# 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;

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 if $ENV{'form.catalogmode'} eq 'interactive';
<input type='hidden' name='catalogmode' value='interactive'>
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\-]+)/;
	    $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\-]+)/;
	    $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\-]+)/;
	    $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.=<<CREATIONDATESTART;
<p>
<font color="#800000" face="helvetica"><b>LIMIT BY CREATION DATE RANGE:</b>
</font>
<br>
between:
CREATIONDATESTART
    $scrout.=&dateboxes('creationdatestart',1,1,1976,
			$ENV{'form.creationdatestart_month'},
			$ENV{'form.creationdatestart_day'},
			$ENV{'form.creationdatestart_year'},
			);
    $scrout.=<<CREATIONDATEEND;
and:
CREATIONDATEEND
    $scrout.=&dateboxes('creationdateend',12,31,2051,
			$ENV{'form.creationdateend_month'},
			$ENV{'form.creationdateend_day'},
			$ENV{'form.creationdateend_year'},
			);
    $scrout.="</p>";

    $scrout.=<<LASTREVISIONDATESTART;
<p>
<font color="#800000" face="helvetica"><b>LIMIT BY LAST REVISION DATE RANGE:
</b></font>
<br>between:
LASTREVISIONDATESTART
    $scrout.=&dateboxes('lastrevisiondatestart',1,1,1976,
			$ENV{'form.lastrevisiondatestart_month'},
			$ENV{'form.lastrevisiondatestart_day'},
			$ENV{'form.lastrevisiondatestart_year'},
			);
    $scrout.=<<LASTREVISIONDATEEND;
and:
LASTREVISIONDATEEND
    $scrout.=&dateboxes('lastrevisiondateend',12,31,2051,
			$ENV{'form.lastrevisiondateend_month'},
			$ENV{'form.lastrevisiondateend_day'},
			$ENV{'form.lastrevisiondateend_year'},
			);
    $scrout.='</p>';

    $scrout.=&searchphrasefield('Limit by publisher/owner','owner',
				$ENV{'form.owner'});
#			$metadatafields{'owner'});

    $ENV{'form.copyright'}='any' unless length($ENV{'form.copyright'});
    $scrout.=&selectbox('Limit by copyright/distribution','copyright',
			$ENV{'form.copyright'},%cprtag);

# ------------------------------------------- Compute customized metadata field
    $scrout.=<<CUSTOMMETADATA;
<p>
<font color="#800000" face="helvetica"><b>LIMIT BY OTHER METADATA FIELDS:</b>
</font>
For author-specific metadata, enter in an expression in the form of 
<i>key</i>=<i>value</i> separated by operators such as AND or OR.<br>
<b>Example:</b> grandmother=75 OR grandfather=85
<br>
CUSTOMMETADATA
$scrout.=&simpletextfield('custommetadata',$ENV{'form.custommetadata'});
$scrout.=' <i>initial users of this system do not need to worry about this option</i>';

# ---------------------------------------------------------------- Print screen
    $r->print(<<ENDDOCUMENT);
<html>
<head>
<title>The LearningOnline Network with CAPA</title>
</head>
<body bgcolor="#FFFFFF">
<img align=right src=/adm/lonIcons/lonlogos.gif>
<h1>Search Catalog</h1>
<form method="post" action="/adm/searchcat">
$hidden
<hr>
<h3>Basic Search</h3>
<p>
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.
<br>
ENDDOCUMENT
    $r->print(&simpletextfield('basicexp',$ENV{'form.basicexp'}));
    $r->print(' ');
    $r->print(&simplecheckbox('titleonly',$ENV{'form.titleonly'}));
    $r->print('<font color="#800000">Title only</font> ');
    $r->print(&simplecheckbox('allversions',$ENV{'form.allversions'}));
    $r->print(<<ENDDOCUMENT);
<font color="#800000">Search historic archives</font>
<br>
<input type="submit" name="basicsubmit" value="SEARCH">
<input type="reset" name="reset" value="RESET">
<input type="button" name="close" value="CLOSE" onClick="self.close()">
</p>
<hr>
<h3>Advanced Search</h3>
$scrout
<p>
<input type="submit" name="advancedsubmit" value="SEARCH">
<input type="reset" name="reset" value="RESET">
<input type="button" name="close" value="CLOSE" onClick="self.close()">
</p>
</form>
</body>
</html>
ENDDOCUMENT
    return OK;
} 

# --------------------------------------------------------- Various form fields

sub simpletextfield {
    my ($name,$value)=@_;
    return '<input type=text name="'.$name.'" size=20 value="'.$value.'">';
}

sub simplecheckbox {
    my ($name,$value)=@_;
    my $checked='';
    $checked="CHECKED" if $value eq 'on';
    return '<input type=checkbox name="'.$name.'" '. $checked . '>';
}

sub searchphrasefield {
    my ($title,$name,$value)=@_;
    my $instruction=<<END;
Enter terms or phrases separated by search operators such
as AND or OR.
END
    my $uctitle=uc($title);
    return "\n<p><font color=\"#800000\" face=\"helvetica\"><b>$uctitle:</b>".
	   "</FONT> $instruction<br>".
           '<input type=text name="'.$name.'" size=80 value="'.$value.'">';
}

sub dateboxes {
    my ($name,$defaultmonth,$defaultday,$defaultyear,
	$currentmonth,$currentday,$currentyear)=@_;
    ($defaultmonth,$defaultday,$defaultyear)=('','','');
    my $month=<<END;
<select name="${name}_month">
<option value='$defaultmonth'> </option>
<option value="1">January</option>
<option value="2">February</option>
<option value="3">March</option>
<option value="4">April</option>
<option value="5">May</option>
<option value="6">June</option>
<option value="7">July</option>
<option value="8">August</option>
<option value="9">September</option>
<option value="10">October</option>
<option value="11">November</option>
<option value="12">December</option>
</select>
END
    $month=~s/(\"$currentmonth\")/$1 SELECTED/ if length($currentmonth);
    my $day=<<END;
<select name="${name}_day">
<option value='$defaultday'> </option>
<option value="1">1</option>
<option value="2">2</option>
<option value="3">3</option>
<option value="4">4</option>
<option value="5">5</option>
<option value="6">6</option>
<option value="7">7</option>
<option value="8">8</option>
<option value="9">9</option>
<option value="10">10</option>
<option value="11">11</option>
<option value="12">12</option>
<option value="13">13</option>
<option value="14">14</option>
<option value="15">15</option>
<option value="16">16</option>
<option value="17">17</option>
<option value="18">18</option>
<option value="19">19</option>
<option value="20">20</option>
<option value="21">21</option>
<option value="22">22</option>
<option value="23">23</option>
<option value="24">24</option>
<option value="25">25</option>
<option value="26">26</option>
<option value="27">27</option>
<option value="28">28</option>
<option value="29">29</option>
<option value="30">30</option>
<option value="31">31</option>
</select>
END
    $day=~s/(\"$currentday\")/$1 SELECTED/ if length($currentday);
    my $year=<<END;
<select name="${name}_year">
<option value='$defaultyear'> </option>
<option value="1976">1976</option>
<option value="1977">1977</option>
<option value="1978">1978</option>
<option value="1979">1979</option>
<option value="1980">1980</option>
<option value="1981">1981</option>
<option value="1982">1982</option>
<option value="1983">1983</option>
<option value="1984">1984</option>
<option value="1985">1985</option>
<option value="1986">1986</option>
<option value="1987">1987</option>
<option value="1988">1988</option>
<option value="1989">1989</option>
<option value="1990">1990</option>
<option value="1991">1991</option>
<option value="1992">1992</option>
<option value="1993">1993</option>
<option value="1994">1994</option>
<option value="1995">1995</option>
<option value="1996">1996</option>
<option value="1997">1997</option>
<option value="1998">1998</option>
<option value="1999">1999</option>
<option value="2000">2000</option>
<option value="2001">2001</option>
<option value="2002">2002</option>
<option value="2003">2003</option>
<option value="2004">2004</option>
<option value="2005">2005</option>
<option value="2006">2006</option>
<option value="2007">2007</option>
<option value="2008">2008</option>
<option value="2009">2009</option>
<option value="2010">2010</option>
<option value="2011">2011</option>
<option value="2012">2012</option>
<option value="2013">2013</option>
<option value="2014">2014</option>
<option value="2015">2015</option>
<option value="2016">2016</option>
<option value="2017">2017</option>
<option value="2018">2018</option>
<option value="2019">2019</option>
<option value="2020">2020</option>
<option value="2021">2021</option>
<option value="2022">2022</option>
<option value="2023">2023</option>
<option value="2024">2024</option>
<option value="2025">2025</option>
<option value="2026">2026</option>
<option value="2027">2027</option>
<option value="2028">2028</option>
<option value="2029">2029</option>
<option value="2030">2030</option>
<option value="2031">2031</option>
<option value="2032">2032</option>
<option value="2033">2033</option>
<option value="2034">2034</option>
<option value="2035">2035</option>
<option value="2036">2036</option>
<option value="2037">2037</option>
<option value="2038">2038</option>
<option value="2039">2039</option>
<option value="2040">2040</option>
<option value="2041">2041</option>
<option value="2042">2042</option>
<option value="2043">2043</option>
<option value="2044">2044</option>
<option value="2045">2045</option>
<option value="2046">2046</option>
<option value="2047">2047</option>
<option value="2048">2048</option>
<option value="2049">2049</option>
<option value="2050">2050</option>
<option value="2051">2051</option>
</select>
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 $selout="\n<p><font color=\"#800000\" face=\"helvetica\"><b>$uctitle:".
	"</b></font><br>".'<select name="'.$name.'">';
    map {
        $selout.='<option value="'.$_.'"';
        if ($_ eq $value) { $selout.=' selected'; }
        $selout.='>'.$options{$_}.'</option>';
    } sort keys %options;
    return $selout.'</select>';
}

# ------------------------------------------------ Performing a advanced search
sub advancedsearch {
    my ($r,$envhash)=@_;
    my %ENV=%{$envhash};

    my $fillflag=0;
    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 $testquery='';
    # Go through logical expression AND/OR/NOT phrase fields.
    foreach my $field ('title','author','subject','notes','abstract') {
	if ($ENV{'form.'.$field}) {
	    $testquery=build_SQL_query($field,$ENV{'form.'.$field});
	}
    }

#    my $concatarg=join(',"    ",',
#		       ('title', 'author', 'subject', 'notes', 'abstract'));

    $query="select * from metadata where $testquery";

    my $reply=&Apache::lonnet::metadata_query($query);

    &output_results('Advanced',$r,$envhash,$query,$reply);
    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};

    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='';
    map {
	if (/^form\./ && !/submit/) {
	    my $name=$_;
	    my $key=$name;
	    $name=~s/^form\.//;
	    $persistent.=<<END;
<INPUT TYPE='hidden' NAME='$name' VALUE='$ENV{$key}'>
END
        }
    } (keys %ENV);

    $r->print(<<BEGINNING);
<html>
<head>
<title>The LearningOnline Network with CAPA</title>
BEGINNING
    $r->print(<<RESULTS);
</head>
<body bgcolor="#ffffff">
<img align=right src=/adm/lonIcons/lonlogos.gif>
<h1>Search Catalog</h1>
<form method="post" action="/adm/searchcat">
$persistent
<input type='button' value='Revise search request'
onClick='this.form.submit();'>
<input type='button' value='CLOSE'
onClick='self.close();'>
<hr>
<h3>Helpful Message</h3>
<p>
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.
</p>
</body>
</html>
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;

	{
	    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 $result (@results) {
	    my ($title,$author,$subject,$url,$keywords,$version,
		$notes,$abstract,$mime,$lang,
		$creationdate,$lastrevisiondate,$owner,$copyright
		)=map {&Apache::lonnet::unescape($_)} (split(/\,/,$result));
	    my $shortabstract=$abstract;
	    $shortabstract=substr($abstract,0,200) if length($abstract)>200;
	    $compiledresult.=<<END;
<p>
END
            $compiledresult.=<<END if $ENV{'form.catalogmode'} eq 'interactive';
<font size='-1'><INPUT TYPE="button" NAME="returnvalues" VALUE="SELECT"
onClick="javascript:select_data('$title','$url')">
</font>
<br>
END
            $compiledresult.=<<END;
<b>URL: </b> <A HREF="http://$ENV{'HTTP_HOST'}$url" TARGET='search_preview'>$url</A>
<br>
<b>Title:</b> $title<br>
<b>Author(s):</b> $author<br>
<b>Subject:</b> $subject<br>
<b>Keyword(s):</b> $keywords<br>
<b>Notes:</b> $notes<br>
<b>Abstract:</b> $shortabstract<br>
<b>MIME Type:</b> $mimetag{$mime}<br>
<b>Language:</b> $language{$lang}<br>
<b>Creation Date:</b> $creationdate<br>
<b>Last Revision Date:</b> $lastrevisiondate<br>
<b>Publisher/Owner:</b> $owner<br>
<b>Copyright/Distribution:</b> $copyright<br>
<b>Repository Location:</b> $hostname
</p>
END
        }

	unless ($compiledresult) {
	    $compiledresult="There were no results that matched your query";
	}

	# make query information persistent to allow for subsequent revision
	my $persistent='';
	map {
	    if (/^form\./ && !/submit/) {
		my $name=$_;
		my $key=$name;
		$name=~s/^form\.//;
		$persistent.=<<END;
<INPUT TYPE='hidden' NAME='$name' VALUE='$ENV{$key}'>
END
            }
	} (keys %ENV);

	$r->print(<<BEGINNING);
<html>
<head>
<title>The LearningOnline Network with CAPA</title>
BEGINNING
        $r->print(<<SCRIPT) if $ENV{'form.catalogmode'} eq 'interactive';
<script>
    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>
SCRIPT
        $r->print(<<RESULTS);
</head>
<body bgcolor="#ffffff">
<img align=right src=/adm/lonIcons/lonlogos.gif>
<h1>Search Catalog</h1>
<form method="post" action="/adm/searchcat">
<input type='button' value='Revise search request'
onClick='this.form.submit();'>
<input type='button' value='CLOSE'
onClick='self.close();'>
$persistent
<hr>
<h3>Search Query</h3>
RESULTS
    if ($mode eq 'Basic') {
	$r->print(<<RESULTS);
<p>
<b>Basic search:</b> $ENV{'form.basicexp'}
</p>
RESULTS
    }
    elsif ($mode eq 'Advanced') {
	$r->print(<<RESULTS);
<p>
<b>Advanced search</b>
$query
</p>
RESULTS
    }
	$r->print(<<RESULTS);
<h3>Search Results</h3>
$compiledresult
</body>
</html>
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;
}

# - 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);
}

1;
__END__