#! /bin/sh
## wwwoffle-find-rerequest-candidates -- Find all the URLs that had a
## 503 WWWOFFLE Remote Host Error [Levis'503 blues] lasttime, etc. that we
## might want to rerequest.
## Copyright       : http://www.fsf.org/copyleft/gpl.html
## Author          : Dan Jacobson -- http://jidanni.org/comp/wwwoffle/
## Created On      : Fri Mar  8 05:42:23 2002
## Last Modified On: Fri Apr 25 13:52:49 2003
## Update Count    : 14
## Status          : look mom what I whipped up in 10 minutes

: ${wwwoffle_cache=/var/spool/wwwoffle} #if not otherwise set by the user
test -d $wwwoffle_cache||{
    echo $0: $wwwoffle_cache: invalid wwwoffle_cache directory 1>&2
    exit 24;}
dir=${1-lasttime}
cd $wwwoffle_cache/$dir||exit #at least root can go there, eh?
wwwoffle-ls $dir|awk '
{i++; dfile[i]=$1; durl[i]=$NF}
END{for (i in dfile){
	getline < dfile[i]; close (dfile[i])
	if ($2 ~ "503|404") {print durl[i]} #503... probably not the only ones
	}
}'|sort|more
#probably most of them you won't want to rerequest anyway, but for the
#ones you do you do wwwoffle -F ...
#Hmm, better yet: do these commands:
# wwwoffle-find-rerequest-candidates|tee rereqlist
# edit rereqlist #so only the ones you really want to reget remain
# wget -i rereqlist; wget -i rereqlist; wget -i rereqlist
#until they all become "404 will get"