jw-pkg/scripts/convert-doc-to-wiki.sh

#!/bin/bash

cat_header()
{
	cat <<-EOT
	====== Dummy-Titel ======
	----

	===== Links =====
	----

	  * {{$namespace:$n, Stand `date +'%d.%m.%Y'`}}

	===== Backlinks =====
	----
	{{backlinks>.}}

	===== Annotierte Version =====
	----
	EOT
}

strip()
{
	sed "s/^ *//; s/ *$//" | grep .
}

field()
{
	grep "| *$2 *\(|\|$\)" $files | grep "pdfwiki\|binary" | cut -d\| -f$1 | strip
}

cmd_name_extern_2_intern()
{
	field 3 "$1"
}

cmd_name_intern_2_extern()
{
	field 2 "$1"
}

cmd_uploads()
{
	cat $files | grep "pdfwiki\|binary" | cut -d\| -f3 | strip
}

cmd_milk_dirs()
{
	local dirs="$@"
	for d in $dirs; do
		ls $d/* 2>/dev/null | while read f; do
			l=`basename "$f"`
			n=`cmd_name_extern_2_intern "$l"`
			[ "$n" ] || {
				echo "ignoring unknown file \"$l\""
				continue
			}
			[ -e "$n" ] && {
				echo "ignoring existing file \"$l\""
				continue
			}
			type=$(field 1 "$l")
			echo type=$type
			case $type in
			pdfwiki)
				t=`echo $n | sed 's/\.[^.]*$//'`
				txt="$t.txt"
				[ -e "$txt" ] && {
					echo "ignoring existing file \"$txt\""
					continue
				}
				cp "$f" "tmp-$n"
				pdftotext "tmp-$n"
				pdfimages -png -j tmp-$n $n
				rm -f $n-*.pnm
				cat_header "$l" "$n" > $txt.tmp
				cat tmp-$txt >> $txt.tmp
				mv $txt.tmp $txt
				mv tmp-$n $n
				rm -f tmp-$txt
				;;
			binary)
				cp "$f" "$n"
				;;
			*)
				echo "ignoring file with unknown extension: \"$f\""
				;;
			esac
		done
	done
}

# ------------------------------------------------ here we go

files=files.txt
namespace=files:doc:extern
eval set -- `getopt -o hn: -l namespace: -- "$@"`
while [ "$1" != -- ]; do
case $1 in
	-n|--namespace)
		namespace="$2"
		shift
		;;
	-h)
		usage 0;;
	*)
		usage 1;;
esac
shift
done
shift

cmd="${1/-/_}"
shift
eval cmd_$cmd "$@"