← Index
NYTProf Performance Profile   « block view • line view • sub view »
For ddd2.pl
  Run on Tue May 25 16:52:24 2010
Reported on Tue May 25 16:56:50 2010

File /project/perl/lib/WWW/Google/PageRank.pm
Statements Executed 368262
Statement Execution Time 4.39s
Subroutines — ordered by exclusive time
Calls P F Exclusive
Time		 Inclusive
Time		 Subroutine
5532211.95s3.00sWWW::Google::PageRank::::_mixWWW::Google::PageRank::_mix
1087961011.16s1.16sWWW::Google::PageRank::::_wsubWWW::Google::PageRank::_wsub
92221637ms3.89sWWW::Google::PageRank::::_compute_chWWW::Google::PageRank::_compute_ch
46111314ms4.31sWWW::Google::PageRank::::_compute_ch_newWWW::Google::PageRank::_compute_ch_new
20650121236ms236msWWW::Google::PageRank::::_waddWWW::Google::PageRank::_wadd
46111106ms153sWWW::Google::PageRank::::getWWW::Google::PageRank::get
9221213.4ms13.4msWWW::Google::PageRank::::CORE:unpackWWW::Google::PageRank::CORE:unpack (opcode)
922228.45ms8.45msWWW::Google::PageRank::::CORE:matchWWW::Google::PageRank::CORE:match (opcode)
461124.29ms4.29msWWW::Google::PageRank::::CORE:packWWW::Google::PageRank::CORE:pack (opcode)
111100µs455µsWWW::Google::PageRank::::newWWW::Google::PageRank::new
0000s0sWWW::Google::PageRank::::BEGINWWW::Google::PageRank::BEGIN
Call graph for these subroutines as a Graphviz dot language file.
Line State
ments		 Time
on line		 Calls Time
in subs		 Code
1package WWW::Google::PageRank;
2
3# -*- perl -*-
4
5395µs128µsuse strict;
# spent 28µs making 1 call to strict::import
63133µs1123µsuse warnings;
# spent 123µs making 1 call to warnings::import
7
8394µs1149µsuse vars qw($VERSION);
# spent 149µs making 1 call to vars::import
9
1037.71ms1144µsuse LWP::UserAgent;
# spent 144µs making 1 call to Exporter::import
1131.60ms1212µsuse URI::Escape;
# spent 212µs making 1 call to Exporter::import
12
1317µs$VERSION = '0.15';
14
15
# spent 455µs (100+355) within WWW::Google::PageRank::new which was called # once (100µs+355µs) by main::RUNTIME at line 6 of ddd2.pl
sub new {
16896µs my $class = shift;
17 my %par = @_;
18 my $self;
19 $self->{ua} = LWP::UserAgent->new(agent => $par{agent} ||
# spent 355µs making 1 call to LWP::UserAgent::new
20 'Mozilla/4.0 (compatible; GoogleToolbar 2.0.111-big; Windows XP 5.1)')
21 or return;
22 $self->{ua}->proxy('http', $par{proxy}) if $par{proxy};
23 $self->{ua}->timeout($par{timeout}) if $par{timeout};
24 $self->{host} = $par{host} || 'toolbarqueries.google.com';
25 bless($self, $class);
26}
27
28
# spent 153s (106ms+152) within WWW::Google::PageRank::get which was called 461 times, avg 331ms/call: # 461 times (106ms+152s) by main::RUNTIME at line 9 of ddd2.pl, avg 331ms/call
sub get {
29276668.5ms my ($self, $url) = @_;
30 return unless defined $url and $url =~ m[^https?://]i;
# spent 5.82ms making 461 calls to WWW::Google::PageRank::CORE:match, avg 13µs/call
31
32 my $ch = '6' . _compute_ch_new('info:' . $url);
# spent 4.31s making 461 calls to WWW::Google::PageRank::_compute_ch_new, avg 9.34ms/call
33 my $query = 'http://' . $self->{host} . '/search?client=navclient-auto&ch=' . $ch .
# spent 120ms making 461 calls to URI::Escape::uri_escape, avg 261µs/call
34 '&ie=UTF-8&oe=UTF-8&features=Rank&q=info:' . uri_escape($url);
35
36 my $resp = $self->{ua}->get($query);
# spent 148s making 461 calls to LWP::UserAgent::get, avg 321ms/call
374612.20ms138343.5ms if ($resp->is_success && $resp->content =~ /Rank_\d+:\d+:(\d+)/) {
# spent 20.5ms making 461 calls to HTTP::Message::content, avg 44µs/call # spent 20.3ms making 461 calls to HTTP::Response::is_success, avg 44µs/call # spent 2.63ms making 461 calls to WWW::Google::PageRank::CORE:match, avg 6µs/call
38 if (wantarray) {
39 return ($1, $resp);
40 } else {
41 return $1;
42 }
43 } else {
444546.71ms if (wantarray) {
45 return (undef, $resp);
46 } else {
47 return;
48 }
49 }
50}
51
52
# spent 4.31s (314ms+3.99) within WWW::Google::PageRank::_compute_ch_new which was called 461 times, avg 9.34ms/call: # 461 times (314ms+3.99s) by WWW::Google::PageRank::get at line 32, avg 9.34ms/call
sub _compute_ch_new {
53184475.7ms my $url = shift;
54
55 my $ch = _compute_ch($url);
# spent 1.62s making 461 calls to WWW::Google::PageRank::_compute_ch, avg 3.51ms/call
56 $ch = (($ch % 0x0d) & 7) | (($ch / 7) << 2);
57
5827660223ms101422.38s return _compute_ch(pack("V20", map {my $t = $ch; _wsub($t, $_*9); $t} 0..19));
# spent 2.27s making 461 calls to WWW::Google::PageRank::_compute_ch, avg 4.93ms/call # spent 101ms making 9220 calls to WWW::Google::PageRank::_wsub, avg 11µs/call # spent 4.29ms making 461 calls to WWW::Google::PageRank::CORE:pack, avg 9µs/call
59}
60
61
# spent 3.89s (637ms+3.25) within WWW::Google::PageRank::_compute_ch which was called 922 times, avg 4.22ms/call: # 461 times (375ms+1.90s) by WWW::Google::PageRank::_compute_ch_new at line 58, avg 4.93ms/call # 461 times (262ms+1.35s) by WWW::Google::PageRank::_compute_ch_new at line 55, avg 3.51ms/call
sub _compute_ch {
6217518232ms my $url = shift;
63
64 my @url = unpack("C*", $url);
# spent 13.4ms making 922 calls to WWW::Google::PageRank::CORE:unpack, avg 15µs/call
65 my ($a, $b, $c, $k) = (0x9e3779b9, 0x9e3779b9, 0xe6359a60, 0);
66 my $len = scalar @url;
67
68 while ($len >= 12) {
6927660361ms461057.1ms _wadd($a, $url[$k+0] | ($url[$k+1] << 8) | ($url[$k+2] << 16) | ($url[$k+3] << 24));
# spent 57.1ms making 4610 calls to WWW::Google::PageRank::_wadd, avg 12µs/call
70 _wadd($b, $url[$k+4] | ($url[$k+5] << 8) | ($url[$k+6] << 16) | ($url[$k+7] << 24));
# spent 51.1ms making 4610 calls to WWW::Google::PageRank::_wadd, avg 11µs/call
71 _wadd($c, $url[$k+8] | ($url[$k+9] << 8) | ($url[$k+10] << 16) | ($url[$k+11] << 24));
# spent 50.7ms making 4610 calls to WWW::Google::PageRank::_wadd, avg 11µs/call
72
73 _mix($a, $b, $c);
# spent 2.50s making 4610 calls to WWW::Google::PageRank::_mix, avg 542µs/call
74
75 $k += 12;
76 $len -= 12;
77 }
78
79 _wadd($c, scalar @url);
# spent 10.6ms making 922 calls to WWW::Google::PageRank::_wadd, avg 11µs/call
80
81 _wadd($c, $url[$k+10] << 24) if $len > 10;
82 _wadd($c, $url[$k+9] << 16) if $len > 9;
83 _wadd($c, $url[$k+8] << 8) if $len > 8;
84 _wadd($b, $url[$k+7] << 24) if $len > 7;
# spent 5.10ms making 461 calls to WWW::Google::PageRank::_wadd, avg 11µs/call
85 _wadd($b, $url[$k+6] << 16) if $len > 6;
# spent 4.97ms making 461 calls to WWW::Google::PageRank::_wadd, avg 11µs/call
86 _wadd($b, $url[$k+5] << 8) if $len > 5;
# spent 4.98ms making 461 calls to WWW::Google::PageRank::_wadd, avg 11µs/call
87 _wadd($b, $url[$k+4]) if $len > 4;
# spent 9.15ms making 831 calls to WWW::Google::PageRank::_wadd, avg 11µs/call
88 _wadd($a, $url[$k+3] << 24) if $len > 3;
# spent 10.4ms making 918 calls to WWW::Google::PageRank::_wadd, avg 11µs/call
89 _wadd($a, $url[$k+2] << 16) if $len > 2;
# spent 10.1ms making 922 calls to WWW::Google::PageRank::_wadd, avg 11µs/call
90 _wadd($a, $url[$k+1] << 8) if $len > 1;
# spent 12.0ms making 922 calls to WWW::Google::PageRank::_wadd, avg 13µs/call
91 _wadd($a, $url[$k]) if $len > 0;
# spent 9.93ms making 922 calls to WWW::Google::PageRank::_wadd, avg 11µs/call
92
93 _mix($a, $b, $c);
# spent 501ms making 922 calls to WWW::Google::PageRank::_mix, avg 543µs/call
94
95 return $c; # integer is positive always
96}
97
98
# spent 3.00s (1.95+1.05) within WWW::Google::PageRank::_mix which was called 5532 times, avg 543µs/call: # 4610 times (1.62s+877ms) by WWW::Google::PageRank::_compute_ch at line 73, avg 542µs/call # 922 times (324ms+177ms) by WWW::Google::PageRank::_compute_ch at line 93, avg 543µs/call
sub _mix {
991604281.76s my ($a, $b, $c) = @_;
100
101 _wsub($a, $b); _wsub($a, $c); $a ^= $c >> 13;
# spent 119ms making 11064 calls to WWW::Google::PageRank::_wsub, avg 11µs/call
102 _wsub($b, $c); _wsub($b, $a); $b ^= ($a << 8) % 4294967296;
# spent 119ms making 11064 calls to WWW::Google::PageRank::_wsub, avg 11µs/call
103 _wsub($c, $a); _wsub($c, $b); $c ^= $b >>13;
# spent 117ms making 11064 calls to WWW::Google::PageRank::_wsub, avg 11µs/call
104 _wsub($a, $b); _wsub($a, $c); $a ^= $c >> 12;
# spent 116ms making 11064 calls to WWW::Google::PageRank::_wsub, avg 11µs/call
105 _wsub($b, $c); _wsub($b, $a); $b ^= ($a << 16) % 4294967296;
# spent 117ms making 11064 calls to WWW::Google::PageRank::_wsub, avg 11µs/call
106 _wsub($c, $a); _wsub($c, $b); $c ^= $b >> 5;
# spent 117ms making 11064 calls to WWW::Google::PageRank::_wsub, avg 11µs/call
107 _wsub($a, $b); _wsub($a, $c); $a ^= $c >> 3;
# spent 116ms making 11064 calls to WWW::Google::PageRank::_wsub, avg 11µs/call
108 _wsub($b, $c); _wsub($b, $a); $b ^= ($a << 10) % 4294967296;
# spent 117ms making 11064 calls to WWW::Google::PageRank::_wsub, avg 11µs/call
109 _wsub($c, $a); _wsub($c, $b); $c ^= $b >> 15;
# spent 116ms making 11064 calls to WWW::Google::PageRank::_wsub, avg 11µs/call
110
111 @_[0 .. $#_] = ($a, $b, $c);
112}
113
11420650285ms
# spent 236ms within WWW::Google::PageRank::_wadd which was called 20650 times, avg 11µs/call: # 4610 times (57.1ms+0s) by WWW::Google::PageRank::_compute_ch at line 69, avg 12µs/call # 4610 times (51.1ms+0s) by WWW::Google::PageRank::_compute_ch at line 70, avg 11µs/call # 4610 times (50.7ms+0s) by WWW::Google::PageRank::_compute_ch at line 71, avg 11µs/call # 922 times (12.0ms+0s) by WWW::Google::PageRank::_compute_ch at line 90, avg 13µs/call # 922 times (10.6ms+0s) by WWW::Google::PageRank::_compute_ch at line 79, avg 11µs/call # 922 times (10.1ms+0s) by WWW::Google::PageRank::_compute_ch at line 89, avg 11µs/call # 922 times (9.93ms+0s) by WWW::Google::PageRank::_compute_ch at line 91, avg 11µs/call # 918 times (10.4ms+0s) by WWW::Google::PageRank::_compute_ch at line 88, avg 11µs/call # 831 times (9.15ms+0s) by WWW::Google::PageRank::_compute_ch at line 87, avg 11µs/call # 461 times (5.10ms+0s) by WWW::Google::PageRank::_compute_ch at line 84, avg 11µs/call # 461 times (4.98ms+0s) by WWW::Google::PageRank::_compute_ch at line 86, avg 11µs/call # 461 times (4.97ms+0s) by WWW::Google::PageRank::_compute_ch at line 85, avg 11µs/call
sub _wadd { $_[0] = int(($_[0] + $_[1]) % 4294967296);}
1151087961.37s
# spent 1.16s within WWW::Google::PageRank::_wsub which was called 108796 times, avg 11µs/call: # 11064 times (119ms+0s) by WWW::Google::PageRank::_mix at line 102, avg 11µs/call # 11064 times (119ms+0s) by WWW::Google::PageRank::_mix at line 101, avg 11µs/call # 11064 times (117ms+0s) by WWW::Google::PageRank::_mix at line 103, avg 11µs/call # 11064 times (117ms+0s) by WWW::Google::PageRank::_mix at line 105, avg 11µs/call # 11064 times (117ms+0s) by WWW::Google::PageRank::_mix at line 106, avg 11µs/call # 11064 times (117ms+0s) by WWW::Google::PageRank::_mix at line 108, avg 11µs/call # 11064 times (116ms+0s) by WWW::Google::PageRank::_mix at line 104, avg 11µs/call # 11064 times (116ms+0s) by WWW::Google::PageRank::_mix at line 109, avg 11µs/call # 11064 times (116ms+0s) by WWW::Google::PageRank::_mix at line 107, avg 11µs/call # 9220 times (101ms+0s) by WWW::Google::PageRank::_compute_ch_new at line 58, avg 11µs/call
sub _wsub { $_[0] = int(($_[0] - $_[1]) % 4294967296);}
116
117112µs1;
118
119
120__END__
121
122=head1 NAME
123
124WWW::Google::PageRank - Query google pagerank of page
125
126=head1 SYNOPSIS
127
128 use WWW::Google::PageRank;
129 my $pr = WWW::Google::PageRank->new;
130 print scalar($pr->get('http://www.yahoo.com/')), "\n";
131
132=head1 DESCRIPTION
133
134The C<WWW::Google::PageRank> is a class implementing a interface for
135querying google pagerank.
136
137To use it, you should create C<WWW::Google::PageRank> object and use its
138method get(), to query page rank of URL.
139
140It uses C<LWP::UserAgent> for making request to Google.
141
142=head1 CONSTRUCTOR METHOD
143
144=over 4
145
146=item $gpr = WWW::Google::PageRank->new(%options);
147
148This method constructs a new C<WWW::Google::PageRank> object and returns it.
149Key/value pair arguments may be provided to set up the initial state.
150The following options correspond to attribute methods described below:
151
152 KEY DEFAULT
153 ----------- --------------------
154 agent "Mozilla/4.0 (compatible; GoogleToolbar 2.0.111-big; Windows XP 5.1)"
155 proxy undef
156 timeout undef
157 host "toolbarqueries.google.com"
158
159C<agent> specifies the header 'User-Agent' when querying Google. If
160the C<proxy> option is passed in, requests will be made through
161specified poxy. C<proxy> is the host which serve requests from Googlebar.
162
163=back
164
165=head1 QUERY METHOD
166
167=over 4
168
169=item $pr = $gpr->get('http://www.yahoo.com');
170
171Queries Google for a specified pagerank URL and returns pagerank. If
172query successfull, integer value from 0 to 10 returned. If query fails
173for some reason (google unreachable, url does not begin from
174'http://', undefined url passed) it return C<undef>.
175
176In list context this function returns list from two elements where
177first is the result as in scalar context and the second is the
178C<HTTP::Response> object (returned by C<LWP::UserAgent::get>). This
179can be usefull for debugging purposes and for querying failure
180details.
181
182=back
183
184=head1 BUGS
185
186If you find any, please report ;)
187
188=head1 AUTHOR
189
190Yuri Karaban F<E<lt>tech@askold.netE<gt>>.
191
192Algorithm of computing checksum taken from mozilla module
193pagerankstatus F<http://pagerankstatus.mozdev.org> by
194Stephane Queraud F<E<lt>squeraud@toteme.comE<gt>>.
195
196Algorithm was modified (15-09-2004) according to new algorithm of
197computingchecksum in googlebar.
198
199=head1 COPYRIGHT
200
201Copyright 2004-2006, Yuri Karaban, All Rights Reserved.
202
203You may use, modify, and distribute this package under the
204same terms as Perl itself.
# spent 8.45ms within WWW::Google::PageRank::CORE:match which was called 922 times, avg 9µs/call: # 461 times (5.82ms+0s) by WWW::Google::PageRank::get at line 30 of WWW/Google/PageRank.pm, avg 13µs/call # 461 times (2.63ms+0s) by WWW::Google::PageRank::get at line 37 of WWW/Google/PageRank.pm, avg 6µs/call
sub WWW::Google::PageRank::CORE:match; # xsub
# spent 4.29ms within WWW::Google::PageRank::CORE:pack which was called 461 times, avg 9µs/call: # 461 times (4.29ms+0s) by WWW::Google::PageRank::_compute_ch_new at line 58 of WWW/Google/PageRank.pm, avg 9µs/call
sub WWW::Google::PageRank::CORE:pack; # xsub
# spent 13.4ms within WWW::Google::PageRank::CORE:unpack which was called 922 times, avg 15µs/call: # 922 times (13.4ms+0s) by WWW::Google::PageRank::_compute_ch at line 64 of WWW/Google/PageRank.pm, avg 15µs/call
sub WWW::Google::PageRank::CORE:unpack; # xsub
Report produced by the NYTProf 3.11 Perl profiler, developed by Tim Bunce and Adam Kaplan.