Optimize search and add guards.

3 files changed, 111 insertions, 79 deletions

diff --git a/README.txt b/README.txt
index ebd6d3f..b744290 100644
--- a/README.txt
+++ b/README.txt
@@ -12,12 +12,13 @@ $ cd /var/www/htdocs
 # find . -type d -exec chmod 554 {} +
 # chmod 554 cgi-bin/find.cgi
 
-Checking CGI script errors in chroot:
+Search lock setup in chroot
 
-# chroot /var/www/ htdocs/www.asciimx.com/cgi-bin/find.cgi
+# mkdir -p /var/www/tmp/search_locks
+# chown www:www /var/www/tmp/search_locks
+# chmod 700 /var/www/tmp/search_locks
 
-Search Perl deps:
+Checking CGI script errors in chroot:
 
-# doas pkg_add p5-File-Slurper
-# doas pkg_add p5-HTML-TreeBuilder-XPath
+# chroot /var/www/ htdocs/www.asciimx.com/cgi-bin/find.cgi
 
diff --git a/cgi-bin/find.cgi b/cgi-bin/find.cgi
index fc1d4af..3e5c8a2 100644
--- a/cgi-bin/find.cgi
+++ b/cgi-bin/find.cgi
@@ -1,86 +1,124 @@
 #!/usr/bin/perl
 
+use strict;
+use warnings;
+use Storable qw(retrieve);
 use Encode qw(decode_utf8);
 use HTML::Escape qw(escape_html);
 
-my $search_text = '';
+# Configuration
+my $max_parallel   = 50;                    # max no. of parallel searches         
+my $lock_timeout   = 30;                    # drop stale locks after this many seconds
+my $max_results    = 20;                    # max search results
+my $min_query_len  = 3;                     # min query length to avoid matching 'a', 'e'
+my $index_file     = 'search_index.dat';    # index file
+my $lock_dir       = '/tmp/search_locks';   # lock file directory
+
+# Concurrency control
+mkdir $lock_dir, 0777 unless -d $lock_dir;
+my $active_count = 0;
+my $now = time();
+
+opendir(my $dh, $lock_dir);
+while (my $file = readdir($dh)) {
+    next unless $file =~ /\.lock$/;
+    my $path = "$lock_dir/$file";
+    my $mtime = (stat($path))[9] || 0;
+    ( $now - $mtime > $lock_timeout ) ? unlink($path) : $active_count++;
+}
+closedir($dh);
+
+# Too many search requests
+if ($active_count >= $max_parallel) {
+    print "Content-Type: text/html\n\n";
+    render_html("<p>Server busy. Please try again in a few seconds.</p>", "", (localtime)[5]+1900);
+    exit;
+}
 
-if ($ENV{QUERY_STRING} =~ /^q=([^&]*)/) {
+my $lock_file = "$lock_dir/$$.lock";
+open(my $fh_lock, '>', $lock_file);
+
+# Decode search text as utf-8, toss non-printable chars, trim
+my $search_text = '';
+if (($ENV{QUERY_STRING} || '') =~ /^q=([^&]*)/) {
     $search_text = decode_utf8($1 // "");
-    $search_text =~ s/\P{Print}//g;  # toss any non-printable utf-8 characters
+    $search_text =~ s/\P{Print}//g; 
     $search_text = substr($search_text, 0, 64);
     $search_text =~ s/^\s+|\s+$//g;
 }
 
-my @results;
+# Pre-prepare common template variables
+my $safe_search_text = escape_html($search_text);
+my $year = (localtime)[5] + 1900;
 
-# Search only index.html files inside the first level of subdirectories
-my $start_dir = '../log'; 
-my @files = glob("$start_dir/*/index.html");
+print "Content-Type: text/html\n\n";
 
-foreach my $path (@files) {
-    # Skip if the path is a symlink or not a file
-    next if -l $path || ! -f $path;
+# Input validation
+if ($search_text eq '') {
+    final_output("<p>Please enter a search term above.</p>");
+}
 
-    next unless open(my $fh, '<:utf8', $path);
-    my $html = do { local $/; <$fh> };
-    close($fh);
-  
-    my ($text) = $html =~ m|<main>(.*?)</main>|is;
-    $text =~ s|<[^>]+>| |g;
-    $text =~ s|\s+| |g;
+if (length($search_text) < $min_query_len) {
+    final_output("<p>Search term is too short. Please enter at least $min_query_len characters.</p>");
+}
+
+if (!-f $index_file) {
+    final_output("<p>Search temporarily unavailable.</p>");
+}
 
-    next unless $text =~ /(.{0,40})(\Q$search_text\E)(.{0,40})/is;
+my $index = retrieve($index_file);
+my @results;
+my $found = 0;
+
+foreach my $url (sort keys %$index) {
+    last if $found >= $max_results;
+    my $data = $index->{$url};
+
+    # Grab 80 char snippet to chop at a word boundary later
+    next unless $data->{c} =~ /(.{0,40})(\Q$search_text\E)(.{0,40})/is;
     my ($before, $actual, $after) = ($1, $2, $3);
+    $found++;
 
-    # Trim if we cut into the middle of a sentence
+    # Chop at 25 or word boundary
     $after =~ s/\s\S*$// if length($after) > 25;
     $before =~ s/^.*?\s// if length($before) > 25;
 
-    if ($before =~ /\S/) { # If before has non-whitespace characters
-        $before = ucfirst($before);
-    } else {
-        $before = "";      # Clear any stray spaces
-        $actual = ucfirst($actual);
-    }
-
-    my $safe_before = escape_html($before);
-    my $safe_actual = escape_html($actual);
-    my $safe_after  = escape_html($after);
-    my $snippet = "${safe_before}<b>${safe_actual}</b>${safe_after}...";
+    $before = ($before =~ /\S/) ? ucfirst($before) : "";
+    $actual = ($before eq "") ? ucfirst($actual) : $actual;
 
-    my ($title) = $html =~ m|<title>(.*?)</title>|is;
-    my $safe_title  = escape_html($title);
-
-    $path =~ s|^\.\./||;
+    my $snippet = escape_html($before) . "<b>" . escape_html($actual) . "</b>" . escape_html($after) . "...";
 
     push @results, { 
-        path    => $path,
-        title   => $safe_title, 
+        path    => $url,
+        title   => escape_html($data->{t}), 
         snippet => $snippet 
     };
 }
 
-print "Content-Type: text/html\n\n";
-
-my $list;
-if ($search_text eq '') {
-    $list = "<p>Please enter a search term above.</p>";
-} elsif (@results == 0) {
-    $list = "<p>No results found for \"<b>$search_text</b>\".</p>";
+# Format results list
+my $list_html = "";
+if (@results == 0) {
+    $list_html = "<p>No results found for \"<b>$safe_search_text</b>\".</p>";
 } else {
-    $list = "<ul>";
-    foreach my $res (@results) {
-        my $url = $res->{path};
-        $list .= "<li><a href=\"/$url\">$res->{title}</a><br><small>$res->{snippet}</small></li>";
-    }
-    $list .= "</ul>";
+    $list_html = "<ul>" . join('', map { 
+        "<li><a href=\"/$_->{path}\">$_->{title}</a><br><small>$_->{snippet}</small></li>" 
+    } @results) . "</ul>";
 }
 
-my $safe_search_text = escape_html($search_text);
-my $year = (localtime)[5] + 1900;
+final_output($list_html);
 
-print <<"HTML";
+# Helper to ensure layout is always preserved
+sub final_output {
+    my ($content) = @_;
+    render_html($content, $safe_search_text, $year);
+    close($fh_lock) if $fh_lock;
+    unlink($lock_file) if -f $lock_file;
+    exit;
+}
+
+sub render_html {
+    my ($content, $q_val, $yr) = @_;
+    print <<"HTML";
 <!DOCTYPE html>
 <html lang="en-us">
 <head>
@@ -105,19 +143,21 @@ print <<"HTML";
         <div class="container">
             <h2>Search</h2>
             <form action="" method="GET">
-                <input id="search-box" type="text" name="q" value="$safe_search_text">
+                <input id="search-box" type="text" name="q" value="$q_val">
                 <input id="search-btn" type="submit" value="Search">
             </form>
-            $list
+            $content
         </div>
     </main>
     <div class="footer">
         <div class="container">
             <div class="twelve columns right container-2">
-                <p id="footer-text">&copy; ASCIIMX - $year</p>
+                <p id="footer-text">&copy; ASCIIMX - $yr</p>
             </div>
         </div>
     </div>
 </body>
 </html>
 HTML
+}
+
diff --git a/cgi-bin/indexer.pl b/cgi-bin/indexer.pl
index d0314a1..38a918e 100644
--- a/cgi-bin/indexer.pl
+++ b/cgi-bin/indexer.pl
@@ -5,13 +5,13 @@ use warnings;
 use Storable qw(nstore);
 use HTML::Entities qw(decode_entities);
 
-my $built_site_dir = '../_site/log';
-my $output_file    = 'search_index.dat';
+# --- Configuration ---
+my $built_site_dir = '../_site/log'; 
+my $output_file    = '../_site/cgi-bin/search_index.dat';
 my %index;
 
 print "Building search index from $built_site_dir...\n";
 
-# glob finds every index.html in subdirectories of /log/
 foreach my $path (glob("$built_site_dir/*/index.html")) {
     next unless open(my $fh, '<:utf8', $path);
     my $html = do { local $/; <$fh> };
@@ -22,23 +22,19 @@ foreach my $path (glob("$built_site_dir/*/index.html")) {
     my ($main)  = $html =~ m|<main>(.*?)</main>|is;
     $main //= '';
 
-    # Remove code and pre blocks to keep index prose-only
+    # Strip HTML and clean prose
     $main =~ s|<pre[^>]*>.*?</pre>| |gs;
     $main =~ s|<code[^>]*>.*?</code>| |gs;
-
-    # Strip all remaining HTML tags
     $main =~ s|<[^>]+>| |g;
-
-    # Decode entities (e.g., &amp; -> &) for accurate searching
     $main = decode_entities($main);
-
-    # Normalize whitespace (squash multiple spaces/newlines)
     $main =~ s|\s+| |g;
     $main =~ s/^\s+|\s+$//g;
 
-    # Map file path to the final web URL
-    # Example: ../_site/log/arduino/index.html -> /log/arduino/index.html
-    (my $url = $path) =~ s|^\.\./_site/|/|;
+    # Normalize path
+    my $url = $path;
+    $url =~ s|^\.\./_site/||; # Remove local build directory
+    $url =~ s|^\.\./||;       # Remove any leading dots
+    $url =~ s|^/+||;          # Remove leading slashes
 
     $index{$url} = {
         t => $title || "Untitled",
@@ -46,10 +42,5 @@ foreach my $path (glob("$built_site_dir/*/index.html")) {
     };
 }
 
-# Save using network-order binary (nstore) for portability
 nstore(\%index, $output_file);
-
-my $count = scalar(keys %index);
-my $size  = -s $output_file;
-printf("Index complete: %d files (%.2f KB)\n", $count, $size / 1024);
-
+printf("Index complete: %d files (%.2f KB)\n", scalar(keys %index), (-s $output_file) / 1024);