH 1/1] Bypass expensive content comparsion during rename detection","fromName":"Shawn Pearce","fromEmail":"spearce@spearce.org","sentAt":"2006-12-14T12:13:36Z","receivedAt":"2006-12-14T12:13:36Z","isPatch":true,"sender":{"key":"spearce@spearce.org","avatar":"https://avatars.githubusercontent.com/u/34844?v=4"},"body":"Junio C Hamano <junio@hera.kernel.org> wrote:\n> estimate_similarity() is called after it has determined that one\n> and two do not match exactly, and it relies on populate_filespec\n> with at least size-only has been called on them so that it can\n> reject filepair with vastly different sizes, still without\n> looking at (thus without loading) the contents.\n\n*ouch* That hurts.  OK, it makes a lot of sense, but its crazy.\nAbout as crazy as the comment I put in my revised patch...  ;-)\n\n-- \n"}]}