#
# http://www.fabiankeil.de/sourcecode/privoxy-log-parser/
#
-# $Id: privoxy-log-parser.pl,v 1.111 2011/02/20 13:17:16 fabiankeil Exp $
+# $Id: privoxy-log-parser.pl,v 1.125 2011/12/31 14:53:47 fabiankeil Exp $
#
# TODO:
# - LOG_LEVEL_CGI, LOG_LEVEL_ERROR, LOG_LEVEL_WRITE content highlighting
or $c =~ m/^keep-alive support is disabled/
or $c =~ m/^Continue hack in da house/
or $c =~ m/^Merged multiple header lines to:/
+ or $c =~ m/^Added header: /
)
{
# XXX: Some of these may need highlighting
# keep-alive support is disabled. Crunching: Keep-Alive: 300.
# Continue hack in da house.
# Merged multiple header lines to: 'X-FORWARDED-PROTO: http X-HOST: 127.0.0.1'
+ # Added header: Content-Encoding: deflate
} elsif ($c =~ m/^scanning headers for:/) {
# Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
# to filter client-ip-address succeeded.
+ } elsif ($c =~ m/^Compressed content from /) {
+
+ # Compressed content from 29258 to 8630 bytes. Compression level: 3
+ $content =~ s@(?<=from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
+ $content =~ s@(?<=to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
+ $content =~ s@(?<=level: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
+
} elsif ($c =~ m/^Reading in filter/) {
return '' unless SHOW_FILTER_READIN_IN;
# XXX: assume the same?
$c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
+ } elsif ($c =~ m/^Percent-encoding redirect/) {
+
+ # Percent-encoding redirect URL: http://www.example.org/\x02
+ $c = highlight_matched_url($c, '(?<=redirect URL: ).*');
+
} else {
found_unknown_content($c);
} elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
# No GIF header found (XXX: Did I ever commit this?)
- # failed while parsing 195 134747048 (XXX: never commited)
+ # failed while parsing 195 134747048 (XXX: never committed)
# Ignore these for now
# Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
# Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
+ # 3.0.18 and later:
+ # Request from 10.0.0.1 marked for blocking. limit-connect{0} doesn't allow CONNECT requests to www.example.org:443
# Malformed server response detected. Downgrading to HTTP/1.0 impossible.
$c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
$c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
$c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
$c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
+ $c =~ s@(?<=to )([^\s]+)@$h{'request_'}$1$h{'Standard'}@;
} elsif ($c =~ m/^Status code/) {
my $c = shift;
- if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
+ if ($c =~ m/^(?:Empty|No) server or forwarder response received on socket \d+\./) {
# Empty server or forwarder response received on socket 4.
# Empty server or forwarder response received on socket 3. \
# Closing client socket 15 without sending data.
+ # Used by Privoxy 3.0.18 and later:
+ # No server or forwarder response received on socket 8. \
+ # Closing client socket 10 without sending data.
+
$c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
$c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
+
+ } elsif ($c =~ m/^Didn't receive data in time:/) {
+
+ # Didn't receive data in time: a.fsdn.com:443
+ $c =~ s@(?<=in time: )(.*)@$h{'destination'}$1$h{'Standard'}@;
}
+
# XXX: There are probably more messages that deserve highlighting.
return $c;
# scan: HTTP/1.1 200 OK
$stats{'method'}{$2}++;
- $stats{'ressource'}{$3}++;
+ $stats{'resource'}{$3}++;
$stats{'http-version'}{$4}++;
} elsif ($c =~ m/^scan: Host: ([^\s]+)/) {
print "URL statistics are disabled. Increase --url-statistics-threshold to enable them.\n";
} else {
print "Requested URLs:\n";
- foreach my $ressource (sort {$stats{'ressource'}{$b} <=> $stats{'ressource'}{$a}} keys %{$stats{'ressource'}}) {
- if ($stats{'ressource'}{$ressource} < $cli_options{'url-statistics-threshold'}) {
+ foreach my $resource (sort {$stats{'resource'}{$b} <=> $stats{'resource'}{$a}} keys %{$stats{'resource'}}) {
+ if ($stats{'resource'}{$resource} < $cli_options{'url-statistics-threshold'}) {
print "Skipped statistics for URLs below the treshold.\n";
last;
}
- printf "%d : %s\n", $stats{'ressource'}{$ressource}, $ressource;
+ printf "%d : %s\n", $stats{'resource'}{$resource}, $resource;
}
}
# Log level other than LOG_LEVEL_CLF?
if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/ or
# LOG_LEVEL_CLF?
- m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
+ m/^((?:\d+\.\d+\.\d+\.\d+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
$log_messages_reached = 1;
print "\n";
} else {
# Wrapped message
+ $_ = "\n". $_ if /^(?:\d+\.\d+\.\d+\.\d+)/;
+ $_ = " " . $_;
}
s@<BR>$@@;
print;
print "\n" unless $log_messages_reached;
}
+ print "\n";
}
sub VersionMessage {
'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
'statistics' => \$cli_options{'statistics'},
'unbreak-lines-only' => \$cli_options{'unbreak-lines-only'},
- 'url-statistics-threshold=s'=> \$cli_options{'url-statistics-threshold'},
- 'host-statistics-threshold=s'=> \$cli_options{'host-statistics-threshold'},
+ 'url-statistics-threshold=i'=> \$cli_options{'url-statistics-threshold'},
+ 'host-statistics-threshold=i'=> \$cli_options{'host-statistics-threshold'},
'version' => sub { VersionMessage && exit(0) },
'help' => \&help,
) or exit(1);
they very well might be. Also note that the results are pretty much guaranteed
to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
-[B<--unbreak-lines-only] Tries to fix lines that got messed up by a broken or
+[B<--unbreak-lines-only>] Tries to fix lines that got messed up by a broken or
interestingly configured mail client and thus are no longer recognized properly.
Only fixes some breakage, but may be good enough or at least better than nothing.
Doesn't do anything else, so you probably want to pipe the output into
B<privoxy-log-parser> again.
-[B<--url-statistics-threshold>] Only show the request count for a ressource
+[B<--url-statistics-threshold>] Only show the request count for a resource
if it's above or equal to the given threshold. If the threshold is 0, URL
statistics are disabled.