#!/usr/bin/perl 
use File::Copy;
use Time::Local;

# moreTranscriptOCR

# jody DeRidder, 6/9/10
## Copyright (c) 2010, The University of Alabama Libraries. 
## Contributed by Jody DeRidder, 6/10/10.
## All rights reserved.

## Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

##    * Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
##    * Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in 
##       the documentation and/or other materials provided with the distribution.
##    * Neither the name of The University of Alabama Libraries nor the names of its contributors may be used to endorse or promote products 
##       derived from this software without specific prior written permission.

##THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, 
##THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR 
##CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, 
##PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 
##LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE,
##EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.


# should ocr not exist for existing transcripts in the
# archive, acumen, or the deposits directory
# this script will provide the option to create OCR for all of 
# the transcripts for any collection indicated, or 
# for only the content provided in an ocrList.txt file 
#  -- which it will look for in the completed directory on the share drive
#  ocr files will be deposited directly into Acumen, and will NOT be 
# made for anything besides transcript images, and will NOT 
# be made if an existing ocr or txt file exists in acumen in a transcripts
# directory, for the image.


&timestamp;
# location hardcoded here:
$inbase = "/cifs-mount/Digital_Coll_Complete/";
$acumen = "/srv/www/htdocs/content/";
$archive = "/srv/archive/";
$depositArea = "/srv/deposits/content/";


print "\n\nDo you want to create OCR for all the transcripts in a particular collection?\n\n";
$ans = <STDIN>;
chomp ($ans);
if ($ans =~ /Y/i){
GETNUM:     print "\n\nPlease enter the collection number:\n\n";
     $collnum = <STDIN>;
     chomp $collnum;
     print "\n\nIs this the correct number?  -->$collnum<-- \n\n";
     $ans = <STDIN>;
     chomp ($ans);
     if ($ans =~ /n/i){
       goto GETNUM;
       }
     }
else{
  print "\n\nI\'m Looking for an ocrList.txt file in the Completed folder.\n  Please help me locate it.\n\n";


  opendir(BASE, $inbase) or die "can't open $inbase\n";
  while ($file = readdir(BASE)){
  #  print "looking at $file\n";
    if ($file =~ /^\./){next;}  # skip dot files
    $path = $inbase.$file;
    if (-d $path){
      push (@dirs, $file);
      }
    }
  close(BASE);
 
  print "\n\nI'm looking through $inbase. \n Here are the directories there.  \n\n";

  $dircount = scalar @dirs;
  for ($i = 1; $i <= $dircount; $i ++){
    print $i.")  ".$dirs[$i-1]."\n";
    }

  print "\nWhich directory do you want? \n Type the number and press enter:\n\n";

  $num = <STDIN>;
  chop $num;
  $num --;
  $inbase = $inbase.$dirs[$num]."/";
  push (@findFile, $inbase);

  undef @dirs;
  foreach $dir (@findFile){
    print "\nI'm looking through $dir; \nis this the exact directory with the ocrList.txt?\n  Y or N: \n\n";
    $ans = <STDIN>;
    chomp ($ans);
    if ($ans =~ /Y/i){
      $ocrlist = $dir."ocrList.txt";
      last;
      }
    else{
      print "\n\nOkay, then, I'll collect the subdirectories, and we'll look through those\n";
      opendir(DIR, $dir) or die "can't look through $dir\n";
      while ($file = readdir(DIR)){
        if ($file =~ /^\./){next;}  # skip dot files
        $path = $dir.$file;
        if (-d $path){
          push (@findFile, $path."/");
          }
        }
     close(DIR);
     }
   } 
 
  if (! -e $ocrlist){
    print "ERROR!  I did not find $ocrlist.  Please start over.\n\n";
    exit;
    }
  else{
TRYAGAIN:
    open (OCR, $ocrlist) or die "can't open $ocrlist\n";
    while ($line = <OCR>){
      if ($line =~ /(([a-z]{1}\d{4}\_\d{7})\_\d{7}.*?)\t *1/){
       #  print "Found line $line\n";
         push (@ocrMe, $1);
         $coll = $2;
         undef $found;
         foreach $c (@collList){
            if ($c eq $coll){ $found = 1;}
            }
         if (!$found){ push (@collList, $coll);}  # compile list of all collections we are working on
         }
      elsif (!($line =~ /[a-z]{1}\d{4}\_\d{7}\_\d{7}/)){
         if ($line =~ /\w/){  # there's something there, but I don't know what
           print OUT "NO OCR performed for $line in ocrList.txt file \n";
           print "NO OCR performed for $line in ocrList.txt file \n";
           }
         }
      }
   close(OCR);
   if (! @ocrMe){
      print "ERROR!  I could not locate any items in ocrList.txt to OCR\n";
      print "Please check file names in the list, and encoding of the file\n";
      print "ANSI text file is best.  I can't read UTF-16.\n\n";
      print "When you have replaced the file, please hit enter\n";
      $ans = <STDIN>;
      goto TRYAGAIN;
      }
    }
  }

# at this point we either have a list in @ocrMe or we have a collection number to OCR


# output file will be based on collection number and timestamp
$output = "/home/jeremiah/AUDIO/output/TranscriptOCRmaker_".$timestamp.".txt";
$out = "AUDIO/output/TranscriptOCRmaker_".$timestamp.".txt";
open (OUT, ">".$output) or die "can't open $output\n";

# call another process to make the ocr so it won't stop when the ssh window closes.
print "\n\nWe are starting to make the OCR now\n";

unless (fork){
   # the child process does this stuff

# if we have to OCR content already uploaded but not in the archive, we'd best find those files
    opendir(DIR, $depositArea) or die "can't look in $depositArea\n";
    while ($file = readdir(DIR)){
       if ($file =~ /^\./){ next; } #skip dot files
       $path = $dir.$file;
       if (-d $path){
         if ($collnum && $file =~ /$collnum/){   # top level directory should contain collection number
            push (@hunting, $path."/");
            }
         else{   # check if it matches any of the collections in our ocrList
            undef $found;
            foreach $c (@collList){
              if ($file =~ /$c/){
                 $found = 1;
                 }
               }
            if ($found){push (@hunting, $path."/");}
            }
         }
       }
     close(DIR);
     foreach $dir (@hunting){
        opendir(DIR, $dir) or die "can't look in $dir\n";
        while ($file = readdir(DIR)){
                 # this hunts for transcripts only
          if ($file =~ /^\./ || $file =~ /Metadata/ || $file =~ /Admin/ || $file =~ /Scans/i){ next; } #skip dot files
          $path = $dir.$file;
          if (-d $path){
             push (@hunting, $path."/");
             }
          elsif ($file =~ /(.*?)\/.tif/){
             $which = $1;
             $deposits{$which} = $path;  # keep it for reference later
             }
          }
        close(DIR);
       }           
     @inDeposits = keys (%deposits); 


# we don't want to create OCR for anything not online.

      if (@ocrMe){   # use the filenumbers to find the files.
        foreach $val (@ocrMe){
            ($mydirs = $val) =~ s,\_,\/,g;  # substitute slashes for underscores in filename
            $mybase = $acumen.$mydirs."/";
            if (-e $mybase){
              push (@dothese, $mybase);
              }
            }
          }
      else{
         ($mydirs = $collnum) =~ s,\_,\/,g;  # substitute slashes for underscores in filename
          $mybase = $acumen.$mydirs."/";
          if (-e $mybase){
              push (@dothese, $mybase);
              }
          }


      foreach $val (@dothese){
          opendir(ITEM, $val) or die "can't open $val\n";
          while ($file = readdir(ITEM)){
             if ($file =~ /^\./ || $file =~ /Metadata/){ next;}
             $path = $val.$file;
             if (-d $path){ push (@dothese, $path."/"); next;}
             elsif ($file =~ /^(.*?)\_2048\.jpg/){
               $thisnum = $1;

          #     print "found $file -- need an ocr for $thisnum\n";
               # are we in a transcripts directory already?
               if ($val =~ /Transcripts/){
                  # if so, then the text would be in this same directory
                  $anocr = $val.$thisnum.".ocr.txt";
                  $atext = $val.$thisnum.".txt"; 
                  $ocrOut = $val.$thisnum.".ocr";
                  undef $newtrans;
                  }
               else {  # text would be in a Transcripts subdirectory
 
                  $anocr = $val."/Transcripts/".$thisnum.".ocr.txt";
                  $atext = $val."/Transcripts/".$thisnum.".txt"; 
                  $ocrOut = $val."/Transcripts/".$thisnum.".ocr";
                  $newtrans = $val."/Transcripts";
                  }

          #     print "anocr:  $anocr\natext:  $atext\nocrOut: $ocrOut\n newtrans:  $newtrans\n";
 
               if ( (-e $anocr) || (-e $atext)){ next;}  # no need to ocr this
               else{
                  
               # we need  to find the tiff.
                 
                  ($mydirs = $thisnum) =~ s,\_,\/,g;  # substitute slashes for underscores in filename

          # check for transcript tiff in archive first
          # if it exists, use that -- don't use the other for this script.  
                  $tiff = $archive.$mydirs."/Transcripts/".$thisnum.".tif";

               #   print "the tiff we need is $tiff\n";
                  if ( ! -e $tiff){ # let's look in the deposits directory
                     foreach $d (@inDeposits){
                       if ($d eq $thisnum){
                         $tiff = $deposits{$d};
                         }
                       }
                    }
 # try again
                 if ( ! -e $tiff){
                     print OUT "ERROR:  Can't create OCR for $thisnum; $tiff does not exist in a Transcripts directory\n";}
                  else{
                     if ($newtrans){  # need a transcripts directory
                       if (! -e $newtrans){ `mkdir -m 0755 $newtrans`;}   # create a directory 
                       }
                     $check = `tesseract $tiff $ocrOut`;
                     if ($check){ print OUT "ERROR $check creating ocr for $filenum\n";}
                     }
                 }
             }   # ignore thumbnails and everything else
           }
         close(ITEM);
         }
       
  print OUT "OCR creation is complete.  Please check Acumen.\n";
  print OUT "\n\nGREAT WORK!!!  :-)\n\n";
  close(OUT);
  exit;
  } 


# parent process gives a little more info, then exits
print "Please check back in a few hours\n";
print "Any problems found will be in the \n$out file.\n";
print "\n\nThank you!!\n\n";
exit;


sub timestamp{
#following for Windows
#print "hit enter twice please\n";
#$date = `date`;
#$time = `time`;

#print $date."\n";
#if ($date =~ /.*? (\d*)\/(\d*)\/(\d*)/){
 #    $date = $3.$1.$2;
  #    }
#print $date."\n";
#if ($time =~ /.*? (\d*)\:(\d*)\:(\d*)\./){
 #    $time = $1.$2.$3;
  #   }
#print $time."\n";
#$timestamp = $date."T".$time;
#print $timestamp."\n";

# following for unix
($sec, $min, $hour, $mday, $mon, $year, $wday, $yday, $isdst) = gmtime();
##$year += 1900;
$mon ++;
if ($mon < 10){ $mon="0".$mon;}  #need 2 digits
if ($sec < 10){ $sec="0".$sec;}
if ($min < 10){ $min="0".$min;}
if ($hour < 10){ $hour="0".$hour;}
if ($mday < 10){ $mday="0".$mday;}

$year = $year + 1900;
$timestamp= "$year-$mon-$mday\T$hour:$min:$sec\Z";
}

sub by_number {$a <=> $b;}