On Wed, 10 May 2000, Gilles Detillieux wrote:
> in the code. Actually, I think the main reason the local device/inode
> duplicate detection had been taken out of the code to begin with in the
> early 3.2 development was because it wasn't optional, and it was feared
> it would cause problems for some. There are probably enough people who
> do use it that it would be worth putting back in, even if we do add MD5
> support as well, because for local files the device/inode lookup is much
> quicker.
One reason for taking it out was that it wasn't optional. Another reason
was because it only worked for local files. This seemed unfair to say the
least--you only would get duplicate elimination if you were using
local_urls. It made for an inconsistent feature--if local_urls switched to
HTTP (e.g. for directory listings) then you would no longer get duplicate
detection.
Yes, it probably makes sense to ultimately have a few possible signature
methods. Inode lookup is fast, but it won't catch copies of files.
--
-Geoff Hutchison
Williams Students Online
http://wso.williams.edu/
------------------------------------
To unsubscribe from the htdig mailing list, send a message to
[EMAIL PROTECTED]
You will receive a message to confirm this.
