cerca con perl...

**biocco** · 18-08-2005, 10:48

Ciao raga io di perl ci capisco poco!!!.

Vorrei chiedervi dove posso avere una guida efficente gratuita e se per favore in questo script per un motore di ricerca si possa escludere l'analisi di alcune directory. Se si come si fa???? il problema è che mi tira fuori i file vti di frontpage.... grazie molte!!!

codice:

#!c:\perl


# this is the url of your site
$base_url = "http://$ENV{'HTTP_HOST'}";

# this the path equivalent to your url
#$base_dir = $ENV{'DOCUMENT_ROOT'}; 
$base_dir = "e:/wwwroot/www/miosito.it/Online/";

# this is the url of this script
$search_url = "$base_url/cgi-bin/IecSrcEng.pl";

# this is the path to header.shtml
$header = "Search/headerEng.htm";

# this is the path to footer.shtml
$footer = "Search/footerEng.htm";

# this is the background color of the search
# you can also use an image by using background=image/ha.gif
#$bg = 'bgcolor=blue';

# this is every directory you want to search in, if you choose
# yes to search in subdirectories, you only need to list the
# main directories here... but if you only want to search some
# subdirectories, you can do so like this:
# @dirs = ('','news'); would search in news and the
# subdirectory oldnews
#@dirs = ('testsite');
#@dirs = ('inglese');

# search all directories within directories? yes/no
$subdirs = "yes";

# what files type do you want to search in? note that htm works for 
# shtml, html, htm, etc...
@searchtypes = ('htm','html');

use CGI qw(:standard);
$search = new CGI;

$result_num = $search->param('result_num');
$show_search = $search->param('show_search');
$show_summary = $search->param('show_summary');
$keyword = $search->param('keyword');
$page = $search->param('page');
$connector = $search->param('connector');
$case = $search->param('case');
#$header = $search->param('header');
#$footer = $search->param('footer');
$mydirs = $search->param('dirs');
#$bg = $search->param('bkg');
@dirs = ($mydirs);

&get_header_footer;
&display_page unless $show_search eq "no";


@keywords = split(/\s+/,$keyword);

foreach $type (@dirs) {
	chomp($type);
	$dir = "$base_dir/$type";

	chdir($dir);
	opendir (DIR, $dir) || &cgiError ("Opening $dir Failed:", "$!");
	@allfiles = readdir(DIR);
	$totalnum=@allfiles;
	closedir (DIR);

	for($i=1; $i<$totalnum; $i++) {

		if (-d $allfiles[$i]) { 
			if ($allfiles[$i] ne "." && $allfiles[$i] ne ".." && $type ne "" && $subdirs eq "yes") { 
				if ($type ne ".") {
					push(@dirs,"$type/$allfiles[$i]");
				}
				else {
					push(@dirs,"$allfiles[$i]");
				}
			}
		}

		if (-T $allfiles[$i]) 
			{
			($file,$ext) = split(/\./, $allfiles[$i]);

			foreach $searchtype (@searchtypes) 
				{
				if ($ext =~ /$searchtype/i) 
					{
					open (SEARCH,"$allfiles[$i]") || &cgiError ("Open $allfiles[$i] Failed:", "$!");
					@search = <SEARCH>;
					close(SEARCH);

					if ($type ne ".") 
						{
						$temp_file = "$type/$allfiles[$i]";
						}
					else
						{
						$temp_file = "$allfiles[$i]";
						}

					foreach $text (@search) 
						{
						chomp($text);

						$text =~ s/<([^>]|\n)*>//g;
						$text =~ s/&([^;])*;//g;
						$text =~ s/])*-->//g;

						if ($connector eq "or" && $case eq "insensitive") {
							foreach $word (@keywords) {
								if (!($text =~ /$word/i)) { $match = 0;} 
								else { $matches{$temp_file} = $text; last;}
							}
						}

						if ($connector eq "and" && $case eq "insensitive") {
							if (!($text =~ /$keyword/i)) { $match = 0;} 
							else { $matches{$temp_file} = $text; last;}
						}

						if ($connector eq "or" && $case eq "sensitive") {
							foreach $word (@keywords) {
								if (!($text =~ /$word/)) { $match = 0;} 
								else { $matches{$temp_file} = $text; last;}
							}
						}

						if ($connector eq "and" && $case eq "sensitive") {
							if (!($text =~ /$keyword/)) { $match = 0;} 
							else { $matches{$temp_file} = $text; last;}
						}
					}
				}
			}
		}
	}
}

foreach $key (sort(keys %matches)) { push(@matches,$key); push(@summary,$matches{$key}); }

&print_header;

print <<SEARCH_PAGE;

      <table width="100%" border="0" cellpadding="10" cellspacing="0">
      <tr bgcolor="#000000">
      <td valign="middle" align="left"><font face="Verdana, Arial, Helvetica" color="#FFFFFF" size="3">Search result</font></td>
      <td valign="bottom" align="right"> 

SEARCH_PAGE

	$total = @matches;
	$real_total = $total;
	eval { $totalnum = int (($total + $result_num)/ $result_num); };

	if ($page ) { ($liststart,$listend) = split(/ /,$page); }
	
	else { 
		if ($total < $result_num) { $liststart = 0; $listend = $total; }
		else { $liststart = 0; $listend = $result_num; }
	}
	
if ($total > $result_num) 
	{
	print "<form action=\"$search_url\" method=\"POST\" name=\"cerca\">\n";
	print "<input type=\"hidden\" name=\"header\" value=\"$header\">\n";
	print "<input type=\"hidden\" name=\"footer\" value=\"$footer\">\n";
	print "<input type=\"hidden\" name=\"dirs\" value=\"$mydirs\">\n";
	print "<input type=\"hidden\" name=\"dirs\" value=\"$mydirs\">\n";

	print "<font face=\"Verdana, Arial, Helvetica\" color=\"#FFFFFF\" size=\"1\">Show documents:[/b]</font>";
	print "<select name=\"page\" onChange=\"javascript:document.cerca.submit();\">\n";
	
	for($i=0; $i<$totalnum; $i++) 
		{
		eval { $num = ($i * $result_num); };
		eval { $end = ($num + $result_num); };
		if ($matches[$num] ) 
			{
			if ($total<$end) { $end = $total; }
		
			$real_num = $num;
			eval { $num = ($num + 1); };
			print " <option value=\"$real_num $end\">$num-$end</option>\n";
			eval { $num = ($num - 1); };
			}
		}

	print "</select><input type=\"hidden\" name=\"show_search\" value=\"no\">\n";
	print "<input type=\"hidden\" name=\"show_summary\" value=\"$show_summary\">\n";
	print "<input type=\"hidden\" name=\"result_num\" value=\"$result_num\">\n";
	print "<input type=\"hidden\" name=\"case\" value=\"$case\">\n";
	print "<input type=\"hidden\" name=\"connector\" value=\"$connector\">\n";
	print "<input type=\"hidden\" name=\"keyword\" value=\"$keyword\"></form></p>\n";
	}
else
	{
	print "\n"
	}

eval { $first = ($liststart + 1); };
$last = $listend;
if ($listend<=$result_num && $end<=$result_num) { $last = $total; }

print "</td></tr></table>\n";

#
print "<table width=\"100%\" border=\"0\" cellpadding=\"10\" cellspacing=\"0\"><tr><td colspan=\"2\">\n";

if ($total eq "0") { print "<font face=\"Verdana, Arial, Helvetica\" size=\"2\" color=\"#FFFFFF\">Sorry, No Matches Found!
We did not find anything that matched your search criteria.
Please try searching again with a different combination of keywords.</font>\n";}

else {

      print "

<font face=\"Verdana, Arial, Helvetica\" size=\"2\" color=\"#FFFFFF\">Below are the results of your search for: <font color=\"#333300\">$keyword</font>
Documents  from $first to $last of $real_total Matches

\n";
	if ($liststart eq "$listend") { eval { $liststart = ($liststart - 1); }; }

	for($liststart; $liststart<$listend; $liststart++) 
		{

		if ($matches[$liststart]) 
			{
			$match = $matches[$liststart];
			$summary = $summary[$liststart];

#			if ($summary eq "") { $summary = "Nessun sommario disponibile."; }

			print "<a href=\"$base_url/$match\">$match</a>
\n";
#			print "URL: $base_url/$match
\n";
#			if ($show_summary eq "yes") { print "Sommario: $summary
"; }
#			print "
\n\n";
			}
		}
	print "</font><font face=\"Verdana, Arial, Helvetica\" size=\"1\" color=\"#FFFFFF\">

";
	print "
Didn't find what you were looking for? Try searching again with a more specific keyword.

\n\n";
	}

print "<form action=\"$search_url\" method=\"POST\" name=\"show_search\">\n";
print "<font color=\"#FFFFFF\" face=\"Verdana, Arial, Helvetica\" size=\"2\">Search for: <input type=\"text\" size=\"36\" name=\"keyword\" MAXLENGTH=\"35\"></font>

\n";
print "<input type=\"hidden\" name=\"show_search\" value=\"no\">\n";
print "<input type=\"hidden\" name=\"result_num\" value=\"$result_num\">\n";
print "<input type=\"hidden\" name=\"case\" value=\"$case\">\n";
print "<input type=\"hidden\" name=\"connector\" value=\"$connector\">\n";
print "<input type=\"hidden\" name=\"show_summary\" value=\"$show_summary\">\n";
print "<input type=\"hidden\" name=\"header\" value=\"$header\">\n";
print "<input type=\"hidden\" name=\"footer\" value=\"$footer\">\n";
print "<input type=\"hidden\" name=\"dirs\" value=\"$mydirs\">
\n";
#print "<input type=\"submit\" value=\"Cerca\"></form>
\n";

print "</td></tr></table>\n";

&print_footer;


# sub routines #

sub display_page {
&print_header;

print "<form action=\"$search_url\" method=\"POST\" name=\"show_search\">";
print <<END_SEARCH_PAGE;
          <table width="100%" border="0" cellspacing="0" cellpadding="10">
            <tr bgcolor="#000000"> 
              <td colspan="3"><font face="Verdana, Arial, Helvetica" color="#FFFFFF" size="2">
Enter a few words to search for and
you will get a list of every file that contains a match</font></td>
            </tr>
            <tr> 
              <td bgcolor="#666666" width="143" align="right"><font color="#FFFFFF" face="Verdana, Arial, Helvetica" size="1">
             Search as </font></td>
              <td width="218" bgcolor="#666666"> <font face="Verdana, Arial, Helvetica" size="1"> 
                <input type="radio" name="connector" value="or" checked>
                <font color="#FFFFFF">Keywords </font></font> </td>
              <td width="139" bgcolor="#333333"> <font face="Verdana, Arial, Helvetica" size="1"> 
                <input type="radio" name="connector" value="and">
                <font color="#FFFFFF">Phrase</font></font></td>
            </tr>
            <tr bgcolor="#7F7F2A"> 
              <td width="143" align="right" bgcolor="#91826E"><font color="#FFFFFF" face="Verdana, Arial, Helvetica" size="1">Result per page</font></td>
              <td colspan="2" bgcolor="#91826E"> 
                <table width="100%" border="0" cellspacing="0" cellpadding="5">
                  <tr bgcolor="#91826E">
                    <td><font face="Verdana, Arial, Helvetica" size="1"> 
                      <input type="radio" name="result_num" value="10">
                      <font color="#FFFFFF">10</font></font></td>
                    <td><font face="Verdana, Arial, Helvetica" size="1"> 
                      <input type="radio" name="result_num" value="20" checked>
                      <font color="#FFFFFF">20</font></font></td>
                    <td><font face="Verdana, Arial, Helvetica" size="1"> 
                      <input type="radio" name="result_num" value="30">
                      <font color="#FFFFFF">30</font></font></td>
                    <td><font face="Verdana, Arial, Helvetica" size="1"> 
                      <input type="radio" name="result_num" value="40">
                      <font color="#FFFFFF">40</font></font></td>
                    <td><font face="Verdana, Arial, Helvetica" size="1"> 
                      <input type="radio" name="result_num" value="50">
                      <font color="#FFFFFF">50</font></font>
		<input type="hidden" name="show_search" value="no">
END_SEARCH_PAGE

print "<input type=\"hidden\" name=\"header\" value=\"$header\">\n";
print "<input type=\"hidden\" name=\"footer\" value=\"$footer\">\n";
print "<input type=\"hidden\" name=\"dirs\" value=\"$mydirs\">\n";

print <<END_SEARCH_PAGE;
		     </td>
                  </tr>
                </table>
              </td>
            </tr>
          </table>
END_SEARCH_PAGE

&print_footer;

exit;
}

sub print_header {
    print "Content-type: text/html\n\n";

	foreach $linez (@dataa) {
	chomp($linez);
	print "$linez\n";
	}
}

sub print_footer {
	foreach $liney (@datab) {
	chomp($liney);
	print "$liney\n";
	}
}

sub get_header_footer {
	open (FILEA,"$base_dir/$header") || &cgiError ("Reading Header Failed:", "$!");
	@dataa = <FILEA>;
	close(FILEA);

	open (FILEB,"$base_dir/$footer") || &cgiError ("Reading Footer Failed:", "$!");
	@datab = <FILEB>;
	close(FILEB);
}

sub cgiError {
	my ($error_cause,$error) = @_;

	if ($error_cause eq "") { $error_cause = "Error:"; }
	if ($error eq "") { $error = "The script encountered problems and terminated"; }

	&print_header;
	print "<h3>$error_cause</h3>$error

";
	&print_footer;

exit;
}

**shishii** · 18-08-2005, 16:16

un metodo è inserire dopo

foreach $type (@dirs) {

questa istruzione:

next if ($type ~= /vti/);

al posto di vti metti il nome che vuoi evitare.

**biocco** · 19-08-2005, 11:20

scivendo cosi (ho agginto solo questo):
foreach $type (@dirs) {
next if ($type ~= /_vti_cnf/);
chomp($type);
$dir = "$base_dir/$type";

ottengo cosi:

CGI Error
The specified CGI application misbehaved by not returning a complete set of HTTP headers.

**shishii** · 19-08-2005, 17:28

fai la prova disattivando la linea così:

foreach $type (@dirs) {
# next if ($type ~= /_vti_cnf/);
chomp($type);
$dir = "$base_dir/$type";

se ti funziona di nuovo allora la cosa va approfondita, se continua a non funzionare vuol dire che c'è qualche altro errore.

Discussione: cerca con perl...

Strumenti discussione

Ricerca discussione

Visualizza

cerca con perl...

Permessi di invio