#!/bin/bash
#    xsane2sandwich - pdf with ocr directly from xsane
#    Copyright (C) 2012, 2013 Heinrich Schwietering
# 
#    This program is free software: you can redistribute it and/or modify
#    it under the terms of the GNU General Public License as published by
#    the Free Software Foundation, either version 3 of the License, or
#    (at your option) any later version.
#
#    This program is distributed in the hope that it will be useful,
#    but WITHOUT ANY WARRANTY; without even the implied warranty of
#    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#    GNU General Public License for more details.
#
#    You should have received a copy of the GNU General Public License
#    along with this program.  If not, see <http://www.gnu.org/licenses/>.
#
##############################################################################
#
#                           xsane2sandwich 0.2
#
#                *** pdf with searchable text layer ***
#
#
##############################################################################
# 
# xsane2sandwich is a wrapper script to produce ocred pdf files with xsane.
# It produces pdf files with a hidden, searchable text layer. tesseract-ocr is
# used to produce a hOCR/html-file. convert is needed to produce a readable
# for hocr2pdf. 
# The scan is converted into a pdf file using hocr2pdf, a part of exact image. 
# Multipage pdf files are produced using pdftdk, a title value is added, too.
# 

TEMP_DIR=/tmp/    # folder for temporary files (all files)
ERRORLOG="xsane2sandwich.log"  # file where STDERR goes 

if [[ -z "$1"  ]]
  then
  echo "Usage: $0 [OPTIONS]
  You need to run the script as OCR-application for xsane!
  xsane2sandwich uses a .tif file created with convert, and produces 
  a 'split' pdf file. The scan result is ocred using tesseract and processed
  with hocr2pdf to add a hidden text layer. Multipage pdf files are possible, 
  using pdftdk.
  
  OPTIONS:
    -i <filename> - temporary XSANE file
    -o <filename> - name for the output
    -l <language> - defines the language used for recognition
    -a <-hocr2pdf-option> - additional hocr2pdf options, optional
 
  Progress- & error-messages will be stored in this logfile:
     $TEMP_DIR$ERRORLOG

  xsane2sandwich depends on
    - XSane, http://www.xsane.org/
    - convert, part of ImageMagick, http://www.imagemagick.org/script/index.php
    - tesseract-ocr 3.0x, https://github.com/tesseract-ocr
    - hocr2pfd, part of ExactImage, http://www.exactcode.de/site/open_source/exactimage/
    - pdftk, http://www.pdflabs.com/tools/pdftk-the-pdf-toolkit/

  Some coding was stolen from 'ocube'
  http://www.geocities.com/thierryguy/ocube.html

  This pdfsandwich adaption is based on xsane2tess  
  http://doc.ubuntu-fr.org/xsane2tess, 

  Hints always welcome! heinrich (dot) schwietering (at) gmx (dot) de
"
  exit
fi

# get options...
while getopts ":i:o:l:a:" OPTION
  do
  case $OPTION in 
    i)  # input filename (with path)
      FILE_PATH="$OPTARG"
    ;;
    o)  # output filename
      FILE_OUT_LONG="$OPTARG"
    ;;
    l)  # recognition language
      LANG="$OPTARG"
    ;;
    a)  # additional hocr2pdf option  
      HOCROPT="$OPTARG"
    ;;
  esac
done

# redirects STDERR to ERRORLOG
exec 2>>$TEMP_DIR$ERRORLOG

# strips extension from FILE_OUT_LONG
FILE_OUT="${FILE_OUT_LONG%.*}"

echo "~~~+++~~~~+++~~~" 1>&2
#produces a tif file for hocr2pdf
TIF_FILE="$TEMP_DIR""xsane2tessResult".tif
convert "$FILE_PATH" -compress none  "$TIF_FILE" 1>&2

#create pdfmark
echo "InfoKey: Title" > $FILE_OUT.pdfmark
echo "InfoValue:`basename $FILE_OUT`" >> $FILE_OUT.pdfmark
echo "InfoKey: Producer" >> $FILE_OUT.pdfmark
echo "InfoValue: xsane2sandwich" >> $FILE_OUT.pdfmark

# produce hocr-file file with tesseract
tesseract "$FILE_PATH" "$FILE_OUT" -l "$LANG" "hocr" 1>&2
echo "tesseract -l $LANG $ENCOPT hocr used" 1>&2


# combine hocr-result with image and produce a pdf file
hocr2pdf $HOCROPT -i "$TIF_FILE" -o "$FILE_OUT.new.pdf" < "$FILE_OUT.html"

# check if outputfile exists
if [[ -a "$FILE_OUT.pdf" ]]

     #add new page to existing file using pdftk
     then pdftk "$FILE_OUT.pdf" "$FILE_OUT.new.pdf" cat output "$FILE_OUT.2.pdf"
     mv "$FILE_OUT.2.pdf" "$FILE_OUT.pdf"
     echo "file added to $FILE_OUT.pdf" 1>&2
     rm "$FILE_OUT.new.pdf"

     #if not start new output file
     else mv "$FILE_OUT.new.pdf" "$FILE_OUT.pdf"
     echo "$FILE_OUT.pdf started" 1>&2

fi

# set pdf title
     pdftk $FILE_OUT.pdf update_info $FILE_OUT.pdfmark output $FILE_OUT-neu.pdf 1>&2
     mv "$FILE_OUT-neu.pdf" "$FILE_OUT.pdf"

# Remove temporary files
rm "$FILE_OUT.txt"
rm "$FILE_OUT.html"
rm "$TIF_FILE"
rm "$FILE_OUT.pdfmark"
echo "~~~+++~~~~+++~~~"$(date +%c) 1>&2

#!/bin/bash
#    xsane2OCRmyPDF - pdf with ocr directly from xsane
#    Copyright (C) 2014 Heinrich Schwietering
# 
#    This program is free software: you can redistribute it and/or modify
#    it under the terms of the GNU General Public License as published by
#    the Free Software Foundation, either version 3 of the License, or
#    (at your option) any later version.
#
#    This program is distributed in the hope that it will be useful,
#    but WITHOUT ANY WARRANTY; without even the implied warranty of
#    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#    GNU General Public License for more details.
#
#    You should have received a copy of the GNU General Public License
#    along with this program.  If not, see <http://www.gnu.org/licenses/>.
#
##############################################################################
#
#                           xsane2OCRmyPDF 0.3.1
#
#                *** pdf with searchable text layer ***
#
#
##############################################################################
# 
# xsane2OCRmyPDF is a wrapper script to produce ocred pdf files with xsane.
# It produces pdf files with a hidden, searchable text layer. tiff2pdf 
# (libtiff) is used to produce an initial pdf file from a tif file, multipage 
# pdf files are combined using pdftdk.
# OCRmyPDF, utilizing tesseract-ocr, is used to  produce a pdf file with a 
# text layer.
# 

TEMP_DIR=/tmp/    # folder for temporary files (all files)
ERRORLOG="xsane2OCRmyPDF.log"  # file where STDERR goes 

if [[ -z "$1"  ]]
  then
  echo "Usage: $0 [OPTIONS]
  You need to run the script as OCR-application for xsane!
  xsane2OCRmyPDF uses a pdf file created with convert and tiff2pdf. The result is
  processed with OCRmyPDF to add a hidden text layer. Multipage pdf files are possible, 
  using pdftdk.
  
  OPTIONS:
    -i <filename> - temporary XSANE file
    -o <filename> - name for the output
    -l <language> - defines the language used for recognition
    -a <options> - additional OCRmyPDF option e.g. to use convert or unpaper, optional
 
  Progress- & error-messages will be stored in this logfile:
     $TEMP_DIR$ERRORLOG

  xsane2OCRmyPDF depends on
    - XSane, http://www.xsane.org/
    - convert, part of ImageMagick, http://www.imagemagick.org/script/index.php
    - tiff2pdf, part of libtiff, http://www.remotesensing.org/libtiff/
    - OCRmyPDF, https://https://github.com/jbarlow83/OCRmyPDF
    - unpaper, https://www.flameeyes.eu/projects/unpaper
    - tesseract-ocr 3.0x, https://github.com/tesseract-ocr
    - pdftk, http://www.pdflabs.com/tools/pdftk-the-pdf-toolkit/
    - yad, http://sourceforge.net/projects/yad-dialog/

  Some coding was stolen from 'ocube'
  http://www.geocities.com/thierryguy/ocube.html

  This OCRmyPDF adaption is based on xsane2tess  
  http://doc.ubuntu-fr.org/xsane2tess, 

  Hints always welcome! heinrich (dot) schwietering (at) gmx (dot) de
"
  exit
fi

# get options...
while getopts ":i:o:l:a:" OPTION
  do
  case $OPTION in 
    i)  # input filename (with path)
      FILE_PATH="$OPTARG"
    ;;
    o)  # output filename
      FILE_OUT_LONG="$OPTARG"
    ;;
    l)  # recognition language
      LAN="$OPTARG"
    ;;
    a)  # additional OCRmyPDF option  
      OCROPT="$OPTARG"
    ;;
  esac
done

# redirects STDERR to ERRORLOG
exec 2>>$TEMP_DIR$ERRORLOG

# strips extension from FILE_OUT_LONG
FILE_OUT="${FILE_OUT_LONG%.*}"

echo "~~~+++~~~~+++~~~" 1>&2

#produces a pdf file for OCRmyPDF
convert "$FILE_PATH" "$FILE_OUT.1.tif" 1>&2
tiff2pdf -o "$FILE_OUT.1.pdf" "$FILE_OUT.1.tif"  1>&2
# check if output file exists
    { if [[ -a "$FILE_OUT.pdf" ]]

     #add new page to existing file using pdftk
     then pdftk "$FILE_OUT.pdf" "$FILE_OUT.1.pdf" cat output "$FILE_OUT.2.pdf"
     mv "$FILE_OUT.2.pdf" "$FILE_OUT.pdf"
     echo "file added to $FILE_OUT.pdf" 1>&2

     #if not start new output file
     else cp "$FILE_OUT.1.pdf" "$FILE_OUT.pdf"
     echo "$FILE_OUT.pdf started" 1>&2
     fi }

# Asks for further scans to be added
$(yad --geometry -0-0 --title="xsane2OCRmyPDF" --button=Ja:2 --button=Nein:4 --text="Sollen weitere Seiten gescannt werden?")
ret=$?
{ if [[ $ret -eq 4 ]] ; then
# produce a pdf file with OCRmyPDF
     ocrmypdf -l $LAN $OCROPT "$FILE_OUT.pdf" "$FILE_OUT.new.pdf"
     mv "$FILE_OUT.new.pdf" "$FILE_OUT.pdf"
     echo "$FILE_OUT.pdf ocred using OCRmyPDF" 1>&2
fi }


# Remove temporary files
rm "$FILE_OUT.1.pdf"
rm "$FILE_OUT.1.tif"
rm "$FILE_OUT"
echo "~~~+++~~~~+++~~~"$(date +%c) 1>&2