On Sun, Nov 30, 2003 at 01:02:51PM +0200, Udi Mottelo wrote:
On Fri, 28 Nov 2003, Dallman Ross wrote:
* ? test -x "$GREEN/$RP"
WELL DONE dman !
It is fast, elegant and cool. I like it.
There is an assemble - where searching for inode is more expensive
then searching in file, but no problem, I guess that you don't have
more then thousand files.
Yup. Thank you. I have 233 in there right now, and that number stays
pretty close. The expiration of addresses is a good feature, too; I
won't have cruft lying around for years from when people change email
addresses, or from one-time or short-term correspondents who then
disappear.
In belated response to Michael Wise in re. shoving the entirety of the
headers into fgrep, I believe that searching a hundred-line header for
a thousand regexes is roughly like a hundred thousand compares; while
searching a thousand lines for only one regex is roughly like a thousand
compares. The extra ninety-nine thousand compares can take some time.
Also, I'm not entirely sure about this part, but I think Michael may
have already unknowingly incurred a subshell or $HOME wouldn't have
worked.
As Sean pointed out, fgrep doesn't take regexes. But OK, let's change
that f to an e. I take "simple" to mean easy to put in place and
update. I think it's far easier to update a thousand addresses than a
thousand regexes, not all in the same exact form. Arguably, a single
search over a thousand names is easier to wrap the mind around than is a
thousand searches over even a small search space like the headers.
I don't want to get religious about this. If Michael finds his
algorithm useful, and his servers don't mind the effort, well, okay,
then. I don't see the method as "simple" or "simplest," however,
as the thread subject requests.
--
dman
_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail