#!/bin/sh
# html2page
# scottm, 2009/07/30
# Extracts data from .html files to .page files for webgen.
# This allows OpenOffice.org presentations to be converted to a webgen site.
# Contents of .page files could also be used for a content management site.
# run from cygwin, or linux
# .page files are overwritten
# Read .html files that OpenOffice.org Impress generated for each slide
#for file in `ls img*.html text*.html|awk -F\. '{print $1}'`
echo "Generating .page files..."
for file in `ls *.html|sed -e 's/.html//'`
do
#generate title section from slide title
export title=`grep "" ${file}.html | sed -e 's#<title>##g'| sed -e 's###g'`
echo "---" > ${file}.page
echo "title: ${title}" >> ${file}.page
echo "---" >> ${file}.page
#strip out extra html lines from top an bottom
cat ${file}.html |grep -v DOCTYPE |grep -v "html>" | grep -v "transitional.dtd"| grep -v "" |grep -v ""| grep -v ""|grep -v ">${file}.page
done
echo "cleaning .page files..."
# get rid of the closing body tag (may not be on a line by itself)
sed -e 's###' -i *.page
echo "adding tags..."
# add some line breaks to the img version to make the menu look better
sed -e 's#\#br/>