Annotation of pandia/crawler, revision 1.5
1.1 snw 1: #!/usr/bin/env perl
2:
3: #
1.5 ! snw 4: # $Id: crawler,v 1.4 2025/06/27 16:20:30 snw Exp $
1.1 snw 5: # Copyright (C) 2025 Coherent Logic Development LLC
6: #
7: # Author: Serena Willis <snw@coherent-logic.com>
8: #
9: # Licensed AGPL-3.0
10: #
1.2 snw 11: # $Log: crawler,v $
1.5 ! snw 12: # Revision 1.4 2025/06/27 16:20:30 snw
! 13: # Add blacklist
! 14: #
1.4 snw 15: # Revision 1.3 2025/06/27 02:14:47 snw
16: # Initial operational capability
17: #
1.3 snw 18: # Revision 1.2 2025/06/25 19:38:48 snw
19: # Add indexer
20: #
1.2 snw 21: # Revision 1.1 2025/06/25 13:44:37 snw
22: # Renaming
23: #
1.1 snw 24: # Revision 1.2 2025/06/25 03:10:01 snw
25: # Initial working crawler
26: #
27: # Revision 1.1.1.1 2025/06/23 23:17:08 snw
28: # Initial commit
29: #
30: #
31:
32: use Getopt::Long;
33: use HTTP::Tiny;
34: use HTML::TreeBuilder;
35: use URI;
36: use DBI;
37: use WWW::RobotRules;
1.4 snw 38: use Fcntl qw(:flock);
1.1 snw 39: use LWP::Simple qw(get);
40:
1.4 snw 41: my $rules = WWW::RobotRules->new('pandia-crawler/0.0.1');
1.1 snw 42: my $dbh = "";
43: my $dsn = "";
44: my $skips = 0;
45: my $inserts = 0;
46: my $seed = "";
47: my $depth = 0;
1.2 snw 48: my $blacklist_matches = 0;
49: my $robots_txt_denies = 0;
50: my $invalid_scheme_skips = 0;
1.1 snw 51:
52: sub store_url {
1.2 snw 53: my ($url, $parent) = @_;
54:
1.4 snw 55: if($url ne "" && length($url) <= 255 && substr($url, 0, 6) ne "mailto" && substr($url, 0, 4) eq "http") {
56:
1.2 snw 57: my $u = URI->new($url);
1.4 snw 58: my $domain = $u->host;
1.2 snw 59: my $scheme = $u->scheme;
60:
61: my $sth = $dbh->prepare("INSERT INTO url_domains (url_domain) VALUES (?)");
62: $sth->execute($domain);
63:
64: my $ins = $dbh->prepare("INSERT INTO crawl_queue (url, parent_url, url_domain, scheme) VALUES (?, ?, ?, ?)");
1.1 snw 65:
1.2 snw 66: if(not $ins->execute($url, $parent, $domain, $scheme)) {
67: $skips = $skips + 1;
68: print "d";
69: }
70: else {
71: print ".";
72: $inserts = $inserts + 1;
73: if($depth < $maxdepth) {
74: $depth = $depth + 1;
75: crawl_url($url);
76: }
77: else {
78: print "l";
79: }
1.1 snw 80: }
1.2 snw 81: }
82: else {
83: print "x";
84: }
1.1 snw 85: }
86:
87: sub crawl_url {
88: my ($url) = @_;
89:
90: my $u = URI->new($url);
91:
92: if ($u->scheme ne "http" && $u->scheme ne "https") {
1.2 snw 93: $invalid_scheme_skips = $invalid_scheme_skips + 1;
94: print "s";
1.1 snw 95: return;
96: }
97:
98: my $sth = $dbh->prepare("SELECT url_domain FROM blacklist WHERE url_domain=?");
99: $sth->execute($u->host);
100: if($sth->rows > 0) {
1.2 snw 101: print "b";
102: $blacklist_matches = $blacklist_matches + 1;
1.1 snw 103: return;
104: }
105:
106: my $robots_url = $u->scheme . '://' . $u->host . "/robots.txt";
107:
108: my $robots_txt = get $robots_url;
109: $rules->parse($robots_url, $robots_txt) if defined $robots_txt;
110:
111: if(!$rules->allowed($url)) {
1.2 snw 112: print "r";
113: $robots_txt_denies = $robots_txt_denies + 1;
1.1 snw 114: return;
115: }
116:
117: my $origurl = $url;
118: ($baseurl) = $origurl =~ m! (.+?\w) (?: /|\z) !x;
119:
120: my $http = HTTP::Tiny->new(agent => "pandia-crawler/0.0.1");
121: my $tree = HTML::TreeBuilder->new();
122:
123: my $response = $http->get($url);
1.3 snw 124:
125:
1.1 snw 126: $tree->parse($response->{content});
127:
128: my @links = $tree->find_by_tag_name('a');
129:
130: my $href = "";
131: my $firstchar = "";
132: my $final = "";
133:
134: foreach my $link (@links) {
135: $href = $link->attr('href');
136: $firstchar = substr($href, 0, 1);
137: $final = "";
138:
139: if($firstchar eq '/') {
140: $final = $baseurl . $href;
141: }
142: elsif($href eq '##') {
143: $final = $baseurl;
144: }
145: elsif($firstchar eq '#') {
146: $final = $baseurl . '/' . $href;
147: }
148: else {
149: $final = $href;
150: }
151:
1.2 snw 152: store_url($final, $url);
1.1 snw 153: }
154:
155: $depth = $depth - 1;
156: }
157:
158: $| = 1;
159: print "pandia crawler v0.0.1\n";
160: print " Copyright (C) 2025 Coherent Logic Development LLC\n\n";
161:
162: GetOptions("dbhost=s" => \$dbhost,
163: "dbname=s" => \$dbname,
164: "dbusername=s" => \$dbusername,
165: "dbpw=s" => \$dbpw,
166: "seed=s" => \$seed,
167: "maxdepth=n" =>\$maxdepth)
168: or die("error in command line arguments");
169:
170: print "pandia: connecting to $dbname database at $dbhost...";
171:
172: $dsn = "DBI:mysql:database=$dbname;host=$dbhost;port=3306;mysql_connect_timeout=5;";
173: $dbh = DBI->connect($dsn, $dbusername, $dbpw, {RaiseError => 0, PrintError => 0});
174: die "pandia: failed to connect to MySQL database: DBI->errstr()" unless $dbh;
175:
176: print "[OK]\n";
177:
178: if($seed ne "") {
179: print "pandia: crawling seed $seed to a maximum depth of $maxdepth";
180: sleep 1;
181: crawl_url($seed);
182: print "[OK]\n";
183: }
184: else {
1.5 ! snw 185: open my $file, ">", "/tmp/pandia_crawler.lock" or die $!;
1.4 snw 186: flock $file, LOCK_EX|LOCK_NB or die "Unable to lock file $!";
187:
1.1 snw 188: my $sth = $dbh->prepare("SELECT url FROM crawl_queue");
189: $sth->execute();
190: my $qlen = $sth->rows;
191:
192:
193: print "pandia: crawling queue with length of $qlen to a maximum depth of $maxdepth";
194: sleep 1;
195: while (my @row = $sth->fetchrow_array()) {
196: my $url = @row[0];
197: crawl_url($url);
198: }
199: print "[OK]\n";
200: }
201:
202:
203: my $total = $inserts + $skips;
204:
205: print "pandia: $inserts URL(s) enqueued for analysis; $skips skipped [$total URL(s) seen this run]\n";
1.2 snw 206: print " - $blacklist_matches blacklist matches\n";
207: print " - $invalid_scheme_skips URLs skipped due to invalid scheme\n";
208: print " - $robots_txt_denies URLs skipped due to robots.txt\n";
209:
210:
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to crawler CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Coherent Logic Development] / pandia