downloads a djvu file, splits it into pages (djvmcvt), parses out ocr'd text (djvused) and also creates an XML of the ocr'd text (djvutoxml)

oicr

#!/bin/bash

[[ -n "${1}" ]] || { echo "Usage: oicr.sh IA_BOOK_TITLE"; exit 0 ; }

# sample record ids: catalogueoflepid02briti electronicnaviga00unit halfhoursinfarno00newy nachrichtsblattd3234190012deut

BOOK=${1}
#BASEURL=http://cluster.biodiversitylibrary.org
BASEURL=http://www.archive.org/download

mkdir -p  ${BOOK}/txt ${BOOK}/xml ${BOOK}/djvu ${BOOK}/src

if [ -f ${BOOK}.djvu ]; then
	echo "${BOOK}.djvu found, continuing without download..."
else
	wget --user-agent="Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.2; Trident/4.0; SLCC2; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0)" --tries=20 --span-hosts --recursive --level=1 --continue --no-parent --no-host-directories --reject html --cut-dirs=2 --execute robots=off ${BASEURL}/${BOOK}/${BOOK}.djvu
fi

#mv $1.djvu $BOOK/src
cp ${1}.djvu ${BOOK}/src
djvmcvt -i ${BOOK}/src/${BOOK}.djvu ${BOOK}/djvu index.djvu

ls -1 ${BOOK}/djvu/*.djvu | grep -v index.djvu | cut -d'_' -f2 | cut -d'.' -f1 | while read PAGE_NUM

do
	echo -n "writing ${BOOK} page ${PAGE_NUM}"; echo -n "		txt..." 
	djvused ${BOOK}/djvu/${BOOK}_${PAGE_NUM}.djvu -e 'output-all' > ${BOOK}/txt/${BOOK}_${PAGE_NUM}.txt
	echo -n "done"; echo -n "		xml..." 
	djvutoxml ${BOOK}/djvu/${BOOK}_${PAGE_NUM}.djvu > ${BOOK}/xml/${BOOK}_${PAGE_NUM}.xml
	echo "done"
done

echo "done"
exit 0