)]}' { "commit": "7da41f48c8acea834e8204917fe59da2b975903b", "tree": "4a07966ead29f783b0350a51c4c5c6338910f176", "parents": [ "155bd0ce23144e5c7067965a22646523f1a38b51" ], "author": { "name": "Shawn O. Pearce", "email": "spearce@spearce.org", "time": "Thu Dec 14 05:07:46 2006 -0500" }, "committer": { "name": "Junio C Hamano", "email": "junkio@cox.net", "time": "Thu Dec 14 02:40:33 2006 -0800" }, "message": "Bypass expensive content comparsion during rename detection.\n\nWhen comparing file contents during the second loop through a rename\ndetection attempt we can skip the expensive byte-by-byte comparsion\nif both source and destination files have valid SHA1 values. This\nimproves performance by avoiding either an expensive open/mmap to\nread the working tree copy, or an expensive inflate of a blob object.\n\nUnfortunately we still have to at least initialize the sizes of the\nsource and destination files even if the SHA1 values don\u0027t match.\nFailing to initialize the sizes causes a number of test cases to fail\nand start reporting different copy/rename behavior than was expected.\n\nSigned-off-by: Shawn O. Pearce \u003cspearce@spearce.org\u003e\nSigned-off-by: Junio C Hamano \u003cjunkio@cox.net\u003e\n", "tree_diff": [ { "type": "modify", "old_id": "57a74b6bb83e14af5f7336f397f79fbfcdc8b5f7", "old_mode": 33188, "old_path": "diffcore-rename.c", "new_id": "91fa2bea519a39ee65dcf018b7a72885b9eeec70", "new_mode": 33188, "new_path": "diffcore-rename.c" } ] }