{"thread":{"id":"12485","subject":"[RFC] Fsck while fetching","startedAt":"2008-03-04T07:55:45Z","lastAt":"2008-03-04T07:55:45Z","messageCount":1,"participants":["Martin Koegler"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"70900","messageId":"20080304075545.GA31793@auto.tuwien.ac.at","threadId":"12485","inReplyTo":null,"subject":"[RFC] Fsck while fetching","fromName":"Martin Koegler","fromEmail":"mkoegler@auto.tuwien.ac.at","sentAt":"2008-03-04T07:55:45Z","receivedAt":"2008-03-04T07:55:45Z","isPatch":false,"sender":{"key":"mkoegler@auto.tuwien.ac.at","avatar":null},"body":"As the infrastructure for checking the objects and their connectivity\nwhile pushing is in place, it is easy to do the same while fetching on\nnon shallow repositories.\n\nOn shallow repositories this is currently only possible, if the list\nof shallow commits does not change (eg. no depth option).\n\nThe neccessary changes are:\n* Read a config option in fetch-pack to disable it (default: true)\n* Add in fetch-pack --strict to the command lines of index-pack and \n  unpack-objects, if the config option is true and args.depth is 0.\n\nThis leads to the question:\n* How should the config option be named? fetch.fsckObjects? Should\n  receive.fsckObjects be reused? Should receive.fsckObjects be renamed\n  to a new name, which is used in both cases?\n\nOn additional question:\n* fsck has a strict option, which looks for files with 0664 mode. How\n  should such files be handled with the checking while pushing/fetching:\n  ignored, rejected or config option for the behaviour?\n\nmfg Martin Kögler\n"}]}