Making Perl Script Make Two Calls to Search Engine

kyoko · Mar 30, 2000

I have a perl script that parses an index for search results. Is there a way to make the script make a call to two separate indices and display the results as one search.  I want to avoid merging the indices, but instead want to  have a divider between the two sets of results.  ( I have done this with another script, but this one eludes me.) The script: #!/usr/local/bin/perl # REQUIRED # Firstly check the path to PERL executable in the first line of the script. $swish_path =  '/path/to/swish-e'; # path to SWISH executable # You should carefully set configuration options in your `swish.conf' file. # Search imports `IndexName' and `IndexPointer' values and # its behaviour considerably depends on `ReplaceRules' settings. ################################################################################# # YOUR CHOICE %form = ( 'ipath', '/path/index.swish');#default `Search path' $bgcolor =     '"#ffffff"'; # background color $textcolor =   'black'; # text color on `bgcolor' background $atextcolor =  'white'; # text color on `linkcolor' background and # activ link color $headercolor = 'maroon'; # headers color $linkcolor =   'blue'; # links color and alternate background $alinkcolor =  'lime'; # links color on `linkcolor' background $vlinkcolor =  'navy'; # visited links color ################################################################################# # YOU SHOULD NOT ALTER THE REST OF THE SCRIPT read( STDIN, $buffer, $ENV{ 'CONTENT_LENGTH' } ); @pairs = split( /&/, $buffer ); foreach $pair( @pairs ) {   ( $name, $value ) = split( /=/, $pair );   $value =~ tr/+/ /;   $value =~ s/%([a-fA-F0-9][a-fA-F0-9])/pack("C",hex($1))/eg;   $form{ $name } = $value; } open( RES, ( $form{ 'terms' } ) ?   "$swish_path -f $form{ 'ipath' } -m 50 -t HBthe -w $form{ 'terms' } ¦" :   "$swish_path -f $form{ 'ipath' } -m 1 -w 0 ¦" ); _1: while( <RES> ) {   $Name = $1 if /^# Name: (.*)$/o;   $Pointer = $1 if /^# Pointer: (.*)$/o;   last _1 if defined( $Name ) && defined( $Pointer );   $error = $1 if /^err: (.*)$/o;   $error = 'Invalid search words' if /^options:/o; } $hc = "FONT COLOR=$headercolor"; $atc = "FONT COLOR=$atextcolor"; $alc = "FONT COLOR=$alinkcolor"; $_ = <<EOP; Content-type: text/html <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> <html> <head> <title></title> </head> </head> <table border="0" cellpadding="0" cellspacing="0" width="100%">   <tr>     <td width="50%" valign="top"><img SRC="<A HREF="

http://www.site.logo.jpg&quot"

TARGET="_new">

http://www.site.logo.jpg&quot</A>;

ALT=" " width="442" height="103" BORDER="0">     </td>     <td width="50%" style="padding-left: 0" bordercolor="#808080">              Tell Your Suppliers You Found Them on       PetroPages</td>   </tr> </table> EOP s/VALUE=.$form{ 'ipath' }./$& CHECKED/o; s/VALUE=.$form2{ 'ipath' }./$& CHECKED/o; print; if( $error ) {   print " <$hc>Error: $error \n"; } elsif( $form{ 'terms' } ) {   print " <$hc>Documents found: (red ¶ sign shows confidence in the match, best match first) \n";   while( <RES> )   {     print " 99 ? 'red' : 'black' ).         "\">¶ <A HREF=\"$2\">$3</A> "       if /^([0-9]*) +([^ ]*) +.([^\"]*). +([0-9]*)/o;     print "No documents found \n" if /^err/o;   }   print " \n"; } else {   print <<EOP;     <UL> EOP } print <<EOP; * <table border="0" cellpadding="0" cellspacing="0" width="100%">   <tr>     <td width="100%" colspan="2">TEXT Display: </td>   </tr>  </table>                <table>  <tr>  <td width="50%"><form METHOD="POST" ACTION="<A HREF="

http://www.site.com/search.cgi&quot"

TARGET="_new">

http://www.site.com/search.cgi&quot...      <input

TYPE="text"       NAME="terms" SIZE="21" VALUE><input TYPE="submit" VALUE="  Product Categories "       ALIGN="top"       style="background-color: #000000; color: #FFFFFF; font-family: sans-serif; font-size: 10pt; font-weight: normal; background-image: url('/images/DoSearch.jpg')"><input TYPE="text"       NAME="terms" SIZE="21" VALUE><input TYPE="submit" VALUE="  Product Categories "       ALIGN="top"       style="background-color: #000000; color: #FFFFFF; font-family: sans-serif; font-size: 10pt; font-weight: normal; background-image: url('/images/DoSearch.jpg')"></td>      </tr> </table> EOP __END__

goBoating · Mar 30, 2000

a not very sophisticated approach would be simply duplicate the sections which query the index and print the results.  A better approach would be to catch, from the calling process, a list of indeces to be queried and loop through them.  This above code is fairly straight forward.  Consequently, I will assume you are new to PERL (or you would not be asking).  So, here are a few notes about the code. # This opens a pipe or handle to another process which  querys your index. # roughly translated...... # open HANDLE, someName, if one of these following things works (returns non-zero) open( RES, ( $form{ 'terms' } ) ?   "$swish_path -f $form{ 'ipath' } -m 50 -t HBthe -w $form{ 'terms' } ¦" :   "$swish_path -f $form{ 'ipath' } -m 1 -w 0 ¦" ); #  the 'while (<RES&gt

' is listening to the handle opened above # while it gets stuff from the index via the pipe/handle 'RES', catch some stuff # or jump out of the while. _1: while( <RES> ) {   $Name = $1 if /^# Name: (.*)$/o;   $Pointer = $1 if /^# Pointer: (.*)$/o;  # bail out if you have name  and pointer   last _1 if defined( $Name ) && defined( $Pointer );                                                                              $error = $1 if /^err: (.*)$/o;   $error = 'Invalid search words' if /^options:/o; } # while you get results, in '$_', back from the opened handle (RES) # print them to the client via STDOUT # note that the conditional statement follows the action taken # if the condition is true.  PERL lets you turn this around and it can # be confusing if you are not used to looking at it.   while( <RES> )   {     # print contents of '$_' with a little tweaking on the fly     print " 99 ? 'red' : 'black' ).         "\">¶ <A HREF=\"$2\">$3</A> "       if /^([0-9]*) +([^ ]*) +.([^\"]*). +([0-9]*)/o; # print error statement          if error is returned.     print "No documents found \n" if /^err/o;   } close RES; # do this prior to next iteration # do next iteration and then print the end stuff for the html.

kyoko · Apr 6, 2000

Thanks!  I got it figured out.  %form=('ipath1', '/path/index', 'ipath2', '/path/index); Plus duplicating the sections.  That was what I wanted, even if apparently unsophisticated. But out of curiosity, can you give me a clue as what this is about? "A better approach would be to catch, from the calling process, a list of indeces to be queried and loop through them."

Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

Making Perl Script Make Two Calls to Search Engine

kyoko

Technical User

goBoating

Programmer

kyoko

Technical User

Similar threads

Part and Inventory Search

Sponsor