File:  [LON-CAPA] / loncom / metadata_database / searchcat.pl
Revision 1.47: download - view: text, annotated - select for diffs
Thu Dec 25 04:06:52 2003 UTC (20 years, 6 months ago) by www
Branches: MAIN
CVS tags: HEAD
Saving my work - this produces and SQL error on Insert.

    1: #!/usr/bin/perl
    2: # The LearningOnline Network
    3: # searchcat.pl "Search Catalog" batch script
    4: #
    5: # $Id: searchcat.pl,v 1.47 2003/12/25 04:06:52 www Exp $
    6: #
    7: # Copyright Michigan State University Board of Trustees
    8: #
    9: # This file is part of the LearningOnline Network with CAPA (LON-CAPA).
   10: #
   11: # LON-CAPA is free software; you can redistribute it and/or modify
   12: # it under the terms of the GNU General Public License as published by
   13: # the Free Software Foundation; either version 2 of the License, or
   14: # (at your option) any later version.
   15: #
   16: # LON-CAPA is distributed in the hope that it will be useful,
   17: # but WITHOUT ANY WARRANTY; without even the implied warranty of
   18: # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   19: # GNU General Public License for more details.
   20: #
   21: # You should have received a copy of the GNU General Public License
   22: # along with LON-CAPA; if not, write to the Free Software
   23: # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
   24: #
   25: # /home/httpd/html/adm/gpl.txt
   26: #
   27: # http://www.lon-capa.org/
   28: #
   29: ###
   30: 
   31: =pod
   32: 
   33: =head1 NAME
   34: 
   35: B<searchcat.pl> - put authoritative filesystem data into sql database.
   36: 
   37: =head1 SYNOPSIS
   38: 
   39: Ordinarily this script is to be called from a loncapa cron job
   40: (CVS source location: F<loncapa/loncom/cron/loncapa>; typical
   41: filesystem installation location: F</etc/cron.d/loncapa>).
   42: 
   43: Here is the cron job entry.
   44: 
   45: C<# Repopulate and refresh the metadata database used for the search catalog.>
   46: C<10 1 * * 7    www    /home/httpd/perl/searchcat.pl>
   47: 
   48: This script only allows itself to be run as the user C<www>.
   49: 
   50: =head1 DESCRIPTION
   51: 
   52: This script goes through a loncapa resource directory and gathers metadata.
   53: The metadata is entered into a SQL database.
   54: 
   55: This script also does general database maintenance such as reformatting
   56: the C<loncapa:metadata> table if it is deprecated.
   57: 
   58: This script evaluates dynamic metadata from the authors'
   59: F<nohist_resevaldata.db> database file in order to store it in MySQL, as
   60: well as to compress the filesize (add up all "count"-type metadata).
   61: 
   62: This script is playing an increasingly important role for a loncapa
   63: library server.  The proper operation of this script is critical for a smooth
   64: and correct user experience.
   65: 
   66: =cut
   67: 
   68: use strict;
   69: 
   70: use lib '/home/httpd/lib/perl/';
   71: use LONCAPA::Configuration;
   72: 
   73: use IO::File;
   74: use HTML::TokeParser;
   75: use DBI;
   76: use GDBM_File;
   77: use POSIX qw(strftime mktime);
   78: 
   79: require "find.pl";
   80: 
   81: my @metalist;
   82: 
   83: my $simplestatus='';
   84: my %countext=();
   85: 
   86: # ----------------------------------------------------- write out simple status
   87: sub writesimple {
   88:     open(SMP,'>/home/httpd/html/lon-status/mysql.txt');
   89:     print SMP $simplestatus."\n";
   90:     close(SMP);
   91: }
   92: 
   93: sub writecount {
   94:     open(RSMP,'>/home/httpd/html/lon-status/rescount.txt');
   95:     foreach (keys %countext) {
   96: 	print RSMP $_.'='.$countext{$_}.'&';
   97:     }
   98:     print RSMP 'time='.time."\n";
   99:     close(RSMP);
  100: }
  101: 
  102: # -------------------------------------- counts files with different extensions
  103: sub count {
  104:     my $file=shift;
  105:     $file=~/\.(\w+)$/;
  106:     my $ext=lc($1);
  107:     if (defined($countext{$ext})) {
  108: 	$countext{$ext}++;
  109:     } else {
  110: 	$countext{$ext}=1;
  111:     }
  112: }
  113: # ----------------------------------------------------- Un-Escape Special Chars
  114: 
  115: sub unescape {
  116:     my $str=shift;
  117:     $str =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C",hex($1))/eg;
  118:     return $str;
  119: }
  120: 
  121: # -------------------------------------------------------- Escape Special Chars
  122: 
  123: sub escape {
  124:     my $str=shift;
  125:     $str =~ s/(\W)/"%".unpack('H2',$1)/eg;
  126:     return $str;
  127: }
  128: 
  129: # ------------------------------------------- Code to evaluate dynamic metadata
  130: 
  131: sub dynamicmeta {
  132:     my $url=&declutter(shift);
  133:     $url=~s/\.meta$//;
  134:     my %returnhash=();
  135:     my ($adomain,$aauthor)=($url=~/^(\w+)\/(\w+)\//);
  136:     my $prodir=&propath($adomain,$aauthor);
  137: 
  138: # Get metadata except counts
  139:     if (tie(my %evaldata,'GDBM_File',
  140:             $prodir.'/nohist_resevaldata.db',&GDBM_READER(),0640)) {
  141: 	my %sum=();
  142: 	my %cnt=();
  143: 	my %concat=();
  144: 	my %listitems=(
  145: 		       'course'       => 'add',
  146: 		       'goto'         => 'add',
  147: 		       'comefrom'     => 'add',
  148: 		       'avetries'     => 'avg',
  149: 		       'stdno'        => 'add',
  150: 		       'difficulty'   => 'avg',
  151: 		       'clear'        => 'avg',
  152: 		       'technical'    => 'avg',
  153: 		       'helpful'      => 'avg',
  154: 		       'correct'      => 'avg',
  155: 		       'depth'        => 'avg',
  156: 		       'comments'     => 'app',
  157: 		       'usage'        => 'cnt'
  158: 		       );
  159: 	
  160: 	my $regexp=$url;
  161: 	$regexp=~s/(\W)/\\$1/g;
  162: 	$regexp='___'.$regexp.'___([a-z]+)$';
  163: 	while (my ($esckey,$value)=each %evaldata) {
  164: 	    my $key=&unescape($esckey);
  165: 	    if ($key=~/$regexp/) {
  166: 		my ($item,$purl,$cat)=split(/___/,$key);
  167: 		if (defined($cnt{$cat})) { $cnt{$cat}++; } else { $cnt{$cat}=1; }
  168: 		unless ($listitems{$cat} eq 'app') {
  169: 		    if (defined($sum{$cat})) {
  170: 			$sum{$cat}+=$evaldata{$esckey};
  171: 			$concat{$cat}.=','.$item;
  172: 		    } else {
  173: 			$sum{$cat}=$evaldata{$esckey};
  174: 			$concat{$cat}=$item;
  175: 		    }
  176: 		} else {
  177: 		    if (defined($sum{$cat})) {
  178: 			if ($evaldata{$esckey}=~/\w/) {
  179: 			    $sum{$cat}.='<hr>'.$evaldata{$esckey};
  180: 			}
  181: 		    } else {
  182: 			$sum{$cat}=''.$evaldata{$esckey};
  183: 		    }
  184: 		}
  185: 	    }
  186: 	}
  187: 	untie(%evaldata);
  188: # transfer gathered data to returnhash, calculate averages where applicable
  189: 	while (my $cat=each(%cnt)) {
  190: 	    if ($listitems{$cat} eq 'avg') {
  191: 		$returnhash{$cat}=int(($sum{$cat}/$cnt{$cat})*100.0+0.5)/100.0;
  192: 	    } elsif ($listitems{$cat} eq 'cnt') {
  193: 		$returnhash{$cat}=$cnt{$cat};
  194: 	    } else {
  195: 		$returnhash{$cat}=$sum{$cat};
  196: 	    }
  197: 	    $returnhash{$cat.'_list'}=$concat{$cat};
  198: 	}
  199:     }
  200: # get count
  201:     if (tie(my %evaldata,'GDBM_File',
  202:             $prodir.'/nohist_accesscount.db',&GDBM_READER(),0640)) {
  203: 	my $escurl=&escape($url);
  204: 	if (! exists($evaldata{$escurl})) {
  205: 	    $returnhash{'count'}=0;
  206: 	} else {
  207: 	    $returnhash{'count'}=$evaldata{$escurl};
  208: 	}
  209: 	untie %evaldata;
  210:     }
  211:     return %returnhash;
  212: }
  213:   
  214: # ---------------  Read loncapa_apache.conf and loncapa.conf and get variables
  215: my $perlvarref=LONCAPA::Configuration::read_conf('loncapa.conf');
  216: my %perlvar=%{$perlvarref};
  217: undef $perlvarref; # remove since sensitive and not needed
  218: delete $perlvar{'lonReceipt'}; # remove since sensitive and not needed
  219: 
  220: # ------------------------------------- Only run if machine is a library server
  221: exit unless $perlvar{'lonRole'} eq 'library';
  222: 
  223: # ----------------------------- Make sure this process is running from user=www
  224: 
  225: my $wwwid=getpwnam('www');
  226: if ($wwwid!=$<) {
  227:     my $emailto="$perlvar{'lonAdmEMail'},$perlvar{'lonSysEMail'}";
  228:     my $subj="LON: $perlvar{'lonHostID'} User ID mismatch";
  229:     system("echo 'User ID mismatch. searchcat.pl must be run as user www.' |\
  230:  mailto $emailto -s '$subj' > /dev/null");
  231:     exit 1;
  232: }
  233: 
  234: 
  235: # ---------------------------------------------------------- We are in business
  236: 
  237: open(LOG,'>'.$perlvar{'lonDaemons'}.'/logs/searchcat.log');
  238: print LOG '==== Searchcat Run '.localtime()."====\n\n";
  239: $simplestatus='time='.time.'&';
  240: my $dbh;
  241: # ------------------------------------- Make sure that database can be accessed
  242: {
  243:     unless (
  244: 	    $dbh = DBI->connect("DBI:mysql:loncapa","www",$perlvar{'lonSqlAccess'},{ RaiseError =>0,PrintError=>0})
  245: 	    ) { 
  246: 	print LOG "Cannot connect to database!\n";
  247: 	$simplestatus.='mysql=defunct';
  248: 	&writesimple();
  249: 	exit;
  250:     }
  251: 
  252: # Make temporary table
  253:     $dbh->do("DROP TABLE IF EXISTS newmetadata");
  254:     my $make_metadata_table = "CREATE TABLE IF NOT EXISTS newmetadata (".
  255:         "title TEXT, author TEXT, subject TEXT, url TEXT, keywords TEXT, ".
  256:         "version TEXT, notes TEXT, abstract TEXT, mime TEXT, language TEXT, ".
  257:         "creationdate DATETIME, lastrevisiondate DATETIME, owner TEXT, ".
  258:         "copyright TEXT, ".
  259:         "count INTEGER UNSIGNED, ".
  260:         "course INTEGER UNSIGNED, course_list TEXT, ".
  261:         "goto INTEGER UNSIGNED, goto_list TEXT, ".
  262:         "comefrom INTEGER UNSIGNED, comefrom_list TEXT, ".
  263:         "fusage INTEGER UNSIGNED, fusage_list TEXT, ".
  264:         "stdno INTEGER UNSIGNED, stdno_list TEXT, ".
  265: 	"avetries FLOAT, avetries_list TEXT, ".
  266:         "difficulty FLOAT, difficulty_list TEXT, ".
  267:         "FULLTEXT idx_title (title), ".
  268:         "FULLTEXT idx_author (author), FULLTEXT idx_subject (subject), ".
  269:         "FULLTEXT idx_url (url), FULLTEXT idx_keywords (keywords), ".
  270:         "FULLTEXT idx_version (version), FULLTEXT idx_notes (notes), ".
  271:         "FULLTEXT idx_abstract (abstract), FULLTEXT idx_mime (mime), ".
  272:         "FULLTEXT idx_language (language), FULLTEXT idx_owner (owner), ".
  273:         "FULLTEXT idx_copyright (copyright)) ".
  274:         "TYPE=MyISAM";
  275:     # It would sure be nice to have some logging mechanism.
  276:     unless ($dbh->do($make_metadata_table)) {
  277: 	print LOG "\nMySQL Error Create: ".$dbh->errstr."\n";
  278: 	die $dbh->errstr;
  279:     }
  280: }
  281: 
  282: # ------------------------------------------------------------- get .meta files
  283: opendir(RESOURCES,"$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}");
  284: my @homeusers = grep {
  285:     &ishome("$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}/$_")
  286:     } grep {!/^\.\.?$/} readdir(RESOURCES);
  287: closedir RESOURCES;
  288: 
  289: #
  290: # Create the statement handlers we need
  291: 
  292: my $insert_sth = $dbh->prepare
  293:     ("INSERT INTO newmetadata VALUES (".
  294:      "?,".   # title
  295:      "?,".   # author
  296:      "?,".   # subject
  297:      "?,".   # declutter url
  298:      "?,".   # version
  299:      "?,".   # current
  300:      "?,".   # notes
  301:      "?,".   # abstract
  302:      "?,".   # mime
  303:      "?,".   # language
  304:      "?,".   # creationdate
  305:      "?,".   # revisiondate
  306:      "?,".   # owner
  307:      "?,".   # copyright
  308:      "?,".   # count
  309:      "?,".   # course
  310:      "?,".   # course_list
  311:      "?,".   # goto
  312:      "?,".   # goto_list
  313:      "?,".   # comefrom
  314:      "?,".   # comefrom_list
  315:      "?,".   # usage
  316:      "?,".   # usage_list
  317:      "?,".   # stdno
  318:      "?,".   # stdno_list
  319:      "?,".   # avetries
  320:      "?,".   # avetries_list
  321:      "?,".   # difficulty
  322:      "?,"    # difficulty_list
  323:      );
  324: 
  325: foreach my $user (@homeusers) {
  326:     print LOG "\n=== User: ".$user."\n\n";
  327: 
  328:     my $prodir=&propath($perlvar{'lonDefDomain'},$user);
  329:     # Use find.pl
  330:     undef @metalist;
  331:     @metalist=();
  332:     &find("$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}/$user");
  333:     # -- process each file to get metadata and put into search catalog SQL
  334:     # database.  Also, check to see if already there.
  335:     # I could just delete (without searching first), but this works for now.
  336:     foreach my $m (@metalist) {
  337:         print LOG "- ".$m."\n";
  338:         my $ref=&metadata($m);
  339:         my $m2='/res/'.&declutter($m);
  340:         $m2=~s/\.meta$//;
  341:  	if ($ref->{'obsolete'}) { print LOG "obsolete\n"; next; }
  342: 	if ($ref->{'copyright'} eq 'private') { print LOG "private\n"; next; }
  343:         my %dyn=&dynamicmeta($m2);
  344: 	&count($m2);
  345:         unless ($insert_sth->execute(
  346: 			     $ref->{'title'},
  347:                              $ref->{'author'},
  348:                              $ref->{'subject'},
  349:                              $m2,
  350:                              $ref->{'keywords'},
  351:                              'current',
  352:                              $ref->{'notes'},
  353:                              $ref->{'abstract'},
  354:                              $ref->{'mime'},
  355:                              $ref->{'language'},
  356:                              sqltime($ref->{'creationdate'}),
  357:                              sqltime($ref->{'lastrevisiondate'}),
  358:                              $ref->{'owner'},
  359:                              $ref->{'copyright'},
  360: 			     $dyn{'count'},
  361: 			     $dyn{'course'},
  362: 			     $dyn{'course_list'},
  363: 			     $dyn{'goto'},
  364: 			     $dyn{'goto_list'},
  365: 			     $dyn{'comefrom'},
  366: 			     $dyn{'comefrom_list'},
  367: 			     $dyn{'usage'},
  368: 			     $dyn{'usage_list'},
  369: 			     $dyn{'stdno'},
  370: 			     $dyn{'stdno_list'},
  371: 			     $dyn{'avetries'},
  372: 			     $dyn{'avetries_list'},
  373: 			     $dyn{'difficulty'},
  374: 			     $dyn{'difficulty_list'}			     
  375: 			     )) {
  376: 	    print LOG "\nMySQL Error Insert: ".$dbh->errstr."\n";
  377: 	    die $dbh->errstr;
  378: 	}
  379:         $ref = undef;
  380:     }
  381: }
  382: # --------------------------------------------------- Close database connection
  383: $dbh->do("DROP TABLE IF EXISTS metadata");
  384: unless ($dbh->do("RENAME TABLE newmetadata TO metadata")) {
  385:     print LOG "\nMySQL Error Rename: ".$dbh->errstr."\n";
  386:     die $dbh->errstr;
  387: }
  388: unless ($dbh->disconnect) {
  389:     print LOG "\nMySQL Error Disconnect: ".$dbh->errstr."\n";
  390:     die $dbh->errstr;
  391: }
  392: print LOG "\n==== Searchcat completed ".localtime()." ====\n";
  393: close(LOG);
  394: &writesimple();
  395: &writecount();
  396: exit 0;
  397: 
  398: 
  399: 
  400: # =============================================================================
  401: 
  402: # ---------------------------------------------------------------- Get metadata
  403: # significantly altered from subroutine present in lonnet
  404: sub metadata {
  405:     my ($uri,$what)=@_;
  406:     my %metacache=();
  407:     $uri=&declutter($uri);
  408:     my $filename=$uri;
  409:     $uri=~s/\.meta$//;
  410:     $uri='';
  411:     unless ($metacache{$uri.'keys'}) {
  412:         unless ($filename=~/\.meta$/) { $filename.='.meta'; }
  413: 	my $metastring=&getfile($perlvar{'lonDocRoot'}.'/res/'.$filename);
  414:         my $parser=HTML::TokeParser->new(\$metastring);
  415:         my $token;
  416:         while ($token=$parser->get_token) {
  417:             if ($token->[0] eq 'S') {
  418:                 my $entry=$token->[1];
  419:                 my $unikey=$entry;
  420:                 if (defined($token->[2]->{'part'})) { 
  421:                     $unikey.='_'.$token->[2]->{'part'}; 
  422:                 }
  423:                 if (defined($token->[2]->{'name'})) { 
  424:                     $unikey.='_'.$token->[2]->{'name'}; 
  425:                 }
  426:                 if ($metacache{$uri.'keys'}) {
  427:                     $metacache{$uri.'keys'}.=','.$unikey;
  428:                 } else {
  429:                     $metacache{$uri.'keys'}=$unikey;
  430:                 }
  431:                 map {
  432:                     $metacache{$uri.''.$unikey.'.'.$_}=$token->[2]->{$_};
  433:                 } @{$token->[3]};
  434:                 unless (
  435:                         $metacache{$uri.''.$unikey}=$parser->get_text('/'.$entry)
  436:                         ) { $metacache{$uri.''.$unikey}=
  437:                                 $metacache{$uri.''.$unikey.'.default'};
  438:                         }
  439:             }
  440:         }
  441:     }
  442:     return \%metacache;
  443: }
  444: 
  445: # ------------------------------------------------------------ Serves up a file
  446: # returns either the contents of the file or a -1
  447: sub getfile {
  448:     my $file=shift;
  449:     if (! -e $file ) { return -1; };
  450:     my $fh=IO::File->new($file);
  451:     my $a='';
  452:     while (<$fh>) { $a .=$_; }
  453:     return $a;
  454: }
  455: 
  456: # ------------------------------------------------------------- Declutters URLs
  457: sub declutter {
  458:     my $thisfn=shift;
  459:     $thisfn=~s/^$perlvar{'lonDocRoot'}//;
  460:     $thisfn=~s/^\///;
  461:     $thisfn=~s/^res\///;
  462:     return $thisfn;
  463: }
  464: 
  465: # --------------------------------------- Is this the home server of an author?
  466: # (copied from lond, modification of the return value)
  467: sub ishome {
  468:     my $author=shift;
  469:     $author=~s/\/home\/httpd\/html\/res\/([^\/]*)\/([^\/]*).*/$1\/$2/;
  470:     my ($udom,$uname)=split(/\//,$author);
  471:     my $proname=propath($udom,$uname);
  472:     if (-e $proname) {
  473: 	return 1;
  474:     } else {
  475:         return 0;
  476:     }
  477: }
  478: 
  479: # -------------------------------------------- Return path to profile directory
  480: # (copied from lond)
  481: sub propath {
  482:     my ($udom,$uname)=@_;
  483:     $udom=~s/\W//g;
  484:     $uname=~s/\W//g;
  485:     my $subdir=$uname.'__';
  486:     $subdir =~ s/(.)(.)(.).*/$1\/$2\/$3/;
  487:     my $proname="$perlvar{'lonUsersDir'}/$udom/$subdir/$uname";
  488:     return $proname;
  489: } 
  490: 
  491: # ---------------------------- convert 'time' format into a datetime sql format
  492: sub sqltime {
  493:     my ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
  494: 	localtime(&unsqltime(@_[0]));
  495:     $mon++; $year+=1900;
  496:     return "$year-$mon-$mday $hour:$min:$sec";
  497: }
  498: 
  499: sub maketime {
  500:     my %th=@_;
  501:     return POSIX::mktime(($th{'seconds'},$th{'minutes'},$th{'hours'},
  502:                           $th{'day'},$th{'month'}-1,
  503:                           $th{'year'}-1900,0,0,$th{'dlsav'}));
  504: }
  505: 
  506: 
  507: #########################################
  508: #
  509: # Retro-fixing of un-backward-compatible time format
  510: 
  511: sub unsqltime {
  512:     my $timestamp=shift;
  513:     if ($timestamp=~/^(\d+)\-(\d+)\-(\d+)\s+(\d+)\:(\d+)\:(\d+)$/) {
  514:         $timestamp=&maketime('year'=>$1,'month'=>$2,'day'=>$3,
  515:                              'hours'=>$4,'minutes'=>$5,'seconds'=>$6);
  516:     }
  517:     return $timestamp;
  518: }
  519: 
  520: # ----------------- Code to enable 'find' subroutine listing of the .meta files
  521: 
  522: no strict "vars";
  523: 
  524: sub wanted {
  525:     (($dev,$ino,$mode,$nlink,$uid,$gid) = lstat($_)) &&
  526:          -f _ &&
  527:          /^.*\.meta$/ && !/^.+\.\d+\.[^\.]+\.meta$/ &&
  528:          push(@metalist,"$dir/$_");
  529: }

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