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

# makeAudioJpegs

# jody DeRidder, 5/3/10
## 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.


# expects share drive content under /cifs-mount/
# requests collection directory on share drive
# locates all Scans directories and Transcript directories
# creates jpegs and dumps them into /home/jeremiah/UploadArea/jpegs
# for testing and distribution into Acumen.

# will make another version of this to pull content from share
# into deposits directory, verify the copies, and then delete on share.

&timestamp;
# location hardcoded here:
$inbase = "/cifs-mount/Digital_Coll_Complete/";
$outbase = "/home/jeremiah/AUDIO/";

# testing
#$outbase = "/home/jlderidder/UploadArea/";



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]."/";

undef $here;
while (!$here){
  print "\nI'm looking through $inbase; \nis this the collection directory?\n  Y or N: \n\n";
  $ans = <STDIN>;
  chomp ($ans);
  if ($ans =~ /Y/i){
     $here = 1;
     last;
     }
  else{
     print "\nShould I go back up a directory level?\n  Y or N: \n\n";
     $ans = <STDIN>;
     chomp ($ans);
     if ($ans =~ /Y/i){
       $inbase =~ s,^(.*\/).*?\/$,\1,;
       next;
       }
   else{
      print "\nPlease choose one of the following directories: \n\n";
      undef @dirs;
      opendir(BASE, $inbase) or die "can't open $inbase\n";
      while ($file = readdir(BASE)){
        if ($file =~ /^\./){next;}  # skip dot files
        $path = $inbase.$file;
        if (-d $path){
           push (@dirs, $file);
           }
         }
       close(BASE);
       $dircount = scalar @dirs;
       for ($i = 1; $i <= $dircount; $i ++){
         print $i.")  ".$dirs[$i-1]."\n";
         }

       print "\n\n Type the number and press enter:\n";

       $num = <STDIN>;
       chop $num;
       $num --;
       $inbase = $inbase.$dirs[$num]."/";
       }
    }
 }
  

# now we are in the correct collection directory
# check for the collection number first

undef $gotit;
if ($inbase =~ /([a-z]{1}[\d]{4}\_[\d]{7})/){
    $collnum = $1;
    }
else{ 
   print "\nPlease enter the collection number\n";
   $collnum = <STDIN>;
   chomp ($collnum);
   }

while (!$gotit){ 
   print "\nIs $collnum  the correct collection number? Y or N:\n";
   $ans = <STDIN>;
   chomp ($ans);
   if ($ans =~ /N/i){
      print "\nPlease enter the collection number\n";
      $collnum = <STDIN>;
      chomp ($collnum);
      }
   else{
      $gotit = 1;
      last;
      }
    }

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

 
# first, let's do some QC on filenames.  A sanity check.
# not checking numbering -- sequences -- at item level.
# not checking Admin or Metadata directories.

print "\n\nShould we check item sequence? Y or N\n";
print "Answer Y only if it starts with item 1 and \nshould be consecutively numbered.\n";
$ans = <STDIN>;
chomp ($ans);
if ($ans =~ /Y/i){
  $ItemNumCheck = 1;
  }


print "\n\nPlease wait while we do a little quality control checking ...\n\n";

# let's find the Scans next.
opendir(BASE, $inbase) or die "can't open $inbase\n";
while ($file = readdir(BASE)){
   $path = $inbase.$file."/";
   if ($file =~ /^\./){ next;}
   elsif ($file =~ /Transcripts/){
     push (@transdirs, $path);
     }
   elsif ($file =~ /Scans/i){
     push (@scansdirs, $path);  # use these to check MODS, not to make jpegs
     }
   }
close(BASE);

# now we have all scans directories in @scansdirs and all transcript directories in @transdirs
# to avoid confusion in the later scripts, we'll duplicate them for this test.
@testScans = @scansdirs;
@testTrans = @transdirs;

$coll = $collnum;

# first layer in each is assumed to be item level
foreach $dir (@testScans){
   $count = 1;    # checking sequence count of files in directories
   opendir (DIR, $dir) or die "can't read files in $dir\n";
    while ($adir = readdir(DIR)){
       if ($adir =~ /^\./ || $adir =~ /Thumbs\.db/i){next;}  #skip dot files and thumbs.db
       ($short = $dir) =~ s,\/cifs\-mount\/Digital\_Coll\_Complete\/,,;
       ($parent = $short) =~ s,.*\/,,;   # parent directory is what follows the last slash, yes?
                                                              # we want to match for this
       $thispath = $dir."/".$adir;
       if ($adir =~ /$coll/i){  # passes first test, it matches collname
          if ($adir =~ /.*\_(\d{7})(\..{4})?$/) {  # check for bad numbering  
              $mynum = $1 + 0;  # get rid of leading zeros
            #  print "I think that the number for $adir is $mynum\n";
               push (@itemnums, $mynum);
               }

         elsif ($adir =~ /.*\_(\d{4})(\..{3})?$/ ) {  # check for bad numbering  
              $mynum = $1 + 0;  # get rid of leading zeros
            #  print "$parent is parent: I think that the number for $adir is $mynum\n";
              push (@{$itempages{$parent}}, $mynum);
              }

         elsif ($ItemNumCheck && $adir =~ /.*\_(\d{4})(\..{3})?$/ ) {  # check for bad numbering  
             $mynum = $1 + 0;  # get rid of leading zeros
           #  print "$thispath: I think that the number for $adir is $mynum\n";
               if ($mynum != $count){
                   $diff = $mynum - $count;
                   if ($diff < 0){ print "Error:  diff:  $diff;  mynum $mynum;  count $count;  $thispath $adir\n";}
                   $lastmissing = $mynum -1;
                   if ($lastmissing != $count){
                      push (@badcount,  "$short $adir --> $diff missing files:  $count through $lastmissing");
                      }
                   else{ push (@badcount,  "$short $adir --> $count is missing");}
                   }
               $count  = $mynum + 1;
             }


         if ( -d $thispath){
             if (!( $adir =~ /^[a-z]{1}\d{4}\_\d{7}\_\d{7}(\_\d{4}(\_\d{3})?)?$/)){
                  push (@badform, $adir);
                  }
            push (@testScans, $thispath); # collect subdirectories for further investigation    
            }
         else{  # not a directory;  must be a file. Does it match its parent directory?

            if ((!($parent =~ /Scans/i))  && (!($adir =~  /$parent/))){
                push (@wrongdir, $adir."   parent ->$parent<- in ".$short);
                }
            if (!( $adir =~ /^[a-z]{1}\d{4}\_\d{7}\_\d{7}(\_\d{4}(\_\d{3})?)?\.\w{3}$/)){
                push (@badform, $adir);
                }
        # pull off the last set of numbers before the extension   
            elsif ($parent =~ /Scans/i  || (!$parent)){ next;}  # don't check non-compound objects for sequence
            }
          }
       else { push (@wrongdir, $adir."   ".$short);}

       }
     close(DIR);
     }


     $count = 1;
     @sorteditems = sort by_number(@itemnums);
     if ($ItemNumCheck){
       foreach $mynum (@sorteditems){

             if ($mynum != $count){
                   $diff = $mynum - $count;
                   if ($diff < 0){ print "error:  $mynum item, expected count is $count\n";}
                   $lastmissing = $mynum -1;
                   if ($lastmissing != $count){
                      push (@badcount,  "item level --> $diff missing files:  $count through $lastmissing");
                      }
                   else{ push (@badcount,  "item level --> $count is missing");}
                   }
             $count  = $mynum + 1;
             }
          }
      @itemmoms = sort by_number(keys (%itempages));
      foreach $item (@itemmoms){
         $count = 1;
         @pages = @{$itempages{$item}};
        # $numpages = scalar (@pages);
        # print "$item has $numpages before and";
        # @sorted = sort by_number(@pages);
        # $numpages = scalar (@sorted);
        # print "$numpages after\n";
         foreach $mynum (@sorted){
            if (! $mynum){ print "no page number for something in $item\n";}
            if ($mynum != $count){
                   $diff = $mynum - $count;
                   if ($diff < 0){ print "error:  $mynum page in $item item, expected count is $count\n";}
                   $lastmissing = $mynum -1;
                   if ($lastmissing != $count){
                      push (@badcount,  "pages of $item --> $diff missing files:  $count through $lastmissing");
                      }
                   else{ push (@badcount,  "pages of $item --> $count is missing");}
                   }
             $count  = $mynum + 1;
             }
          undef @{$itempages{$item}};
          }

    undef @itemnums;
    undef @itempages;


   if (@badform || @badcount || @wrongdir || @missing ){
       print print "TROUBLE in QC:  please check the $out file\nUnable to continue\n\n";
       print OUT "\nTROUBLE: $topdir \n";
       print OUT "-------------------------------------------\n";
     if (@missing){
        print OUT "\nThe following files are missing or badly named \n";
        foreach (@missing){ print OUT "    $_\n";}
        }

     if (@wrongdir){
        print OUT "\nThe following files or directories do NOT reflect the name of their parent directory\n";
        print OUT "Are they in the right place?  Please check:\n";
        foreach (@wrongdir){ print OUT "    $_\n";}
        }

     if (@badform){
        print OUT "\nThe following filenames or directories are not in the correct format:\n";
        foreach (@badform){ print OUT "    $_\n";}
        }

      if (@badcount){
        print OUT "\nSome files seem to be missing, according to sequence gaps\n";
        foreach (@badcount){ print OUT "    $_\n";}
    
       }
     print "\n\nAny problems found have been output to the command line \nand also will be found in the \n$out file\n";
     exit;
     }

 
   print "\n\nQuality control checking completed;  no problems found so far.\n\n";

   

   # then look through directories
   # need to make sure there's a MODS for each item...?

   # expect item at first level within scans folders
   # expect MODS in MODS directory within Metadata

undef $doMods;
print "\n\nAre you uploading MODS?  Y or N:\n";
$ans = <STDIN>;
chomp ($ans);
if ($ans =~ /Y/i){
   $doMods = 1;

   $modsdir = $inbase."Metadata/MODS/";
   if (! -e $modsdir){
REMODS:
     print "Unable to locate $modsdir. \n Please enter the MODS directory within $inbase:\n";
     $dir = <STDIN>;
     chomp ($collnum);
     $modsdir = $inbase.$dir;
     if (! -e $modsdir){ goto REMODS;}
     }

  opendir(MODS, $modsdir) or die "cannot open $modsdir\n";
  while ($file = readdir(MODS)){
     if ($file =~ /^\./){ next;}
     if ($file =~ /^(.*)\.mods\.xml/){
       # pick up number  
        push (@modsnums, $1);
        }
      }

  if (!@modsnums){ 
     print "Unable to find MODS in $modsdir\n Are you uploading MODS?  Y or N:\n";
     $ans = <STDIN>;
     chomp ($ans);
     if ($ans =~ /Y/i){
       $doMods = 1;
       print "Please enter the MODS directory within $inbase\n";
       $dir = <STDIN>;
       chomp ($collnum);
       $modsdir = $inbase.$dir;
       if (! -e $modsdir){ goto REMODS;}
       }
     }
  }

# so if $doMods exists, compare list of numbers in @modsnums to items in scans, and 
#  also copy the MODS to the working directory.

if ($doMods){   # no sense comparing if there aren't any

  print "\n\nChecking for one-to-one correspondence between items and MODS\n\n";
  print "\n\nPlease wait...  \n\n";

  foreach $dir (@scansdirs){
    opendir(DIR, $dir) or die "can't open $dir\n";
    while ($file = readdir(DIR)){
       
       if ($file =~ /^\./ || $file =~ /thumbs\.db/i){ next;}
       elsif ($file =~ /^(.*?)(\.tif)?$/ ){  # we assume here that the item number is at the top level
          $item = $1;
          $path = $dir.$file;
          if ( -d $path){  # then this is a directory 
            $path = $dir.$file."/".$item."_0001.tif";   # look for page 1
            if (! -e $path){
               print "ERROR:  No tiffs in $dir$file?\n"; 
               print OUT "ERROR:  No tiffs in $dir$file?\n";
               next;
               }  # sanity check;  is there a tiff?
            }
          push (@scannums, $item);   
          }
       }
    close(DIR);
    }

  foreach $s (@scannums){
     
     undef $found;
     foreach $m (@modsnums){
        if ($m eq $s){ $found = 1;}
        }
     if (!$found){ push (@missingMods, $s);}
     }

  foreach $m (@modsnums){
     undef $found;
     foreach $s (@scannums){
        if ($m eq $s){ $found = 1;}
        }
     if (!$found){ push (@missingScans, $m);}
     }

  if (@missingScans || @missingMods){
     print "ERROR!  Cannot continue until you have a MODS for each item, and an item for each MODS\n";
     print OUT "\nERROR!  Cannot continue until you have a MODS for each item, and an item for each MODS\n";
     if (@missingScans){  
       print "\nThese  MODS files do not have scanned items:\n";
       print OUT "\nThese  MODS files do not have scanned items:\n";
       foreach (@missingScans){ print "  $_\n";  print OUT "  $_\n";}
       }
    if (@missingMods){
       print "\nThese scanned items do not have MODS:\n";
       print OUT "\nThese scanned items do not have MODS:\n";
       foreach (@missingMods){ print "  $_\n"; print OUT "  $_\n";}
       } 
    print "\n\nAny problems found have been output to the command line \nand also will be found in the \n$out file\n";
    exit;
    }

  else{     # copy the MODS over

    print "\n\nMODS check complete!  Copying the MODS to the server now.\n\n";

    $outmods = $outbase."MODS/.";
    $inmods = $modsdir."*";
    print " copying $inmods $outmods\n";
    `cp $inmods $outmods`;
    $val = `diff -r $inmods $outmods`;
    if (! $val){ `rm -r $inmods`;}
    else{ 
        print "ERROR in copying MODS:  $val\n";
        print OUT "ERROR in copying MODS:  $val\n";
        }
    }
  } 



# call another process to make the jpegs, using nohup so it won't stop when the ssh window closes.
print "\n\nWe are starting to make the JPEGS now\n";

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


# we have all scans directories in @scansdirs and all transcript directories in @transdirs

# let's move the mp3 files to the server
  $mp3dir = $outbase."mp3/";
  undef (@dirs);
  foreach $dir (@scansdirs){
     opendir(DIR, $dir) or die "can't open $dir\n";
      while ($file = readdir(DIR)){
         if ($file =~ /^\./ || $file =~ /thumbs\.db/i){ next;}
         $path = $dir.$file;
    
         if (-d $path){
           push (@scansdirs, $path."/");
           }
         elsif ($file =~ /\.mp3/i){
           $new = $mp3dir.$file;
           copy ($path, $new);
           if (-e $new){ unlink $path;}  # deletes the copy on the share drive
           }
        }  # ignore other files
    }

  # for audio, transcript jpegs go into jpegs.
  # ocr and corrected ocr goes into ocr.
  $newdir= $outbase."jpegs/";
  $ocrdir = $outbase."ocr/";

# now, let's make the jpegs for the scans, and then the transcripts
  foreach $dir (@transdirs){
      opendir(DIR, $dir) or die "can't open $dir\n";
      while ($file = readdir(DIR)){
         if ($file =~ /^\./ || $file =~ /thumbs\.db/i){ next;}
         $path = $dir.$file;
     #    print "looking at $file\n";
         if ($file =~ /^(.*?)(\.tif)/ ){   # if a tiff, make the jpegs and drop them on the server

      #     print "file number is $1\n";
            $filenum = $1;
            $big = $newdir."/".$filenum."_2048.jpg";
  #          $mid = $newdir."/".$filenum."_512.jpg";
            $small = $newdir."/".$filenum."_128.jpg";

        # create the derivatives if they don't exist

       #  print "convert $path -strip -density 96 -resample 96x96  -resize 2048x2048 -filter Cubic -quiet $big\n\n";
           if (! -e $big){
              `convert $path -strip -density 96 -resample 96x96  -resize 2048x2048 -filter Cubic -quiet $big`;
              }
   #      if (! -e $mid){
    #        `convert $path -strip -density 96 -resample 96x96  -resize 512x512 -filter Cubic -quiet $mid`;
     #       }
           if (! -e $small){
              `convert $path -strip -density 96 -resample 96x96  -resize 128x128 -filter Cubic -quiet $small`;
              }

            # look for OCR or hand-corrected transcription.
            $ocrfile = $dir.$filenum.".ocr.txt";
            $txtfile = $dir.$filenum.".txt";
            if (-e $txtfile){   # copy it over.
              $new = $ocrdir.$filenum.".txt";
              copy ($txtfile, $new);   # don't delete share copy, we want it in the archive
              }
            elsif (-e $ocrfile){
              $new = $ocrdir.$filenum.".ocr.txt";
              copy ($ocrfile, $new);
              if (-e $new){ unlink $ocrfile;}  # delete share copy
              }
            else{   # ocr it
              $new = $ocrdir.$filenum.".ocr";
              $val = `tesseract $path $new`;
              if ($val){ print OUT "ERROR $val creating OCR for $filenum\n";} 
              }  
            }
        elsif (-d $path){
           push (@transdirs, $path."/");
           }
        else{ print OUT "Not a tif;  ignoring:  $file\n";}
        }
     close(DIR);
     }
# these for repair, later
  push (@dirlist, $newdir);



# now repair the problems caused by thumbs in the tiffs.
# goes through  the directories
# finds jpg files with -1.jpg and -0.jpg
# renames the -0 ones, removing the -0, and 
# deletes the -1 files

  foreach $dir (@dirlist){
    opendir (DIR, $dir) or die "can't look through $dir\n";
    while ($file = readdir(DIR)){
#     print "looking at $file in $dir\n";
     if ($file =~ /^\./ ){ next;  }  # skip dot files
       $path = $dir."/".$file;
       if ( -d $path){   # must be a directory
         push (@dirlist, $path);
         }
       elsif ($file =~ /^(.*)\-1\.jpg/){  # jpg file
         `rm $path`;
          }
       elsif ($file =~ /^(.*)\-0\.jpg/){
       $filenum = $1.".jpg";
       $good = $dir."/".$filenum;
       `mv $path $good`;
        }
     }

   }

  print OUT "Upload of JPEGs and any transcript material is complete.\n";
  print OUT "Please check content in UploadArea, then run relocate_all\n to distribute into 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;}