On Mon, Jan 26, 2009 at 01:37:23AM -0700, LuKreme wrote:
On 25-Jan-2009, at 17:59, Dallman Ross wrote:
[LuKreme cites man page and addes emphasis:]
Names an rcfile (relative to the current directory) to which
processing will be switched. If the named rcfile doesn't
exist or is not a normal file or /dev/null then an error
will be logged ***and processing will continue in the current
rcfile.***
You read it, but you didn't grok it. The case you have presented
does not meet the criteria described in the first clause of the
second quoted sentence: (a) there is *no* named rcfile; and
thus, by definition, (b) there likewise is nothing to discern as
either a normal file or /dev/null. There is no "ARG," as you put
it, to the SWITCHRC statement in your syntax.
"No named rcfile" is no[t] the same as "a named rcfile that doesn't
exist."
OK, it says right there, "If the named rcfile doesn't exist" and "or is
not a normal file"
I can't tell if you're agreeing with me finally or if you're still
disagreeing. If the latter, run my challenge code.
################ start rcfile ################
SWITCHRC
LOG = "Kilroy was here.
"
################# end rcfile #################
Typing just
SWITCHRC
performs the act of unsetting the value (if there was any)
assigned to that variable. I.e., SWITCHRC will now be unset.
That is not the case the man page is describing. The man
page is saying that if I have something like:
SWITCHRC = LuKreme
but I have no rcfile named "LuKreme", then "the named rcfile
doesn't exist". In that case, as per the man page, "processing
will continue".
Dallman
____________________________________________________________
procmail mailing list Procmail homepage: http://www.procmail.org/
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail