Ron Johnson:
> On 10/28/06 16:26, Jochen Schulz wrote:
> >
> > But one should mention that there are few automatically generated
> > filenames that are as ugly (and mostly meaningless) as those of
> > maildirs.
>
> Ugly? yes.
>
> Meaningless? Definitely not. For example, the first "field" in a
> Maildir name is the timestamp of the message's arrival time, another
> field in the node name, and another field is the file size. I'm
> sure that in an hour or so we could deduce the whole name's meaning.
Ah, you're (partly) right. I only had in mind that the mail's status
(new, read, unread) may be part of the name. The corresponding Wikipedia
article sounds like the file name scheme is not standardized, though:
"The filename can be almost any unique series of filename characters
(except a colon), but a typical implementation might use the current
time in seconds, the hostname, the process ID and some random numbers."
And DJB (inventor of maildir) writes: "Do not try to extract information
from unique names." (http://cr.yp.to/proto/maildir.html)
> Besides, How often do you poke around Maildir/cur?
I don't, but it sounded like the OP does.
J.
--
Every day in every way I am getting better and better.
[Agree] [Disagree]
<http://www.slowlydownward.com/NODATA/data_enter2.html>
Attachment:
signature.asc
Description: Digital signature