ds-rss/gen.sh

#!/usr/bin/env sh

#~ set -x
assert_tools () {
	err=0
	while test $# -gt 0; do
		which $1  >/dev/null 2>/dev/null || {
			>&2 printf "tool missing: $1\n"
			err=$(( $err + 1 ))
		}
		shift
	done
	test $err -eq 0 || exit $err
}
assert_source () {
	dependencies="test printf"
	assert_tools ${dependencies}
	test $# -gt 0 && {
		datafn="$1";
	} || {
		>&2 printf "option for data source missing.\n";
		datafn="data.xml";
		>&2 printf "assuming fallback '${datafn}'.\n";
	}
	test -f "$datafn" && { >&2 printf "from ${datafn} …\n"; exit 0; } || { >&2 printf "failed.\nfile '${datafn}' does not exist.\n"; exit 1; }
	printf "$datafn"
}
build () {
	datafn="$1"
	dependencies="xsltproc"
	assert_tools ${dependencies}
	>&2 printf "building ds-feed.xml … "
	xsltproc --encoding utf-8 -o ds-feed.xml rss.xsl $datafn && { >&2 printf "done.\n"; } || { >&2 printf "failed on rss.xsl.\n"; exit 1; }
	>&2 printf "building download.html … "
	xsltproc --encoding iso-8859-1 -o download.html download.xsl $datafn && { >&2 printf "done.\n"; } || { >&2 printf "failed on download.xsl.\n"; exit 1; }
	>&2 printf "building index.html … "
	xsltproc --encoding utf-8 -o index.html index.xsl $datafn && { >&2 printf "done.\n"; } || { >&2 printf "failed on index.xsl.\n"; exit 1; }
	exit 0
}
sortxml () {
#~ test "$1" = "sortxml" && {
	dependencies="xsltproc"
	assert_tools ${dependencies}
	xsltproc --encoding utf-8 -o sorted.xml sort.xsl data.xml
	mv -i sorted.xml data.xml
}
linkcheck () {
	dependencies="curl uuid"
	assert_tools ${dependencies}
	tmpfile="/tmp/"$(uuid); curl -k https://ds.ccc.de/download.html > $tmpfile
	for f in download.html ${tmpfile} ; do
		test -e $f || continue
		for e in $(cat ${f}|sed 's/<a /\n<a /g'|grep -i " href="|cut -d"=" -f2|cut -d'"' -f2); do
			status=$(curl -k --max-time 1 --head --silent --output /dev/null --write-out "%{http_code}" $e )
			test $status -eq 200 && >&2 printf ${e}" – "${status}" OK\n" || >&2 printf ${e}" – "${status}" FAILED\n"  
		done
	done
}
mirror_website () {
	dependencies="wget"
	assert_tools ${dependencies}
	local url=$@
	local domain=`expr "$url" : '^http[s]*://\([^/?]*\)'`
	wget \
		--no-check-certificate \
		--recursive \
		--no-clobber \
		--page-requisites \
		--html-extension \
		--convert-links \
		--restrict-file-names=windows \
		--domains $domain \
		--no-parent \
		$url 
}
newdataentry () {
	dependencies="cat grep head cut rev date printf"
	assert_tools ${dependencies}

	#~ datafn=$(assert_source $1) && { shift; } || { >&2 printf "file '${datafn}' not found."; err=1; exit 1; }
	datafn="data.xml"
	
	err=0
	customdir=$(pwd)
	fl=""

	>&2 printf "determining issue number based on '${datafn}'.\n" 
	lastentry=$(grep -i "<schleuder id=" ${datafn} | head -1 | cut -d'"' -f2)
	#~ lastentry=$(grep "<schleuder id=" ${datafn} | cut -d'"' -f2 | sort | tail -1)
	newentry=$(( $lastentry + 1 ))

	# todo: switches
	#~ for o in "$@"; do
		#~ on=$(printf $o|cut -d"=" -f1)
		#~ ov=$(printf $o|cut -d"=" -f2-)
		#~ case $on in
		# custom data directory for availability test and filesize, default pwd
			#~ c)
			#~ customdir) ;;
		# custom issue, default next after top entry in ${datafn}
			#~ i)
			#~ issue) ;;
		# custom timestamp, default: current time
			#~ d)
			#~ datetime) ;;
		# custom teaser xml-text, default: none
			#~ t)
			#~ teaser) ;;
		# custom fs for files
			#~ p)
			#~ pdffs) ;;
			#~ e)
			#~ epubfs) ;;
		# help text
			#~ h)
			#~ help)
			#~ *) ;;
		#~ esac
	#~ done
	test -d $1 && {
		customdir=$1; shift
		>&2 printf "custom base directory for files such as jpg, pdf, epub: "${customdir}".\n\n" 
	}
	test "$1" = "help" && {
		printf "prints a new data entry for a release of datenschleuder\n\n"
		printf "release [options]\n\n"
		printf "option\tdescription\n\n"
		printf "directory\tcustom data directory for availability test and filesize, default pwd\n\n"
		printf "number\tprint entry for release with custom number, default is the increment of the top entry in ${datafn}\n\n"
		printf "date\tuse a unix timestamp for a custom date\n\n"
		printf "teaser\tinclude teaser xml-message as last argument(s)\n\n"
		printf "full example: './gen.sh release 99 1568505600 Die Ausgabe 99 ist fertig.'\n"
		exit 1
	}
	2>/dev/null test $(( $1 * 1 )) -eq $1 && {
		newentry=$1; shift;
		>&2 printf "custom issue number '${newentry}' for '${datafn}'.\n" 
	} || {
		printf "next argument not a number. assuming issue based on top entry.\n"
	}
	2>/dev/null test $(( $1 * 1 )) -eq $1 && {
		>&2 printf "custom datetime.\n\n" 
		datetime="$1"; shift;
	} || { printf "next argument not a number. using actual system time.\n"; datetime=0; } # remaining arguments should be teaser xml-message.

	printf "\nschleuder id=\"$newentry\">\n"
	# todo: switch for dtg, integrate custom datetime input early
	#       or simply implement xslt for handling ISO and nothing else for input
	# RFC-5322-Format without seconds and TZ
	dtg="TZ=:UTC LANG=en_US.utf8 date -u -R |rev|cut -d\":\" -f2-|rev"; tz=" UT"
	# ISO dtg as issues may provide datetime-data on months but mostly not on days …
	#~ dtg="TZ=:UTC LANG=en_US.utf8 date +%Y-%m-%d %H:%M %Z"; tz=""
	test $datetime -eq 0 && {
		datetime=$(TZ=:UTC LANG=en_US.utf8 date -u -R |rev|cut -d":" -f2-|rev) # 
	} || {
		#~ dtg="TZ=:UTC LANG=en_US.utf8 date -d @${datetime} +\"%Y-%m-%d %H:%M %Z\""; tz=""
		dtg="TZ=:UTC LANG=en_US.utf8 date -d @${datetime} -u -R |rev|cut -d\":\" -f2-|rev"; tz=" UT"
		#~ datetime="$(TZ=:UTC LANG=en_US.utf8 date -d @${datetime} -u -R|rev|cut -d":" -f2-|rev)"
		# TZ=:UTC LANG=en_US.utf8 date -d @${datetime} +"%Y-%m-%d %H:%M %Z"
	}
	datetime="$(sh -c "${dtg}")"
	printf "\t<date>${datetime}${tz}</date>\n" # custom format universal time
	fn="ds"$(printf "%03d" $newentry)
	for d in covers pdfs epubs ; do
		case "$d" in
			covers) ffn=${d}"/"${fn}".jpg"; cover=$ffn ;;
			pdfs)   ffn=${d}"/"${fn}".pdf"  ;;
			epubs)  ffn=${d}"/"${fn}".epub" ;;
			#~ *) printf "file "${ffn}" not found, in "${d}".\n" ;;
		esac
		#~ printf "$ffn\n"
		test -e ${customdir}"/"${ffn} && {
			test "$d" = "covers" && {
				printf "\t<image>"${cover}"</image>\n"
			} || {
				fs=$(stat -c %s ${customdir}"/"${ffn})
				printf "\t<link filesize=\""${fs}"\">"${ffn}"</link>\n"
			}
		} || {
			mfl=${mfl}" "${ffn}"\n"
			#~ printf "file "${ffn}" not found.\n"
			err=3
		}
	done
    #~ printf "\t<preface></preface>\n" # not used on server yet
	test $# -eq 0 && { teaser="" ; } || {
		teaser="$@";
		printf "\t<teaser>"${teaser}"</teaser>\n"
	}
	printf "</schleuder>\n"
	test "$teaser" = "" && { >&2 printf "\nmissing teaser message.\n"; }
	test -e ${customdir}"/"$cover || { >&2 printf "\nmissing cover file: "${cover}"\n"; }
	test $err -gt 0 && { >&2 printf "\nmissing files:\n"${mfl}"\n"; }
}
newinfoentry () {
	>&2 printf "not implemented yet.\n\nwill be similar to 'release'.\n"
}

echo $# $@

if test $# -lt 1 ; then
	printf "missing parameter.\n"
	exit 1
else
	# a mirror is useful for local testing with files
	case "$1" in
		build)
			f=$(assert_source $@) && { shift; } || { exit 1; }
			build "${f}"
			;;
		#~ sortxml)
			#~ f=$(assert_source $@) && { shift; } || { exit 1; }
			#~ sortxml "${f}"
			#~ ;;
		mirror)
			>&2 printf "downloading the website may take quite some time.\n"
			mirror_website http://ds.ccc.de/
			;;
		linkcheck)
			linkcheck http://ds.ccc.de/
			;;
		release)
			shift
			newdataentry $@
			;;
		info)
			shift
			newinfoentry $@
			;;
		*)
			printf "this is not implemented yet.\n"
			;;
	esac
fi

# generate small size version of pdf, might help e.g. for email attachments
#gs -sDEVICE=pdfwrite -dPDFSETTINGS=/ebook -q -o ds100-smaller.pdf ds100.pdf