Diff for /loncom/metadata_database/searchcat.pl between versions 1.30 and 1.33

version 1.30, 2003/02/03 17:01:55 version 1.33, 2003/06/19 19:34:27
Line 26 Line 26
 #  #
 # http://www.lon-capa.org/  # http://www.lon-capa.org/
 #  #
 # YEAR=2001  
 # 04/14/2001, 04/16/2001 Scott Harrison  
 #  
 # YEAR=2002  
 # 05/11/2002 Scott Harrison  
 #  
 # YEAR=2003  
 # Scott Harrison  
 #  
 ###  ###
   
 =pod  =pod
Line 52  filesystem installation location: F</etc Line 43  filesystem installation location: F</etc
 Here is the cron job entry.  Here is the cron job entry.
   
 C<# Repopulate and refresh the metadata database used for the search catalog.>  C<# Repopulate and refresh the metadata database used for the search catalog.>
   
 C<10 1 * * 7    www    /home/httpd/perl/searchcat.pl>  C<10 1 * * 7    www    /home/httpd/perl/searchcat.pl>
   
 This script only allows itself to be run as the user C<www>.  This script only allows itself to be run as the user C<www>.
Line 65  The metadata is entered into a SQL datab Line 55  The metadata is entered into a SQL datab
 This script also does general database maintenance such as reformatting  This script also does general database maintenance such as reformatting
 the C<loncapa:metadata> table if it is deprecated.  the C<loncapa:metadata> table if it is deprecated.
   
 This script also builds dynamic temporal metadata and stores this inside  This script evaluates dynamic metadata from the authors'
 a F<nohist_resevaldata.db> database file.  F<nohist_resevaldata.db> database file in order to store it in MySQL, as
   well as to compress the filesize (add up all "count"-type metadata).
   
 This script is playing an increasingly important role for a loncapa  This script is playing an increasingly important role for a loncapa
 library server.  The proper operation of this script is critical for a smooth  library server.  The proper operation of this script is critical for a smooth
Line 74  and correct user experience. Line 65  and correct user experience.
   
 =cut  =cut
   
 # ========================================================== Setting things up.  
   
 # ------------------------------------------------------  Use external modules.  
   
 use lib '/home/httpd/lib/perl/';  use lib '/home/httpd/lib/perl/';
 use LONCAPA::Configuration;  use LONCAPA::Configuration;
   
Line 87  use DBI; Line 74  use DBI;
 use GDBM_File;  use GDBM_File;
 use POSIX qw(strftime mktime);  use POSIX qw(strftime mktime);
   
 # ----------------- Code to enable 'find' subroutine listing of the .meta files  
 use File::Find;  
   
 # List of .meta files (used on a per-user basis).  
 my @metalist;  my @metalist;
   
 # ---------------  Read loncapa_apache.conf and loncapa.conf and get variables.  
 my $perlvarref = LONCAPA::Configuration::read_conf('loncapa.conf');  
 my %perlvar = %{$perlvarref};  
 undef($perlvarref); # Remove since sensitive and not needed.  
 delete($perlvar{'lonReceipt'}); # Remove since sensitive and not needed.  
   
 # ------------------------------------- Only run if machine is a library server  
 if ($perlvar{'lonRole'} ne 'library')  
   {  
     exit(0);  
   }  
   
 # ------------------------------ Make sure this process is running as user=www.  
 my $wwwid = getpwnam('www');  
 if ($wwwid != $<)  
   {  
     $emailto = "$perlvar{'lonAdmEMail'},$perlvar{'lonSysEMail'}";  
     $subj = "LON: $perlvar{'lonHostID'} User ID mismatch";  
     system("echo 'User ID mismatch. searchcat.pl must be run as user www.' | ".  
    "mailto $emailto -s '$subj' > /dev/null");  
     exit(1);  
   }  
   
 # ------------------------------------------------------ Initialize log output.  
 open(LOG,'>'.$perlvar{'lonDaemons'}.'/logs/searchcat.log');  
 print(LOG '==== Searchcat Run '.localtime().' ===='."\n\n");  
   
 my $dbh; # Database object reference handle.  
   
 # ----------------------------- Verify connection to loncapa:metadata database.  
 unless (  
  $dbh = DBI->connect('DBI:mysql:loncapa','www',  
     $perlvar{'lonSqlAccess'},  
     { RaiseError => 0,PrintError => 0})  
  )  
   {   
     print(LOG '**** ERROR **** Cannot connect to database!'."\n");  
     exit(0);  
   }  
   
 # ------------------------------ Create loncapa:metadata table if non-existent.  
 my $make_metadata_table = 'CREATE TABLE IF NOT EXISTS metadata ('.  
     'title TEXT, author TEXT, subject TEXT, url TEXT, keywords TEXT, '.  
     'version TEXT, notes TEXT, abstract TEXT, mime TEXT, language TEXT, '.  
     'creationdate DATETIME, lastrevisiondate DATETIME, owner TEXT, '.  
     'copyright TEXT, utilitysemaphore BOOL, FULLTEXT idx_title (title), '.  
     'FULLTEXT idx_author (author), FULLTEXT idx_subject (subject), '.  
     'FULLTEXT idx_url (url), FULLTEXT idx_keywords (keywords), '.  
     'FULLTEXT idx_version (version), FULLTEXT idx_notes (notes), '.  
     'FULLTEXT idx_abstract (abstract), FULLTEXT idx_mime (mime), '.  
     'FULLTEXT idx_language (language), FULLTEXT idx_owner (owner), '.  
     'FULLTEXT idx_copyright (copyright)) TYPE=MYISAM';  
   
 $dbh->do($make_metadata_table); # Generate the table.  
   
 # ----------------------------- Verify format of the loncapa:metadata database.  
 #                               (delete and recreate database if necessary).  
   
 # Make a positive control for verifying table structure.  
 my $make_metadata_table_CONTROL = $make_metadata_table;  
 $make_metadata_table_CONTROL =~  
     s/^(CREATE TABLE IF NOT EXISTS) metadata/$1 CONTROL_metadata/;  
   
 $dbh->do('DROP TABLE IF EXISTS CONTROL_metadata');  
 $dbh->do($make_metadata_table_CONTROL);  
   
 my $table_description; # selectall reference to the table description.  
   
 my $CONTROL_table_string; # What the table description should look like.  
 my $table_string; # What the table description does look like.  
   
 # Calculate the CONTROL table description (what it should be).  
 $table_description = $dbh->selectall_arrayref('describe CONTROL_metadata');  
 foreach my $table_row (@{$table_description})  
   {  
     $CONTROL_table_string .= join(',',@{$table_row})."\n";  
   }  
   
 # Calculate the current table description (what it currently looks like).  
 $table_description = $dbh->selectall_arrayref('describe metadata');  
 foreach my $table_row (@{$table_description})  
   {  
     $table_string .= join(',',@{$table_row})."\n";  
   }  
   
 if ($table_string ne $CONTROL_table_string)  
   {  
     # Log this incident.  
     print(LOG '**** WARNING **** Table structure mismatch, need to regenerate'.  
   '.'."\n");  
     # Delete the table.  
     $dbh->do('DROP TABLE IF EXISTS metadata');  
     # Generate the table.  
     $dbh->do($make_metadata_table);  
   }  
   
 $dbh->do('DROP TABLE IF EXISTS CONTROL_metadata'); # Okay.  Done with control.  
   
 # ----------------------------------------------- Set utilitysemaphore to zero.  
 $dbh->do('UPDATE metadata SET utilitysemaphore = 0');  
   
 # ========================================================= Main functionality.  
   
 # - Determine home authors on this server based on resources dir and user tree.  
   
 # RESOURCES: the resources directory (subdirs correspond to author usernames).  
 opendir(RESOURCES,"$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}") or  
     (print(LOG '=== /res/--lonDefDomain-- directory is not accessible'."\n")  
      and exit(0));  
   
 # query_home_server_status will look for user home directories on this machine.  
 my @homeusers =  
     grep {&query_home_server_status($perlvar{'lonDocRoot'}.'/res/'.  
     $perlvar{'lonDefDomain'}.'/'.$_)  
   } grep {!/^\.\.?$/} readdir(RESOURCES);  
 closedir(RESOURCES);  
   
 unless (@homeusers)  
   {  
     print(LOG '=== No home users found on this server.'."\n");  
   }  
   
 # Consider each author individually.  
 foreach my $user (@homeusers)  
   {  
     # Make a log entry.  
     print(LOG "\n".'=== User: '.$user."\n\n");  
   
     # Get filesystem path to this user's directory.  
     my $user_directory =  
  &construct_path_to_user_directory($perlvar{'lonDefDomain'},$user);  
   
     # Remove left-over db-files from a potentially crashed searchcat run.  
     unlink($user_directory.'/nohist_new_resevaldata.db');  
   
     # Cleanup the metalist array.  
     undef(@metalist);  
     @metalist = ();  
   
     # This will add entries to the @metalist array.  
     &File::Find::find(\&wanted,  
       $perlvar{'lonDocRoot'}.'/res/'.  
       $perlvar{'lonDefDomain'}.'/'.$user);  
   
     # -- process file to get metadata and put into search catalog SQL database  
     # Also, build and store dynamic metadata.  
     # Also, delete record entries before refreshing.  
     foreach my $m (@metalist)  
       {  
  # Log this action.  
  print(LOG "- ".$m."\n");  
   
  # Get metadata from the file.  
  my $ref = get_metadata_from_file($m);  
   
  # Make a datarecord identifier for this resource.  
  my $m2 = '/res/'.declutter($m);  
  $m2 =~ s/\.meta$//;  
   
  # Build and store dynamic metadata inside nohist_resevaldata.db.  
  build_on_the_fly_dynamic_metadata($m2);  
   
  # Delete record if it already exists.  
  my $q2 = 'select * from metadata where url like binary '."'".$m2."'";  
  my $sth = $dbh->prepare($q2);  
  $sth->execute();  
  my $r1 = $sth->fetchall_arrayref;  
  if (@$r1)  
   {  
     $sth =   
  $dbh->prepare('delete from metadata where url like binary '.  
       "'".$m2."'");  
     $sth->execute();  
   }  
   
  # Add new/replacement record into the loncapa:metadata table.  
  $sth = $dbh->prepare('insert into metadata values ('.  
      '"'.delete($ref->{'title'}).'"'.','.  
      '"'.delete($ref->{'author'}).'"'.','.  
      '"'.delete($ref->{'subject'}).'"'.','.  
      '"'.$m2.'"'.','.  
      '"'.delete($ref->{'keywords'}).'"'.','.  
      '"'.'current'.'"'.','.  
      '"'.delete($ref->{'notes'}).'"'.','.  
      '"'.delete($ref->{'abstract'}).'"'.','.  
      '"'.delete($ref->{'mime'}).'"'.','.  
      '"'.delete($ref->{'language'}).'"'.','.  
      '"'.sql_formatted_time(  
        delete($ref->{'creationdate'})).'"'.','.  
      '"'.sql_formatted_time(  
    delete($ref->{'lastrevisiondate'})).'"'.','.  
      '"'.delete($ref->{'owner'}).'"'.','.  
      '"'.delete($ref->{'copyright'}).'"'.','.  
      '1'.')');  
  $sth->execute();  
       }  
   
 # ----------------------- Clean up database, remove stale SQL database records.  
     $dbh->do('DELETE FROM metadata WHERE utilitysemaphore = 0');  
   
 # -------------------------------------------------- Copy over the new db-files  
     system('mv '.$user_directory.'/nohist_new_resevaldata.db '.  
          $user_directory.'/nohist_resevaldata.db');  
   }  
   
 # --------------------------------------------------- Close database connection  
 $dbh->disconnect;  
 print LOG "\n==== Searchcat completed ".localtime()." ====\n";  
 close(LOG);  
 exit(0);  
   
 # ================================================================ Subroutines.  # ----------------------------------------------------- Un-Escape Special Chars
   
 =pod  sub unescape {
       my $str=shift;
 =head1 SUBROUTINES  
   
 =cut  
   
 =pod  
   
 B<unescape> - translate to unstrange escaped syntax to strange characters.  
   
 =over 4  
   
 Parameters:  
   
 =item I<$str> - string with unweird characters.  
   
 =back  
   
 =over 4  
   
 Returns:  
   
 =item C<string> - string with potentially weird characters.  
   
 =back  
   
 =cut  
   
 sub unescape ($)  
   {  
     my $str = shift(@_);  
     $str =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C",hex($1))/eg;      $str =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C",hex($1))/eg;
     return($str);      return $str;
   }  }
   
 =pod  
   
 B<escape> - translate strange characters to unstrange escaped syntax.  
   
 =over 4  
   
 Parameters:  
   
 =item I<$str> - string with potentially weird characters to unweird-ify.  
   
 =back  
   
 =over 4  
   
 Returns:  
   
 =item C<string> - unweird-ified string.  
   
 =back  
   
 =cut  # -------------------------------------------------------- Escape Special Chars
   
 sub escape ($)  sub escape {
   {      my $str=shift;
     my $str = shift(@_);  
     $str =~ s/(\W)/"%".unpack('H2',$1)/eg;      $str =~ s/(\W)/"%".unpack('H2',$1)/eg;
     return($str);      return $str;
   }  }
   
 =pod  
   
 B<build_on_the_fly_dynamic_metadata> - evaluate and store dynamic metadata.  
   
 Returns the dynamic metadata for an author, which will later be added to the  
 MySQL database (not yet implemented).  
   
 The vast majority of entries in F<nohist_resevaldata.db>, which contains  
 the dynamic metadata for an author's resources, are "count", which make  
 the file really large and evaluation really slow.  
   
 While computing the current value of all dynamic metadata  
 for later insertion into the MySQL metadata cache (not yet implemented),  
 this routine also simply adds up all "count" type fields and replaces them by  
 one new field with the to-date count.  
   
 Only after successful completion of working with one author, copy new file to  
 original file. Copy to tmp-"new"-db-file was necessary since db-file size   
 would not shrink after "delete" of key.  
   
 =over 4  
   
 Parameters:  
   
 =item I<$url> - the filesystem path (url may be a misnomer...)  
   
 =back  
   
 =over 4  
   
 Returns:  
   
 =item C<hash> - key-value table of dynamically evaluated metadata.  
   
 =back  
   
 =cut  # ------------------------------------------- Code to evaluate dynamic metadata
   
 sub build_on_the_fly_dynamic_metadata {  sub dynamicmeta {
   
     # Need to compute the user's directory.  
     my $url=&declutter(shift);      my $url=&declutter(shift);
     $url=~s/\.meta$//;      $url=~s/\.meta$//;
     my %returnhash=();      my %returnhash=();
     my ($adomain,$aauthor)=($url=~/^(\w+)\/(\w+)\//);      my ($adomain,$aauthor)=($url=~/^(\w+)\/(\w+)\//);
     my $user_directory=&construct_path_to_user_directory($adomain,$aauthor);      my $prodir=&propath($adomain,$aauthor);
   
     # Attempt a GDBM database instantiation inside users directory and proceed.  
     if ((tie(%evaldata,'GDBM_File',      if ((tie(%evaldata,'GDBM_File',
             $user_directory.               $prodir.'/nohist_resevaldata.db',&GDBM_READER(),0640)) &&
      '/nohist_resevaldata.db',&GDBM_READER(),0640)) &&  
         (tie(%newevaldata,'GDBM_File',          (tie(%newevaldata,'GDBM_File',
      $user_directory.               $prodir.'/nohist_new_resevaldata.db',&GDBM_WRCREAT(),0640))) {
      '/nohist_new_resevaldata.db',&GDBM_WRCREAT(),0640))) {          my %sum=();
  # For different variables, track the running sum and counts.          my %cnt=();
  my %sum=();          my %listitems=('count'        => 'add',
  my %cnt=();                         'course'       => 'add',
                          'avetries'     => 'avg',
  # Define computed items as a sum (add) or an average (avg) or a raw                         'stdno'        => 'add',
  # count (cnt) or append (app)?                         'difficulty'   => 'avg',
  my %listitems=('count'        => 'add',                         'clear'        => 'avg',
        'course'       => 'add',                         'technical'    => 'avg',
        'avetries'     => 'avg',                         'helpful'      => 'avg',
        'stdno'        => 'add',                         'correct'      => 'avg',
        'difficulty'   => 'avg',                         'depth'        => 'avg',
        'clear'        => 'avg',                         'comments'     => 'app',
        'technical'    => 'avg',                         'usage'        => 'cnt'
        'helpful'      => 'avg',                         );
        'correct'      => 'avg',          my $regexp=$url;
        'depth'        => 'avg',          $regexp=~s/(\W)/\\$1/g;
        'comments'     => 'app',          $regexp='___'.$regexp.'___([a-z]+)$';
        'usage'        => 'cnt'          foreach (keys %evaldata) {
        );              my $key=&unescape($_);
               if ($key=~/$regexp/) {
  # Untaint the url and use as part of a regular expression.                  my $ctype=$1;
  my $regexp=$url;                  if (defined($cnt{$ctype})) { 
  $regexp=~s/(\W)/\\$1/g;                      $cnt{$ctype}++; 
  $regexp='___'.$regexp.'___([a-z]+)$'; #' emacs                  } else { 
                       $cnt{$ctype}=1; 
  # Check existing database for this author.                  }
         # this is modifying the 'count' entries                  unless ($listitems{$ctype} eq 'app') {
         # and copying all other entries over                      if (defined($sum{$ctype})) {
                           $sum{$ctype}+=$evaldata{$_};
  foreach (keys %evaldata) {                      } else {
     my $key=&unescape($_);                          $sum{$ctype}=$evaldata{$_};
     if ($key=~/$regexp/) { # If url-based entry exists.                      }
  my $ctype=$1; # Set to specific category type.                  } else {
                       if (defined($sum{$ctype})) {
  # Do an increment for this category type.                          if ($evaldata{$_}) {
  if (defined($cnt{$ctype})) {                              $sum{$ctype}.='<hr>'.$evaldata{$_};
     $cnt{$ctype}++;                           }
  } else {                      } else {
     $cnt{$ctype}=1;                           $sum{$ctype}=''.$evaldata{$_};
  }                      }
                 unless ($listitems{$ctype} eq 'app') { # append comments                  }
     # Increment the sum based on the evaluated data in the db.                  if ($ctype ne 'count') {
     if (defined($sum{$ctype})) {                      $newevaldata{$_}=$evaldata{$_};
  $sum{$ctype}+=$evaldata{$_};                  }
     } else {              }
  $sum{$ctype}=$evaldata{$_};          }
     }          foreach (keys %cnt) {
   } else { # 'app' mode, means to use '<hr />' as a separator              if ($listitems{$_} eq 'avg') {
     if (defined($sum{$ctype})) {                  $returnhash{$_}=int(($sum{$_}/$cnt{$_})*100.0+0.5)/100.0;
  if ($evaldata{$_}) {              } elsif ($listitems{$_} eq 'cnt') {
     $sum{$ctype}.='<hr />'.$evaldata{$_};                  $returnhash{$_}=$cnt{$_};
  }              } else {
     } else {                  $returnhash{$_}=$sum{$_};
  $sum{$ctype}=''.$evaldata{$_};              }
     }          }
         }          if ($returnhash{'count'}) {
  if ($ctype ne 'count') {              my $newkey=$$.'_'.time.'_searchcat___'.&escape($url).'___count';
                     # this is copying all data except 'count' attributes              $newevaldata{$newkey}=$returnhash{'count'};
     $newevaldata{$_}=$evaldata{$_};          }
         }          untie(%evaldata);
     }          untie(%newevaldata);
  }  
   
         # these values will be returned (currently still unused)  
  foreach (keys %cnt) {  
     if ($listitems{$_} eq 'avg') {  
  $returnhash{$_}=int(($sum{$_}/$cnt{$_})*100.0+0.5)/100.0;  
     } elsif ($listitems{$_} eq 'cnt') {  
  $returnhash{$_}=$cnt{$_};  
     } else {  
  $returnhash{$_}=$sum{$_};  
     }  
  }  
   
         # generate new count key in resevaldata, insert sum  
  if ($returnhash{'count'}) {  
     my $newkey=$$.'_'.time.'_searchcat___'.&escape($url).'___count';  
     $newevaldata{$newkey}=$returnhash{'count'};  
  }  
   
  untie(%evaldata); # Close/release the original nohist database.  
  untie(%newevaldata); # Close/release the new nohist database.  
     }      }
     return %returnhash;      return %returnhash;
 }  }
   
 =pod  # ----------------- Code to enable 'find' subroutine listing of the .meta files
   require "find.pl";
 B<wanted> - used by B<File::Find::find> subroutine.  sub wanted {
   
 This evaluates whether a file is wanted, and pushes it onto the  
 I<@metalist> array.  This subroutine was, for the most part, auto-generated  
 by the B<find2perl> command.  
   
 =over 4  
   
 Parameters:  
   
 =item I<$file> - a path to the file.  
   
 =back  
   
 =over 4  
   
 Returns:  
   
 =item C<boolean> - true or false based on logical statement.  
   
 =back  
   
 =cut  
   
 sub wanted ($)  
   {  
     (($dev,$ino,$mode,$nlink,$uid,$gid) = lstat($_)) &&      (($dev,$ino,$mode,$nlink,$uid,$gid) = lstat($_)) &&
     -f $_ &&          -f _ &&
     /^.*\.meta$/ && !/^.+\.\d+\.[^\.]+\.meta$/ &&          /^.*\.meta$/ && !/^.+\.\d+\.[^\.]+\.meta$/ &&
     push(@metalist,$File::Find::dir.'/'.$_);          push(@metalist,"$dir/$_");
   }  }
   
 =pod  
   
 B<get_metadata_from_file> - read xml-tagged file and return parsed metadata.  
   
 I<Note that this is significantly altered from a subroutine present in lonnet.>  
   
 =over 4  
   
 Parameters:  
   
 =item I<$file> - a path.to the file.  
   
 =back  
   
 =over 4  
   
 Returns:  
   
 =item C<hash reference> - a hash array (keys and values).  
   
 =back  
   
 =cut  
   
 sub get_metadata_from_file ($)  
   {  
     my ($filename) = @_;  
     my %metatable; # Used to store return value of hash-tabled metadata.  
     $filename = &declutter($filename); # Remove non-identifying filesystem info  
     my $uri = ''; # The URI is not relevant in this scenario.  
     unless ($filename =~ m/\.meta$/) # Unless ending with .meta.  
       {  
  $filename .= '.meta'; # Append a .meta suffix.  
       }  
     # Get the file contents.  
     my $metadata_string =  
  &get_file_contents($perlvar{'lonDocRoot'}.'/res/'.$filename);  
   
     # Parse the file based on its XML tags.  
     my $parser = HTML::TokeParser->new(\$metadata_string);  
     my $token;  
     while ($token = $parser->get_token) # Loop through tokens.  
       {  
  if ($token->[0] eq 'S') # If it is a start token.  
   {  
     my $entry = $token->[1];  
     my $unikey = $entry; # A unique identifier for this xml tag key.  
     if (defined($token->[2]->{'part'}))  
       {   
  $unikey .= '_'.$token->[2]->{'part'};   
       }  
     if (defined($token->[2]->{'name'}))  
       {   
  $unikey .= '_'.$token->[2]->{'name'};   
       }  
     # Append $unikey to metatable's keys entry.  
     if ($metatable{$uri.'keys'})  
       {  
  $metatable{$uri.'keys'} .= ','.$unikey;  
       }  
     else  
       {  
  $metatable{$uri.'keys'} = $unikey;  
       }  
     # Insert contents into metatable entry for the unikey.  
     foreach my $t3 (@{$token->[3]})  
       {  
  $metatable{$uri.''.$unikey.'.'.$_} = $token->[2]->{$t3};  
       }  
     # If there was no text contained inside the tags, set = default.  
     unless  
       (  
         $metatable{$uri.''.$unikey} = $parser->get_text('/'.$entry)  
       )  
       {  
  $metatable{$uri.''.$unikey} =  
     $metatable{$uri.''.$unikey.'.default'};  
       }  
   }  
       }  
     # Return with a key-value table of XML tags and their tag contents.  
     return(\%metatable);  
   }  
   
 =pod  
   
 B<get_file_contents> - returns either the contents of the file or a -1.  
   
 =over 4  
   
 Parameters:  
   
 =item I<$file> - a complete filesystem path.to the file.  
   
 =back  
   
 =over 4  
   
 Returns:  
   
 =item C<string> - file contents or a -1.  
   
 =back  
   
 =cut  
   
 sub get_file_contents ($)  
   {  
     my $file = shift(@_);  
   
     # If file does not exist, then return a -1 value.  
     unless (-e $file)  
       {  
  return(-1);  
       }  
   
     # Read in file contents.  
     my $file_handle = IO::File->new($file);  
     my $file_contents = '';  
     while (<$file_handle>)  
       {  
  $file_contents .= $_;  
       }  
   
     # Return file contents.  
     return($file_contents);  
   }  
   
 =pod  
   
 B<declutter> - Declutters URLs (remove extraneous prefixed filesystem path).  
   
 =over 4  
   
 Parameters:  
   
 =item I<$filesystem_path> - a complete filesystem path.  
   
 =back  
   
 =over 4  
   
 Returns:  
   
 =item C<string> - remnants of the filesystem path (beginning portion removed).  
   
 =back  
   
 =cut  
   
 sub declutter  
   {  
     my $filesystem_path = shift(@_);  
   
     # Remove beginning portions of the filesystem path.  
     $filesystem_path =~ s/^$perlvar{'lonDocRoot'}//;  
     $filesystem_path =~ s!^/!!;  
     $filesystem_path =~ s!^res/!!;  
   
     # Return what is remaining for the filesystem path.  
     return($filesystem_path);  
   }  
   
 =pod  
   
 B<query_home_server_status> - Is this the home server of an author's directory?  
   
 =over 4  
   
 Parameters:  
   
 =item I<$author_filesystem_path> - directory path for a user.  
   
 =back  
   
 =over 4  
   
 Returns:  
   
 =item C<boolean> - 1 if true; 0 if false.  
   
 =back  
   
 =cut  
   
 sub query_home_server_status ($)  
   {  
     my $author_filesystem_path = shift(@_);  
   
     # Remove beginning portion of this filesystem path.  
     $author_filesystem_path =~ s!/home/httpd/html/res/([^/]*)/([^/]*).*!$1/$2!;  
   
     # Construct path to the author's ordinary user directory.  
     my ($user_domain,$username) = split(m!/!,$author_filesystem_path);  
     my $user_directory_path = construct_path_to_user_directory($user_domain,  
        $username);  
   
     # Return status of whether the user directory path is defined.  
     if (-e $user_directory_path)  
       {  
  return(1); # True.  
       }  
     else  
       {  
         return(0); # False.  
       }  
   }  
   
 =pod  
   
 B<construct_path_to_user_directory> ($$) - makes a filesystem path to user dir.  
   
 =over 4  
   
 Parameters:  
   
 =item I<$user_domain> - the loncapa domain of the user.  
   
 =item I<$username> - the unique username (user id) of the user.  
   
 =back  
   
 =over 4  
   
 Returns:  # ---------------  Read loncapa_apache.conf and loncapa.conf and get variables
   my $perlvarref=LONCAPA::Configuration::read_conf('loncapa.conf');
   my %perlvar=%{$perlvarref};
   undef $perlvarref; # remove since sensitive and not needed
   delete $perlvar{'lonReceipt'}; # remove since sensitive and not needed
   
 =item C<string> - representing the path on the filesystem.  # ------------------------------------- Only run if machine is a library server
   exit unless $perlvar{'lonRole'} eq 'library';
   
 =back  # ----------------------------- Make sure this process is running from user=www
   
 =cut  my $wwwid=getpwnam('www');
   if ($wwwid!=$<) {
       $emailto="$perlvar{'lonAdmEMail'},$perlvar{'lonSysEMail'}";
       $subj="LON: $perlvar{'lonHostID'} User ID mismatch";
       system("echo 'User ID mismatch. searchcat.pl must be run as user www.' |\
    mailto $emailto -s '$subj' > /dev/null");
       exit 1;
   }
   
 sub construct_path_to_user_directory ($$)  
   {  
     my ($user_domain,$username) = @_;  
   
     # Untaint.  
     $user_domain =~ s/\W//g;  
     $username =~ s/\W//g;  
   
     # Create three levels of sub-directoried filesystem path  
     # based on the first three characters of the username.  
     my $sub_filesystem_path = $username.'__';  
     $sub_filesystem_path =~ s!(.)(.)(.).*!$1/$2/$3/!;  
   
     # Use the sub-directoried levels and other variables to generate  
     # the complete filesystem path.  
     my $complete_filesystem_path =  
  join('/',($perlvar{'lonUsersDir'},  
   $user_domain,  
   $sub_filesystem_path,  
   $username));  
   
     # Return the complete filesystem path.  
     return($complete_filesystem_path);  
   }  
   
 =pod  # ---------------------------------------------------------- We are in business
   
 B<sql_formatted_time> (@) - turns seconds since epoch into datetime sql format.  open(LOG,'>'.$perlvar{'lonDaemons'}.'/logs/searchcat.log');
   print LOG '==== Searchcat Run '.localtime()."====\n\n";
   my $dbh;
   # ------------------------------------- Make sure that database can be accessed
   {
       unless (
       $dbh = DBI->connect("DBI:mysql:loncapa","www",$perlvar{'lonSqlAccess'},{ RaiseError =>0,PrintError=>0})
       ) { 
    print LOG "Cannot connect to database!\n";
    exit;
       }
       my $make_metadata_table = "CREATE TABLE IF NOT EXISTS metadata (".
           "title TEXT, author TEXT, subject TEXT, url TEXT, keywords TEXT, ".
           "version TEXT, notes TEXT, abstract TEXT, mime TEXT, language TEXT, ".
           "creationdate DATETIME, lastrevisiondate DATETIME, owner TEXT, ".
           "copyright TEXT, FULLTEXT idx_title (title), ".
           "FULLTEXT idx_author (author), FULLTEXT idx_subject (subject), ".
           "FULLTEXT idx_url (url), FULLTEXT idx_keywords (keywords), ".
           "FULLTEXT idx_version (version), FULLTEXT idx_notes (notes), ".
           "FULLTEXT idx_abstract (abstract), FULLTEXT idx_mime (mime), ".
           "FULLTEXT idx_language (language), FULLTEXT idx_owner (owner), ".
           "FULLTEXT idx_copyright (copyright)) TYPE=MYISAM";
       # It would sure be nice to have some logging mechanism.
       $dbh->do($make_metadata_table);
   }
   
 =over 4  # ------------------------------------------------------------- get .meta files
   opendir(RESOURCES,"$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}");
   my @homeusers = grep {
       &ishome("$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}/$_")
       } grep {!/^\.\.?$/} readdir(RESOURCES);
   closedir RESOURCES;
   foreach my $user (@homeusers) {
       print LOG "\n=== User: ".$user."\n\n";
       # Remove left-over db-files from potentially crashed searchcat run
       my $prodir=&propath($perlvar{'lonDefDomain'},$user);
       unlink($prodir.'/nohist_new_resevaldata.db');
       # Use find.pl
       undef @metalist;
       @metalist=();
       &find("$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}/$user");
       # -- process each file to get metadata and put into search catalog SQL
       # database.  Also, check to see if already there.
       # I could just delete (without searching first), but this works for now.
       foreach my $m (@metalist) {
           print LOG "- ".$m."\n";
           my $ref=&metadata($m);
           my $m2='/res/'.&declutter($m);
           $m2=~s/\.meta$//;
           &dynamicmeta($m2);
           my $q2="select * from metadata where url like binary '$m2'";
           my $sth = $dbh->prepare($q2);
           $sth->execute();
           my $r1=$sth->fetchall_arrayref;
           if (@$r1) {
               $sth=$dbh->prepare("delete from metadata where url like binary '$m2'");
               $sth->execute();
           }
           $sth=$dbh->prepare('insert into metadata values ('.
                              '"'.delete($ref->{'title'}).'"'.','.
                              '"'.delete($ref->{'author'}).'"'.','.
                              '"'.delete($ref->{'subject'}).'"'.','.
                              '"'.$m2.'"'.','.
                              '"'.delete($ref->{'keywords'}).'"'.','.
                              '"'.'current'.'"'.','.
                              '"'.delete($ref->{'notes'}).'"'.','.
                              '"'.delete($ref->{'abstract'}).'"'.','.
                              '"'.delete($ref->{'mime'}).'"'.','.
                              '"'.delete($ref->{'language'}).'"'.','.
                              '"'.sqltime(delete($ref->{'creationdate'})).'"'.','.
                              '"'.sqltime(delete($ref->{'lastrevisiondate'})).'"'.','.
                              '"'.delete($ref->{'owner'}).'"'.','.
                              '"'.delete($ref->{'copyright'}).'"'.')');
           $sth->execute();
       }
       
       # --------------------------------------------------- Clean up database
       # Need to, perhaps, remove stale SQL database records.
       # ... not yet implemented
           
       # ------------------------------------------- Copy over the new db-files
       system('mv '.$prodir.'/nohist_new_resevaldata.db '.
              $prodir.'/nohist_resevaldata.db');
   }
   # --------------------------------------------------- Close database connection
   $dbh->disconnect;
   print LOG "\n==== Searchcat completed ".localtime()." ====\n";
   close(LOG);
   exit 0;
   
 Parameters:  
   
 =item I<$epochtime> - time in seconds since epoch (may need to be sanitized).  
   
 =back  # =============================================================================
   
 =over 4  # ---------------------------------------------------------------- Get metadata
   # significantly altered from subroutine present in lonnet
   sub metadata {
       my ($uri,$what)=@_;
       my %metacache;
       $uri=&declutter($uri);
       my $filename=$uri;
       $uri=~s/\.meta$//;
       $uri='';
       unless ($metacache{$uri.'keys'}) {
           unless ($filename=~/\.meta$/) { $filename.='.meta'; }
    my $metastring=&getfile($perlvar{'lonDocRoot'}.'/res/'.$filename);
           my $parser=HTML::TokeParser->new(\$metastring);
           my $token;
           while ($token=$parser->get_token) {
               if ($token->[0] eq 'S') {
                   my $entry=$token->[1];
                   my $unikey=$entry;
                   if (defined($token->[2]->{'part'})) { 
                       $unikey.='_'.$token->[2]->{'part'}; 
                   }
                   if (defined($token->[2]->{'name'})) { 
                       $unikey.='_'.$token->[2]->{'name'}; 
                   }
                   if ($metacache{$uri.'keys'}) {
                       $metacache{$uri.'keys'}.=','.$unikey;
                   } else {
                       $metacache{$uri.'keys'}=$unikey;
                   }
                   map {
                       $metacache{$uri.''.$unikey.'.'.$_}=$token->[2]->{$_};
                   } @{$token->[3]};
                   unless (
                           $metacache{$uri.''.$unikey}=$parser->get_text('/'.$entry)
                           ) { $metacache{$uri.''.$unikey}=
                                   $metacache{$uri.''.$unikey.'.default'};
                           }
               }
           }
       }
       return \%metacache;
   }
   
 Returns:  # ------------------------------------------------------------ Serves up a file
   # returns either the contents of the file or a -1
   sub getfile {
       my $file=shift;
       if (! -e $file ) { return -1; };
       my $fh=IO::File->new($file);
       my $a='';
       while (<$fh>) { $a .=$_; }
       return $a;
   }
   
 =item C<string> - datetime sql formatted string.  # ------------------------------------------------------------- Declutters URLs
   sub declutter {
       my $thisfn=shift;
       $thisfn=~s/^$perlvar{'lonDocRoot'}//;
       $thisfn=~s/^\///;
       $thisfn=~s/^res\///;
       return $thisfn;
   }
   
 =back  # --------------------------------------- Is this the home server of an author?
   # (copied from lond, modification of the return value)
   sub ishome {
       my $author=shift;
       $author=~s/\/home\/httpd\/html\/res\/([^\/]*)\/([^\/]*).*/$1\/$2/;
       my ($udom,$uname)=split(/\//,$author);
       my $proname=propath($udom,$uname);
       if (-e $proname) {
    return 1;
       } else {
           return 0;
       }
   }
   
 =cut  # -------------------------------------------- Return path to profile directory
   # (copied from lond)
   sub propath {
       my ($udom,$uname)=@_;
       $udom=~s/\W//g;
       $uname=~s/\W//g;
       my $subdir=$uname.'__';
       $subdir =~ s/(.)(.)(.).*/$1\/$2\/$3/;
       my $proname="$perlvar{'lonUsersDir'}/$udom/$subdir/$uname";
       return $proname;
   } 
   
 sub sql_formatted_time ($)  # ---------------------------- convert 'time' format into a datetime sql format
   {  sub sqltime {
     # Sanitize the time argument and convert to localtime array.  
     my ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =      my ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
  localtime(&sanitize_time(shift(@_)));   localtime(&unsqltime(@_[0]));
       $mon++; $year+=1900;
     # Convert month from (0..11) to (1..12).      return "$year-$mon-$mday $hour:$min:$sec";
     $mon += 1;  }
   
     # Make the year compatible with A.D. specification.  
     $year += 1900;  
   
     # Return a date which is compatible with MySQL's "DATETIME" format.  
     return(join('-',($year,$mon,$mday)).  
    ' '.  
    join(':',($hour,$min,$sec))  
    );  
   }  
   
   
 # ==================================== The following two subroutines are needed  
 #                 for accommodating incorrect time formats inside the metadata.  
   
 =pod  
   
 B<make_seconds_since_epoch> (@) - turns time metadata into seconds since epoch.  
   
 =over 4  
   
 Parameters:  
   
 =item I<%time_metadata> - a key-value listing characterizing month, year, etc.  
   
 =back  
   
 =over 4  
   
 Returns:  
   
 =item C<integer> - seconds since epoch.  
   
 =back  
   
 =cut  
   
 sub make_seconds_since_epoch (@)  
   {  
     # Keytable of time metadata.  
     my %time_metadata = @_;  
   
     # Return seconds since the epoch (January 1, 1970, 00:00:00 UTC).  
     return(POSIX::mktime(  
  ($time_metadata{'seconds'},  
   $time_metadata{'minutes'},  
   $time_metadata{'hours'},  
   $time_metadata{'day'},  
   $time_metadata{'month'}-1,  
   $time_metadata{'year'}-1900,  
   0,  
   0,  
   $time_metadata{'dlsav'})  
  )  
    );  
   }  
   
 =pod  
   
 B<sanitize_time> - if time looks sql-formatted, make it seconds since epoch.  
   
 Somebody described this subroutine as  
 "retro-fixing of un-backward-compatible time format".  
   
 What this means, is that a part of this code expects to get UTC seconds  
 since the epoch (beginning of 1970).  Yet, some of the .meta files have  
 sql-formatted time strings (2001-04-01, etc.) instead of seconds-since-epoch  
 integers (e.g. 1044147435).  These time strings do not encode the timezone  
 and, in this sense, can be considered "un-backwards-compatible".  
   
 =over 4  
   
 Parameters:  
   
 =item I<$potentially_badformat_string> - string to "retro-fix".  
   
 =back  
   
 =over 4  
   
 Returns:  
   
 =item C<integer> - seconds since epoch.  
   
 =back  
   
 =cut  
   
 sub sanitize_time ($)  
   {  
     my $timestamp = shift(@_);  
     # If timestamp is in this unexpected format....  
     if ($timestamp =~ /^(\d+)\-(\d+)\-(\d+)\s+(\d+)\:(\d+)\:(\d+)$/)  
       {  
  # then convert into seconds since epoch (the expected format).  
  $timestamp = &make_seconds_since_epoch(  
        'year' => $1,  
        'month' => $2,  
        'day' => $3,  
        'hours' => $4,  
        'minutes' => $5,  
        'seconds' => $6  
        );  
       }  
     # Otherwise we assume timestamp to be as expected.  
     return($timestamp);  
   }  
   
 =pod  
   
 =head1 AUTHOR  sub maketime {
       my %th=@_;
       return POSIX::mktime(($th{'seconds'},$th{'minutes'},$th{'hours'},
                             $th{'day'},$th{'month'}-1,
                             $th{'year'}-1900,0,0,$th{'dlsav'}));
   }
   
 Written to help the loncapa project.  
   
 Scott Harrison, sharrison@users.sourceforge.net  #########################################
   #
   # Retro-fixing of un-backward-compatible time format
   
 This is distributed under the same terms as loncapa (i.e. "freeware").  sub unsqltime {
       my $timestamp=shift;
       if ($timestamp=~/^(\d+)\-(\d+)\-(\d+)\s+(\d+)\:(\d+)\:(\d+)$/) {
           $timestamp=&maketime('year'=>$1,'month'=>$2,'day'=>$3,
                                'hours'=>$4,'minutes'=>$5,'seconds'=>$6);
       }
       return $timestamp;
   }
   
 =cut  

Removed from v.1.30  
changed lines
  Added in v.1.33


FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>