# The LearningOnline Network # Search Catalog # # 03/08/2001 Scott Harrison # Scott Harrison: 03/12/2001, 03/13/2001, 03/14/2001, 03/15/2001, 03/19/2001 # Scott Harrison: 03/20/2001 # # Functions # # handler(server reference) : interacts with the Apache server layer # (for /adm/searchcat URLs) # simpletextfield(name,value) : returns HTML formatted string for simple text # field # simplecheckbox(name,value) : returns HTML formatted string for simple # checkbox # searchphrasefield(title,name,value) : returns HTML formatted string for # a search expression phrase field # dateboxes(name, defaultmonth, defaultday, defaultyear) : returns HTML # formatted string # for a calendar date # selectbox(title,name,value,%HASH=options) : returns HTML formatted string for # a selection box field # advancedsearch(server reference, environment reference) : perform a complex # multi-field logical query # filled(field) : determines whether a given field has been filled # basicsearch(server reference, environment reference) : perform a simple # single-field logical query # output_blank_field_error(server reference) : outputs a message saying that # more fields need to be filled in # output_results(output mode, # server reference, # environment reference, # reply list reference) : outputs results from search # build_SQL_query(field name, logic) : builds a SQL query string from a # logical expression with AND/OR keywords # recursive_SQL_query_build(field name, reverse notation expression) : # builds a SQL query string from a reverse notation expression # logical expression with AND/OR keywords package Apache::lonsearchcat; use strict; use Apache::Constants qw(:common); use Apache::lonnet(); use Apache::File(); use CGI qw(:standard); use Text::Query; my %language; my $scrout; my %metadatafields; my %cprtag; my %mimetag; my $closebutton; my $basicviewselect=< END my $advancedviewselect=< END sub handler { my $r = shift; # -------------------------------------- see if called from an interactive mode map { my ($name, $value) = split(/=/,$_); $value =~ tr/+/ /; $value =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C",hex($1))/eg; if ($name eq 'catalogmode') { $ENV{'form.'.$name}=$value; } } (split(/&/,$ENV{'QUERY_STRING'})); $r->content_type('text/html'); $r->send_http_header; return OK if $r->header_only; %metadatafields=(); my $hidden=''; $hidden=< END $closebutton=< END # ------------------------------------------------ First, check out environment $metadatafields{'owner'}=$ENV{'user.name'}.'@'.$ENV{'user.domain'}; # --------------------------------- Compute various listings of metadata values %language=(); $language{'any'}='Any language'; { my $fh=Apache::File->new($r->dir_config('lonTabDir').'/language.tab'); map { $_=~/(\w+)\s+([\w\s\-]+)/; chomp; $language{$1}=$2; } <$fh>; } %cprtag=(); $cprtag{'any'}='Any copyright/distribution'; { my $fh=Apache::File->new($r->dir_config('lonIncludes').'/copyright.tab'); map { $_=~/(\w+)\s+([\w\s\-]+)/; chomp; $cprtag{$1}=$2; } <$fh>; } %mimetag=(); $mimetag{'any'}='Any type'; { my $fh=Apache::File->new($r->dir_config('lonTabDir').'/filetypes.tab'); map { $_=~/(\w+)\s+(\w+)\s+([\w\s\-]+)/; chomp; $mimetag{$1}=".$1 $3"; } <$fh>; } if ($ENV{'form.basicsubmit'} eq 'SEARCH') { return &basicsearch($r,\%ENV); } elsif ($ENV{'form.advancedsubmit'} eq 'SEARCH') { return &advancedsearch($r,\%ENV); } $scrout=''; # building a part of screen output $scrout.=&searchphrasefield('Limit by title','title', $ENV{'form.title'}); $scrout.=&searchphrasefield('Limit by author','author', $ENV{'form.author'}); $scrout.=&searchphrasefield('Limit by subject','subject', $ENV{'form.subject'}); $scrout.=&searchphrasefield('Limit by keywords','keywords', $ENV{'form.keywords'}); $scrout.=&searchphrasefield('Limit by URL','url', $ENV{'form.url'}); $scrout.=&searchphrasefield('Limit by version','version', $ENV{'form.version'}); $scrout.=&searchphrasefield('Limit by notes','notes', $ENV{'form.notes'}); $scrout.=&searchphrasefield('Limit by abstract','abstract', $ENV{'form.abstract'}); $ENV{'form.mime'}='notxxx' unless length($ENV{'form.mime'}); $scrout.=&selectbox('Limit by MIME type','mime', $ENV{'form.mime'},%mimetag); $ENV{'form.language'}='any' unless length($ENV{'form.language'}); $scrout.=&selectbox('Limit by language','language', $ENV{'form.language'},%language); # ------------------------------------------------ Compute date selection boxes $scrout.=< LIMIT BY CREATION DATE RANGE:
between: CREATIONDATESTART $scrout.=&dateboxes('creationdatestart',1,1,1976, $ENV{'form.creationdatestart_month'}, $ENV{'form.creationdatestart_day'}, $ENV{'form.creationdatestart_year'}, ); $scrout.=< LIMIT BY LAST REVISION DATE RANGE:
between: LASTREVISIONDATESTART $scrout.=&dateboxes('lastrevisiondatestart',1,1,1976, $ENV{'form.lastrevisiondatestart_month'}, $ENV{'form.lastrevisiondatestart_day'}, $ENV{'form.lastrevisiondatestart_year'}, ); $scrout.=< LIMIT BY OTHER METADATA FIELDS: For author-specific metadata, enter in an expression in the form of key=value separated by operators such as AND or OR.
Example: grandmother=75 OR grandfather=85
CUSTOMMETADATA $scrout.=&simpletextfield('custommetadata',$ENV{'form.custommetadata'}); $scrout.=' initial users of this system do not need to worry about this option'; # ---------------------------------------------------------------- Print screen $r->print(< The LearningOnline Network with CAPA

Search Catalog

$hidden

Basic Search

Enter terms or phrases separated by search operators such as AND or OR then press SEARCH below. Terms should be specific to the title, author, subject, notes, or abstract information associated with a resource.
ENDDOCUMENT $r->print(&simpletextfield('basicexp',$ENV{'form.basicexp'})); $r->print(' '); $r->print(&simplecheckbox('titleonly',$ENV{'form.titleonly'})); $r->print('Title only '); $r->print(&simplecheckbox('allversions',$ENV{'form.allversions'})); $r->print(<Search historic archives
$closebutton $basicviewselect


Advanced Search

$scrout

$closebutton $advancedviewselect

ENDDOCUMENT return OK; } # --------------------------------------------------------- Various form fields sub simpletextfield { my ($name,$value)=@_; return ''; } sub simplecheckbox { my ($name,$value)=@_; my $checked=''; $checked="CHECKED" if $value eq 'on'; return ''; } sub searchphrasefield { my ($title,$name,$value)=@_; my $instruction=<$uctitle:". " $instruction
". ''; } sub dateboxes { my ($name,$defaultmonth,$defaultday,$defaultyear, $currentmonth,$currentday,$currentyear)=@_; ($defaultmonth,$defaultday,$defaultyear)=('','',''); my $month=< END $month=~s/(\"$currentmonth\")/$1 SELECTED/ if length($currentmonth); my $day=< END $day=~s/(\"$currentday\")/$1 SELECTED/ if length($currentday); my $year=< END $year=~s/(\"$currentyear\")/$1 SELECTED/ if length($currentyear); return "$month$day$year"; } sub selectbox { my ($title,$name,$value,%options)=@_; my $uctitle=uc($title); my $selout="\n

$uctitle:". "
".''; } # ----------------------------------------------- Performing an advanced search sub advancedsearch { my ($r,$envhash)=@_; my %ENV=%{$envhash}; my $fillflag=0; # Clean up fields for safety for my $field ('title','author','subject','keywords','url','version', 'creationdatestart_month','creationdatestart_day', 'creationdatestart_year','creationdateend_month', 'creationdateend_day','creationdateend_year', 'lastrevisiondatestart_month','lastrevisiondatestart_day', 'lastrevisiondatestart_year','lastrevisiondateend_month', 'lastrevisiondateend_day','lastrevisiondateend_year', 'notes','abstract','mime','language','owner', 'custommetadata') { $ENV{"form.$field"}=~s/[^\w\s\(\)\=\-\"\']//g; } for my $field ('title','author','subject','keywords','url','version', 'notes','abstract','mime','language','owner', 'custommetadata') { if (&filled($ENV{"form.$field"})) { $fillflag++; } } unless ($fillflag) { &output_blank_field_error($r); return OK; } my $query=''; my @queries; # Go through logical expression AND/OR/NOT phrase fields. foreach my $field ('title','author','subject','notes','abstract','url', 'keywords','version','owner') { if ($ENV{'form.'.$field}) { push @queries,&build_SQL_query($field,$ENV{'form.'.$field}); } } if ($ENV{'form.language'} and $ENV{'form.language'} ne 'any') { push @queries,"(language like \"\%$ENV{'form.language'}\%\")"; } if ($ENV{'form.mime'} and $ENV{'form.mime'} ne 'any') { push @queries,"(mime like \"\%$ENV{'form.mime'}\%\")"; } if ($ENV{'form.copyright'} and $ENV{'form.copyright'} ne 'any') { push @queries,"(copyright like \"\%$ENV{'form.copyright'}\%\")"; } my $datequery=&build_date_queries( $ENV{'form.creationdatestart_month'}, $ENV{'form.creationdatestart_day'}, $ENV{'form.creationdatestart_year'}, $ENV{'form.creationdateend_month'}, $ENV{'form.creationdateend_day'}, $ENV{'form.creationdateend_year'}, $ENV{'form.lastrevisiondatestart_month'}, $ENV{'form.lastrevisiondatestart_day'}, $ENV{'form.lastrevisiondatestart_year'}, $ENV{'form.lastrevisiondateend_month'}, $ENV{'form.lastrevisiondateend_day'}, $ENV{'form.lastrevisiondateend_year'}, ); if ($datequery=~/^Incorrect/) { &output_date_error($r,$datequery); return OK; } elsif ($datequery) { push @queries,$datequery; } my $customquery; if ($ENV{'form.custommetadata'}) { $customquery=&build_custommetadata_query('custommetadata', $ENV{'form.custommetadata'}); } if (@queries) { $query=join(" AND ",@queries); $query="select * from metadata where $query"; my $reply=&Apache::lonnet::metadata_query($query); &output_results('Advanced',$r,$envhash,$customquery,$reply); } else { &output_results('Advanced',$r,$envhash,$query); } return OK; } # ---------------------------------------------------- see if a field is filled sub filled { my ($field)=@_; if ($field=~/\S/) { return 1; } else { return 0; } } # --------------------------------------------------- Performing a basic search sub basicsearch { my ($r,$envhash)=@_; my %ENV=%{$envhash}; # Clean up fields for safety for my $field ('basicexp') { $ENV{"form.$field"}=~s/[^\w\s\(\)\-]//g; } unless (&filled($ENV{'form.basicexp'})) { &output_blank_field_error($r); return OK; } my $query=''; my $concatarg=join('," ",', ('title', 'author', 'subject', 'notes', 'abstract')); $query="select * from metadata where concat($concatarg) like '\%$ENV{'form.basicexp'}\%'"; my $reply=&Apache::lonnet::metadata_query($query); &output_results('Basic',$r,$envhash,$query,$reply); return OK; } # ---------------- Message to output when there are not enough fields filled in sub output_blank_field_error { my ($r)=@_; # make query information persistent to allow for subsequent revision my $persistent=&make_persistent(); $r->print(< The LearningOnline Network with CAPA BEGINNING $r->print(<

Search Catalog

$persistent $closebutton

Helpful Message

Incorrect search query due to blank entry fields. You need to fill in the relevant fields on the search page in order for a query to be processed.

RESULTS } # ----------------------------- format and output results based on a reply list sub output_results { my ($mode,$r,$envhash,$query,@replylist)=@_; my %ENV=%{$envhash}; my $compiledresult=''; foreach my $reply (@replylist) { my @results; my $replyfile=''; $reply=~/^([\.\w]+)$/; # must do since 'use strict' checks for tainting $replyfile=$r->dir_config('lonDaemons').'/tmp/'.$1; $reply=~/(.*?)\_/; my $hostname=$1; sleep 3; # temporary fix, need to check for completion and status { while (1) { last if -e $replyfile; sleep 1; } # QUESTION: how should I handle this error condition.. # I'm sure there is syntax elsewhere I can use.. my $fh=Apache::File->new($replyfile) or ($r->print('file cannot be opened') and return OK); @results=<$fh>; } foreach my $result (@results) { my @fields=map {&Apache::lonnet::unescape($_)} (split(/\,/,$result)); my ($title,$author,$subject,$url,$keywords,$version, $notes,$abstract,$mime,$lang, $creationdate,$lastrevisiondate,$owner,$copyright)=@fields; my $shortabstract=$abstract; $shortabstract=substr($abstract,0,200) if length($abstract)>200; $fields[7]=$shortabstract; $compiledresult.=< END $compiledresult.=<
END my $httphost=$ENV{'HTTP_HOST'}; my $viewselect; if ($mode eq 'Basic') { $viewselect=$ENV{'form.basicviewselect'}; } elsif ($mode eq 'Advanced') { $viewselect=$ENV{'form.advancedviewselect'}; } if ($viewselect eq 'Detailed Citation View') { $compiledresult.=&detailed_citation_view(@fields, $hostname,$httphost); } elsif ($viewselect eq 'Summary View') { $compiledresult.=&summary_view(@fields,$hostname,$httphost); } elsif ($viewselect eq 'Fielded Format') { $compiledresult.=&fielded_format_view(@fields,$hostname, $httphost); } elsif ($viewselect eq 'XML/SGML') { $compiledresult.=&xml_sgml_view(@fields,$hostname,$httphost); } } unless ($compiledresult) { $compiledresult="There were no results that matched your query"; } # make query information persistent to allow for subsequent revision my $persistent=&make_persistent(); $r->print(< The LearningOnline Network with CAPA BEGINNING $r->print(< function select_data(title,url) { changeTitle(title); changeURL(url); } function changeTitle(val) { if (opener.inf.document.forms.resinfo.elements.t) { opener.inf.document.forms.resinfo.elements.t.value=val; } } function changeURL(val) { if (opener.inf.document.forms.resinfo.elements.u) { opener.inf.document.forms.resinfo.elements.u.value=val; } } SCRIPT $r->print(<

Search Catalog

$closebutton $persistent

Search Query

RESULTS if ($mode eq 'Basic') { $r->print(< Basic search: $ENV{'form.basicexp'}

RESULTS } elsif ($mode eq 'Advanced') { $r->print(< Advanced search $query

RESULTS } $r->print(<Search Results $compiledresult RESULTS } } # ------------------------------------------------------------- build_SQL_query sub build_SQL_query { my ($field_name,$logic_statement)=@_; my $q=new Text::Query('abc', -parse => 'Text::Query::ParseAdvanced', -build => 'Text::Query::Build'); $q->prepare($logic_statement); my $matchexp=${$q}{'matchexp'}; chomp $matchexp; my $sql_query=&recursive_SQL_query_build($field_name,$matchexp); return $sql_query; } # ------------------------------------------------- build custom metadata query sub build_custommetadata_query { my ($field_name,$logic_statement)=@_; my $q=new Text::Query('abc', -parse => 'Text::Query::ParseAdvanced', -build => 'Text::Query::BuildAdvancedString'); $q->prepare($logic_statement); my $matchexp=${$q}{'-parse'}{'-build'}{'matchstring'}; # quick fix to change literal into xml tag-matching s/\(\?\:([^\)]*)\\\=([^\)]*)\)/\(\?\:\<$1\>\[\^\\\<\]\*$2\[\^\\\<\]\*\<\/$1\>)/g; return $matchexp; } # - Recursively parse a reverse notation expression into a SQL query expression sub recursive_SQL_query_build { my ($dkey,$pattern)=@_; my @matches=($pattern=~/(\[[^\]|\[]*\])/g); return $pattern unless @matches; foreach my $match (@matches) { $match=~/\[ (\w+)\s(.*) \]/; my ($key,$value)=($1,$2); my $replacement=''; if ($key eq 'literal') { $replacement="($dkey like \"\%$value\%\")"; } elsif ($key eq 'and') { $value=~/(.*[\"|\)]) ([|\(|\^].*)/; $replacement="($1 AND $2)"; } elsif ($key eq 'or') { $value=~/(.*[\"|\)]) ([|\(|\^].*)/; $replacement="($1 OR $2)"; } substr($pattern, index($pattern,$match), length($match), $replacement ); } &recursive_SQL_query_build($dkey,$pattern); } # ------------------------------------------------------ Detailed Citation View sub detailed_citation_view { my ($title,$author,$subject,$url,$keywords,$version, $notes,$shortabstract,$mime,$lang, $creationdate,$lastrevisiondate,$owner,$copyright, $hostname,$httphost)=@_; my $result=<$owner, last revised $lastrevisiondate

$title

$author

Subject: $subject
Keyword(s): $keywords
Notes: $notes
MIME Type: $mimetag{$mime}
Language: $language{$lang}
Copyright/Distribution: $cprtag{$copyright}
$shortabstract

END return $result; } # ---------------------------------------------------------------- Summary View sub summary_view { my ($title,$author,$subject,$url,$keywords,$version, $notes,$shortabstract,$mime,$lang, $creationdate,$lastrevisiondate,$owner,$copyright, $hostname,$httphost)=@_; my $result=<$author
$title
$owner -- $lastrevisiondate
$cprtag{$copyright}

END return $result; } # -------------------------------------------------------------- Fielded Format sub fielded_format_view { my ($title,$author,$subject,$url,$keywords,$version, $notes,$shortabstract,$mime,$lang, $creationdate,$lastrevisiondate,$owner,$copyright, $hostname,$httphost)=@_; my $result=<URL: $url
Title: $title
Author(s): $author
Subject: $subject
Keyword(s): $keywords
Notes: $notes
MIME Type: $mimetag{$mime}
Language: $language{$lang}
Creation Date: $creationdate
Last Revision Date: $lastrevisiondate
Publisher/Owner: $owner
Copyright/Distribution: $cprtag{$copyright}
Repository Location: $hostname
Abstract: $shortabstract

END return $result; } # -------------------------------------------------------------------- XML/SGML sub xml_sgml_view { my ($title,$author,$subject,$url,$keywords,$version, $notes,$shortabstract,$mime,$lang, $creationdate,$lastrevisiondate,$owner,$copyright, $hostname,$httphost)=@_; my $result=< <LonCapaResource> <url>$url</url> <title>$title</title> <author>$author</author> <subject>$subject</subject> <keywords>$keywords</keywords> <notes>$notes</notes> <mimeInfo> <mime>$mime</mime> <mimetag>$mimetag{$mime}</mimetag> </mimeInfo> <languageInfo> <language>$lang</language> <languagetag>$language{$lang}</languagetag> </languageInfo> <creationdate>$creationdate</creationdate> <lastrevisiondate>$lastrevisiondate</lastrevisiondate> <owner>$owner</owner> <copyrightInfo> <copyright>$copyright</copyright> <copyrighttag>$cprtag{$copyright}</copyrighttag> </copyrightInfo> <repositoryLocation>$hostname</repositoryLocation> <shortabstract>$shortabstract</shortabstract> </LonCapaResource> END return $result; } sub build_date_queries { my ($cmonth1,$cday1,$cyear1,$cmonth2,$cday2,$cyear2, $lmonth1,$lday1,$lyear1,$lmonth2,$lday2,$lyear2)=@_; my @queries; if ($cmonth1 or $cday1 or $cyear1 or $cmonth2 or $cday2 or $cyear2) { unless ($cmonth1 and $cday1 and $cyear1 and $cmonth2 and $cday2 and $cyear2) { return "Incorrect entry for the creation date. You must specify ". "a starting month, day, and year and an ending month, ". "day, and year."; } my $cnumeric1=sprintf("%d%2d%2d",$cyear1,$cmonth1,$cday1); $cnumeric1+=0; my $cnumeric2=sprintf("%d%2d%2d",$cyear2,$cmonth2,$cday2); $cnumeric2+=0; if ($cnumeric1>$cnumeric2) { return "Incorrect entry for the creation date. The starting ". "date must occur before the ending date."; } my $cquery="(creationdate BETWEEN '$cyear1-$cmonth1-$cday1' AND '". "$cyear2-$cmonth2-$cday2 23:59:59')"; push @queries,$cquery; } if ($lmonth1 or $lday1 or $lyear1 or $lmonth2 or $lday2 or $lyear2) { unless ($lmonth1 and $lday1 and $lyear1 and $lmonth2 and $lday2 and $lyear2) { return "Incorrect entry for the last revision date. You must ". "specify a starting month, day, and year and an ending ". "month, day, and year."; } my $lnumeric1=sprintf("%d%2d%2d",$lyear1,$lmonth1,$lday1); $lnumeric1+=0; my $lnumeric2=sprintf("%d%2d%2d",$lyear2,$lmonth2,$lday2); $lnumeric2+=0; if ($lnumeric1>$lnumeric2) { return "Incorrect entry for the last revision date. The ". "starting date must occur before the ending date."; } my $lquery="(lastrevisiondate BETWEEN '$lyear1-$lmonth1-$lday1' AND '". "$lyear2-$lmonth2-$lday2 23:59:59')"; push @queries,$lquery; } if (@queries) { return join(" AND ",@queries); } return ''; } sub output_date_error { my ($r,$message)=@_; # make query information persistent to allow for subsequent revision my $persistent=&make_persistent(); $r->print(< The LearningOnline Network with CAPA BEGINNING $r->print(<

Search Catalog

$persistent $closebutton

Helpful Message

$message

RESULTS } sub make_persistent { my $persistent=''; map { if (/^form\./ && !/submit/) { my $name=$_; my $key=$name; $ENV{$key}=~s/\'//g; # do not mess with html field syntax $name=~s/^form\.//; $persistent.=< END } } (keys %ENV); return $persistent; } 1; __END__