#!/bin/bash
# By Linc 10/1/2004
# Find the latest script at http://linc.homeunix.org:8080/scripts/bashpodder
# Last revision 07/01/2005 - Many Contributers!
# If you use this and have made improvements or have comments
# drop me an email at linc dot fessenden at gmail dot com
# I'd appreciate it!
# Amended by Nik Butler 2005 10 12
# Dropped creation of date in favour of downloading podcasts by title.
# using wgets -N timestamp and -c Continue feature to grab the files.
# wget knows what the filename is and if we have seen it or not 
# of course the above screws up if the file names change !
# also we accept 1 argument ... the count of the shows to download from the feed.
# this stops the files going MAD trying to grab everything from the feed.
# ive split the xsl files up to spot the title in the feed.
#  

# Make script crontab friendly:
cd $(dirname $0)
rm current_shows temp_shows 
du -a | grep -i mp3 | awk '{FS=" "; print $2}' >temp_shows
# Read the PodCast for Titles and Grab CurrentList into $title with wget  
while read podcast
	do

	echo $podcast

	title=$(wget -q $podcast -O - | xsltproc parse_enclosure_title.xsl - 2>/dev/null) 
	# drop spaces and annoying characters from the name.
	title=${title//\ /_}
	title=${title//\'/}
	title=${title//\\/}
	title=${title//\//}
	# then create the directory.

	if ! test  -d "$title" 
        then
        	mkdir $title
	fi

	# the file is the rss file from the site. and we seek order !
	file=$(wget -q $podcast -O - | xsltproc parse_enclosure_url.xsl - 2>/dev/null )|| file=$(wget -q $podcast -O - | tr '\r' '\n' | tr \' \" | sed -n 's/.*url="\([^"]*\)".*/\1/p') 

	# count the files we are downloading.
	fileCount=0

	for url in $file
        do

		# increment by one and if a arguement was passed and 
		# we match the amount of files downloaded drop out of this
		# go find more files other wise.
		# 2005 11 01 -- Lugradio seems to adda query string to the end of the url ...
		# so we want to strip it from the url before downloading the file.
		# 
		#$filetmp = ${file//\\?.*/}
		infile=${url//\?[a-z]*/}
		# find the file name time..
		outfile=$(echo $infile | awk 'BEGIN { FS = "/" } ; { print $(NF) }')
		if [ -n "$1" ]
			then
			if [ $fileCount -eq $1  ]
			then
				break
			fi
		fi
		echo "fetching $url"
                wget -c -N -P $title -O ./${title}/${outfile} "$infile"  
		fileCount=$(($fileCount+1))
        done
done < bp.conf
du -a | grep -i mp3 | awk '{FS=" "; print $2}' >>temp_shows
datenow=$(date +%Y-%m-%d)
sort temp_shows | uniq -u >> PodCasts-$datenow