version 1.35, 2003/06/25 14:42:00
|
version 1.56, 2004/04/09 22:04:53
|
Line 56 This script also does general database m
|
Line 56 This script also does general database m
|
the C<loncapa:metadata> table if it is deprecated. |
the C<loncapa:metadata> table if it is deprecated. |
|
|
This script evaluates dynamic metadata from the authors' |
This script evaluates dynamic metadata from the authors' |
F<nohist_resevaldata.db> database file in order to store it in MySQL, as |
F<nohist_resevaldata.db> database file in order to store it in MySQL. |
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 65 and correct user experience.
|
Line 64 and correct user experience.
|
|
|
=cut |
=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 |
|
|
|
sub unescape { |
|
my $str=shift; |
|
$str =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C",hex($1))/eg; |
|
return $str; |
|
} |
|
|
|
# -------------------------------------------------------- Escape Special Chars |
|
|
|
sub escape { |
|
my $str=shift; |
|
$str =~ s/(\W)/"%".unpack('H2',$1)/eg; |
|
return $str; |
|
} |
|
|
|
|
|
# ------------------------------------------- Code to evaluate dynamic metadata |
|
|
|
sub dynamicmeta { |
|
|
|
my $url=&declutter(shift); |
# |
$url=~s/\.meta$//; |
# Set up configuration options |
my %returnhash=(); |
my ($simulate,$oneuser,$help,$verbose,$logfile,$debug); |
my ($adomain,$aauthor)=($url=~/^(\w+)\/(\w+)\//); |
GetOptions ( |
my $prodir=&propath($adomain,$aauthor); |
'help' => \$help, |
if ((tie(%evaldata,'GDBM_File', |
'simulate' => \$simulate, |
$prodir.'/nohist_resevaldata.db',&GDBM_READER(),0640)) && |
'only=s' => \$oneuser, |
(tie(%newevaldata,'GDBM_File', |
'verbose=s' => \$verbose, |
$prodir.'/nohist_new_resevaldata.db',&GDBM_WRCREAT(),0640))) { |
'debug' => \$debug, |
my %sum=(); |
); |
my %cnt=(); |
|
my %listitems=('count' => 'add', |
if ($help) { |
'course' => 'add', |
print <<"ENDHELP"; |
'avetries' => 'avg', |
$0 |
'stdno' => 'add', |
Rebuild and update the LON-CAPA metadata database. |
'difficulty' => 'avg', |
Options: |
'clear' => 'avg', |
-help Print this help |
'technical' => 'avg', |
-simulate Do not modify the database. |
'helpful' => 'avg', |
-only=user Only compute for the given user. Implies -simulate |
'correct' => 'avg', |
-verbose=val Sets logging level, val must be a number |
'depth' => 'avg', |
-debug Turns on debugging output |
'comments' => 'app', |
ENDHELP |
'usage' => 'cnt' |
exit 0; |
); |
} |
my $regexp=$url; |
|
$regexp=~s/(\W)/\\$1/g; |
if (! defined($debug)) { |
$regexp='___'.$regexp.'___([a-z]+)$'; |
$debug = 0; |
foreach (keys %evaldata) { |
} |
my $key=&unescape($_); |
|
if ($key=~/$regexp/) { |
if (! defined($verbose)) { |
my $ctype=$1; |
$verbose = 0; |
if (defined($cnt{$ctype})) { |
} |
$cnt{$ctype}++; |
|
} else { |
if (defined($oneuser)) { |
$cnt{$ctype}=1; |
$simulate=1; |
} |
} |
unless ($listitems{$ctype} eq 'app') { |
|
if (defined($sum{$ctype})) { |
## |
$sum{$ctype}+=$evaldata{$_}; |
## Use variables for table names so we can test this routine a little easier |
} else { |
my $oldname = 'metadata'; |
$sum{$ctype}=$evaldata{$_}; |
my $newname = 'newmetadata'; |
} |
|
} else { |
|
if (defined($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 |
# |
|
# Read loncapa_apache.conf and loncapa.conf |
my $perlvarref=LONCAPA::Configuration::read_conf('loncapa.conf'); |
my $perlvarref=LONCAPA::Configuration::read_conf('loncapa.conf'); |
my %perlvar=%{$perlvarref}; |
my %perlvar=%{$perlvarref}; |
undef $perlvarref; # remove since sensitive and not needed |
undef $perlvarref; |
delete $perlvar{'lonReceipt'}; # remove since sensitive and not needed |
delete $perlvar{'lonReceipt'}; # remove since sensitive (really?) & not needed |
|
# |
# ------------------------------------- Only run if machine is a library server |
# Only run if machine is a library server |
exit unless $perlvar{'lonRole'} eq 'library'; |
exit if ($perlvar{'lonRole'} ne 'library'); |
|
# |
# ----------------------------- Make sure this process is running from user=www |
# Make sure this process is running from user=www |
|
|
my $wwwid=getpwnam('www'); |
my $wwwid=getpwnam('www'); |
if ($wwwid!=$<) { |
if ($wwwid!=$<) { |
$emailto="$perlvar{'lonAdmEMail'},$perlvar{'lonSysEMail'}"; |
my $emailto="$perlvar{'lonAdmEMail'},$perlvar{'lonSysEMail'}"; |
$subj="LON: $perlvar{'lonHostID'} User ID mismatch"; |
my $subj="LON: $perlvar{'lonHostID'} User ID mismatch"; |
system("echo 'User ID mismatch. searchcat.pl must be run as user www.' |\ |
system("echo 'User ID mismatch. searchcat.pl must be run as user www.' |\ |
mailto $emailto -s '$subj' > /dev/null"); |
mailto $emailto -s '$subj' > /dev/null"); |
exit 1; |
exit 1; |
} |
} |
|
# |
|
# Let people know we are running |
# ---------------------------------------------------------- We are in business |
|
|
|
open(LOG,'>'.$perlvar{'lonDaemons'}.'/logs/searchcat.log'); |
open(LOG,'>'.$perlvar{'lonDaemons'}.'/logs/searchcat.log'); |
print LOG '==== Searchcat Run '.localtime()."====\n\n"; |
&log(0,'==== Searchcat Run '.localtime()."===="); |
|
if ($debug) { |
|
&log(0,'simulating') if ($simulate); |
|
&log(0,'only processing user '.$oneuser) if ($oneuser); |
|
&log(0,'verbosity level = '.$verbose); |
|
} |
|
# |
|
# Connect to database |
my $dbh; |
my $dbh; |
# ------------------------------------- Make sure that database can be accessed |
if (! ($dbh = DBI->connect("DBI:mysql:loncapa","www",$perlvar{'lonSqlAccess'}, |
{ |
{ RaiseError =>0,PrintError=>0}))) { |
unless ( |
&log(0,"Cannot connect to database!"); |
$dbh = DBI->connect("DBI:mysql:loncapa","www",$perlvar{'lonSqlAccess'},{ RaiseError =>0,PrintError=>0}) |
die "MySQL Error: Cannot connect to database!\n"; |
) { |
} |
print LOG "Cannot connect to database!\n"; |
# This can return an error and still be okay, so we do not bother checking. |
exit; |
# (perhaps it should be more robust and check for specific errors) |
} |
$dbh->do('DROP TABLE IF EXISTS '.$newname); |
my $make_metadata_table = "CREATE TABLE IF NOT EXISTS metadata (". |
# |
"title TEXT, author TEXT, subject TEXT, url TEXT, keywords TEXT, ". |
# Create the new table |
"version TEXT, notes TEXT, abstract TEXT, mime TEXT, language TEXT, ". |
my $request = &LONCAPA::lonmetadata::create_metadata_storage($newname); |
"creationdate DATETIME, lastrevisiondate DATETIME, owner TEXT, ". |
$dbh->do($request); |
"copyright TEXT, FULLTEXT idx_title (title), ". |
if ($dbh->err) { |
"FULLTEXT idx_author (author), FULLTEXT idx_subject (subject), ". |
$dbh->disconnect(); |
"FULLTEXT idx_url (url), FULLTEXT idx_keywords (keywords), ". |
&log(0,"MySQL Error Create: ".$dbh->errstr); |
"FULLTEXT idx_version (version), FULLTEXT idx_notes (notes), ". |
die $dbh->errstr; |
"FULLTEXT idx_abstract (abstract), FULLTEXT idx_mime (mime), ". |
} |
"FULLTEXT idx_language (language), FULLTEXT idx_owner (owner), ". |
# |
"FULLTEXT idx_copyright (copyright)) TYPE=MYISAM"; |
# find out which users we need to examine |
# It would sure be nice to have some logging mechanism. |
my $dom = $perlvar{'lonDefDomain'}; |
$dbh->do($make_metadata_table); |
opendir(RESOURCES,"$perlvar{'lonDocRoot'}/res/$dom"); |
} |
my @homeusers = |
|
grep { |
# ------------------------------------------------------------- get .meta files |
&ishome("$perlvar{'lonDocRoot'}/res/$dom/$_"); |
opendir(RESOURCES,"$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}"); |
} grep { |
my @homeusers = grep { |
!/^\.\.?$/; |
&ishome("$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}/$_") |
} readdir(RESOURCES); |
} grep {!/^\.\.?$/} readdir(RESOURCES); |
|
closedir RESOURCES; |
closedir RESOURCES; |
|
|
# |
# |
# Create the statement handlers we need |
if ($oneuser) { |
my $delete_sth = $dbh->prepare |
@homeusers=($oneuser); |
("DELETE FROM metadata WHERE url LIKE BINARY ?"); |
} |
|
# |
my $insert_sth = $dbh->prepare |
# Loop through the users |
("INSERT INTO metadata VALUES (". |
|
"?,". # title |
|
"?,". # author |
|
"?,". # subject |
|
"?,". # m2??? |
|
"?,". # version |
|
"?,". # current |
|
"?,". # notes |
|
"?,". # abstract |
|
"?,". # mime |
|
"?,". # language |
|
"?,". # creationdate |
|
"?,". # revisiondate |
|
"?,". # owner |
|
"?)" # copyright |
|
); |
|
|
|
foreach my $user (@homeusers) { |
foreach my $user (@homeusers) { |
print LOG "\n=== User: ".$user."\n\n"; |
&log(0,"=== User: ".$user); |
# Remove left-over db-files from potentially crashed searchcat run |
&process_dynamic_metadata($user,$dom); |
my $prodir=&propath($perlvar{'lonDefDomain'},$user); |
# |
unlink($prodir.'/nohist_new_resevaldata.db'); |
# Use File::Find to get the files we need to read/modify |
# Use find.pl |
find( |
undef @metalist; |
{preprocess => \&only_meta_files, |
@metalist=(); |
# wanted => \&print_filename, |
&find("$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}/$user"); |
# wanted => \&log_metadata, |
# -- process each file to get metadata and put into search catalog SQL |
wanted => \&process_meta_file, |
# database. Also, check to see if already there. |
}, |
# I could just delete (without searching first), but this works for now. |
"$perlvar{'lonDocRoot'}/res/$perlvar{'lonDefDomain'}/$user"); |
foreach my $m (@metalist) { |
} |
print LOG "- ".$m."\n"; |
# |
my $ref=&metadata($m); |
# Rename the table |
my $m2='/res/'.&declutter($m); |
if (! $simulate) { |
$m2=~s/\.meta$//; |
$dbh->do('DROP TABLE IF EXISTS '.$oldname); |
&dynamicmeta($m2); |
if (! $dbh->do('RENAME TABLE '.$newname.' TO '.$oldname)) { |
$delete_sth->execute($m2); |
&log(0,"MySQL Error Rename: ".$dbh->errstr); |
$insert_sth->execute($ref->{'title'}, |
die $dbh->errstr; |
$ref->{'author'}, |
} else { |
$ref->{'subject'}, |
&log(1,"MySQL table rename successful."); |
$m2, |
} |
$ref->{'keywords'}, |
} |
'current', |
|
$ref->{'notes'}, |
if (! $dbh->disconnect) { |
$ref->{'abstract'}, |
&log(0,"MySQL Error Disconnect: ".$dbh->errstr); |
$ref->{'mime'}, |
die $dbh->errstr; |
$ref->{'language'}, |
} |
sqltime($ref->{'creationdate'}), |
## |
sqltime($ref->{'lastrevisiondate'}), |
## Finished! |
$ref->{'owner'}, |
&log(0,"==== Searchcat completed ".localtime()." ===="); |
$ref->{'copyright'}); |
|
# if ($dbh->err()) { |
|
# print STDERR "Error:".$dbh->errstr()."\n"; |
|
# } |
|
$ref = undef; |
|
} |
|
|
|
# --------------------------------------------------- Clean up database |
|
# Need to, perhaps, remove stale SQL database records. |
|
# ... not yet implemented |
|
|
|
# ------------------------------------------- Copy over the new db-files |
|
system('mv '.$prodir.'/nohist_new_resevaldata.db '. |
|
$prodir.'/nohist_resevaldata.db'); |
|
} |
|
# --------------------------------------------------- Close database connection |
|
$dbh->disconnect; |
|
print LOG "\n==== Searchcat completed ".localtime()." ====\n"; |
|
close(LOG); |
close(LOG); |
|
|
|
&write_type_count(); |
|
&write_copyright_count(); |
|
|
exit 0; |
exit 0; |
|
|
|
## |
|
## Status logging routine. Inputs: $level, $message |
|
## |
|
## $level 0 should be used for normal output and error messages |
|
## |
|
## $message does not need to end with \n. In the case of errors |
|
## the message should contain as much information as possible to |
|
## help in diagnosing the problem. |
|
## |
|
sub log { |
|
my ($level,$message)=@_; |
|
$level = 0 if (! defined($level)); |
|
if ($verbose >= $level) { |
|
print LOG $message.$/; |
|
} |
|
} |
|
|
|
######################################################## |
|
######################################################## |
|
### ### |
|
### File::Find support routines ### |
|
### ### |
|
######################################################## |
|
######################################################## |
|
## |
|
## &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; |
|
} |
|
|
|
## |
|
## |
|
## 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; |
|
} |
|
|
|
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; |
|
} |
|
|
|
|
# ============================================================================= |
## |
|
## process_meta_file |
|
## Called by File::Find. |
|
## Only input is the filename in $_. |
|
sub process_meta_file { |
|
my ($file) = $_; |
|
my $filename = $File::Find::name; # full filename |
|
return if (-d $filename); # No need to do anything here for directories |
|
# |
|
&log(3,$filename) if ($debug); |
|
# |
|
my $ref=&metadata($filename); |
|
# |
|
# $url is the original file url, not the metadata file |
|
my $url='/res/'.&declutter($filename); |
|
$url=~s/\.meta$//; |
|
&log(3," ".$url) if ($debug); |
|
# |
|
# Ignore some files based on their metadata |
|
if ($ref->{'obsolete'}) { |
|
&log(3,"obsolete") if ($debug); |
|
return; |
|
} |
|
&count_copyright($ref->{'copyright'}); |
|
if ($ref->{'copyright'} eq 'private') { |
|
&log(3,"private") if ($debug); |
|
return; |
|
} |
|
# |
|
# Find the dynamic metadata |
|
my %dyn; |
|
if ($url=~ m:/default$:) { |
|
$url=~ s:/default$:/:; |
|
&log(3,"Skipping dynamic data") if ($debug); |
|
} else { |
|
&log(3,"Retrieving dynamic data") if ($debug); |
|
%dyn=&get_dynamic_metadata($url); |
|
&count_type($url); |
|
} |
|
# |
|
$ref->{'creationdate'} = &sqltime($ref->{'creationdate'}); |
|
$ref->{'lastrevisiondate'} = &sqltime($ref->{'lastrevisiondate'}); |
|
my %Data = ( |
|
%$ref, |
|
%dyn, |
|
'url'=>$url, |
|
'version'=>'current'); |
|
if (! $simulate) { |
|
my ($count,$err) = &LONCAPA::lonmetadata::store_metadata($dbh,$newname, |
|
\%Data); |
|
if ($err) { |
|
&log(0,"MySQL Error Insert: ".$err); |
|
die $err; |
|
} |
|
if ($count < 1) { |
|
&log(0,"Unable to insert record into MySQL database for $url"); |
|
die "Unable to insert record into MySQl database for $url"; |
|
} |
|
} |
|
# |
|
# Reset $_ before leaving |
|
$_ = $file; |
|
} |
|
|
# ---------------------------------------------------------------- Get metadata |
######################################################## |
# significantly altered from subroutine present in lonnet |
######################################################## |
|
### ### |
|
### &metadata($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 ($metacache{$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.''.$unikey.'.default'}; |
|
} |
|
} |
} |
} |
if (defined($token->[2]->{'name'})) { |
|
$unikey.='_'.$token->[2]->{'name'}; |
|
} |
|
if ($metacache{$uri.'keys'}) { |
|
$metacache{$uri.'keys'}.=','.$unikey; |
|
} else { |
|
$metacache{$uri.'keys'}=$unikey; |
|
} |
|
foreach ( @{$token->[3]}) { |
|
$metacache{$uri.''.$unikey.'.'.$_}=$token->[2]->{$_}; |
|
} |
|
if (! ($metacache{$uri.''.$unikey}=$parser->get_text('/'.$entry))){ |
|
$metacache{$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 ) { |
|
return undef; |
|
} |
my $fh=IO::File->new($file); |
my $fh=IO::File->new($file); |
my $a=''; |
my $contents = ''; |
while (<$fh>) { $a .=$_; } |
while (<$fh>) { |
return $a; |
$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', |
|
$prodir.'/nohist_resevaldata.db',&GDBM_READER(),0640)) { |
|
return 0; |
|
} |
|
# |
|
# Process every stored element |
|
while (my ($storedkey,$value) = each(%evaldata)) { |
|
my ($source,$file,$type) = split('___',$storedkey); |
|
$source = &unescape($source); |
|
$file = &unescape($file); |
|
$value = &unescape($value); |
|
" got ".$file."\n ".$type." ".$source."\n"; |
|
if ($type =~ /^(avetries|count|difficulty|stdno|timestamp)$/) { |
|
# |
|
# Statistics: $source is course id |
|
$DynamicData{$file}->{'statistics'}->{$source}->{$type}=$value; |
|
} elsif ($type =~ /^(clear|comments|depth|technical|helpful)$/){ |
|
# |
|
# Evaluation $source is username, check if they evaluated it |
|
# more than once. If so, pad the entry with a space. |
|
while(exists($DynamicData{$file}->{'evaluation'}->{$type}->{$source})) { |
|
$source .= ' '; |
|
} |
|
$DynamicData{$file}->{'evaluation'}->{$type}->{$source}=$value; |
|
} elsif ($type =~ /^(course|comefrom|goto|usage)$/) { |
|
# |
|
# Context $source is course id or resource |
|
push(@{$DynamicData{$file}->{$type}},&unescape($source)); |
|
} else { |
|
&log(0," ".$user."@".$dom.":Process metadata: Unable to decode ".$type); |
|
} |
|
} |
|
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; |
|
my $resdata = $DynamicData{$url}; |
|
# |
|
# Get the statistical data |
|
foreach my $type (qw/avetries difficulty stdno/) { |
|
my $count; |
|
my $sum; |
|
my @Values; |
|
foreach my $coursedata (values(%{$resdata->{'statistics'}})) { |
|
if (ref($coursedata) eq 'HASH' && exists($coursedata->{$type})) { |
|
$count++; |
|
$sum += $coursedata->{$type}; |
|
push(@Values,$coursedata->{$type}); |
|
} |
|
} |
|
if ($count) { |
|
$data{$type} = $sum/$count; |
|
$data{$type.'_list'} = join(',',@Values); |
|
} |
|
} |
|
# find the count |
|
$data{'count'} = $Counts{$url}; |
|
# |
|
# Get the context data |
|
foreach my $type (qw/course goto comefrom/) { |
|
if (defined($resdata->{$type}) && |
|
ref($resdata->{$type}) eq 'ARRAY') { |
|
$data{$type} = scalar(@{$resdata->{$type}}); |
|
$data{$type.'_list'} = join(',',@{$resdata->{$type}}); |
|
} |
|
} |
|
if (defined($resdata->{'usage'}) && |
|
ref($resdata->{'usage'}) eq 'ARRAY') { |
|
$data{'sequsage'} = scalar(@{$resdata->{'usage'}}); |
|
$data{'sequsage_list'} = join(',',@{$resdata->{'usage'}}); |
|
} |
|
# |
|
# Get the evaluation data |
|
foreach my $type (qw/clear technical correct helpful depth/) { |
|
my $count; |
|
my $sum; |
|
foreach my $evaluator (keys(%{$resdata->{'evaluation'}->{$type}})){ |
|
$sum += $resdata->{'evaluation'}->{$type}->{$evaluator}; |
|
$count++; |
|
} |
|
if ($count > 0) { |
|
$data{$type}=$sum/$count; |
|
} |
|
} |
|
# |
|
# put together comments |
|
my $comments = '<div class="LCevalcomments">'; |
|
foreach my $evaluator (keys(%{$resdata->{'evaluation'}->{'comments'}})){ |
|
$comments .= $evaluator.':'. |
|
$resdata->{'evaluation'}->{'comments'}->{$evaluator}.'<hr />'; |
|
} |
|
$comments .= '</div>'; |
|
# |
|
# Log the dynamic metadata |
|
if ($debug) { |
|
while (my($k,$v)=each(%data)) { |
|
&log(8," ".$k." => ".$v); |
|
} |
|
} |
|
# |
|
return %data; |
} |
} |
|
|
# --------------------------------------- Is this the home server of an author? |
} # End of %DynamicData and %Counts scope |
# (copied from lond, modification of the return value) |
|
|
######################################################## |
|
######################################################## |
|
### ### |
|
### 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); |
|
} |
|
|
|
} # end of scope for %copyrights |
|
|
|
######################################################## |
|
######################################################## |
|
### ### |
|
### 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\/$2/; |
Line 395 sub ishome {
|
Line 692 sub ishome {
|
} |
} |
} |
} |
|
|
# -------------------------------------------- Return path to profile directory |
## |
# (copied from lond) |
## &propath($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 407 sub propath {
|
Line 706 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(($th{'seconds'},$th{'minutes'},$th{'hours'}, |
## Given a filename, returns a url for the filename. |
$th{'day'},$th{'month'}-1, |
sub declutter { |
$th{'year'}-1900,0,0,$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; |
|
} |