From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: Date: Wed, 7 Mar 2007 11:00:14 -0500 From: "Russ Cox" To: "Fans of the OS Plan 9 from Bell Labs" <9fans@cse.psu.edu> Subject: Re: [9fans] replica/updatedb In-Reply-To: <07f076c821c5ff2efca16357f198075a@coraid.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <07f076c821c5ff2efca16357f198075a@coraid.com> Topicbox-Message-UUID: 19b59232-ead2-11e9-9d60-3106f5b1d025 On 3/7/07, erik quanstrom wrote: > i'm using updatedb on two successive days in a dump > where $path on day n-1 was a file and $path on day n > was a directory. updatedb generates an 'm' (modified > stat) line for this case. this seems wrong to me. > i would think that it should generate a 'd' (delete) followed > by a 'c' (create) pair. although i'm sure an argument could > be made the other way. > > i'm leaning toward changing updatedb. is there a better > argument for fixing the consumer? you should fix updatedb. if this had happened on sources then you would need to fix applylog too, since lots of people would need to be able to apply the existing log. russ