The Molecular Universe

Making a Book from HTML Pages

I spent a little time recently in figuring out how to compile various pages into a pdf book. The key to doing this, I found, is to use 'htmldoc' - which can be downloaded for various platforms easily.

Here is a summary of the scripts that I used.

make_pdf.sh controls the overall processing, it uses a list of files stored in 'chapters.txt' to determine the files to process.

#!/bin/sh

rm -f missing_files
rm -f image_files 

i=0
while read CHAPTER
do
  i=`expr $i + 1`
  TMP=`echo $i | awk '{printf "tmp%04d.html", $1}'`
  echo $TMP 
  ./txttohtml.sh $CHAPTER | sed -f greek.sed > $TMP
  if [ $i -eq "500" ]
  then
    break
  fi
done < chapters.txt

htmldoc --book --linkstyle plain --toctitle "The Molecular Universe"  -f output.pdf --no-title --headfootfont times --headfootsize 10 --charset iso-8859-7 --embedfonts --size letter tmp*.html --titlefile title.html

./make_index.sh output.pdf 12

# make a report on the current status to append to the book
WORDS=`wc output.txt | awk '{print $2}'`
MISSING=`wc missing_files | awk '{print $1}'`
IMAGES=`wc image_files | awk '{print $1}'`
echo "<pre>" > status.html
echo "PDF file created:" >> status.html
date >> status.html
echo "" >> status.html
echo "Current word total: " $WORDS >> status.html
echo "" >> status.html
echo "Number of image files that should be enlarged: " $MISSING >> status.html
echo "" >> status.html
echo "Missing file names follow: " >> status.html
cat missing_files >> status.html
echo "" >> status.html
echo "Current image file count: " $IMAGES >> status.html
echo "All image file names follow: " >> status.html
cat image_files >> status.html
echo "</pre>" >> status.html

htmldoc --webpage -f status.pdf --no-title --size letter status.html

pdftk A=output.pdf B=output.index.pdf C=status.pdf output output.pdf

rm -f output.pdf output.txt output.index.pdf output.data.txt status.html status.pdf

if [ -f missing_files ]
then
  echo "THERE ARE " `wc missing_files | awk '{print $1}'` " MISSING FILES"
  cat missing_files
fi

txttohtml.sh is a very crude script that converts the raw nanoblogger txt file into a crude html file which can be used as the input to the htmldoc processor.

#!/bin/sh

awk '{
  if(match($0,"TITLE:")){
    title=substr($0,7)
    print "<h1> " title " </h1>"
  }
  if(match($0,"<blockquote>")){
    print "<table border=\"1\" cellpadding=\"10\"><tr><td>"
    getline
    sub("Note:","<b>Note:</b>");
    print $0
    next
  }
  if(match($0,"</blockquote>")){
    print "</td></tr></table>"
    next
  }
  if(match($0,"BODY:")){
    intext=1
    next
  }
  if(!intext)next
  if(match($0,"END-----")){
    intext=0
    exit
  }
  if(match($0,"<table class=\"image")){
    imagetable=1 
  }
  if(match($0,"\"left\"") && imagetable){
    sub("\"left\"","\"center\"");
  }
  if(match($0,"\"right\"") && imagetable){
    sub("\"right\"","\"center\"");
  }
  if(match($0,"<img src=")){
    record=$0
    sub("^.*<img src=","",record);
    sub(" .*$","",record);
    sub("^.*/","",record);
    sub("\"","",record);
    sub("_scale","",record);
    largefile=record
    largefile="../../../images/" largefile
    located=0
    # look for file with original extension
    line=""
    getline line < largefile
    close largefile
    if(length(line)>0){
      largefile="\""largefile"\""
      sub("\".*\"",largefile)
      located=1
    }
    # look for file with .png extension
    line=""
    sub(".gif",".png",largefile)
    getline line < largefile
    close largefile
    if(length(line)>0){
      largefile="\""largefile"\""
      sub("\".*\"",largefile)
      located=1
    }
    if(!located){
      print "COULD NOT LOCATE " record > "/dev/tty"
      print record >> "missing_files"
    }
    print largefile >> "image_files"
  }
  if(match($0,"</table") && imagetable){
    imagetable=0
    print $0
    print "<table width=\"100%\" summary=\"\"><tr><td> </td></tr></table>"
    next
  }
  print $0
}' $1

make_index.sh comes from pdftk and uses two tools from the pdftk site to create a crude book-like index for the the book. I modified the make_index.sh script a little - so I am including it here.

#!/bin/sh
# make_index.sh, version 1.0
# usage: make_index.sh <PDF filename> <page window>
# requires: pdftk, kw_catcher, page_refs,
#           pdftotext, enscript, ps2pdf
#
# by Ross Presser, Imtek.com
# adapted by Sid Steward
# http://www.pdfhacks.com/kw_index/

# modified somewhat from the original distributed version to correct
# problems encountered in initial testing

export PATH=/opt/local/bin:/opt/local/sbin:$PATH
LANG=C

fname=`basename $1 .pdf`
pdftk ${fname}.pdf dump_data output ${fname}.data.txt && \
sed 's/LowercaseRomanNumerals/DecimalArabicNumerals/' ${fname}.data.txt > j && \
mv j ${fname}.data.txt && \
pdftotext ${fname}.pdf ${fname}.txt && \
  page_refs ${fname}.txt index-terms.dat ${fname}.data.txt \
| sed 's/PageLabelNumStyle://g' \
| enscript --columns 2 --font 'Times-Roman@10' \
  --header '|Index' --header-font 'ArialBold@20' \
  --margins 54:54:36:54 --word-wrap --output - \
| ps2pdf - ${fname}.index.pdf