Diff for /loncom/metadata_database/searchcat.pl between versions 1.26 and 1.57

version 1.26, 2003/01/04 15:04:12 version 1.57, 2004/04/12 21:11:45
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  
 #  
 ###  ###
   
 # This script goes through a LON-CAPA resource  =pod
 # directory and gathers metadata.  
 # The metadata is entered into a SQL database.  =head1 NAME
   
   B<searchcat.pl> - put authoritative filesystem data into sql database.
   
   =head1 SYNOPSIS
   
   Ordinarily this script is to be called from a loncapa cron job
   (CVS source location: F<loncapa/loncom/cron/loncapa>; typical
   filesystem installation location: F</etc/cron.d/loncapa>).
   
   Here is the cron job entry.
   
   C<# Repopulate and refresh the metadata database used for the search catalog.>
   C<10 1 * * 7    www    /home/httpd/perl/searchcat.pl>
   
   This script only allows itself to be run as the user C<www>.
   
   =head1 DESCRIPTION
   
   This script goes through a loncapa resource directory and gathers metadata.
   The metadata is entered into a SQL database.
   
   This script also does general database maintenance such as reformatting
   the C<loncapa:metadata> table if it is deprecated.
   
   This script evaluates dynamic metadata from the authors'
   F<nohist_resevaldata.db> database file in order to store it in MySQL.
   
   This script is playing an increasingly important role for a loncapa
   library server.  The proper operation of this script is critical for a smooth
   and correct user experience.
   
   =cut
   
   use strict;
   
   use DBI;
 use lib '/home/httpd/lib/perl/';  use lib '/home/httpd/lib/perl/';
 use LONCAPA::Configuration;  use LONCAPA::Configuration;
   use LONCAPA::lonmetadata;
   
   use Getopt::Long;
 use IO::File;  use IO::File;
 use HTML::TokeParser;  use HTML::TokeParser;
 use DBI;  
 use GDBM_File;  use GDBM_File;
 use POSIX qw(strftime mktime);  use POSIX qw(strftime mktime);
   
 my @metalist;  use File::Find;
   
   
 # ----------------------------------------------------- Un-Escape Special Chars  #
   # Set up configuration options
   my ($simulate,$oneuser,$help,$verbose,$logfile,$debug);
   GetOptions (
               'help'     => \$help,
               'simulate' => \$simulate,
               'only=s'   => \$oneuser,
               'verbose=s'  => \$verbose,
               'debug' => \$debug,
               );
   
   if ($help) {
       print <<"ENDHELP";
   $0
   Rebuild and update the LON-CAPA metadata database. 
   Options:
       -help          Print this help
       -simulate      Do not modify the database.
       -only=user     Only compute for the given user.  Implies -simulate   
       -verbose=val   Sets logging level, val must be a number
       -debug         Turns on debugging output
   ENDHELP
       exit 0;
   }
   
   if (! defined($debug)) {
       $debug = 0;
   }
   
   if (! defined($verbose)) {
       $verbose = 0;
   }
   
   if (defined($oneuser)) {
       $simulate=1;
   }
   
   ##
   ## Use variables for table names so we can test this routine a little easier
   my $oldname = 'metadata';
   my $newname = 'newmetadata';
   
 sub unescape {  #
     my $str=shift;  # Read loncapa_apache.conf and loncapa.conf
     $str =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C",hex($1))/eg;  my $perlvarref=LONCAPA::Configuration::read_conf('loncapa.conf');
     return $str;  my %perlvar=%{$perlvarref};
   undef $perlvarref;
   delete $perlvar{'lonReceipt'}; # remove since sensitive (really?) & not needed
   #
   # Only run if machine is a library server
   exit if ($perlvar{'lonRole'} ne 'library');
   #
   #  Make sure this process is running from user=www
   my $wwwid=getpwnam('www');
   if ($wwwid!=$<) {
       my $emailto="$perlvar{'lonAdmEMail'},$perlvar{'lonSysEMail'}";
       my $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;
 }  }
   #
   # Let people know we are running
   open(LOG,'>'.$perlvar{'lonDaemons'}.'/logs/searchcat.log');
   &log(0,'==== Searchcat Run '.localtime()."====");
   
 # -------------------------------------------------------- Escape Special Chars  
   
 sub escape {  if ($debug) {
     my $str=shift;      &log(0,'simulating') if ($simulate);
     $str =~ s/(\W)/"%".unpack('H2',$1)/eg;      &log(0,'only processing user '.$oneuser) if ($oneuser);
     return $str;      &log(0,'verbosity level = '.$verbose);
   }
   #
   # Connect to database
   my $dbh;
   if (! ($dbh = DBI->connect("DBI:mysql:loncapa","www",$perlvar{'lonSqlAccess'},
                             { RaiseError =>0,PrintError=>0}))) {
       &log(0,"Cannot connect to database!");
       die "MySQL Error: Cannot connect to database!\n";
   }
   # This can return an error and still be okay, so we do not bother checking.
   # (perhaps it should be more robust and check for specific errors)
   $dbh->do('DROP TABLE IF EXISTS '.$newname);
   #
   # Create the new table
   my $request = &LONCAPA::lonmetadata::create_metadata_storage($newname);
   $dbh->do($request);
   if ($dbh->err) {
       $dbh->disconnect();
       &log(0,"MySQL Error Create: ".$dbh->errstr);
       die $dbh->errstr;
   }
   #
   # find out which users we need to examine
   my $dom = $perlvar{'lonDefDomain'};
   opendir(RESOURCES,"$perlvar{'lonDocRoot'}/res/$dom");
   my @homeusers = 
       grep {
           &ishome("$perlvar{'lonDocRoot'}/res/$dom/$_");
       } grep { 
           !/^\.\.?$/;
       } readdir(RESOURCES);
   closedir RESOURCES;
   #
   if ($oneuser) {
       @homeusers=($oneuser);
   }
   #
   # Loop through the users
   foreach my $user (@homeusers) {
       &log(0,"=== User: ".$user);
       &process_dynamic_metadata($user,$dom);
       #
       # Use File::Find to get the files we need to read/modify
       find(
            {preprocess => \&only_meta_files,
   #          wanted     => \&print_filename,
   #          wanted     => \&log_metadata,
             wanted     => \&process_meta_file,
             }, 
            "$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}/$user");
   }
   #
   # Rename the table
   if (! $simulate) {
       $dbh->do('DROP TABLE IF EXISTS '.$oldname);
       if (! $dbh->do('RENAME TABLE '.$newname.' TO '.$oldname)) {
           &log(0,"MySQL Error Rename: ".$dbh->errstr);
           die $dbh->errstr;
       } else {
           &log(1,"MySQL table rename successful.");
       }
 }  }
   
   if (! $dbh->disconnect) {
       &log(0,"MySQL Error Disconnect: ".$dbh->errstr);
       die $dbh->errstr;
   }
   ##
   ## Finished!
   &log(0,"==== Searchcat completed ".localtime()." ====");
   close(LOG);
   
 # ------------------------------------------- Code to evaluate dynamic metadata  &write_type_count();
   &write_copyright_count();
   
 sub dynamicmeta {  exit 0;
   
     my $url=&declutter(shift);  ##
     $url=~s/\.meta$//;  ## Status logging routine.  Inputs: $level, $message
     my %returnhash=();  ## 
     my ($adomain,$aauthor)=($url=~/^(\w+)\/(\w+)\//);  ## $level 0 should be used for normal output and error messages
     my $prodir=&propath($adomain,$aauthor);  ##
     if ((tie(%evaldata,'GDBM_File',  ## $message does not need to end with \n.  In the case of errors
             $prodir.'/nohist_resevaldata.db',&GDBM_READER(),0640)) &&  ## the message should contain as much information as possible to
         (tie(%newevaldata,'GDBM_File',  ## help in diagnosing the problem.
             $prodir.'/nohist_new_resevaldata.db',&GDBM_WRCREAT(),0640))) {  ##
        my %sum=();  sub log {
        my %cnt=();      my ($level,$message)=@_;
        my %listitems=('count'        => 'add',      $level = 0 if (! defined($level));
                       'course'       => 'add',      if ($verbose >= $level) {
                       'avetries'     => 'avg',          print LOG $message.$/;
                       'stdno'        => 'add',      }
                       'difficulty'   => 'avg',  
                       'clear'        => 'avg',  
                       'technical'    => 'avg',  
                       'helpful'      => 'avg',  
                       'correct'      => 'avg',  
                       'depth'        => 'avg',  
                       'comments'     => 'app',  
                       'usage'        => 'cnt'  
                       );  
        my $regexp=$url;  
        $regexp=~s/(\W)/\\$1/g;  
        $regexp='___'.$regexp.'___([a-z]+)$';  
        foreach (keys %evaldata) {  
  my $key=&unescape($_);  
  if ($key=~/$regexp/) {  
     my $ctype=$1;  
             if (defined($cnt{$ctype})) {   
                $cnt{$ctype}++;   
             } else {   
                $cnt{$ctype}=1;   
             }  
             unless ($listitems{$ctype} eq 'app') {  
                if (defined($sum{$ctype})) {  
                   $sum{$ctype}+=$evaldata{$_};  
           } else {  
                   $sum{$ctype}=$evaldata{$_};  
        }  
             } else {  
                if (defined($sum{$ctype})) {  
                   if ($evaldata{$_}) {  
                      $sum{$ctype}.='<hr>'.$evaldata{$_};  
           }  
         } else {  
              $sum{$ctype}=''.$evaldata{$_};  
        }  
     }  
     if ($ctype ne 'count') {  
        $newevaldata{$_}=$evaldata{$_};  
    }  
  }  
       }  
       foreach (keys %cnt) {  
          if ($listitems{$_} eq 'avg') {  
      $returnhash{$_}=int(($sum{$_}/$cnt{$_})*100.0+0.5)/100.0;  
          } elsif ($listitems{$_} eq 'cnt') {  
              $returnhash{$_}=$cnt{$_};  
          } else {  
              $returnhash{$_}=$sum{$_};  
          }  
      }  
      if ($returnhash{'count'}) {  
          my $newkey=$$.'_'.time.'_searchcat___'.&escape($url).'___count';  
          $newevaldata{$newkey}=$returnhash{'count'};  
      }  
      untie(%evaldata);  
      untie(%newevaldata);  
    }  
    return %returnhash;  
 }  
     
 # ----------------- Code to enable 'find' subroutine listing of the .meta files  
 require "find.pl";  
 sub wanted {  
     (($dev,$ino,$mode,$nlink,$uid,$gid) = lstat($_)) &&  
     -f _ &&  
     /^.*\.meta$/ && !/^.+\.\d+\.[^\.]+\.meta$/ &&  
     push(@metalist,"$dir/$_");  
 }  }
   
 # ---------------  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  ###          File::Find support routines             ###
 delete $perlvar{'lonReceipt'}; # remove since sensitive and not needed  ###                                                  ###
   ########################################################
   ########################################################
   ##
   ## &only_meta_files
   ##
   ## Called by File::Find.
   ## Takes a list of files/directories in and returns a list of files/directories
   ## to search.
   sub only_meta_files {
       my @PossibleFiles = @_;
       my @ChosenFiles;
       foreach my $file (@PossibleFiles) {
           if ( ($file =~ /\.meta$/ &&            # Ends in meta
                 $file !~ /\.\d+\.[^\.]+\.meta$/  # is not for a prior version
                ) || (-d $file )) { # directories are okay
                    # but we do not want /. or /..
               push(@ChosenFiles,$file);
           }
       }
       return @ChosenFiles;
   }
   
 # ------------------------------------- Only run if machine is a library server  ##
 exit unless $perlvar{'lonRole'} eq 'library';  ##
   ## Debugging routines, use these for 'wanted' in the File::Find call
   ##
   sub print_filename {
       my ($file) = $_;
       my $fullfilename = $File::Find::name;
       if ($debug) {
           if (-d $file) {
               &log(5," Got directory ".$fullfilename);
           } else {
               &log(5," Got file ".$fullfilename);
           }
       }
       $_=$file;
   }
   
 # ---------------------------------------------------------- We are in business  sub log_metadata {
       my ($file) = $_;
       my $fullfilename = $File::Find::name;
       return if (-d $fullfilename); # No need to do anything here for directories
       if ($debug) {
           &log(6,$fullfilename);
           my $ref=&metadata($fullfilename);
           if (! defined($ref)) {
               &log(6,"    No data");
               return;
           }
           while (my($key,$value) = each(%$ref)) {
               &log(6,"    ".$key." => ".$value);
           }
           &count_copyright($ref->{'copyright'});
       }
       $_=$file;
   }
   
 open(LOG,'>'.$perlvar{'lonDaemons'}.'/logs/searchcat.log');  ##
 print LOG '==== Searchcat Run '.localtime()."====\n\n";  ## process_meta_file
 my $dbh;  ##   Called by File::Find. 
 # ------------------------------------- Make sure that database can be accessed  ##   Only input is the filename in $_.  
 {  sub process_meta_file {
     unless (      my ($file) = $_;
     $dbh = DBI->connect("DBI:mysql:loncapa","www",$perlvar{'lonSqlAccess'},{ RaiseError =>0,PrintError=>0})      my $filename = $File::Find::name; # full filename
     ) {       return if (-d $filename); # No need to do anything here for directories
  print LOG "Cannot connect to database!\n";      #
  exit;      &log(3,$filename) if ($debug);
     }      #
     my $make_metadata_table = "CREATE TABLE IF NOT EXISTS metadata (".      my $ref=&metadata($filename);
         "title TEXT, author TEXT, subject TEXT, url TEXT, keywords TEXT, ".      #
         "version TEXT, notes TEXT, abstract TEXT, mime TEXT, language TEXT, ".      # $url is the original file url, not the metadata file
         "creationdate DATETIME, lastrevisiondate DATETIME, owner TEXT, ".      my $url='/res/'.&declutter($filename);
         "copyright TEXT, FULLTEXT idx_title (title), ".      $url=~s/\.meta$//;
         "FULLTEXT idx_author (author), FULLTEXT idx_subject (subject), ".      &log(3,"    ".$url) if ($debug);
         "FULLTEXT idx_url (url), FULLTEXT idx_keywords (keywords), ".      #
         "FULLTEXT idx_version (version), FULLTEXT idx_notes (notes), ".      # Ignore some files based on their metadata
         "FULLTEXT idx_abstract (abstract), FULLTEXT idx_mime (mime), ".      if ($ref->{'obsolete'}) { 
         "FULLTEXT idx_language (language), FULLTEXT idx_owner (owner), ".          &log(3,"obsolete") if ($debug);
         "FULLTEXT idx_copyright (copyright)) TYPE=MYISAM";          return; 
     # It would sure be nice to have some logging mechanism.      }
     $dbh->do($make_metadata_table);      &count_copyright($ref->{'copyright'});
 }      if ($ref->{'copyright'} eq 'private') { 
           &log(3,"private") if ($debug);
 # ------------------------------------------------------------- get .meta files          return; 
 opendir(RESOURCES,"$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}");      }
 my @homeusers=grep      #
           {&ishome("$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}/$_")}      # Find the dynamic metadata
           grep {!/^\.\.?$/} readdir(RESOURCES);      my %dyn;
 closedir RESOURCES;      if ($url=~ m:/default$:) {
 foreach my $user (@homeusers) {          $url=~ s:/default$:/:;
     print LOG "\n=== User: ".$user."\n\n";          &log(3,"Skipping dynamic data") if ($debug);
 # Remove left-over db-files from potentially crashed searchcat run      } else {
     my $prodir=&propath($perlvar{'lonDefDomain'},$user);          &log(3,"Retrieving dynamic data") if ($debug);
     unlink($prodir.'/nohist_new_resevaldata.db');          %dyn=&get_dynamic_metadata($url);
 # Use find.pl          &count_type($url);
     undef @metalist;      }
     @metalist=();      #
     &find("$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}/$user");      $ref->{'creationdate'}     = &sqltime($ref->{'creationdate'});
       $ref->{'lastrevisiondate'} = &sqltime($ref->{'lastrevisiondate'});
 # -- process each file to get metadata and put into search catalog SQL database      my %Data = (
 # Also, check to see if already there.                  %$ref,
 # I could just delete (without searching first), but this works for now.                  %dyn,
 foreach my $m (@metalist) {                  'url'=>$url,
     print LOG "- ".$m."\n";                  'version'=>'current');
     my $ref=&metadata($m);      if (! $simulate) {
     my $m2='/res/'.&declutter($m);          my ($count,$err) = &LONCAPA::lonmetadata::store_metadata($dbh,$newname,
     $m2=~s/\.meta$//;                                                                   \%Data);
     &dynamicmeta($m2);          if ($err) {
     my $q2="select * from metadata where url like binary '$m2'";              &log(0,"MySQL Error Insert: ".$err);
     my $sth = $dbh->prepare($q2);          }
     $sth->execute();          if ($count < 1) {
     my $r1=$sth->fetchall_arrayref;              &log(0,"Unable to insert record into MySQL database for $url");
     if (@$r1) {          }
  $sth=$dbh->prepare("delete from metadata where url like binary '$m2'");      }
         $sth->execute();      #
     }      # Reset $_ before leaving
     $sth=$dbh->prepare('insert into metadata values ('.      $_ = $file;
   '"'.delete($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 '.  
          $prodir.'/nohist_resevaldata.db');  
     system('chown www:www '.$prodir.'/nohist_resevaldata.db');  
 }  
 # --------------------------------------------------- Close database connection  
 $dbh->disconnect;  
 print LOG "\n==== Searchcat completed ".localtime()." ====\n";  
 close(LOG);  
 exit 0;  
 # =============================================================================  
   
 # ---------------------------------------------------------------- Get metadata  ########################################################
 # significantly altered from subroutine present in lonnet  ########################################################
   ###                                                  ###
   ###  &metadata($uri)                                 ###
   ###   Retrieve metadata for the given file           ###
   ###                                                  ###
   ########################################################
   ########################################################
 sub metadata {  sub metadata {
     my ($uri,$what)=@_;      my ($uri)=@_;
     my %metacache;      my %metacache=();
     $uri=&declutter($uri);      $uri=&declutter($uri);
     my $filename=$uri;      my $filename=$uri;
     $uri=~s/\.meta$//;      $uri=~s/\.meta$//;
     $uri='';      $uri='';
     unless ($metacache{$uri.'keys'}) {      if ($filename !~ /\.meta$/) { 
         unless ($filename=~/\.meta$/) { $filename.='.meta'; }          $filename.='.meta';
  my $metastring=&getfile($perlvar{'lonDocRoot'}.'/res/'.$filename);      }
         my $parser=HTML::TokeParser->new(\$metastring);      my $metastring=&getfile($perlvar{'lonDocRoot'}.'/res/'.$filename);
         my $token;      return undef if (! defined($metastring));
         while ($token=$parser->get_token) {      my $parser=HTML::TokeParser->new(\$metastring);
            if ($token->[0] eq 'S') {      my $token;
       my $entry=$token->[1];      while ($token=$parser->get_token) {
               my $unikey=$entry;          if ($token->[0] eq 'S') {
               if (defined($token->[2]->{'part'})) {               my $entry=$token->[1];
                  $unikey.='_'.$token->[2]->{'part'};               my $unikey=$entry;
       }              if (defined($token->[2]->{'part'})) { 
               if (defined($token->[2]->{'name'})) {                   $unikey.='_'.$token->[2]->{'part'}; 
                  $unikey.='_'.$token->[2]->{'name'};               }
       }              if (defined($token->[2]->{'name'})) { 
               if ($metacache{$uri.'keys'}) {                  $unikey.='_'.$token->[2]->{'name'}; 
                  $metacache{$uri.'keys'}.=','.$unikey;              }
               } else {              if ($metacache{$uri.'keys'}) {
                  $metacache{$uri.'keys'}=$unikey;                  $metacache{$uri.'keys'}.=','.$unikey;
       }              } else {
               map {                  $metacache{$uri.'keys'}=$unikey;
   $metacache{$uri.''.$unikey.'.'.$_}=$token->[2]->{$_};              }
               } @{$token->[3]};              foreach ( @{$token->[3]}) {
               unless (                  $metacache{$uri.''.$unikey.'.'.$_}=$token->[2]->{$_};
                  $metacache{$uri.''.$unikey}=$parser->get_text('/'.$entry)              } 
       ) { $metacache{$uri.''.$unikey}=              if (! ($metacache{$uri.''.$unikey}=$parser->get_text('/'.$entry))){
       $metacache{$uri.''.$unikey.'.default'};                  $metacache{$uri.''.$unikey} = 
       }                      $metacache{$uri.''.$unikey.'.default'};
           }              }
        }          } # End of ($token->[0] eq 'S')
     }      }
     return \%metacache;      return \%metacache;
 }  }
   
 # ------------------------------------------------------------ Serves up a file  ##
 # returns either the contents of the file or a -1  ## &getfile($filename)
   ##   Slurps up an entire file into a scalar.  
   ##   Returns undef if the file does not exist
 sub getfile {  sub getfile {
   my $file=shift;      my $file = shift();
   if (! -e $file ) { return -1; };      if (! -e $file ) { 
   my $fh=IO::File->new($file);          return undef; 
   my $a='';      }
   while (<$fh>) { $a .=$_; }      my $fh=IO::File->new($file);
   return $a      my $contents = '';
       while (<$fh>) { 
           $contents .= $_;
       }
       return $contents;
 }  }
   
 # ------------------------------------------------------------- Declutters URLs  ########################################################
 sub declutter {  ########################################################
     my $thisfn=shift;  ###                                                  ###
     $thisfn=~s/^$perlvar{'lonDocRoot'}//;  ###    Dynamic Metadata                              ###
     $thisfn=~s/^\///;  ###                                                  ###
     $thisfn=~s/^res\///;  ########################################################
     return $thisfn;  ########################################################
   ##
   ## Dynamic metadata description
   ##
   ##   Field             Type
   ##-----------------------------------------------------------
   ##   count             integer
   ##   course            integer
   ##   course_list       comma seperated list of course ids
   ##   avetries          real                                
   ##   avetries_list     comma seperated list of real numbers
   ##   stdno             real
   ##   stdno_list        comma seperated list of real numbers
   ##   usage             integer   
   ##   usage_list        comma seperated list of resources
   ##   goto              scalar
   ##   goto_list         comma seperated list of resources
   ##   comefrom          scalar
   ##   comefrom_list     comma seperated list of resources
   ##   difficulty        real
   ##   difficulty_list   comma seperated list of real numbers
   ##   sequsage          scalar
   ##   sequsage_list     comma seperated list of resources
   ##   clear             real
   ##   technical         real
   ##   correct           real
   ##   helpful           real
   ##   depth             real
   ##   comments          html of all the comments made
   ##
   {
   
   my %DynamicData;
   my %Counts;
   
   sub process_dynamic_metadata {
       my ($user,$dom) = @_;
       undef(%DynamicData);
       undef(%Counts);
       #
       my $prodir = &propath($dom,$user);
       #
       # Read in the dynamic metadata
       my %evaldata;
       if (! tie(%evaldata,'GDBM_File',
                 $prodir.'/nohist_resevaldata.db',&GDBM_READER(),0640)) {
           return 0;
       }
       #
       %DynamicData = &LONCAPA::lonmetadata::process_reseval_data(\%evaldata);
       untie(%evaldata);
       #
       # Read in the access count data
       &log(7,'Reading access count data') if ($debug);
       my %countdata;
       if (! tie(%countdata,'GDBM_File',
                 $prodir.'/nohist_accesscount.db',&GDBM_READER(),0640)) {
           return 0;
       }
       while (my ($key,$count) = each(%countdata)) {
           next if ($key !~ /^$dom/);
           $key = &unescape($key);
           &log(8,'    Count '.$key.' = '.$count) if ($debug);
           $Counts{$key}=$count;
       }
       untie(%countdata);
       if ($debug) {
           &log(7,scalar(keys(%Counts)).
                " Counts read for ".$user."@".$dom);
           &log(7,scalar(keys(%DynamicData)).
                " Dynamic metadata read for ".$user."@".$dom);
       }
       #
       return 1;
   }
   
   sub get_dynamic_metadata {
       my ($url) = @_;
       $url =~ s:^/res/::;
       if (! exists($DynamicData{$url})) {
           &log(7,'    No dynamic data for '.$url) if ($debug);
           return ();
       }
       my %data = &LONCAPA::lonmetadata::process_dynamic_metadata($url,
                                                                  \%DynamicData);
       # find the count
       $data{'count'} = $Counts{$url};
       #
       # Log the dynamic metadata
       if ($debug) {
           while (my($k,$v)=each(%data)) {
               &log(8,"    ".$k." => ".$v);
           }
       }
       return %data;
   }
   
   } # End of %DynamicData and %Counts scope
   
   ########################################################
   ########################################################
   ###                                                  ###
   ###   Counts                                         ###
   ###                                                  ###
   ########################################################
   ########################################################
   {
   
   my %countext;
   
   sub count_type {
       my $file=shift;
       $file=~/\.(\w+)$/;
       my $ext=lc($1);
       $countext{$ext}++;
   }
   
   sub write_type_count {
       open(RESCOUNT,'>/home/httpd/html/lon-status/rescount.txt');
       while (my ($extension,$count) = each(%countext)) {
    print RESCOUNT $extension.'='.$count.'&';
       }
       print RESCOUNT 'time='.time."\n";
       close(RESCOUNT);
   }
   
   } # end of scope for %countext
   
   {
   
   my %copyrights;
   
   sub count_copyright {
       $copyrights{@_[0]}++;
   }
   
   sub write_copyright_count {
       open(COPYCOUNT,'>/home/httpd/html/lon-status/copyrightcount.txt');
       while (my ($copyright,$count) = each(%copyrights)) {
    print COPYCOUNT $copyright.'='.$count.'&';
       }
       print COPYCOUNT 'time='.time."\n";
       close(COPYCOUNT);
 }  }
   
 # --------------------------------------- Is this the home server of an author?  } # end of scope for %copyrights
 # (copied from lond, modification of the return value)  
   ########################################################
   ########################################################
   ###                                                  ###
   ###   Miscellanous Utility Routines                  ###
   ###                                                  ###
   ########################################################
   ########################################################
   ##
   ## &ishome($username)
   ##   Returns 1 if $username is a LON-CAPA author, 0 otherwise
   ##   (copied from lond, modification of the return value)
 sub ishome {  sub ishome {
     my $author=shift;      my $author=shift;
     $author=~s/\/home\/httpd\/html\/res\/([^\/]*)\/([^\/]*).*/$1\/$2/;      $author=~s/\/home\/httpd\/html\/res\/([^\/]*)\/([^\/]*).*/$1\/$2/;
Line 337  sub ishome { Line 611  sub ishome {
     }      }
 }  }
   
 # -------------------------------------------- Return path to profile directory  ##
 # (copied from lond)  ## &propath($udom,$uname)
   ##   Returns the path to the users LON-CAPA directory
   ##   (copied from lond)
 sub propath {  sub propath {
     my ($udom,$uname)=@_;      my ($udom,$uname)=@_;
     $udom=~s/\W//g;      $udom=~s/\W//g;
Line 349  sub propath { Line 625  sub propath {
     return $proname;      return $proname;
 }   } 
   
 # ---------------------------- convert 'time' format into a datetime sql format  ##
   ## &sqltime($timestamp)
   ##
   ## Convert perl $timestamp to MySQL time.  MySQL expects YYYY-MM-DD HH:MM:SS
   ##
 sub sqltime {  sub sqltime {
     my ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =      my ($time) = @_;
  localtime(&unsqltime(@_[0]));      my $mysqltime;
     $mon++; $year+=1900;      if ($time =~ 
     return "$year-$mon-$mday $hour:$min:$sec";          /(\d+)-(\d+)-(\d+) # YYYY-MM-DD
           \s                 # a space
           (\d+):(\d+):(\d+)  # HH:MM::SS
           /x ) { 
           # Some of the .meta files have the time in mysql
           # format already, so just make sure they are 0 padded and
           # pass them back.
           $mysqltime = sprintf('%04d-%02d-%02d %02d:%02d:%02d',
                                $1,$2,$3,$4,$5,$6);
       } elsif ($time =~ /^\d+$/) {
           my @TimeData = gmtime($time);
           # Alter the month to be 1-12 instead of 0-11
           $TimeData[4]++;
           # Alter the year to be from 0 instead of from 1900
           $TimeData[5]+=1900;
           $mysqltime = sprintf('%04d-%02d-%02d %02d:%02d:%02d',
                                @TimeData[5,4,3,2,1,0]);
       } elsif (! defined($time) || $time == 0) {
           $mysqltime = 0;
       } else {
           &log(0,"    sqltime:Unable to decode time ".$time);
           $mysqltime = 0;
       }
       return $mysqltime;
 }  }
   
 sub maketime {  ##
     my %th=@_;  ## &declutter($filename)
     return POSIX::mktime(  ##   Given a filename, returns a url for the filename.
         ($th{'seconds'},$th{'minutes'},$th{'hours'},  sub declutter {
          $th{'day'},$th{'month'}-1,$th{'year'}-1900,0,0,$th{'dlsav'}));      my $thisfn=shift;
       $thisfn=~s/^$perlvar{'lonDocRoot'}//;
       $thisfn=~s/^\///;
       $thisfn=~s/^res\///;
       return $thisfn;
 }  }
   
   ##
 #########################################  ## Escape / Unescape special characters
 #  sub unescape {
 # Retro-fixing of un-backward-compatible time format      my $str=shift;
       $str =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C",hex($1))/eg;
 sub unsqltime {      return $str;
     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;  
 }  }
   
   sub escape {
       my $str=shift;
       $str =~ s/(\W)/"%".unpack('H2',$1)/eg;
       return $str;
   }

Removed from v.1.26  
changed lines
  Added in v.1.57


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