shithub: barf

ref: 255eea6b9e7a886ccfcea5e19bbc95a773367085
dir: /bin/gz.tumblr/

#!/bin/rc
# Parse XML RSS feeds into BARF blog posts for the specified site.
# If a post with a matching _link_ already exists, no new post will
# be created for that item. Tags will be created according to the rules
# defined in the get_tags() function.
#
# The file argument should point to a file containing one line per feed,
# with fields separated by the | character, in the following format:
#
#	http://feeds.feedburner.com/ImNotReallyStanleyLieber|stanleylieber
#
# where the first field is the feed URL and each addition field is a tag.
#
# Requires 20h's xmlpull and rssread.
rfork en
switch($1){
case /*
	feeds=$1
	site=tumblr.stanleylieber.com
	tags=()
case *
	echo 'Usage: gz file' >[1=2]
	exit usage
}

file=/tmp/gz.$pid
werc=/usr/sl/www/werc

if(test -f /boot/factotum)
	cmd=hget
if not
	cmd='curl -s'

fn cram{
	ssam '
,s/^link:.*$/HJDIVIDER&HJDIVIDER/g
,s/^title:.*$/HJDIVIDER&HJDIVIDER/g
,s/\n//g
,s/HJDIVIDERtitle:/\n&/g
	'
}

fn scape{
	ssam '
,s/
//g
,s/\"/\"/g
,s/\"/\"/g
,s/\&/\&/g
,s/\&/\&/g
,s/\'/''/g
,s/\,/,/g
,s/\-/-/g
,s/\./\./g
,s/\//\//g
,s/\:/:/g
,s/\&#59;/;/g
,s/\&lt;/</g
,s/\&#60;/</g
,s/\&#61;/=/g
,s/\&gt;/>/g
,s/\&#62;/>/g
,s/\&#95;/_/g
,s/\|/\&#124;/g
	'
}

fn get_posts{
	$"cmd $"feed | rssread | cram | scape >$file
	echo >>$file
}

fn parse_posts{
	ifs='
' {
		posts=`{cat $file}
		for(i in `{seq 1 $#posts | sort -nr}){
			post=$posts($i)
			if(! ~ $post ''){
				a_title=`{echo $post | sed 's/^.*HJDIVIDERtitle: //g; s/HJDIVIDER.*$//g'}
				a_date=`{date}
				a_link=`{echo $post | sed 's/^.*HJDIVIDERlink:  //g; s/HJDIVIDER.*$//g'}
				a_body=`{echo $post | sed 's/^.*HJDIVIDER//g; s/^.*HJDIVIDER//g; s/^.*HJDIVIDER//g'}
				a_id=`{echo `{ls -p $werc/sites/$site/src | sort -n | tail -1}^+1 | bc}
				if(~ $#a_id 0)
					a_id=1
				while(test -d $werc/sites/$site/src/$a_id)
					a_id=`{echo $a_id^+1 | bc}
				if(! ~ $"a_link '' && ! grep -s `{echo $"a_link | md5sum} $werc/sites/$site/links){
					mkdir -p $werc/sites/$site/src/$a_id/tags # big fat race
					echo $"a_title >$werc/sites/$site/src/$a_id/title
					echo $"a_date >$werc/sites/$site/src/$a_id/date
					echo $a_link(1) >$werc/sites/$site/src/$a_id/link
					echo $a_link(1) | md5sum >>$werc/sites/$site/links
					if(~ $a_link(1) *staticflickr.com*)
						echo '<img src="'$"a_link'">' >$werc/sites/$site/src/$a_id/body
					echo $"a_body '</a></li></ul>' >>$werc/sites/$site/src/$a_id/body
					ifs=' ' {
						for(j in $tags){
							>$werc/sites/$site/src/$a_id/tags/$j
							echo $a_id/tags/$j >>$werc/sites/$site/tags
						}
					}
				}
				if(test -f /boot/factotum && ~ $site *.stanleylieber.com && test -d $werc/sites/$site/src/$a_id)
					chmod +t $werc/sites/$site/src/$a_id $werc/sites/$site/src/$a_id $werc/sites/$site/src/$a_id $werc/sites/$site/src/$a_id/* $werc/sites/$site/src/$a_id $werc/sites/$site/src/$a_id/tags/*
			}
		}
	}
}

if(test -f /boot/factotum && test -f /rc/bin/hget)
	webfs
for(i in `{grep -v -e '^#' $feeds}){
	feed=`{echo $"i | sed 's/\|.*$//g'}
	tags=`{echo $"i | sed 's/\|/ /g'}
	tags=$tags(2-)
	get_posts
	parse_posts
}