#!/usr/bin/perl 

use Fcntl;
use DBI;
use Time::Local;

# teisgm
# this version for translating TEI-Lite tdh files in SGML into OAI
# version 2.0 unqualified Dublin Core
# adds fields to MySQL database tables in Paul Cummin's OAI repository at
# http://diglib.lib.utk.edu/oai/
# datestamp used is seconds from the epoch in GMT time
# as well as to add metadata section to hardcopy file (if this is all you want,
#     comment out "#" all lines referring to the database calls
#   (let me know if you want a script for creating the date for headers)
# type tei (return) and wait
# Jody DeRidder, UTK Libraries, 12/16/02: deridder@aztec.lib.utk.edu
# type teisgm (return) and wait

# NOTE:   THIS IS SET FOR SET 10, TDH!!!

# this software provided free without guarantees


$netid = "netid";  # a field in our database to identify worker
$schema = "http://www.w3.org/2001/XMLSchema";  # then "-instance" if needed
$xmlns = "http://www.openarchives.org/OAI/2.0/OAI_"; #follow with verb
$purl = "http://purl.org/dc/elements/1.1/";
$xsd = "http://www.openarchives.org/OAI/2.0/"; # follow with format, then xsd
$username = "username";  # for access to database
$password = "password";
$us = "oai:oai.sunsite.utk.edu:";  # preface to all ids
$begin_rec = " <record>\n  <header>\n";
$rec_footer = "   </oai_dc:dc>\n  </metadata>\n  </record>\n";

$meta_header = "  <metadata>\n";
$udc_blurb = "   <oai_dc:dc\n        xmlns:oai_dc=\"".$xsd."oai_dc/\"\n        xmlns:dc=\"$purl\"\n        xmlns:xsi=\"$schema-instance\"\n        xsi:schemaLocation=\"".$xsd."oai_dc/\n        $xsd"."oai_dc.xsd\">\n";

$meta_footer = "   </oai_dc:dc>\n  </metadata>\n";
$abt_header = "   <about>\n";
$abt_footer = "   </oai_dc:dc>\n  </about>\n";

#this gets next oai_id

##    This database is
## set up with auto-incrementing identifiers as primary key.
$dbh = DBI->connect ("dbi:mysql:dcedit", $username, $password)
   or die "Can't connect to Mysql database: ",$DBI::errstr,"\n";

$h->{PrintError} = 1;
$h->{RaiseError} = 1;

$netid = $dbh->quote($netid);

  $sth = $dbh->prepare(" select max(id) as id from dc")
    or print "Can't get a count! Kill me and call for help!! : ",$dbh->errstr(),"\n";
  $sth->execute()
    or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
  $lastone = $sth->fetchrow_array();
  warn "Problem in fetchrow_array(): ",$sth->errstr(),"\n" if $sth->err();
  $sth->finish();
  $dbh->disconnect() or warn "Disconnection failed: $DBI::errstr\n";

print "verify that this was the last OAI ID used: $lastone\n";
print "Please enter y or n:\n";
    chop($another = <STDIN>);
    while ($another ne "y" && $another ne "n"){
         print "Please type y or n:\n";
         chop($another = <STDIN>);
         }
    if ($another eq "n"){ print "Fix it !!!   Bye!!\n";  exit;}


#  CHANGE THIS!!!  PUT FILES IN YOUR CHOSEN DIRECTORY
$thisdir =  "/home/firestar/tdhnew/sgm/";
$rightdir=  "/home/firestar/tdhnew/oai";



opendir(DIR,$thisdir);

while (defined($name= readdir (DIR))){
  print "looking at ".$name." in ".$thisdir."\n";
  if ($name =~ /([a-z]{2}[0-9]{4})\.sgm(l)?/){    # check this! 
                                      # this picks up files starting with
                                      # 2 letters, then 4 numbers, followed
                                      # by '.sgm' or '.sgml'
     print " name without .sgm is:  $1\n";
     $spc_id = $1;   #prefix

  $lastone ++;      #increment to next oai_id;  use this for 1st file
  $thisone = $lastone;
  print "opening $name and $spc_id\n";
 

   #hardcode files are named by prefix of sgm file
   sysopen (THIS, $rightdir."/".$spc_id, O_WRONLY | O_CREAT, 0666) or next;
   sysopen (THAT,$name, O_RDONLY) or die "Opening: $!";
   print "opening $thisdir/$spc_id\n"; 



## set input record delimiter to undefined, slurp in whole file
undef $/;
my $line = <THAT>;
  
           #first clear everything
   while (@title){pop(@title);}
   while (@creator){pop(@creator);}
   while (@subject){pop(@subject);}
   while (@descr){pop(@descr);}
   while (@pub){pop(@pub);}
   while (@contrib){pop(@contrib);}
   while (@date){pop(@date);}
   while (@type){pop(@type);}
   while (@format){pop(@format);}
   while (@ids){pop(@ids);}
   while (@source){pop(@source);}
   while (@lang){pop(@lang);}
   while (@rel){pop(@rel);}
   while (@cov){pop(@cov);}
   while (@rts){pop(@rts);}
   while (@all){pop(@all);}

   $creator=$description=$subject=$title=$pub=$morepub=$moretitle=""; 
   $dat=$morep=$morer=$rights=$contrib=$subject=$admin=$moreabs="";
   $descrip2 = "";
   $titl=$dat="";
   $id="";

# this is where you can add a description if you want;  we 
# added our own identifier, and the collection of the file
   push(@descr, "Document ID:  $spc_id");
   push (@descr, "Tennessee Documentary History Collection");

   $flag = 0;
   $found = ""; $done = "";
  #  print "LOOKING AT >\n".$_."\n\n";
     $_ = $line;
    s/\n//g;
    s/\s{3,}/ /g;     # get rid of excess spaces
    s/\&amp\;/and/g;   # ampersands become 'and' 
    s/\&/and/g;

#print "My file is this:\n\n$_\n\n";
    $flag ++;
    if ($flag == 50){print "endless loop\n"; exit(1);}
    if (!@title){
      if ($_ =~ /<titleStmt>(.*)<\/titleStmt>/i){
        $work = $1;

        if ($work =~ /<title>(.*)<\/title>( )*?<author>(.*)<\/author>.*/i){
         $title = $1; 
         $creator = $3;
         if ($title =~ /(.*)<\/title>( )*?<author>(.*)<\/author>.*/i){  # remove next section
           $title = $1;
           $creator = $3;
           }
         if ($creator =~ /(\s)*(.*)(\s)*/){  #remove leading/ending spaces
           $creator = $2;
           }
         if ($creator =~/(.*)<\/author>( )*?<author>(.*)/i){ # multiple authors
         #  print "CREATOR: $creator";
           while ($creator =~/(.*)<\/author>( )*?<author>(.*)/i){
             push (@creator, $3);
             $creator = $1;
             }
           push (@creator, $creator);
           }    
         else {push(@creator, $creator);}
         while ($title =~ /(.*)<\/title>( )*?<title>(.*)/i){
            push (@title, $3);
            $title = $1;
            }
         push (@title, $title);
         $found = "yes"; $done = "yes";
         $find = "pub";
         }
 
# if no authors, this one will still work
      elsif ($work =~ /<title>(.*)<\/title>( )*?<resp.*/i){
         $title = $1;
         while ($title =~ /(.*)<\/title>( )*?<title>(.*)/i){
            push (@title, $3);
            $title = $1;
            }
         push (@title, $1);
         $creator = "";
         $found = "yes"; $done = "yes";
         $find = "pub";
         }
       }
     }


    if (($done eq "yes")&&($find eq "pub")){
      if ($_ =~ /<PUBLICATIONSTMT>(\s)*?(.*)<\/PUBLICATIONSTMT>/i){
        $work = $2;
        if ($work =~ /<publisher>(.*)(\s)*/i){
          $pub = $1;
          $found = "yes"; $done = "no";
          if ($pub =~/(\s)*(.*)<\/publisher>/i){  # take out end tag
            $pub = $2;
            $find = "doc";
        #    print "pub is found, looking for DOC\n";
            if ($pub =~ /(\s)*(.*)(\s)*/){  #remove leading/ending spaces
              $pub = $2;
              }

            $done = "yes";
            push (@pub, $pub);
            }
          }
        $find = "doc";
        }
      }
     if ($find eq "doc"){
         $find = "rights";
         }
     if (($done eq "yes") && ($find eq "rights")){
        if ($_ =~ /<p>(.*)<\/p>( )*?<\/availability>/i){
          push (@rts, $1);
          $rights = $1;
           $done = "yes";
           }
         $find = "date";
         }

     if (($done eq "yes")&&($find eq "date")){
       if($_ =~ /author>( )*?<date value="(.*)">(.*)<\/date>/i){
          $working = $2;
          while ($working =~/(.*)">.*/){
            $working = $1; 
            } 
          $dat = $working;
          }
       if (!$dat){ 
         if ($_ =~ /author>( )*?<date value="(.*)">.*<\/date>/i){
           $working = $2;
            while ($working =~/(.*)">.*/){
              $working = $1;
              }
            $dat = $working;
            }
# this for no authors
          elsif ($_ =~ /title>( )*?<date value="(.*)">.*<\/date>/i){
            $working = $2;
            while ($working =~/(.*)">.*/){
              $working = $1;
              }
            $dat = $working;
            }
          }
       push (@date, $dat); 
       $find = "des";
       }
     

     if (($done eq "yes") && ($find eq "des")){
        $working = $_;
        if ($working =~ /.*<note type="summary">(.*)(\s)*<\/note>( *)?<note.*/i){
          #print "found a match!\n";
          $working = $1;
          while ($working =~ /(.*)(\s)*<\/note>( *)?<note.*/i){
            $working = $1;
            }
          while ($working =~ /(.*)<xref>(.*)<\/xref>(.*)/i){
            $working = $1." ".$2.$3;
             }
          $found = "yes"; $done = "yes";
          #print "found this des:\n$working\n";
          $working =~ s/ {3,}/ /g; # correct for overspacing in document
          $working =~ s/<HI REND="i">/"/ig;
          $working =~ s/<\/HI>/"/ig;
          if ($working){ push (@descr, $working);}
          #next;
          }
        $find = "sub";
        }
       

       if ($find eq "sub"){
          $work = "";
          if ($_ =~ /<profileDesc>(\s)*?<textClass>(\s)*?<keywords scheme="LCSH">(\s)*?<list>(.*)(\s)*?<\/keywords>(\s)*?<\/textClass>(\s)*?<\/profileDesc>/i){
             $work = $4;
             }

          if ($work =~ /<item>(.*)<\/item>/i){
            $wk = $1;
            while ($wk =~ /(.*)<\/item>( )*?<item>(.+)/i){
               push (@subject, $3);
               $wk = $1;
               }
            push (@subject, $wk);
            #next;
            }
          $find = "nada";
          }
        
   
# change this;  we needed to tweak our output on publisher and contributor
   if (!@pub){
         push (@pub, "University of Tennessee Special Collections Library");
         }

   push (@contrib, "University of Tennessee Special Collections Library, Knoxville");
   if ($rights =~ /(.*) McClung Museum(.*)/){
      push (@contrib, "Frank H. McClung Museum (Knoxville, Tenn.)");
      }
   if ($rights =~ /(.*)McClung Historical Collection(.*)/){
      push (@contrib, "Knox County Public Library (Knoxville, Tenn.)");
      }
   if ($rights =~ /.*Memphis Public Library.*/){
      push (@contrib, "Memphis Public Library (Memphis, Tenn.)");
      }

   if ($rights =~ /.*Tennessee State Library.*/){
      push (@contrib, "Tennessee State Library and Archives (Nashville, Tenn.)");
      }

# this is where you put in your identifier link, where you have the item online

   $id = "http://anasazi.lib.utk.edu/cgi/t/text/text-idx?c=tdh;view=text;rgn=main;idno=$spc_id";
  #  print "check identifier link:  $id\n";
   push (@ids, $id);

   # our default language is English, and these files were text/html files with
   # tiff images
   push (@lang , "en");
   @type = ("Image", "Text"); 
   @format = ("Image/tiff", "Text/html");

 
     $check = 0;
     print "\nHere are all the fields for the record that you entered:\n\n";
     if(@title){print "title:"; foreach(@title){print "   ".$_."\n";}}
     else {$check ++;}
     if(@creator){print "\ncreator:"; foreach(@creator){print "   ".$_."\n";}}
     else {$check ++;}
     if(@subject){print "\nsubject:"; foreach(@subject){print "   ".$_."\n";}}
     else {$check ++;}
     if(@descr){print "\ndescription:"; foreach(@descr){print "   ".$_."\n";}}
     else {$check ++;}
     if(@pub){print "\npublisher:"; foreach(@pub){print "   ".$_."\n";}}
     else {$check ++;}
     if(@contrib){print "\ncontributor:"; foreach(@contrib){print "   ".$_."\n";}}
     else {$check ++;}
     if(@date){print "\ndate:"; foreach(@date){print "   ".$_."\n";}}
     else {$check ++;}
     if(@type){print "\ntype:"; foreach(@type){print "   ".$_."\n";}}
     else {$check ++;}
     if(@format){print "\nformat:"; foreach(@format){print "   ".$_."\n";}}
     else {$check ++;}
     if(@ids){print "\nidentifier:"; foreach(@ids){print "   ".$_."\n";}}
     else {$check ++;}
     if(@source){print "\nsource:"; foreach(@source){print "   ".$_."\n";}}
     else {$check ++;}
     if(@lang){print "\nlanguage:"; foreach(@lang){print "   ".$_."\n";}}
     else {$check ++;}
     if(@rel){print "\nrelation:"; foreach(@rel){print "   ".$_."\n";}}
     else {$check ++;}
     if(@cov){print "\ncoverage:"; foreach(@cov){print "   ".$_."\n";}}
     else {$check ++;}
     if(@rts){print "\nrights:"; foreach(@rts){print "   ".$_."\n";}}
     else {$check ++;}

  if ($check == 15){
       print "Sorry, we cannot enter a record with no DC values.\n";
       exit;
       }

     print "\nIf all this is correct, type y\n";
     print "TO ABORT THIS RECORD AND EXIT, type x\n";
     chop($ok = <STDIN>);

     while ($ok ne "y" &&  $ok ne "x"){
         print "Please type y or or x:\n";
         chop($ok = <STDIN>);
         }
     if ($ok eq "x"){
        print "\nBye-Bye!!\n";
        exit;
        }


   print "got here\n";
   $mydate = &getDate;
   $thisone = $dbh->quote($thisone);
   $mydate = $dbh->quote($mydate);




   $dbh = DBI->connect ("dbi:mysql:dcedit", $username, $password)
      or die "Can't connect to Mysql database: ",$DBI::errstr,"\n";

    $h->{PrintError} = 1;
    $h->{RaiseError} = 1;
    print "inserting $name as $thisone\n";
    $sth = $dbh->prepare("insert into dc (id,netid,oai_date) values($thisone,$netid,$mydate)")
    or die "dc: unable to insert $name, $thisone : ",$dbh->errstr(),"\n";
    $sth->execute()
    or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
    $sth->finish();

# NOTE:   THIS IS SET FOR SET 10, TDH!!!
    $sth = $dbh->prepare("insert into set2dc (id,setid,dcid) values(0,'10',$thisone)")
    or print "set2dc:  unable to insert $name, $thisone : ",$dbh->errstr(),"\n";
    $sth->execute()
    or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
    $sth->finish();

# collecting entire xml file for text blob in database
   $all = "";

   print THIS $meta_header.$udc_blurb;
   foreach(@title){
      print THIS "    <dc:title>$_</dc:title>\n";
      $title = $dbh->quote($_);
      $sth = $dbh->prepare("insert into title (id, data, dcid) values (0, $title, $thisone)")
      or print "title: unable to insert $name, $thisone : ",$dbh->errstr(),"\n";
      $sth->execute()
        or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
      $sth->finish();
      }
   foreach(@creator){
       print THIS "    <dc:creator>$_</dc:creator>\n";
       $creator = $dbh->quote($_);
       $sth = $dbh->prepare("insert into creator (id, data, dcid) values (0, $creator, $thisone)")
       or print "creator:  unable to insert $name, $thisone : $dbh->errstr()\n";
       $sth->execute()
       or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
       $sth->finish();
       }
   foreach(@subject){
       print THIS "    <dc:subject>$_</dc:subject>\n";
       $subject = $dbh->quote($_);
       $sth = $dbh->prepare("insert into subject (id, data, dcid) values (0, $subject, $thisone)")
       or print "subject:  unable to insert $name, $thisone : ",$dbh->errstr(),"\n";
       $sth->execute()
       or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
       $sth->finish();
       }
   foreach(@descr){
       print THIS "    <dc:description>$_</dc:description>\n";
       $description = $dbh->quote($_);
       $sth = $dbh->prepare("insert into description (id, data, dcid) values (0, $description, $thisone)")
       or print "description: unable to insert $name, $thisone : ",$dbh->errstr(),"\n";
       $sth->execute()
       or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
       $sth->finish();
       }
   foreach(@pub){
       print THIS "    <dc:publisher>$_</dc:publisher>\n";
       $publisher = $dbh->quote($_);
       $sth = $dbh->prepare("insert into publisher (id, data, dcid) values (0, $publisher, $thisone)")
       or print "publisher:  unable to insert $name, $thisone : ",$dbh->errstr(),"\n";
       $sth->execute()
       or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
       $sth->finish();
       }
   foreach(@contrib){
       print THIS "    <dc:contributor>$_</dc:contributor>\n";
       $contributor = $dbh->quote($_);
       $sth = $dbh->prepare("insert into contributor (id, data, dcid) values (0, $contributor, $thisone)")
       or print "contributor:  unable to insert $name, $thisone : $dbh->errstr()\n";
       $sth->execute()
       or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
       $sth->finish();
       }
   foreach(@date){
       print THIS "    <dc:date>$_</dc:date>\n";
       $date = $dbh->quote($_);
       $sth = $dbh->prepare("insert into date (id, data, dcid) values (0, $date, $thisone)")
       or print "date: unable to insert $name, $thisone : $dbh->errstr()\n";
       $sth->execute()
       or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
       $sth->finish();
       }
   foreach(@type){
       print THIS "    <dc:type>$_</dc:type>\n";
       $type = $dbh->quote($_);
       $sth = $dbh->prepare("insert into type (id, data, dcid) values (0, $type, $thisone)")
       or print "type: unable to insert $name, $thisone : $dbh->errstr()\n";
       $sth->execute()
       or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
       $sth->finish();
       }
   foreach(@format){
      print THIS "    <dc:format>$_</dc:format>\n";
      $format = $dbh->quote($_);
      $sth = $dbh->prepare("insert into format (id, data, dcid) values (0, $format, $thisone)")
      or print "format:  unable to insert $name, $thisone : $dbh->errstr()\n";
      $sth->execute()
      or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
      $sth->finish();
      }
   foreach(@ids){
      print THIS "    <dc:identifier>$_</dc:identifier>\n";
      $id = $dbh->quote($_);
      $sth = $dbh->prepare("insert into identifier (id, data, dcid) values (0, $id, $thisone)")
      or print "identifier:  unable to insert $name, $thisone : $dbh->errstr()\n";
      $sth->execute()
      or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
      $sth->finish();
      }
   foreach(@source){
      print THIS "    <dc:source>$_</dc:source>\n";
      $source = $dbh->quote($_);
      $sth = $dbh->prepare("insert into source (id, data, dcid) values (0, $source, $thisone)")
      or print "source: unable to insert $name, $thisone : $dbh->errstr()\n";
      $sth->execute()
      or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
      $sth->finish();
      }
   foreach(@lang){
      print THIS "    <dc:language>$_</dc:language>\n";
      $language = $dbh->quote($_);
      $sth = $dbh->prepare("insert into language (id, data, dcid) values (0, $language, $thisone)")
      or print "language: unable to insert $name, $thisone : $dbh->errstr()\n";
      $sth->execute()
      or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
      $sth->finish();
      }
   foreach(@rel){
      print THIS "    <dc:relation>$_</dc:relation>\n";
      $relation = $dbh->quote($_);
      $sth = $dbh->prepare("insert into relation (id, data, dcid) values (0, $relation, $thisone)")
      or print "relation:  unable to insert $name, $thisone : $dbh->errstr()\n";
      $sth->execute()
      or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
      $sth->finish();
      }
   foreach(@cov){
      print THIS "    <dc:coverage>$_</dc:coverage>\n";
      $coverage = $dbh->quote($_);
      $sth = $dbh->prepare("insert into coverage (id, data, dcid) values (0, $coverage, $thisone)")
      or print "coverage:  unable to insert $name, $thisone : $dbh->errstr()\n";
      $sth->execute()
      or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
      $sth->finish();
      }
   foreach(@rts){
      print  THIS "    <dc:rights>$_</dc:rights>\n";
      $rights = $dbh->quote($_);
      $sth = $dbh->prepare("insert into rights (id, data, dcid) values (0, $rights,$thisone)")
      or print "rights: unable to insert $name, $thisone : $dbh->errstr()\n";
      $sth->execute()
      or die "Can't execute SQL statement: ", $sth->errstr(),"\n";
      $sth->finish();
      }
   print THIS $meta_footer;

   close (THIS);
   close (THAT);


   $dbh->disconnect
   or warn "Disconnection failed: $DBI::errstr\n";

   }   # end of if this file name fits 2 letters 3 numbers
 }  # end of looking through this directory


closedir(DIR);
exit;


sub getDate{     # gets time from the epoch, GMT time
@when = split(/ +/, `date -u`);

($mymonth, $myday, $time, $myyear) = (@when)[1,2,3,5];
($hour, $min, $sec) = split (':', $time);
chop($myyear);  #lose the newline

# month must be in numbers, 0-11 !!!!!

if ($mymonth eq "Jan"){$mymonth = "00";}
elsif ($mymonth eq "Feb"){$mymonth = "01";}
elsif ($mymonth eq "Mar"){$mymonth = "02";}
elsif ($mymonth eq "Apr"){$mymonth = "03";}
elsif ($mymonth eq "May"){$mymonth = "04";}
elsif ($mymonth eq "Jun"){$mymonth = "05";}
elsif ($mymonth eq "Jul"){$mymonth = "06";}
elsif ($mymonth eq "Aug"){$mymonth = "07";}
elsif ($mymonth eq "Sep"){$mymonth = "08";}
elsif ($mymonth eq "Oct"){$mymonth = "09";}
elsif ($mymonth eq "Nov"){$mymonth = "10";}
else{$mymonth = "11";}
#$olddate = `date -u`;
#print " date: $olddate\n";
#print " month: $mymonth\n day: $myday\n year: $myyear\n hour: $hour\n minute: $min\n second: $sec\n";
$mydate = timegm($sec, $min, $hour, $myday, $mymonth, $myyear);
print "time:  $mydate\n";
$mydate;
}