3 ################################################################################
6 # A parser for Privoxy log messages. For incomplete documentation run
7 # perldoc privoxy-log-parser(.pl), for fancy screenshots see:
9 # http://www.fabiankeil.de/sourcecode/privoxy-log-parser/
11 # $Id: privoxy-log-parser.pl,v 1.135 2012/07/27 17:40:31 fabiankeil Exp $
14 # - LOG_LEVEL_CGI, LOG_LEVEL_ERROR, LOG_LEVEL_WRITE content highlighting
15 # - create fancy statistics
16 # - grep through Privoxy sources to find unsupported log messages
17 # - hunt down substitutions that match content from variables which
18 # can contain stuff like ()?'[]
19 # - replace $h{'foo'} with h('foo') where possible
20 # - hunt down XXX comments instead of just creating them
21 # - add example log lines for every regex and mark them up for
23 # - Handle incomplete input without Perl warning about undefined variables.
24 # - Use generic highlighting function that takes a regex and the
26 # - Add --compress and --decompress options.
28 # Copyright (c) 2007-2010 Fabian Keil <fk@fabiankeil.de>
30 # Permission to use, copy, modify, and distribute this software for any
31 # purpose with or without fee is hereby granted, provided that the above
32 # copyright notice and this permission notice appear in all copies.
34 # THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
35 # WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
36 # MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
37 # ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
38 # WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
39 # ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
40 # OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
41 ################################################################################
48 PRIVOXY_LOG_PARSER_VERSION => '0.7',
49 # Feel free to mess with these ...
50 DEFAULT_BACKGROUND => 'black', # Choose registered colour (like 'black')
51 DEFAULT_TEXT_COLOUR => 'white', # Choose registered colour (like 'black')
52 HEADER_DEFAULT_COLOUR => 'yellow',
53 REGISTER_HEADERS_WITH_THE_SAME_COLOUR => 1,
55 CLI_OPTION_DEFAULT_TO_HTML_OUTPUT => 0,
56 CLI_OPTION_TITLE => 'Privoxy-Log-Parser in da house',
57 CLI_OPTION_NO_EMBEDDED_CSS => 0,
58 CLI_OPTION_NO_MSECS => 0,
59 CLI_OPTION_NO_SYNTAX_HIGHLIGHTING => 0,
60 CLI_OPTION_SHORTEN_THREAD_IDS => 0,
61 CLI_OPTION_SHOW_INEFFECTIVE_FILTERS => 0,
62 CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES => 0,
63 CLI_OPTION_STATISTICS => 0,
64 CLI_OPTION_UNBREAK_LINES_ONLY => 0,
65 CLI_OPTION_URL_STATISTICS_THRESHOLD => 0,
66 CLI_OPTION_HOST_STATISTICS_THRESHOLD => 0,
68 SUPPRESS_SUCCEEDED_FILTER_ADDITIONS => 1,
70 SHOW_FILTER_READIN_IN => 0,
71 SUPPRESS_EMPTY_LINES => 1,
72 SUPPRESS_SUCCESSFUL_CONNECTIONS => 1,
73 SUPPRESS_ACCEPTED_CONNECTIONS => 1,
74 SUPPRESS_GIF_NOT_CHANGED => 1,
75 SUPPRESS_NEED_TO_DE_CHUNK_FIRST => 1,
77 DEBUG_HEADER_REGISTERING => 0,
78 DEBUG_HEADER_HIGHLIGHTING => 0,
81 DEBUG_SUPPRESS_LOG_MESSAGES => 0,
83 PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH => 0,
84 PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH => 1,
86 LOG_UNPARSED_LINES_TO_EXTRA_FILE => 0,
87 ERROR_LOG_FILE => '/var/log/privoxy-log-parser',
89 # You better leave these alone unless you know what you're doing.
90 COLOUR_RESET => "\033[0;0m",
94 # For performance reasons, these are global.
97 my %req; # request data from previous lines
102 my $thread_colour_index = 0;
103 my $header_colour_index = 0;
104 my $time_colour_index = 0;
106 my $no_special_header_highlighting;
109 my $header_highlight_regex = '';
111 my $html_output_mode;
112 my $no_msecs_mode; # XXX: should probably be removed
113 my $shorten_thread_ids;
116 sub prepare_our_stuff () {
118 # Syntax Higlight hash
120 'red', 'green', 'brown', 'blue', 'purple', 'cyan',
121 'light_gray', 'light_red', 'light_green', 'yellow',
122 'light_blue', 'pink', 'light_cyan', 'white'
129 Filter => 'purple', # XXX: Used?
130 'Re-Filter' => 'purple',
132 Request => 'light_cyan',
133 CGI => 'light_green',
135 Error => 'light_red',
137 'Fatal error' => 'light_red',
138 'Gif-Deanimate' => 'blue',
140 Writing => 'light_green',
141 Received => 'yellow',
143 # ----------------------
146 request_ => 'brown', # host+path but no protocol
147 'ip-address' => 'yellow',
150 Truncation => 'light_red',
152 Timestamp => 'brown',
153 Crunching => 'light_red',
154 crunched => 'light_red',
155 'Request-Line' => 'pink',
157 destination => 'yellow',
158 'http-version' => 'pink',
159 'crunch-pattern' => 'pink',
164 'program-name' => 'cyan',
167 warning => 'light_red',
168 debug => 'light_red',
172 'status-message' => 'light_cyan',
173 'status-code' => 'yellow',
174 'invalid-request' => 'light_red',
176 error => 'light_red',
177 'rewritten-URL' => 'light_red',
178 'pcrs-delimiter' => 'light_red',
179 'ignored' => 'light_red',
180 'action-bits-update' => 'light_red',
181 'configuration-line' => 'red',
182 'content-type' => 'yellow',
183 'HOST' => HEADER_DEFAULT_COLOUR,
188 # Header colours need their own hash so the keys can be accessed properly
190 # Prefilled with headers that should not appear with default header colours
191 Cookie => 'light_red',
192 'Set-Cookie' => 'light_red',
193 Warning => 'light_red',
194 Default => HEADER_DEFAULT_COLOUR,
197 # Crunch reasons need their own hash as well
199 'Unsupported HTTP feature' => 'light_red',
200 Blocked => 'light_red',
201 Untrusted => 'light_red',
202 Redirected => 'green',
203 'CGI Call' => 'white',
204 'DNS failure' => 'red',
205 'Forwarding failed' => 'light_red',
206 'Connection failure' => 'light_red',
207 'Out of memory (may mask other reasons)' => 'light_red',
208 'No reason recorded' => 'light_red',
211 @time_colours = ('white', 'light_gray');
213 # Translate highlight strings into highlight code
214 prepare_highlight_hash(\%header_colours);
215 prepare_highlight_hash(\%reason_colours);
216 prepare_highlight_hash(\%h);
217 prepare_colour_array(\@all_colours);
218 prepare_colour_array(\@time_colours);
225 ###############################################################
226 # Takes a colour string and returns an ANSI escape sequence
227 # (unless --no-syntax-highlighting is used).
228 # XXX: The Rolling Stones reference has to go.
229 ###############################################################
233 return "" if cli_option_is_set('no-syntax-highlighting');
275 my $bg_code = get_background();
277 our $default = default_colours();
279 if (defined($text{$colour})) {
280 $colour_code = ESCAPE;
281 $colour_code .= $text{$colour};
283 $colour_code .= $light{$colour} ? "1" : "2";
285 $colour_code .= $bg_code;
287 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
289 } elsif ($colour =~ /reset/) {
291 $colour_code = default_colours();
295 die "What's $colour supposed to mean?\n";
301 sub get_semantic_html_markup ($) {
302 ###############################################################
303 # Takes a string and returns a span element
304 ###############################################################
309 if ($type =~ /Standard/) {
313 $code = '<span title="' . $type . '" class="' . $type . '">';
319 sub cli_option_is_set ($) {
322 my $cli_option = shift;
324 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
326 return $cli_options{$cli_option};
329 sub get_html_title () {
332 return $cli_options{'title'};
336 sub init_css_colours() {
344 purple => "F06", # XXX: wrong
345 cyan => "F09", # XXX: wrong
350 light_green => "33F",
359 sub get_css_colour ($) {
364 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
366 return '#' . $css_colours{$colour};
369 sub get_css_line ($) {
374 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
375 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
376 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
377 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
378 $css_line .= '}' . "\n";
383 sub get_css_line_for_colour ($) {
388 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
389 $css_line .= 'color:' . get_css_colour($colour) . ';';
390 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
391 $css_line .= '}' . "\n";
396 # XXX: Wrong solution
397 sub get_missing_css_lines () {
401 $css_line .= '.' . 'default' . ' {';
402 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
403 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
404 $css_line .= '}' . "\n";
411 our %css_colours; #XXX: Wrong solution
415 $css .= '.privoxy-log {';
416 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
417 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
420 foreach my $key (keys %h_colours) {
422 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
423 $css .= get_css_line($key);
427 foreach my $colour (keys %css_colours) {
429 $css .= get_css_line_for_colour($colour);
433 $css .= get_missing_css_lines(); #XXX: Wrong solution
442 if (cli_option_is_set('html-output')) {
444 my $title = get_html_title();
446 $intro .= '<html><head>';
447 $intro .= '<title>' . $title . '</title>';
448 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
449 $intro .= '</head><body>';
450 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
460 if (cli_option_is_set('html-output')) {
462 $outro = '</p></body></html>';
468 sub get_line_end () {
469 return cli_option_is_set('html-output') ? "<br>\n" : "\n";
472 sub get_colour_html_markup ($) {
473 ###############################################################
474 # Takes a colour string a span element. XXX: WHAT?
475 # XXX: This function shouldn't be necessary, the
476 # markup should always be semantically correct.
477 ###############################################################
482 if ($type =~ /Standard/) {
485 $code = '<span class="' . lc($type) . '">';
491 sub default_colours () {
494 return reset_colours();
497 sub show_colours () {
501 sub reset_colours () {
502 return ESCAPE . "0m";
505 sub set_background ($){
521 if (defined($backgrounds{$colour})) {
522 $bg_code = $backgrounds{$colour};
524 die "Invalid background colour: " . $colour;
528 sub get_background (){
532 sub prepare_highlight_hash ($) {
535 foreach my $key (keys %$ref) {
536 $$ref{$key} = $html_output_mode ?
537 get_semantic_html_markup($key) :
538 paint_it($$ref{$key});
542 sub prepare_colour_array ($) {
545 foreach my $i (0 ... @$ref - 1) {
546 $$ref[$i] = $html_output_mode ?
547 get_colour_html_markup($$ref[$i]) :
552 sub found_unknown_content ($) {
557 return if cli_option_is_set('accept-unknown-messages');
559 return if ($unknown =~ /\[too long, truncated\]$/);
561 $message = "found_unknown_content: Don't know how to highlight: ";
562 # Break line so the log file can later be parsed as Privoxy log file again
563 $message .= '"' . $unknown . '"' . " in:\n";
564 $message .= $req{$t}{'log-message'};
565 debug_message($message);
566 log_parse_error($req{$t}{'log-message'});
568 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
571 sub log_parse_error ($) {
575 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
576 open(my $errorlog_fd, ">>", ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
577 print $errorlog_fd $message;
582 sub debug_message (@) {
585 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
588 ################################################################################
589 # highlighter functions that aren't loglevel-specific
590 ################################################################################
594 # Get highlight marker
595 my $highlight = shift; # XXX: Stupid name;
599 if (defined($highlight)) {
601 $result = $h{$highlight};
605 $message = "h: Don't recognize highlighter $highlight.";
606 debug_message($message);
607 log_parser_error($message);
608 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
614 sub highlight_known_headers ($) {
618 debug_message("Searching $content for things to highlight.") if DEBUG_HEADER_HIGHLIGHTING;
620 if ($content =~ m/(?<=\s)($header_highlight_regex):/) {
622 $content =~ s@(?<=[\s|'])($header)(?=:)@$header_colours{$header}$1$h{'Standard'}@ig;
623 debug_message("Highlighted '$header' in '$content'") if DEBUG_HEADER_HIGHLIGHTING;
629 sub highlight_matched_request_line ($$) {
631 my $result = shift; # XXX: Stupid name;
633 if ($result =~ m@(.*)($regex)(.*)@) {
634 $result = $1 . highlight_request_line($2) . $3
639 sub highlight_request_line ($) {
642 my ($method, $url, $http_version);
644 #GET http://images.sourceforge.net/sfx/icon_warning.gif HTTP/1.1
645 if ($rl =~ m/Invalid request/) {
647 $rl = h('invalid-request') . $rl . h('Standard');
649 } elsif ($rl =~ m/^([-\w]+) (.*) (HTTP\/\d\.\d)/) {
651 # XXX: might not match in case of HTTP method fuzzing.
652 # XXX: save these: ($method, $path, $http_version) = ($1, $2, $3);
653 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
654 if ($rl =~ /http:\/\//) {
655 $rl = highlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
657 $rl = highlight_matched_pattern($rl, 'request_', '[^\s]*(?=\sHTTP)');
660 $rl =~ s@(HTTP\/\d\.\d)$@$h{'http-version'}$1$h{'Standard'}@;
662 } elsif ($rl =~ m/\.\.\. \[too long, truncated\]$/) {
664 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
665 $rl = highlight_matched_url($rl, '[^\s]*(?=\.\.\.)');
667 } elsif ($rl =~ m/^ $/) {
669 $rl = h('error') . "No request line specified!" . h('Standard');
673 debug_message ("Can't parse request line: $rl");
680 sub highlight_response_line ($) {
683 my ($http_version, $status_code, $status_message);
688 # TODO: Mark different status codes differently
690 if ($rl =~ m/((?:HTTP\/\d\.\d|ICY)) (\d+) (.*)/) {
691 ($http_version, $status_code, $status_message) = ($1, $2, $3);
693 debug_message ("Can't parse response line: $rl") and die 'Fix this';
696 # Rebuild highlighted
698 $rl .= h('http-version') . $http_version . h('Standard');
700 $rl .= h('status-code') . $status_code . h('Standard');
702 $rl .= h('status-message') . $status_message . h('Standard');
707 sub highlight_matched_url ($$) {
709 my $result = shift; # XXX: Stupid name;
712 #print "Got $result, regex ($regex)\n";
714 if ($result =~ m@(.*?)($regex)(.*)@) {
715 $result = $1 . highlight_url($2) . $3;
716 #print "Now the result is $result\n";
722 sub highlight_matched_host ($$) {
724 my ($result, $regex) = @_; # XXX: result ist stupid name;
726 if ($result =~ m@(.*?)($regex)(.*)@) {
727 $result = $1 . $h{host} . $2 . $h{Standard} . $3;
733 sub highlight_matched_pattern ($$$) {
735 my $result = shift; # XXX: Stupid name;
739 die "Unknown key $key" unless defined $h{$key};
741 if ($result =~ m@(.*?)($regex)(.*)@) {
742 $result = $1 . h($key) . $2 . h('Standard') . $3;
748 sub highlight_matched_path ($$) {
750 my $result = shift; # XXX: Stupid name;
753 if ($result =~ m@(.*?)($regex)(.*)@) {
754 $result = $1 . h('path') . $2 . h('Standard') . $3;
760 sub highlight_url ($) {
764 if ($html_output_mode) {
766 $url = '<a href="' . $url . '">' . $url . '</a>';
770 $url = h('URL') . $url . h('Standard');
777 sub update_header_highlight_regex ($) {
780 my $headers = join ('|', keys %header_colours);
782 $header_highlight_regex = qr/$headers/;
783 print "Registering '$header'\n" if DEBUG_HEADER_HIGHLIGHTING;
786 ################################################################################
787 # loglevel-specific highlighter functions
788 ################################################################################
790 sub handle_loglevel_header ($) {
794 if ($c =~ /^scan:/) {
796 if ($c =~ m/^scan: ([^: ]+):/) {
798 # Register new headers
799 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
801 if (!defined($header_colours{$header}) and $header =~ /^[\d\w-]*$/) {
802 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
804 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
805 $header_colours{$header} = $header_colours{'Default'};
807 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
808 $header_colour_index++;
810 update_header_highlight_regex($header);
813 } elsif ($c =~ m/^(scan: )(\w+ .+ HTTP\/\d\.\d)/) {
815 # scan: GET http://p.p/ HTTP/1.1
816 $c = $1 . highlight_request_line($2);
818 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
820 # scan: HTTP/1.1 200 OK
821 $req{$t}{'response_line'} = $2;
822 $req{$t}{'status_code'} = $3;
823 $req{$t}{'status_message'} = $4;
824 $c = $1 . highlight_response_line($req{$t}{'response_line'});
827 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
829 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
830 $c =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
831 $c =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
833 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
835 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/!
836 $c = highlight_matched_host($c, '(?<=New host is: )[^\s]+(?=\.)');
837 $c = highlight_matched_url($c, '(?<=Crunching Referer: )[^\s!]+');
839 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
841 # Text mode enabled by force. Take cover!
842 $c =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
844 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
846 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
847 $c = $1 . highlight_request_line($2);
849 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
851 # Adjusted Content-Length to 2132
852 # Adjust Content-Length to 33533
853 $c =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
854 $c = highlight_known_headers($c);
856 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
858 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
859 $c = highlight_matched_url($c, '(?<=New request URL: ).*');
861 } elsif ($c =~ m/^Couldn\'t parse:/) {
863 # XXX: These should probable be logged with LOG_LEVEL_ERROR
864 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
865 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
866 $c =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
868 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
869 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
871 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
872 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
873 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
875 # XXX: Save tag and tagger
877 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
878 $c =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
879 $c =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
880 $no_special_header_highlighting = 1;
882 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
884 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
885 # XXX: Save tag and tagger
887 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
888 $c =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
890 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
891 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
892 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
893 or $c =~ m/^(addh-unique|Referer forged to)/
894 or $c =~ m/^Downgraded answer to HTTP\/1.0/
895 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
896 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
897 or $c =~ m/^Referer crunched!/
898 or $c =~ m/^crunched x-forwarded-for!/
899 or $c =~ m/^crunched From!/
900 or $c =~ m/^ modified$/
901 or $c =~ m/^Content filtering is enabled. Crunching:/
902 or $c =~ m/^force-text-mode overruled the client/
903 or $c =~ m/^Server time in the future\./
904 or $c =~ m/^content-disposition header crunched and replaced with:/i
905 or $c =~ m/^Reducing white space in /
906 or $c =~ m/^Ignoring single quote in /
907 or $c =~ m/^Converting tab to space in /
908 or $c =~ m/A HTTP\/1\.1 response without/
909 or $c =~ m/Disabled filter mode on behalf of the client/
910 or $c =~ m/Keeping the (?:server|client) header /
911 or $c =~ m/Content modified with no Content-Length header set/
912 or $c =~ m/^Appended client IP address to/
913 or $c =~ m/^Removing 'Connection: close' to imply keep-alive./
914 or $c =~ m/^keep-alive support is disabled/
915 or $c =~ m/^Continue hack in da house/
916 or $c =~ m/^Merged multiple header lines to:/
917 or $c =~ m/^Added header: /
918 or $c =~ m/^Enlisting (?:sorted|left-over) header/
921 # XXX: Some of these may need highlighting
923 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
924 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
925 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
926 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
927 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
928 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
929 # Text mode is already enabled
930 # Denied request with NULL byte(s) turned into line break(s)
931 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
932 # addh-unique: Host: people.freebsd.org
933 # Suppressed offer to compress content
934 # Crunched incoming cookie -- yum!
935 # Accepted the client's request to fetch without filtering.
936 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
937 # addh-unique: Host: subkeys.pgp.net:11371
938 # Referer forged to: Referer: http://10.0.0.1/
939 # Downgraded answer to HTTP/1.0
940 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
941 # Referer overwritten with: Referer: pille-palle
942 # Referer replaced with: Referer: pille-palle
943 # crunched x-forwarded-for!
945 # modified # XXX: pretty stupid log message
946 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
947 # force-text-mode overruled the client's request to fetch without filtering!
948 # Server time in the future.
949 # content-disposition header crunched and replaced with: content-disposition: filename=baz
950 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
951 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
952 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
953 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
955 # A HTTP/1.1 response without Connection header implies keep-alive.
956 # Disabled filter mode on behalf of the client.
957 # Keeping the server header 'Connection: keep-alive' around.
958 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
959 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
960 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
961 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
962 # Removing 'Connection: close' to imply keep-alive.
963 # keep-alive support is disabled. Crunching: Keep-Alive: 300.
964 # Continue hack in da house.
965 # Merged multiple header lines to: 'X-FORWARDED-PROTO: http X-HOST: 127.0.0.1'
966 # Added header: Content-Encoding: deflate
967 # Enlisting sorted header User-Agent: Mozilla/5.0 (X11; SunOS i86pc; rv:10.0.3) Gecko/20100101 Firefox/10.0.3
968 # Enlisting left-over header Connection: close
970 } elsif ($c =~ m/^scanning headers for:/) {
972 return '' unless SHOW_SCAN_INTRO;
974 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
975 # crunched User-Agent!
976 # Crunching: Content-Encoding: gzip
978 $c =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
980 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
982 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
984 $c = h('warning') . $c . h('Standard');
986 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
988 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
989 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
991 $c =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
992 $c =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
994 } elsif ($c =~ m/^Removing empty header/) {
996 # Removing empty header
999 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
1001 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
1002 # Enable force-text-mode if you know what you're doing.
1003 # XXX: Could highlight more here.
1004 $c =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
1006 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
1008 # Server keep-alive timeout is 5. Sticking with 10.
1009 # Client keep-alive timeout is 20. Sticking with 10.
1011 $c =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1012 $c =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1014 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1016 # Reducing keep-alive timeout from 60 to 10.
1018 $c =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1019 $c =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1021 } elsif ($c =~ m/^Killed all-caps Host header line: HOST:/) {
1023 # Killed all-caps Host header line: HOST: bestproxydb.com
1024 $c = highlight_matched_host($c, '(?<=HOST: )[^\s]+');
1025 $c = highlight_matched_pattern($c, 'HOST', 'HOST');
1029 found_unknown_content($c);
1033 unless ($c =~ m/^Transforming/) {
1034 $c = highlight_known_headers($c) unless $no_special_header_highlighting;
1040 sub handle_loglevel_re_filter ($) {
1042 my $content = shift;
1046 if ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1048 # XXX: only the second version gets highlighted properly.
1049 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1050 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1051 $req{$t}{'content_source'} = $1;
1052 $req{$t}{'content_size'} = $2;
1053 $req{$t}{'content_filter'} = $3;
1054 $req{$t}{'content_hits'} = $4;
1055 $req{$t}{'new_content_size'} = $5;
1056 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1057 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1058 if ($req{$t}{'content_hits'} == 0 and
1059 not (cli_option_is_set('show-ineffective-filters')
1060 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1064 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1065 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1066 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1068 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1069 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1072 $c .= "(" . $h{'Number'};
1073 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1074 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1077 } elsif ($c =~ /\.{3}$/
1078 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1080 # Used by Privoxy 3.0.5 and 3.0.6:
1082 # Used by Privoxy 3.0.7:
1083 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1085 $req{$t}{'filtered_header'} = $1;
1086 $req{$t}{'old_header_size'} = $2;
1087 $req{$t}{'header_filter_name'} = $3;
1089 unless (cli_option_is_set('show-ineffective-filters') or
1090 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1093 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1094 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1096 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1098 # ...produced 0 hits (new size 23).
1099 #... produced 1 hits (new size 54).
1101 $req{$t}{'header_filter_hits'} = $1;
1102 $req{$t}{'new_header_size'} = $2;
1104 unless (cli_option_is_set('show-ineffective-filters') or
1105 (defined($req{$t}{'header_filter_name'}) and
1106 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1108 if ($req{$t}{'header_filter_hits'} == 0 and
1109 not (defined($req{$t}{'header_filter_name'}) and
1110 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1113 # Reformat including information from the intro
1114 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1116 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1117 # $c .= $req{$t}{'filtered_header'} . " ";
1118 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1119 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1121 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1123 $c .= "changing size from ";
1124 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1126 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1131 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1135 # Highlight from last line (XXX: What?)
1136 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1137 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1142 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1143 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1148 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1150 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1151 # Tagger variable-test has empty joblist. Nothing to do.
1153 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1155 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1157 $req{$t}{'chunked-size'} = $1;
1158 $req{$t}{'dechunked-size'} = $2;
1159 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1161 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1163 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1164 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1166 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1168 # Decompression successful. Old size: 670, new size: 1166.
1170 $req{$t}{'size-compressed'} = $1;
1171 $req{$t}{'size-decompressed'} = $2;
1172 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1174 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1175 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1177 # XXX: Create sub get_percentage()
1178 if ($req{$t}{'size-decompressed'}) {
1179 $req{$t}{'decompression-gain-percent'} =
1180 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1182 $content .= " (saved: ";
1183 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1185 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1189 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1191 # Need to de-chunk first
1192 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1194 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1196 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1198 # Adding re_filter job ...
1199 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1200 # to filter client-ip-address succeeded.
1202 } elsif ($c =~ m/^Compressed content from /) {
1204 # Compressed content from 29258 to 8630 bytes. Compression level: 3
1205 $content =~ s@(?<=from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1206 $content =~ s@(?<=to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1207 $content =~ s@(?<=level: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1209 } elsif ($c =~ m/^Reading in filter/) {
1211 return '' unless SHOW_FILTER_READIN_IN;
1215 found_unknown_content($content);
1222 sub handle_loglevel_redirect ($) {
1226 if ($c =~ m/^Decoding "([^""]*)"/) {
1228 $req{$t}{'original-destination'} = $1;
1229 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1232 } elsif ($c =~ m/^Checking/) {
1234 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1235 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1237 # TODO: Change colour if really url-decoded
1238 $req{$t}{'decoded-original-destination'} = $1;
1239 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1242 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1244 # pcrs command "s@&from=rss@@" changed \
1245 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1246 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1247 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1248 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1249 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1250 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1251 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1252 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1254 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1256 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1257 # change "http://www.example.org/".
1258 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1259 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1261 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1263 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1264 # XXX: Use URL highlighter
1266 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1268 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1269 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1270 # is already properly formatted.
1271 # XXX: assume the same?
1272 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1274 } elsif ($c =~ m/^Percent-encoding redirect/) {
1276 # Percent-encoding redirect URL: http://www.example.org/\x02
1277 $c = highlight_matched_url($c, '(?<=redirect URL: ).*');
1281 found_unknown_content($c);
1288 sub handle_loglevel_gif_deanimate ($) {
1290 my $content = shift;
1292 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1294 my $bytes_from = $1;
1296 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1297 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1298 # XXX: Do we need g in case of ($1 == $2)?
1299 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1301 } elsif ($content =~ m/GIF (not) changed/) {
1303 # Gif-Deanimate: GIF not changed.
1304 return '' if SUPPRESS_GIF_NOT_CHANGED;
1305 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1307 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1309 # failed! (gif parsing)
1310 # XXX: Replace this error message with something less stupid
1311 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1313 } elsif ($content =~ m/^Need to de-chunk first/) {
1315 # Need to de-chunk first
1316 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1318 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1320 # No GIF header found (XXX: Did I ever commit this?)
1321 # failed while parsing 195 134747048 (XXX: never committed)
1323 # Ignore these for now
1327 found_unknown_content($content);
1334 sub handle_loglevel_request ($) {
1336 my $content = shift;
1338 if ($content =~ m/crunch! /) {
1340 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1342 # Highlight crunch reasons
1343 foreach my $reason (keys %reason_colours) {
1344 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1346 # Highlight request URL domain and ditch 'crunch!'
1347 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1348 $content =~ s@ crunch!@@;
1350 } elsif ($content =~ m/\[too long, truncated\]$/) {
1352 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1353 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1355 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1357 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1358 $content = h('request_') . $content . h('Standard');
1362 found_unknown_content($content);
1369 sub handle_loglevel_crunch ($) {
1371 my $content = shift;
1373 # Highlight crunch reason
1374 foreach my $reason (keys %reason_colours) {
1375 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1378 if ($content =~ m/\[too long, truncated\]$/) {
1380 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1381 # [...]&filter... [too long, truncated]
1382 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1386 # Blocked: http://ads.example.org/
1387 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1393 sub handle_loglevel_connect ($) {
1397 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1399 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1401 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1402 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1404 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1406 # connect to: www.example.org.noconnect failed: Operation not permitted
1408 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1410 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1412 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1414 # Connect: to www.nzherald.co.nz successful
1415 # Connect: to archiv.radiotux.de
1417 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1418 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1420 } elsif ($c =~ m/^to ([^\s]*)$/) {
1422 # Connect: to lists.sourceforge.net:443
1424 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1426 } elsif ($c =~ m/^accepted connection from .*/ or
1429 # accepted connection from 10.0.0.1
1430 # Privoxy 3.0.6 and earlier just say:
1432 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1433 $c = highlight_matched_host($c, '(?<=connection from ).*');
1435 } elsif ($c =~ m/^Closing client socket/) {
1437 # Closing client socket 5. Keep-alive: 0, Socket alive: 1. Data available: 0.
1438 $c = highlight_matched_pattern($c, 'Number', '(?<=socket )\d+');
1439 $c = highlight_matched_pattern($c, 'Number', '(?<=Keep-alive: )\d+');
1440 $c = highlight_matched_pattern($c, 'Number', '(?<=Socket alive: )\d+');
1441 $c = highlight_matched_pattern($c, 'Number', '(?<=available: )\d+');
1443 } elsif ($c =~ m/^write header to: .* failed:/) {
1445 # write header to: 10.0.0.1 failed: Broken pipe
1447 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1448 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1450 } elsif ($c =~ m/^write header to client failed:/) {
1452 # write header to client failed: Broken pipe
1454 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1456 } elsif ($c =~ m/^socks4_connect:/) {
1458 # socks4_connect: SOCKS request rejected or failed.
1459 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1461 } elsif ($c =~ m/^Listening for new connections/ or
1462 $c =~ m/^accept connection/) {
1464 # Privoxy versions above 3.0.6 say:
1465 # Listening for new connections ...
1466 # earlier versions say:
1467 # accept connection ...
1470 } elsif ($c =~ m/^accept failed:/) {
1472 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1474 } elsif ($c =~ m/^Overriding forwarding settings/) {
1476 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1477 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1479 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1481 # Denying suspicious CONNECT request from 10.0.0.1
1482 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1484 } elsif ($c =~ m/^socks5_connect:/) {
1486 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1488 } elsif ($c =~ m/^Created new connection to/) {
1490 # Created new connection to www.privoxy.org:80 on socket 11.
1491 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1492 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1494 } elsif ($c =~ m/^Found reusable socket/) {
1496 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1498 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1499 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1500 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1501 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1502 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1503 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1504 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1505 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1507 } elsif ($c =~ m/^Marking open socket/) {
1509 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1510 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1511 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1512 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1514 } elsif ($c =~ m/^No reusable/) {
1516 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1517 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1519 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1521 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1522 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1524 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1525 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1526 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1528 } elsif ($c =~ m/^Socket/) {
1530 # Socket 16 already forgotten or never remembered.
1531 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1533 } elsif ($c =~ m/^The connection to/) {
1535 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1537 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1538 # Timeout is: 1. Assumed latency: 4.
1539 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1540 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1541 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1542 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1543 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1544 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1546 } elsif ($c =~ m/^Stopped waiting for the request line/ or
1547 $c =~ m/^No request line on socket \d received in time/ or
1548 $c =~ m/^The client side of the connection on socket \d/) {
1550 # Stopped waiting for the request line. Timeout: 121.
1551 # Privoxy 3.0.19 and later:
1552 # No request line on socket 5 received in time. Timeout: 1.
1553 # The client side of the connection on socket 5 got closed \
1554 # without sending a complete request line.
1555 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1556 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1558 } elsif ($c =~ m/^Waiting for \d/) {
1560 # Waiting for 1 connections to timeout.
1561 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1563 } elsif ($c =~ m/^Initialized/) {
1565 # Initialized 20 socket slots.
1566 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1568 } elsif ($c =~ m/^Done reading from server/) {
1570 # Done reading from server. Expected content length: 24892. \
1571 # Actual content length: 24892. Most recently received: 4412.
1573 # Done reading from server. Expected content length: 24892. \
1574 # Actual content length: 24892. Bytes most recently read: 4412.
1575 # Done reading from server. Content length: 6018 as expected. \
1576 # Bytes most recently read: 294.
1577 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1578 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1579 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1581 } elsif ($c =~ m/^Continuing buffering headers/) {
1583 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1584 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1585 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1586 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1588 # Continuing buffering headers. Bytes most recently read: %d.
1589 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1591 } elsif ($c =~ m/^Received \d+ bytes while/) {
1593 # Received 206 bytes while expecting 12103.
1594 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1595 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1597 } elsif ($c =~ m/^(Rejecting c|C)onnection from/) {
1599 # Connection from 81.163.28.218 dropped due to ACL
1600 # Rejecting connection from 178.63.152.227. Maximum number of connections reached.
1601 $c =~ s@(?<=onnection from )((?:\d+\.?){3}\d+)@$h{'Number'}$1$h{'Standard'}@;
1603 } elsif ($c =~ m/^(?:Reusing|Closing) server socket \d./ or
1604 $c =~ m/^No additional client request/) {
1606 # Reusing server socket 4. Opened for 10.0.0.1.
1607 # Closing server socket 2. Opened for 10.0.0.1.
1608 # No additional client request received in time. \
1609 # Closing server socket 4, initially opened for 10.0.0.1.
1610 # No additional client request received in time on socket 29.
1612 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1613 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.$)');
1615 } elsif ($c =~ m/^Connected to /) {
1617 # Connected to tor-jail[10.0.0.2]:9050.
1619 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1620 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1621 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1623 } elsif ($c =~ m/^Could not connect to /) {
1625 # Could not connect to [10.0.0.1]:80.
1627 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1628 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1630 } elsif ($c =~ m/^Waiting for the next client request/ or
1631 $c =~ m/^The connection on server socket/ or
1632 $c =~ m/^Client request arrived in time /) {
1634 # Waiting for the next client request on socket 3. Keeping the server \
1635 # socket 12 to a.fsdn.com open.
1636 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1637 # Used by Privoxy 3.0.18 and later:
1638 # Client request arrived in time on socket 21.
1639 # Used by earlier version:
1640 # Client request arrived in time or the client closed the connection on socket 12.
1642 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1643 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1644 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1646 } elsif ($c =~ m/^Marking the server socket/) {
1648 # Marking the server socket 7 tainted.
1650 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1652 } elsif ($c =~ m/^Reduced expected bytes to /) {
1654 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1655 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1656 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1658 } elsif ($c =~ m/^The client closed socket /) {
1660 # The client closed socket 2 while the server socket 4 is still open.
1661 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1662 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1664 } elsif ($c =~ m/^Expected client content length set /) {
1666 # Expected client content length set to 667325411 after reading 4999 bytes.
1667 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1668 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1670 } elsif ($c =~ m/^Reducing expected bytes to /) {
1672 # Reducing expected bytes to 0. Marking the server socket tainted after throwing 4 bytes away.
1673 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1674 $c =~ s@(?<=after throwing )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1676 } elsif ($c =~ m/^Waiting for up to /) {
1678 # Waiting for up to 4999 bytes from the client.
1679 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1681 } elsif ($c =~ m/^Stopping to watch the client socket/) {
1683 # Stopping to watch the client socket. There's already another request waiting.
1684 # Privoxy 3.0.20 and later:
1685 # Stopping to watch the client socket 5. There's already another request waiting.
1686 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1688 } elsif ($c =~ m/^Looks like we / or
1689 $c =~ m/^Unsetting keep-alive flag/ or
1690 $c =~ m/^No connections to wait/ or
1691 $c =~ m/^Complete client request received/ or
1692 $c =~ m/^Possible pipeline attempt detected./ or
1693 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1694 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1695 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1696 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1697 $c =~ m/^Done reading from the client\.$/) {
1699 # Looks like we reached the end of the last chunk. We better stop reading.
1700 # Looks like we read the end of the last chunk together with the server \
1701 # headers. We better stop reading.
1702 # Looks like we got the last chunk together with the server headers. \
1703 # We better stop reading.
1704 # Unsetting keep-alive flag.
1705 # No connections to wait for left.
1706 # Client request arrived in time or the client closed the connection.
1707 # Complete client request received
1708 # Possible pipeline attempt detected. The connection will not be \
1709 # kept alive and we will only serve the first request.
1710 # POST request detected. The connection will not be kept alive.
1711 # The server still wants to talk, but the client hung up on us.
1712 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1713 # There might be a request body. The connection will not be kept alive.
1714 # Done reading from the client\.
1718 found_unknown_content($c);
1726 sub handle_loglevel_info ($) {
1730 if ($c =~ m/^Rewrite detected:/) {
1732 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1733 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1735 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1736 $c =~ m/^Compressed content detected/ or
1737 $c =~ m/^SDCH-compressed content detected/ or
1740 # Decompressing deflated iob: 117
1741 # Decompression didn't result in any content.
1742 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1743 # with zlib support or enable the prevent-compression action.
1744 # SDCH-compressed content detected, content filtering disabled.\
1745 # Consider suppressing SDCH offers made by the client.
1746 # Tagger 'complete-url' created empty tag. Ignored.
1750 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1752 # loading configuration file '/usr/local/etc/privoxy/config':
1753 # Reloading configuration file '/usr/local/etc/privoxy/config'
1754 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1756 } elsif ($c =~ m/^Loading (actions|filter) file: /) {
1758 # Loading actions file: /usr/local/etc/privoxy/default.action
1759 # Loading filter file: /usr/local/etc/privoxy/default.filter
1760 $c =~ s@(?<= file: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1762 } elsif ($c =~ m/^exiting by signal/) {
1764 # exiting by signal 15 .. bye
1765 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1767 } elsif ($c =~ m/^Privoxy version/) {
1769 # Privoxy version 3.0.7
1770 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1772 } elsif ($c =~ m/^Program name: /) {
1774 # Program name: /usr/local/sbin/privoxy
1775 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1777 } elsif ($c =~ m/^Listening on port /) {
1779 # Listening on port 8118 on IP address 10.0.0.1
1780 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1781 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1783 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1785 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1786 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1788 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1790 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1791 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1793 # Request from 10.0.0.1 marked for blocking. limit-connect{0} doesn't allow CONNECT requests to www.example.org:443
1794 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1796 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1797 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1798 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1799 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1800 $c =~ s@(?<=to )([^\s]+)@$h{'request_'}$1$h{'Standard'}@;
1802 } elsif ($c =~ m/^Status code/) {
1804 # Status code 304 implies no body.
1805 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1807 } elsif ($c =~ m/^Method/) {
1809 # Method HEAD implies no body.
1810 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1812 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1814 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1815 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1816 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1818 } elsif ($c =~ m/^File modification detected: /) {
1820 # File modification detected: /usr/local/etc/privoxy/user-agent.action
1821 $c =~ s@(?<= detected: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1823 } elsif ($c =~ m/^No logfile configured/ or
1824 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1825 $c =~ m/^Invalid \"chunked\" transfer/ or
1826 $c =~ m/^Support for/ or
1827 $c =~ m/^Flushing header and buffers/ or
1828 $c =~ m/^Can not resolve/
1831 # No logfile configured. Please enable it before reporting any problems.
1832 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1833 # response or even no response at all.
1834 # No logfile configured. Logging disabled.
1835 # Invalid "chunked" transfer encoding detected and ignored.
1836 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1837 # known not to work properly in some situations.
1838 # Flushing header and buffers. Stepping back from filtering.
1839 # Can not resolve doesnotexist: hostname nor servname provided, or not known
1843 found_unknown_content($c);
1850 sub handle_loglevel_cgi ($) {
1854 if ($c =~ m/^Granting access to/) {
1856 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1858 } elsif ($c =~ m/^Substituting: s(.)/) {
1860 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1861 # XXX: prone to span several lines
1864 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1865 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1866 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1872 sub handle_loglevel_force ($) {
1876 if ($c =~ m/^Ignored force prefix in request:/) {
1878 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1879 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1880 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1882 } elsif ($c =~ m/^Enforcing request:/) {
1884 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1885 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1889 found_unknown_content($c);
1896 sub handle_loglevel_error ($) {
1900 if ($c =~ m/^(?:Empty|No) server or forwarder response received on socket \d+\./) {
1902 # Empty server or forwarder response received on socket 4.
1903 # Empty server or forwarder response received on socket 3. \
1904 # Closing client socket 15 without sending data.
1905 # Used by Privoxy 3.0.18 and later:
1906 # No server or forwarder response received on socket 8. \
1907 # Closing client socket 10 without sending data.
1909 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1910 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1912 } elsif ($c =~ m/^Didn't receive data in time:/) {
1914 # Didn't receive data in time: a.fsdn.com:443
1915 $c =~ s@(?<=in time: )(.*)@$h{'destination'}$1$h{'Standard'}@;
1918 # XXX: There are probably more messages that deserve highlighting.
1924 sub handle_loglevel_ignore ($) {
1928 sub gather_loglevel_request_stats ($$) {
1936 sub gather_loglevel_crunch_stats ($$) {
1944 if ($c =~ m/^Redirected:/) {
1945 # Redirected: http://www.example.org/http://p.p/
1946 $stats{'fast-redirections'}++;
1948 } elsif ($c =~ m/^Blocked:/) {
1949 # Blocked: blogger.googleusercontent.com:443
1950 $stats{'blocked'}++;
1955 sub gather_loglevel_error_stats ($$) {
1962 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1964 # Empty server or forwarder response received on socket 4.
1965 $stats{'empty-responses'}++;
1966 if ($thread_data{$thread}{'new_connection'}) {
1967 $stats{'empty-responses-on-new-connections'}++;
1969 $stats{'empty-responses-on-reused-connections'}++;
1974 sub gather_loglevel_connect_stats ($$) {
1976 my ($c, $thread) = @_;
1980 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
1982 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1983 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
1985 } elsif ($c =~ m/^to ([^\s]*)$/) {
1987 # Connect: to lists.sourceforge.net:443
1989 $thread_data{$thread}{'forwarder'} = 'direct connection';
1991 } elsif ($c =~ m/^Created new connection to/) {
1993 # Created new connection to www.privoxy.org:80 on socket 11.
1995 $thread_data{$thread}{'new_connection'} = 1;
1997 } elsif ($c =~ m/^Reusing server socket \d./ or
1998 $c =~ m/^Found reusable socket/) {
2000 # Reusing server socket 4. Opened for 10.0.0.1.
2001 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
2003 $thread_data{$thread}{'new_connection'} = 0;
2004 $stats{'reused-connections'}++;
2008 sub gather_loglevel_header_stats ($$) {
2010 my ($c, $thread) = @_;
2013 if ($c =~ m/^A HTTP\/1\.1 response without/ or
2014 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
2016 # A HTTP/1.1 response without Connection header implies keep-alive.
2017 # Keeping the server header 'Connection: keep-alive' around.
2018 $stats{'server-keep-alive'}++;
2020 } elsif ($c =~ m/^scan: ((\w+) (.+) (HTTP\/\d\.\d))/) {
2022 # scan: HTTP/1.1 200 OK
2023 $stats{'method'}{$2}++;
2024 $stats{'resource'}{$3}++;
2025 $stats{'http-version'}{$4}++;
2027 } elsif ($c =~ m/^scan: Host: ([^\s]+)/) {
2030 $stats{'hosts'}{$1}++;
2038 'server-keep-alive' => 0,
2039 'reused-connections' => 0,
2040 'empty-responses' => 0,
2041 'empty-responses-on-new-connections' => 0,
2042 'empty-responses-on-reused-connections' => 0,
2043 'fast-redirections' => 0,
2045 'reused-connections' => 0,
2046 'server-keep-alive' => 0,
2050 sub get_percentage ($$) {
2054 # If small is 0 the percentage is always 0%.
2055 # Make sure it works even if big is 0 as well.
2056 return "0.00%" if ($small eq 0);
2058 # Prevent division by zero.
2059 # XXX: Is this still supposed to be reachable?
2060 return "NaN" if ($big eq 0);
2062 return sprintf("%.2f%%", $small / $big * 100);
2065 sub print_stats () {
2069 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
2070 my $outgoing_requests = $stats{requests} - $stats{crunches};
2072 if ($stats{requests} eq 0) {
2073 print "No requests yet.\n";
2077 print "Client requests total: " . $stats{requests} . "\n";
2078 print "Crunches: " . $stats{crunches} . " (" .
2079 get_percentage($stats{requests}, $stats{crunches}) . ")\n";
2080 print "Blocks: " . $stats{'blocked'} . " (" .
2081 get_percentage($stats{requests}, $stats{'blocked'}) . ")\n";
2082 print "Fast redirections: " . $stats{'fast-redirections'} . " (" .
2083 get_percentage($stats{requests}, $stats{'fast-redirections'}) . ")\n";
2084 print "Outgoing requests: " . $outgoing_requests . " (" .
2085 get_percentage($stats{requests}, $outgoing_requests) . ")\n";
2086 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
2087 get_percentage($stats{requests}, $stats{'server-keep-alive'}) . ")\n";
2088 print "New outgoing connections: " . $new_connections . " (" .
2089 get_percentage($stats{requests}, $new_connections) . ")\n";
2090 print "Reused connections: " . $stats{'reused-connections'} . " (" .
2091 get_percentage($stats{requests}, $stats{'reused-connections'}) .
2092 "; server offers accepted: " .
2093 get_percentage($stats{'server-keep-alive'}, $stats{'reused-connections'}) . ")\n";
2094 print "Empty responses: " . $stats{'empty-responses'} . " (" .
2095 get_percentage($stats{requests}, $stats{'empty-responses'}) . ")\n";
2096 print "Empty responses on new connections: "
2097 . $stats{'empty-responses-on-new-connections'} . " (" .
2098 get_percentage($stats{requests}, $stats{'empty-responses-on-new-connections'})
2100 print "Empty responses on reused connections: " .
2101 $stats{'empty-responses-on-reused-connections'} . " (" .
2102 get_percentage($stats{requests}, $stats{'empty-responses-on-reused-connections'}) .
2105 if ($stats{method} eq 0) {
2106 print "No response lines parsed yet yet.\n";
2109 print "Method distribution:\n";
2110 foreach my $method (sort {$stats{'method'}{$b} <=> $stats{'method'}{$a}} keys %{$stats{'method'}}) {
2111 printf "%8d : %-8s\n", $stats{'method'}{$method}, $method;
2113 print "Client HTTP versions:\n";
2114 foreach my $http_version (sort {$stats{'http-version'}{$b} <=> $stats{'http-version'}{$a}} keys %{$stats{'http-version'}}) {
2115 printf "%d : %s\n", $stats{'http-version'}{$http_version}, $http_version;
2118 if ($cli_options{'url-statistics-threshold'} == 0) {
2119 print "URL statistics are disabled. Increase --url-statistics-threshold to enable them.\n";
2121 print "Requested URLs:\n";
2122 foreach my $resource (sort {$stats{'resource'}{$b} <=> $stats{'resource'}{$a}} keys %{$stats{'resource'}}) {
2123 if ($stats{'resource'}{$resource} < $cli_options{'url-statistics-threshold'}) {
2124 print "Skipped statistics for URLs below the treshold.\n";
2127 printf "%d : %s\n", $stats{'resource'}{$resource}, $resource;
2131 if ($cli_options{'host-statistics-threshold'} == 0) {
2132 print "Host statistics are disabled. Increase --host-statistics-threshold to enable them.\n";
2134 print "Requested Hosts:\n";
2135 foreach my $host (sort {$stats{'hosts'}{$b} <=> $stats{'hosts'}{$a}} keys %{$stats{'hosts'}}) {
2136 if ($stats{'hosts'}{$host} < $cli_options{'host-statistics-threshold'}) {
2137 print "Skipped statistics for Hosts below the treshold.\n";
2140 printf "%d : %s\n", $stats{'hosts'}{$host}, $host;
2146 ################################################################################
2147 # Functions that actually print stuff
2148 ################################################################################
2150 sub print_clf_message () {
2152 our ($ip, $timestamp, $request_line, $status_code, $size);
2155 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2157 # Rebuild highlighted
2158 $output .= $h{'Number'} . $ip . $h{'Standard'};
2160 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2162 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2164 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2166 $output .= $h{'Number'} . $size . $h{'Standard'};
2167 $output .= $line_end;
2172 sub print_non_clf_message ($) {
2174 my $content = shift;
2175 my $msec_string = $no_msecs_mode ? '' : '.' . $req{$t}{'msecs'};
2176 my $line_start = $html_output_mode ? '' : $h{"Standard"};
2178 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2181 . $time_colours[$time_colour_index % 2]
2182 . $req{$t}{'time-stamp'}
2184 . $h{Standard} . " "
2185 . $thread_colours{$t}
2189 . $h{$req{$t}{'log-level'}}
2190 . $req{$t}{'log-level'}
2197 sub shorten_thread_id ($) {
2199 my $thread_id = shift;
2201 our %short_thread_ids;
2204 unless (defined $short_thread_ids{$thread_id}) {
2205 $short_thread_ids{$thread_id} = sprintf "%.3d", $max_threadid++;
2208 return $short_thread_ids{$thread_id}
2213 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2215 my $last_thread = 0;
2216 my $last_timestamp = 0;
2217 my $filters_that_did_nothing;
2220 $time_colour = paint_it('white');
2222 my %log_level_handlers = (
2223 'Re-Filter' => \&handle_loglevel_re_filter,
2224 'Header' => \&handle_loglevel_header,
2225 'Connect' => \&handle_loglevel_connect,
2226 'Redirect' => \&handle_loglevel_redirect,
2227 'Request' => \&handle_loglevel_request,
2228 'Crunch' => \&handle_loglevel_crunch,
2229 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2230 'Info' => \&handle_loglevel_info,
2231 'CGI' => \&handle_loglevel_cgi,
2232 'Force' => \&handle_loglevel_force,
2233 'Error' => \&handle_loglevel_error,
2234 'Fatal error' => \&handle_loglevel_ignore,
2235 'Writing' => \&handle_loglevel_ignore,
2236 'Received' => \&handle_loglevel_ignore,
2237 'Actions' => \&handle_loglevel_ignore,
2238 'Unknown log level' => \&handle_loglevel_ignore,
2243 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2244 $thread = $t = ($shorten_thread_ids) ? shorten_thread_id($4) : $4;
2245 $req{$t}{'day'} = $day = $1;
2246 $req{$t}{'time-stamp'} = $time_stamp = $2;
2247 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2248 $req{$t}{'log-level'} = $log_level = $5;
2249 $req{$t}{'content'} = $content = $c = $6;
2250 $req{$t}{'log-message'} = $_;
2251 $no_special_header_highlighting = 0;
2253 if (defined($log_level_handlers{$log_level})) {
2255 $content = $log_level_handlers{$log_level}($content);
2259 die "No handler found for log level \"$log_level\"\n";
2262 # Highlight Truncations
2263 if (length($_) > 4000) {
2264 $content =~ s@(too long, truncated)]$@$h{'Truncation'}$1$h{'Standard'}]@g;
2267 next unless $content;
2269 # Register threads to keep the colour constant
2270 if (!defined($thread_colours{$thread})) {
2271 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2272 $thread_colour_index++;
2275 # Switch timestamp colour if timestamps differ
2276 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2277 debug_message("Tick tack!") if DEBUG_TICKS;
2278 $time_colour = $time_colours[$time_colour_index % 2];
2279 $time_colour_index++;
2280 $last_msecs = $msecs;
2281 $last_timestamp = $time_stamp;
2284 $last_thread = $thread;
2286 print_non_clf_message($content);
2288 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2290 # LOG_LEVEL_CLF lines look like this
2291 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2292 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2293 # HTTP/1.1" 403 1730
2294 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2296 print_clf_message();
2300 # Some Privoxy log messages span more than one line,
2301 # usually to dump lots of content that doesn't need any syntax highlighting.
2302 # XXX: add mechanism to forward these lines to the right handler anyway.
2304 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2305 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2313 my ($day, $time_stamp, $msecs, $thread, $log_level, $content);
2314 my %log_level_handlers = (
2315 'Re-Filter' => \&handle_loglevel_ignore,
2316 'Header' => \&gather_loglevel_header_stats,
2317 'Connect' => \&gather_loglevel_connect_stats,
2318 'Redirect' => \&handle_loglevel_ignore,
2319 'Request' => \&gather_loglevel_request_stats,
2320 'Crunch' => \&gather_loglevel_crunch_stats,
2321 'Gif-Deanimate' => \&handle_loglevel_ignore,
2322 'Info' => \&handle_loglevel_ignore,
2323 'CGI' => \&handle_loglevel_ignore,
2324 'Force' => \&handle_loglevel_ignore,
2325 'Error' => \&gather_loglevel_error_stats,
2326 'Fatal error' => \&handle_loglevel_ignore,
2327 'Writing' => \&handle_loglevel_ignore,
2328 'Received' => \&handle_loglevel_ignore,
2329 'Actions' => \&handle_loglevel_ignore,
2330 'Unknown log level' => \&handle_loglevel_ignore
2334 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2337 $msecs = $3 ? $3 : 0;
2342 if (defined($log_level_handlers{$log_level})) {
2344 $content = $log_level_handlers{$log_level}($content, $thread);
2348 die "No handler found for log level \"$log_level\"\n";
2358 sub unbreak_lines_only_loop() {
2359 my $log_messages_reached = 0;
2363 # Log level other than LOG_LEVEL_CLF?
2364 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/ or
2366 m/^((?:\d+\.\d+\.\d+\.\d+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2367 $log_messages_reached = 1;
2372 $_ = "\n". $_ if /^(?:\d+\.\d+\.\d+\.\d+)/;
2377 print "\n" unless $log_messages_reached;
2382 sub VersionMessage {
2383 my $version_message;
2385 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2386 $version_message .= 'Copyright (C) 2007-2010 Fabian Keil <fk@fabiankeil.de>' . "\n";
2387 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2389 print $version_message;
2392 sub get_cli_options () {
2394 our %cli_options = (
2395 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2396 'title' => CLI_OPTION_TITLE,
2397 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2398 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2399 'no-msecs' => CLI_OPTION_NO_MSECS,
2400 'shorten-thread-ids' => CLI_OPTION_SHORTEN_THREAD_IDS,
2401 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2402 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
2403 'statistics' => CLI_OPTION_STATISTICS,
2404 'url-statistics-threshold' => CLI_OPTION_URL_STATISTICS_THRESHOLD,
2405 'unbreak-lines-only' => CLI_OPTION_UNBREAK_LINES_ONLY,
2406 'host-statistics-threshold'=> CLI_OPTION_HOST_STATISTICS_THRESHOLD,
2410 'html-output' => \$cli_options{'html-output'},
2411 'title' => \$cli_options{'title'},
2412 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2413 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2414 'no-msecs' => \$cli_options{'no-msecs'},
2415 'shorten-thread-ids' => \$cli_options{'shorten-thread-ids'},
2416 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2417 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
2418 'statistics' => \$cli_options{'statistics'},
2419 'unbreak-lines-only' => \$cli_options{'unbreak-lines-only'},
2420 'url-statistics-threshold=i'=> \$cli_options{'url-statistics-threshold'},
2421 'host-statistics-threshold=i'=> \$cli_options{'host-statistics-threshold'},
2422 'version' => sub { VersionMessage && exit(0) },
2426 $html_output_mode = cli_option_is_set('html-output');
2427 $no_msecs_mode = cli_option_is_set('no-msecs');
2428 $shorten_thread_ids = cli_option_is_set('shorten-thread-ids');
2429 $line_end = get_line_end();
2440 Options and their default values if they have any:
2441 [--accept-unknown-messages]
2442 [--host-statistics-threshold $cli_options{'host-statistics-threshold'}]
2446 [--no-syntax-highlighting]
2447 [--shorten-thread-ids]
2448 [--show-ineffective-filters]
2450 [--unbreak-lines-only]
2451 [--url-statistics-threshold $cli_options{'url-statistics-threshold'}]
2452 [--title $cli_options{'title'}]
2454 see "perldoc $0" for more information
2460 ################################################################################
2462 ################################################################################
2466 set_background(DEFAULT_BACKGROUND);
2467 prepare_our_stuff();
2471 # XXX: should explicitly reject incompatible argument combinations
2472 if (cli_option_is_set('unbreak-lines-only')) {
2473 unbreak_lines_only_loop();
2474 } elsif (cli_option_is_set('statistics')) {
2487 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2491 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
2492 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--statistics>]
2493 [B<--shorten-thread-ids>] [B<--show-ineffective-filters>]
2494 [B<--url-statistics-threshold>] [B<--version>]
2498 B<privoxy-log-parser> reads Privoxy log messages and
2500 - syntax-highlights recognized lines,
2502 - reformats some of them for easier comprehension,
2504 - filters out less useful messages, and
2506 - (in some cases) calculates additional information,
2507 like the compression ratio or how a filter affected
2510 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2511 without getting confused by the resulting amount of output. For example for
2512 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2513 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2514 will hide the "filter foo caused 0 hits" message.
2518 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
2519 just don't highlight them.
2521 [B<--host-statistics-threshold>] Only show the request count for a host
2522 if it's above or equal to the given threshold. If the threshold is 0, host
2523 statistics are disabled.
2525 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2526 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2527 This option is only intended to make embedding log excerpts in web pages easier.
2528 It does not escape any input!
2530 [B<--no-msecs>] Don't expect milisecond resolution
2532 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2533 the filtered output is piped into less in which case the ANSI control
2534 codes don't work, or if the terminal itself doesn't support the control
2537 [B<--shorten-thread-ids>] Shorten the thread ids to a three-digit decimal number.
2538 Note that the mapping from thread ids to shortened ids is created at run-time
2539 and thus varies with the input.
2541 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2542 that didn't modify the content.
2544 [B<--statistics>] Gather various statistics instead of syntax highlighting
2545 log messages. This is an experimental feature, if the results look wrong
2546 they very well might be. Also note that the results are pretty much guaranteed
2547 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2549 [B<--unbreak-lines-only>] Tries to fix lines that got messed up by a broken or
2550 interestingly configured mail client and thus are no longer recognized properly.
2551 Only fixes some breakage, but may be good enough or at least better than nothing.
2552 Doesn't do anything else, so you probably want to pipe the output into
2553 B<privoxy-log-parser> again.
2555 [B<--url-statistics-threshold>] Only show the request count for a resource
2556 if it's above or equal to the given threshold. If the threshold is 0, URL
2557 statistics are disabled.
2559 [B<--version>] Print version and exit.
2563 To monitor a log file:
2565 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2567 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2568 or if the log won't get rotated anyway. The log file location depends
2569 on your system (Doh!).
2571 To monitor Privoxy without having it write to a log file:
2573 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2575 Again, the config file location depends on your system. Output redirection
2576 depends on your shell, the above works with bourne shells.
2578 To read a processed Privoxy log file from top to bottom, letting the content
2579 scroll by slightly faster than you can read:
2581 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2583 This is probably only useful to fill screens in the background of haxor movies.
2587 Syntax highlighting with ANSI escape sequences will look strange
2588 if your background color isn't black.
2590 Some messages aren't recognized yet and will not be fully highlighted.
2592 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2593 using earlier Privoxy versions will probably result in an increased amount
2594 of unrecognized log lines.
2596 Privoxy's log files tend to be rather large. If you use HTML
2597 highlighting some browsers can't handle them, get confused and
2598 will eventually crash because of segmentation faults or unexpected
2599 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2604 Many settings can't be controlled through command line options yet.
2612 Fabian Keil <fk@fabiankeil.de>