Dan Jacobson <[EMAIL PROTECTED]> writes:
> What about partly downloaded files that our wwwoffle.conf says not to
> keep; in the future can we set a variable to leave them in some
> temporary place where we can examine them? Perhaps a
> /var/cache/wwwoffle/junk, prevjunk1,2... or perhaps /tmp.
> Yes, don't keep the file, but still, don't throw away those partially
> downloaded carcasses that fast. And, that way one could see directly
> how many failed -- not readily visible in say lastout, and without
> needing to look in a log file -- we might not be root too.
You have the choice of keeping the partially downloaded file or not.
I can't keep adding in finer details to this such as sort of keeping
the files, but not in somewhere that normal users can usefully get to
them. Anything in /tmp would be wiped when rebooting and anything in
/var/spool/wwwoffle might not be readable by an unprivileged user.
> By the way, on the man page OnlineOptions
> If the server connection times out...
> If the browser closes the connection...
> but no mention of if the server closes the connection...
If the server closes the connection then it is the end of the file.
> By the way, if I were in charge, I'd take all the comments out of
> /etc/wwwoffle/wwwoffle.conf, and just refer users to the wwwoffle.conf
> man page. Otherwise you have to do tedious risky surgery to the
> user's /etc/wwwoffle/wwwoffle.conf with each new version.
There is nothing tedious or risky about updating the configuration
file, there is a script provided to do it for you (upgrade-config.pl).
If there is a wwwoffle.conf.install file in the current directory or
the same directory that the config file is in then the comments will
be taken from that. If there isn't then it removes the comments for
the reasons that you suggest.
> P.S. There seems to be certain situations offline where one goes to
> lasttime and clicks on a entry, and sees a fresh 'request recoded',
> with no explanation of why the file wasn't gotten last time. Perhaps
> there is some clue in the logs ... but we might not be root to read
> it.
This is most likely to be caused by some Javascript in the page that
has been fetched that is not activated until viewed in a browser.
This could then redirect the browser to fetch a page that it doesn't
already have. Since WWWOFFLE cannot understand Javascript there is
not much that can be done about this.
--
Andrew.
----------------------------------------------------------------------
Andrew M. Bishop [EMAIL PROTECTED]
http://www.gedanken.demon.co.uk/
WWWOFFLE users page:
http://www.gedanken.demon.co.uk/wwwoffle/version-2.8/user.html
