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 1.101 2010/11/08 17:52:55 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 => '0.6',
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_URL_STATISTICS_THRESHOLD => 0,
65 CLI_OPTION_HOST_STATISTICS_THRESHOLD => 0,
67 SUPPRESS_SUCCEEDED_FILTER_ADDITIONS => 1,
69 SHOW_FILTER_READIN_IN => 0,
70 SUPPRESS_EMPTY_LINES => 1,
71 SUPPRESS_SUCCESSFUL_CONNECTIONS => 1,
72 SUPPRESS_ACCEPTED_CONNECTIONS => 1,
73 SUPPRESS_GIF_NOT_CHANGED => 1,
74 SUPPRESS_NEED_TO_DE_CHUNK_FIRST => 1,
76 DEBUG_HEADER_REGISTERING => 0,
77 DEBUG_HEADER_HIGHLIGHTING => 0,
80 DEBUG_SUPPRESS_LOG_MESSAGES => 0,
82 PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH => 0,
83 PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH => 1,
85 LOG_UNPARSED_LINES_TO_EXTRA_FILE => 0,
86 ERROR_LOG_FILE => '/var/log/privoxy-log-parser',
88 # You better leave these alone unless you know what you're doing.
89 COLOUR_RESET => "\033[0;0m",
93 # For performance reasons, these are global.
96 my %req; # request data from previous lines
101 my $thread_colour_index = 0;
102 my $header_colour_index = 0;
103 my $time_colour_index = 0;
105 my $no_special_header_highlighting;
108 my $header_highlight_regex = '';
110 my $html_output_mode;
111 my $no_msecs_mode; # XXX: should probably be removed
112 my $shorten_thread_ids;
115 sub prepare_our_stuff () {
117 # Syntax Higlight hash
119 'red', 'green', 'brown', 'blue', 'purple', 'cyan',
120 'light_gray', 'light_red', 'light_green', 'yellow',
121 'light_blue', 'pink', 'light_cyan', 'white'
128 Filter => 'purple', # XXX: Used?
129 'Re-Filter' => 'purple',
131 Request => 'light_cyan',
132 CGI => 'light_green',
134 Error => 'light_red',
136 'Fatal error' => 'light_red',
137 'Gif-Deanimate' => 'blue',
139 Writing => 'light_green',
140 Received => 'yellow',
141 # ----------------------
144 request_ => 'brown', # host+path but no protocol
145 'ip-address' => 'yellow',
148 Truncation => 'light_red',
150 Timestamp => 'brown',
151 Crunching => 'light_red',
152 crunched => 'light_red',
153 'Request-Line' => 'pink',
155 destination => 'yellow',
156 'http-version' => 'pink',
157 'crunch-pattern' => 'pink',
162 'program-name' => 'cyan',
165 warning => 'light_red',
166 debug => 'light_red',
170 'status-message' => 'light_cyan',
171 'status-code' => 'yellow',
172 'invalid-request' => 'light_red',
174 error => 'light_red',
175 'rewritten-URL' => 'light_red',
176 'pcrs-delimiter' => 'light_red',
177 'ignored' => 'light_red',
178 'action-bits-update' => 'light_red',
179 'configuration-line' => 'red',
180 'content-type' => 'yellow',
181 'HOST' => HEADER_DEFAULT_COLOUR,
186 # Header colours need their own hash so the keys can be accessed properly
188 # Prefilled with headers that should not appear with default header colours
189 Cookie => 'light_red',
190 'Set-Cookie' => 'light_red',
191 Warning => 'light_red',
192 Default => HEADER_DEFAULT_COLOUR,
195 # Crunch reasons need their own hash as well
197 'Unsupported HTTP feature' => 'light_red',
198 Blocked => 'light_red',
199 Untrusted => 'light_red',
200 Redirected => 'green',
201 'CGI Call' => 'white',
202 'DNS failure' => 'red',
203 'Forwarding failed' => 'light_red',
204 'Connection failure' => 'light_red',
205 'Out of memory (may mask other reasons)' => 'light_red',
206 'No reason recorded' => 'light_red',
209 @time_colours = ('white', 'light_gray');
211 # Translate highlight strings into highlight code
212 prepare_highlight_hash(\%header_colours);
213 prepare_highlight_hash(\%reason_colours);
214 prepare_highlight_hash(\%h);
215 prepare_colour_array(\@all_colours);
216 prepare_colour_array(\@time_colours);
223 ###############################################################
224 # Takes a colour string and returns an ANSI escape sequence
225 # (unless --no-syntax-highlighting is used).
226 # XXX: The Rolling Stones reference has to go.
227 ###############################################################
229 my $colour = shift @_;
231 return "" if cli_option_is_set('no-syntax-highlighting');
273 my $bg_code = get_background();
275 our $default = default_colours();
277 if (defined($text{$colour})) {
278 $colour_code = ESCAPE;
279 $colour_code .= $text{$colour};
281 $colour_code .= $light{$colour} ? "1" : "2";
283 $colour_code .= $bg_code;
285 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
287 } elsif ($colour =~ /reset/) {
289 $colour_code = default_colours();
293 die "What's $colour supposed to mean?\n";
299 sub get_semantic_html_markup ($) {
300 ###############################################################
301 # Takes a string and returns a span element
302 ###############################################################
307 if ($type =~ /Standard/) {
311 $code = '<span title="' . $type . '" class="' . $type . '">';
317 sub cli_option_is_set ($) {
320 my $cli_option = shift;
322 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
324 return $cli_options{$cli_option};
327 sub get_html_title () {
330 return $cli_options{'title'};
334 sub init_css_colours() {
342 purple => "F06", # XXX: wrong
343 cyan => "F09", # XXX: wrong
348 light_green => "33F",
357 sub get_css_colour ($) {
362 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
364 return '#' . $css_colours{$colour};
367 sub get_css_line ($) {
372 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
373 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
374 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
375 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
376 $css_line .= '}' . "\n";
381 sub get_css_line_for_colour ($) {
386 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
387 $css_line .= 'color:' . get_css_colour($colour) . ';';
388 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
389 $css_line .= '}' . "\n";
394 # XXX: Wrong solution
395 sub get_missing_css_lines () {
399 $css_line .= '.' . 'default' . ' {';
400 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
401 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
402 $css_line .= '}' . "\n";
409 our %css_colours; #XXX: Wrong solution
413 $css .= '.privoxy-log {';
414 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
415 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
418 foreach my $key (keys %h_colours) {
420 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
421 $css .= get_css_line($key);
425 foreach my $colour (keys %css_colours) {
427 $css .= get_css_line_for_colour($colour);
431 $css .= get_missing_css_lines(); #XXX: Wrong solution
440 if (cli_option_is_set('html-output')) {
442 my $title = get_html_title();
444 $intro .= '<html><head>';
445 $intro .= '<title>' . $title . '</title>';
446 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
447 $intro .= '</head><body>';
448 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
458 if (cli_option_is_set('html-output')) {
460 $outro = '</p></body></html>';
466 sub get_line_end () {
467 return cli_option_is_set('html-output') ? "<br>\n" : "\n";
470 sub get_colour_html_markup ($) {
471 ###############################################################
472 # Takes a colour string a span element. XXX: WHAT?
473 # XXX: This function shouldn't be necessary, the
474 # markup should always be semantically correct.
475 ###############################################################
480 if ($type =~ /Standard/) {
483 $code = '<span class="' . lc($type) . '">';
489 sub default_colours () {
492 return reset_colours();
495 sub show_colours () {
499 sub reset_colours () {
500 return ESCAPE . "0m";
503 sub set_background ($){
519 if (defined($backgrounds{$colour})) {
520 $bg_code = $backgrounds{$colour};
522 die "Invalid background colour: " . $colour;
526 sub get_background (){
530 sub prepare_highlight_hash ($) {
533 foreach my $key (keys %$ref) {
534 $$ref{$key} = $html_output_mode ?
535 get_semantic_html_markup($key) :
536 paint_it($$ref{$key});
540 sub prepare_colour_array ($) {
543 foreach my $i (0 ... @$ref - 1) {
544 $$ref[$i] = $html_output_mode ?
545 get_colour_html_markup($$ref[$i]) :
550 sub found_unknown_content ($) {
555 return if cli_option_is_set('accept-unknown-messages');
557 return if ($unknown =~ /\[too long, truncated\]$/);
559 $message = "found_unknown_content: Don't know how to highlight: ";
560 # Break line so the log file can later be parsed as Privoxy log file again
561 $message .= '"' . $unknown . '"' . " in:\n";
562 $message .= $req{$t}{'log-message'};
563 debug_message($message);
564 log_parse_error($req{$t}{'log-message'});
566 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
569 sub log_parse_error ($) {
573 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
574 open(ERRORLOG, ">>" . ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
575 print ERRORLOG $message;
580 sub debug_message (@) {
583 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
586 ################################################################################
587 # highlighter functions that aren't loglevel-specific
588 ################################################################################
592 # Get highlight marker
593 my $highlight = shift; # XXX: Stupid name;
597 if (defined($highlight)) {
599 $result = $h{$highlight};
603 $message = "h: Don't recognize highlighter $highlight.";
604 debug_message($message);
605 log_parser_error($message);
606 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
612 sub highlight_known_headers ($) {
616 debug_message("Searching $content for things to highlight.") if DEBUG_HEADER_HIGHLIGHTING;
618 if ($content =~ m/(?<=\s)($header_highlight_regex):/) {
620 $content =~ s@(?<=[\s|'])($header)(?=:)@$header_colours{$header}$1$h{'Standard'}@ig;
621 debug_message("Highlighted '$header' in '$content'") if DEBUG_HEADER_HIGHLIGHTING;
627 sub highlight_matched_request_line ($$) {
629 my $result = shift; # XXX: Stupid name;
631 if ($result =~ m@(.*)($regex)(.*)@) {
632 $result = $1 . highlight_request_line($2) . $3
637 sub highlight_request_line ($) {
640 my ($method, $url, $http_version);
642 #GET http://images.sourceforge.net/sfx/icon_warning.gif HTTP/1.1
643 if ($rl =~ m/Invalid request/) {
645 $rl = h('invalid-request') . $rl . h('Standard');
647 } elsif ($rl =~ m/^([-\w]+) (.*) (HTTP\/\d\.\d)/) {
649 # XXX: might not match in case of HTTP method fuzzing.
650 # XXX: save these: ($method, $path, $http_version) = ($1, $2, $3);
651 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
652 if ($rl =~ /http:\/\//) {
653 $rl = highlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
655 $rl = highlight_matched_pattern($rl, 'request_', '[^\s]*(?=\sHTTP)');
658 $rl =~ s@(HTTP\/\d\.\d)$@$h{'http-version'}$1$h{'Standard'}@;
660 } elsif ($rl =~ m/\.\.\. \[too long, truncated\]$/) {
662 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
663 $rl = highlight_matched_url($rl, '[^\s]*(?=\.\.\.)');
665 } elsif ($rl =~ m/^ $/) {
667 $rl = h('error') . "No request line specified!" . h('Standard');
671 debug_message ("Can't parse request line: $rl");
678 sub highlight_response_line ($) {
681 my ($http_version, $status_code, $status_message);
686 # TODO: Mark different status codes differently
688 if ($rl =~ m/((?:HTTP\/\d\.\d|ICY)) (\d+) (.*)/) {
689 ($http_version, $status_code, $status_message) = ($1, $2, $3);
691 debug_message ("Can't parse response line: $rl") and die 'Fix this';
694 # Rebuild highlighted
696 $rl .= h('http-version') . $http_version . h('Standard');
698 $rl .= h('status-code') . $status_code . h('Standard');
700 $rl .= h('status-message') . $status_message . h('Standard');
705 sub highlight_matched_url ($$) {
707 my $result = shift; # XXX: Stupid name;
710 #print "Got $result, regex ($regex)\n";
712 if ($result =~ m@(.*?)($regex)(.*)@) {
713 $result = $1 . highlight_url($2) . $3;
714 #print "Now the result is $result\n";
720 sub highlight_matched_host ($$) {
722 my ($result, $regex) = @_; # XXX: result ist stupid name;
724 if ($result =~ m@(.*?)($regex)(.*)@) {
725 $result = $1 . $h{host} . $2 . $h{Standard} . $3;
731 sub highlight_matched_pattern ($$$) {
733 my $result = shift; # XXX: Stupid name;
737 die "Unknown key $key" unless defined $h{$key};
739 if ($result =~ m@(.*?)($regex)(.*)@) {
740 $result = $1 . h($key) . $2 . h('Standard') . $3;
746 sub highlight_matched_path ($$) {
748 my $result = shift; # XXX: Stupid name;
751 if ($result =~ m@(.*?)($regex)(.*)@) {
752 $result = $1 . h('path') . $2 . h('Standard') . $3;
758 sub highlight_url ($) {
762 if ($html_output_mode) {
764 $url = '<a href="' . $url . '">' . $url . '</a>';
768 $url = h('URL') . $url . h('Standard');
775 sub update_header_highlight_regex ($) {
778 my $headers = join ('|', keys %header_colours);
780 $header_highlight_regex = qr/$headers/;
781 print "Registering '$header'\n" if DEBUG_HEADER_HIGHLIGHTING;
784 ################################################################################
785 # loglevel-specific highlighter functions
786 ################################################################################
788 sub handle_loglevel_header ($) {
792 if ($c =~ /^scan:/) {
794 if ($c =~ m/^scan: ([^: ]+):/) {
796 # Register new headers
797 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
799 if (!defined($header_colours{$header}) and $header =~ /^[\d\w-]*$/) {
800 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
802 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
803 $header_colours{$header} = $header_colours{'Default'};
805 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
806 $header_colour_index++;
808 update_header_highlight_regex($header);
811 } elsif ($c =~ m/^(scan: )(\w+ .+ HTTP\/\d\.\d)/) {
813 # scan: GET http://p.p/ HTTP/1.1
814 $c = $1 . highlight_request_line($2);
816 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
818 # scan: HTTP/1.1 200 OK
819 $req{$t}{'response_line'} = $2;
820 $req{$t}{'status_code'} = $3;
821 $req{$t}{'status_message'} = $4;
822 $c = $1 . highlight_response_line($req{$t}{'response_line'});
825 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
827 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
828 $c =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
829 $c =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
831 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
833 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/!
834 $c = highlight_matched_host($c, '(?<=New host is: )[^\s]+(?=\.)');
835 $c = highlight_matched_url($c, '(?<=Crunching Referer: )[^\s!]+');
837 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
839 # Text mode enabled by force. Take cover!
840 $c =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
842 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
844 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
845 $c = $1 . highlight_request_line($2);
847 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
849 # Adjusted Content-Length to 2132
850 # Adjust Content-Length to 33533
851 $c =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
852 $c = highlight_known_headers($c);
854 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
856 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
857 $c = highlight_matched_url($c, '(?<=New request URL: ).*');
859 } elsif ($c =~ m/^Couldn\'t parse:/) {
861 # XXX: These should probable be logged with LOG_LEVEL_ERROR
862 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
863 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
864 $c =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
866 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
867 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
869 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
870 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
871 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
873 # XXX: Save tag and tagger
875 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
876 $c =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
877 $c =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
878 $no_special_header_highlighting = 1;
880 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
882 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
883 # XXX: Save tag and tagger
885 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
886 $c =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
888 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
889 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
890 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
891 or $c =~ m/^(addh-unique|Referer forged to)/
892 or $c =~ m/^Downgraded answer to HTTP\/1.0/
893 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
894 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
895 or $c =~ m/^Referer crunched!/
896 or $c =~ m/^crunched x-forwarded-for!/
897 or $c =~ m/^crunched From!/
898 or $c =~ m/^ modified$/
899 or $c =~ m/^Content filtering is enabled. Crunching:/
900 or $c =~ m/^force-text-mode overruled the client/
901 or $c =~ m/^Server time in the future\./
902 or $c =~ m/^content-disposition header crunched and replaced with:/i
903 or $c =~ m/^Reducing white space in /
904 or $c =~ m/^Ignoring single quote in /
905 or $c =~ m/^Converting tab to space in /
906 or $c =~ m/A HTTP\/1\.1 response without/
907 or $c =~ m/Disabled filter mode on behalf of the client/
908 or $c =~ m/Keeping the (?:server|client) header /
909 or $c =~ m/Content modified with no Content-Length header set/
910 or $c =~ m/^Appended client IP address to/
911 or $c =~ m/^Removing 'Connection: close' to imply keep-alive./
912 or $c =~ m/^keep-alive support is disabled/
913 or $c =~ m/^Continue hack in da house/
916 # XXX: Some of these may need highlighting
918 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
919 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
920 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
921 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
922 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
923 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
924 # Text mode is already enabled
925 # Denied request with NULL byte(s) turned into line break(s)
926 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
927 # addh-unique: Host: people.freebsd.org
928 # Suppressed offer to compress content
929 # Crunched incoming cookie -- yum!
930 # Accepted the client's request to fetch without filtering.
931 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
932 # addh-unique: Host: subkeys.pgp.net:11371
933 # Referer forged to: Referer: http://10.0.0.1/
934 # Downgraded answer to HTTP/1.0
935 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
936 # Referer overwritten with: Referer: pille-palle
937 # Referer replaced with: Referer: pille-palle
938 # crunched x-forwarded-for!
940 # modified # XXX: pretty stupid log message
941 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
942 # force-text-mode overruled the client's request to fetch without filtering!
943 # Server time in the future.
944 # content-disposition header crunched and replaced with: content-disposition: filename=baz
945 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
946 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
947 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
948 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
950 # A HTTP/1.1 response without Connection header implies keep-alive.
951 # Disabled filter mode on behalf of the client.
952 # Keeping the server header 'Connection: keep-alive' around.
953 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
954 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
955 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
956 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
957 # Removing 'Connection: close' to imply keep-alive.
958 # keep-alive support is disabled. Crunching: Keep-Alive: 300.
959 # Continue hack in da house.
961 } elsif ($c =~ m/^scanning headers for:/) {
963 return '' unless SHOW_SCAN_INTRO;
965 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
966 # crunched User-Agent!
967 # Crunching: Content-Encoding: gzip
969 $c =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
971 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
973 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
975 $c = h('warning') . $c . h('Standard');
977 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
979 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
980 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
982 $c =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
983 $c =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
985 } elsif ($c =~ m/^Removing empty header/) {
987 # Removing empty header
990 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
992 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
993 # Enable force-text-mode if you know what you're doing.
994 # XXX: Could highlight more here.
995 $c =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
997 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
999 # Server keep-alive timeout is 5. Sticking with 10.
1000 # Client keep-alive timeout is 20. Sticking with 10.
1002 $c =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1003 $c =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1005 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1007 # Reducing keep-alive timeout from 60 to 10.
1009 $c =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1010 $c =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1012 } elsif ($c =~ m/^Killed all-caps Host header line: HOST:/) {
1014 # Killed all-caps Host header line: HOST: bestproxydb.com
1015 $c = highlight_matched_host($c, '(?<=HOST: )[^\s]+');
1016 $c = highlight_matched_pattern($c, 'HOST', 'HOST');
1020 found_unknown_content($c);
1024 unless ($c =~ m/^Transforming/) {
1025 $c = highlight_known_headers($c) unless $no_special_header_highlighting;
1031 sub handle_loglevel_re_filter ($) {
1033 my $content = shift;
1037 if ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1039 # XXX: only the second version gets highlighted properly.
1040 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1041 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1042 $req{$t}{'content_source'} = $1;
1043 $req{$t}{'content_size'} = $2;
1044 $req{$t}{'content_filter'} = $3;
1045 $req{$t}{'content_hits'} = $4;
1046 $req{$t}{'new_content_size'} = $5;
1047 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1048 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1049 if ($req{$t}{'content_hits'} == 0 and
1050 not (cli_option_is_set('show-ineffective-filters')
1051 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1055 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1056 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1057 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1059 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1060 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1063 $c .= "(" . $h{'Number'};
1064 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1065 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1068 } elsif ($c =~ /\.{3}$/
1069 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1071 # Used by Privoxy 3.0.5 and 3.0.6:
1073 # Used by Privoxy 3.0.7:
1074 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1076 $req{$t}{'filtered_header'} = $1;
1077 $req{$t}{'old_header_size'} = $2;
1078 $req{$t}{'header_filter_name'} = $3;
1080 unless (cli_option_is_set('show-ineffective-filters') or
1081 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1084 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1085 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1087 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1089 # ...produced 0 hits (new size 23).
1090 #... produced 1 hits (new size 54).
1092 $req{$t}{'header_filter_hits'} = $1;
1093 $req{$t}{'new_header_size'} = $2;
1095 unless (cli_option_is_set('show-ineffective-filters') or
1096 (defined($req{$t}{'header_filter_name'}) and
1097 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1099 if ($req{$t}{'header_filter_hits'} == 0 and
1100 not (defined($req{$t}{'header_filter_name'}) and
1101 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1104 # Reformat including information from the intro
1105 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1107 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1108 # $c .= $req{$t}{'filtered_header'} . " ";
1109 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1110 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1112 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1114 $c .= "changing size from ";
1115 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1117 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1122 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1126 # Highlight from last line (XXX: What?)
1127 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1128 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1133 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1134 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1139 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1141 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1142 # Tagger variable-test has empty joblist. Nothing to do.
1144 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1146 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1148 $req{$t}{'chunked-size'} = $1;
1149 $req{$t}{'dechunked-size'} = $2;
1150 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1152 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1154 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1155 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1157 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1159 # Decompression successful. Old size: 670, new size: 1166.
1161 $req{$t}{'size-compressed'} = $1;
1162 $req{$t}{'size-decompressed'} = $2;
1163 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1165 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1166 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1168 # XXX: Create sub get_percentage()
1169 if ($req{$t}{'size-decompressed'}) {
1170 $req{$t}{'decompression-gain-percent'} =
1171 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1173 $content .= " (saved: ";
1174 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1176 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1180 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1182 # Need to de-chunk first
1183 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1185 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1187 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1189 # Adding re_filter job ...
1190 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1191 # to filter client-ip-address succeeded.
1193 } elsif ($c =~ m/^Reading in filter/) {
1195 return '' unless SHOW_FILTER_READIN_IN;
1199 found_unknown_content($content);
1206 sub handle_loglevel_redirect ($) {
1210 if ($c =~ m/^Decoding "([^""]*)"/) {
1212 $req{$t}{'original-destination'} = $1;
1213 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1216 } elsif ($c =~ m/^Checking/) {
1218 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1219 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1221 # TODO: Change colour if really url-decoded
1222 $req{$t}{'decoded-original-destination'} = $1;
1223 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1226 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1228 # pcrs command "s@&from=rss@@" changed \
1229 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1230 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1231 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1232 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1233 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1234 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1235 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1236 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1238 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1240 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1241 # change "http://www.example.org/".
1242 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1243 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1245 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1247 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1248 # XXX: Use URL highlighter
1250 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1252 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1253 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1254 # is already properly formatted.
1255 # XXX: assume the same?
1256 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1260 found_unknown_content($c);
1267 sub handle_loglevel_gif_deanimate ($) {
1269 my $content = shift;
1271 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1273 my $bytes_from = $1;
1275 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1276 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1277 # XXX: Do we need g in case of ($1 == $2)?
1278 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1280 } elsif ($content =~ m/GIF (not) changed/) {
1282 # Gif-Deanimate: GIF not changed.
1283 return '' if SUPPRESS_GIF_NOT_CHANGED;
1284 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1286 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1288 # failed! (gif parsing)
1289 # XXX: Replace this error message with something less stupid
1290 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1292 } elsif ($content =~ m/^Need to de-chunk first/) {
1294 # Need to de-chunk first
1295 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1297 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1299 # No GIF header found (XXX: Did I ever commit this?)
1300 # failed while parsing 195 134747048 (XXX: never commited)
1302 # Ignore these for now
1306 found_unknown_content($content);
1313 sub handle_loglevel_request ($) {
1315 my $content = shift;
1317 if ($content =~ m/crunch! /) {
1319 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1321 # Highlight crunch reasons
1322 foreach my $reason (keys %reason_colours) {
1323 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1325 # Highlight request URL domain and ditch 'crunch!'
1326 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1327 $content =~ s@ crunch!@@;
1329 } elsif ($content =~ m/\[too long, truncated\]$/) {
1331 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1332 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1334 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1336 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1337 $content = h('request_') . $content . h('Standard');
1341 found_unknown_content($content);
1348 sub handle_loglevel_crunch ($) {
1350 my $content = shift;
1352 # Highlight crunch reason
1353 foreach my $reason (keys %reason_colours) {
1354 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1357 if ($content =~ m/\[too long, truncated\]$/) {
1359 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1360 # [...]&filter... [too long, truncated]
1361 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1365 # Blocked: http://ads.example.org/
1366 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1372 sub handle_loglevel_connect ($) {
1376 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1378 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1380 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1381 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1383 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1385 # connect to: www.example.org.noconnect failed: Operation not permitted
1387 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1389 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1391 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1393 # Connect: to www.nzherald.co.nz successful
1394 # Connect: to archiv.radiotux.de
1396 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1397 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1399 } elsif ($c =~ m/^to ([^\s]*)$/) {
1401 # Connect: to lists.sourceforge.net:443
1403 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1405 } elsif ($c =~ m/^accepted connection from .*/ or
1408 # accepted connection from 10.0.0.1
1409 # Privoxy 3.0.6 and earlier just say:
1411 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1412 $c = highlight_matched_host($c, '(?<=connection from ).*');
1414 } elsif ($c =~ m/^write header to: .* failed:/) {
1416 # write header to: 10.0.0.1 failed: Broken pipe
1418 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1419 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1421 } elsif ($c =~ m/^write header to client failed:/) {
1423 # write header to client failed: Broken pipe
1425 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1427 } elsif ($c =~ m/^socks4_connect:/) {
1429 # socks4_connect: SOCKS request rejected or failed.
1430 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1432 } elsif ($c =~ m/^Listening for new connections/ or
1433 $c =~ m/^accept connection/) {
1435 # Privoxy versions above 3.0.6 say:
1436 # Listening for new connections ...
1437 # earlier versions say:
1438 # accept connection ...
1441 } elsif ($c =~ m/^accept failed:/) {
1443 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1445 } elsif ($c =~ m/^Overriding forwarding settings/) {
1447 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1448 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1450 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1452 # Denying suspicious CONNECT request from 10.0.0.1
1453 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1455 } elsif ($c =~ m/^socks5_connect:/) {
1457 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1459 } elsif ($c =~ m/^Created new connection to/) {
1461 # Created new connection to www.privoxy.org:80 on socket 11.
1462 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1463 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1465 } elsif ($c =~ m/^Found reusable socket/) {
1467 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1469 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1470 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1471 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1472 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1473 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1474 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1475 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1476 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1478 } elsif ($c =~ m/^Marking open socket/) {
1480 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1481 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1482 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1483 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1485 } elsif ($c =~ m/^No reusable/) {
1487 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1488 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1490 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1492 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1493 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1495 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1496 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1497 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1499 } elsif ($c =~ m/^Socket/) {
1501 # Socket 16 already forgotten or never remembered.
1502 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1504 } elsif ($c =~ m/^The connection to/) {
1506 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1508 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1509 # Timeout is: 1. Assumed latency: 4.
1510 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1511 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1512 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1513 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1514 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1515 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1517 } elsif ($c =~ m/^Stopped waiting for the request line./) {
1519 # Stopped waiting for the request line. Timeout: 121.
1520 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1522 } elsif ($c =~ m/^Waiting for \d/) {
1524 # Waiting for 1 connections to timeout.
1525 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1527 } elsif ($c =~ m/^Initialized/) {
1529 # Initialized 20 socket slots.
1530 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1532 } elsif ($c =~ m/^Done reading from server/) {
1534 # Done reading from server. Expected content length: 24892. \
1535 # Actual content length: 24892. Most recently received: 4412.
1537 # Done reading from server. Expected content length: 24892. \
1538 # Actual content length: 24892. Bytes most recently read: 4412.
1539 # Done reading from server. Content length: 6018 as expected. \
1540 # Bytes most recently read: 294.
1541 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1542 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1543 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1545 } elsif ($c =~ m/^Continuing buffering headers/) {
1547 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1548 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1549 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1550 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1552 # Continuing buffering headers. Bytes most recently read: %d.
1553 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1555 } elsif ($c =~ m/^Received \d+ bytes while/) {
1557 # Received 206 bytes while expecting 12103.
1558 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1559 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1561 } elsif ($c =~ m/^(Rejecting c|C)onnection from/) {
1563 # Connection from 81.163.28.218 dropped due to ACL
1564 # Rejecting connection from 178.63.152.227. Maximum number of connections reached.
1565 $c =~ s@(?<=onnection from )((?:\d+\.?){3}\d+)@$h{'Number'}$1$h{'Standard'}@;
1567 } elsif ($c =~ m/^(?:Reusing|Closing) server socket \d./ or
1568 $c =~ m/^No additional client request/) {
1570 # Reusing server socket 4. Opened for 10.0.0.1.
1571 # Closing server socket 2. Opened for 10.0.0.1.
1572 # No additional client request received in time. \
1573 # Closing server socket 4, initially opened for 10.0.0.1.
1574 # No additional client request received in time on socket 29.
1576 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1577 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.$)');
1579 } elsif ($c =~ m/^Connected to /) {
1581 # Connected to tor-jail[10.0.0.2]:9050.
1583 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1584 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1585 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1587 } elsif ($c =~ m/^Could not connect to /) {
1589 # Could not connect to [10.0.0.1]:80.
1591 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1592 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1594 } elsif ($c =~ m/^Waiting for the next client request/ or
1595 $c =~ m/^The connection on server socket/ or
1596 $c =~ m/^Client request arrived in time or the client closed the connection/) {
1598 # Waiting for the next client request on socket 3. Keeping the server \
1599 # socket 12 to a.fsdn.com open.
1600 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1601 # Client request arrived in time or the client closed the connection on socket 12.
1603 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1604 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1605 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1607 } elsif ($c =~ m/^Marking the server socket/) {
1609 # Marking the server socket 7 tainted.
1611 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1613 } elsif ($c =~ m/^Reduced expected bytes to /) {
1615 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1616 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1617 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1619 } elsif ($c =~ m/^The client closed socket /) {
1621 # The client closed socket 2 while the server socket 4 is still open.
1622 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1623 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1625 } elsif ($c =~ m/^Expected client content length set /) {
1627 # Expected client content length set to 667325411 after reading 4999 bytes.
1628 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1629 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1631 } elsif ($c =~ m/^Waiting for up to /) {
1633 # Waiting for up to 4999 bytes from the client.
1634 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1636 } elsif ($c =~ m/^Looks like we / or
1637 $c =~ m/^Unsetting keep-alive flag/ or
1638 $c =~ m/^No connections to wait/ or
1639 $c =~ m/^Complete client request received/ or
1640 $c =~ m/^Possible pipeline attempt detected./ or
1641 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1642 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1643 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1644 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1645 $c =~ m/^Stopping to watch the client socket. There's already another request waiting./ or
1646 $c =~ m/^Done reading from the client\.$/) {
1648 # Looks like we reached the end of the last chunk. We better stop reading.
1649 # Looks like we read the end of the last chunk together with the server \
1650 # headers. We better stop reading.
1651 # Looks like we got the last chunk together with the server headers. \
1652 # We better stop reading.
1653 # Unsetting keep-alive flag.
1654 # No connections to wait for left.
1655 # Client request arrived in time or the client closed the connection.
1656 # Complete client request received
1657 # Possible pipeline attempt detected. The connection will not be \
1658 # kept alive and we will only serve the first request.
1659 # POST request detected. The connection will not be kept alive.
1660 # The server still wants to talk, but the client hung up on us.
1661 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1662 # There might be a request body. The connection will not be kept alive.
1663 # Stopping to watch the client socket. There's already another request waiting.
1664 # Done reading from the client\.
1668 found_unknown_content($c);
1676 sub handle_loglevel_info ($) {
1680 if ($c =~ m/^Rewrite detected:/) {
1682 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1683 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1685 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1686 $c =~ m/^Compressed content detected/ or
1689 # Decompressing deflated iob: 117
1690 # Decompression didn't result in any content.
1691 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1692 # with zlib support or enable the prevent-compression action.
1693 # Tagger 'complete-url' created empty tag. Ignored.
1697 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1699 # loading configuration file '/usr/local/etc/privoxy/config':
1700 # Reloading configuration file '/usr/local/etc/privoxy/config'
1701 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1703 } elsif ($c =~ m/^Loading (actions|filter) file: /) {
1705 # Loading actions file: /usr/local/etc/privoxy/default.action
1706 # Loading filter file: /usr/local/etc/privoxy/default.filter
1707 $c =~ s@(?<= file: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1709 } elsif ($c =~ m/^exiting by signal/) {
1711 # exiting by signal 15 .. bye
1712 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1714 } elsif ($c =~ m/^Privoxy version/) {
1716 # Privoxy version 3.0.7
1717 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1719 } elsif ($c =~ m/^Program name: /) {
1721 # Program name: /usr/local/sbin/privoxy
1722 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1724 } elsif ($c =~ m/^Listening on port /) {
1726 # Listening on port 8118 on IP address 10.0.0.1
1727 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1728 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1730 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1732 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1733 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1735 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1737 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1738 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1739 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1741 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1742 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1743 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1744 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1746 } elsif ($c =~ m/^Status code/) {
1748 # Status code 304 implies no body.
1749 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1751 } elsif ($c =~ m/^Method/) {
1753 # Method HEAD implies no body.
1754 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1756 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1758 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1759 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1760 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1762 } elsif ($c =~ m/^No logfile configured/ or
1763 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1764 $c =~ m/^Invalid \"chunked\" transfer/ or
1765 $c =~ m/^Support for/ or
1766 $c =~ m/^Flushing header and buffers/ or
1767 $c =~ m/^Can not resolve/
1770 # No logfile configured. Please enable it before reporting any problems.
1771 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1772 # response or even no response at all.
1773 # No logfile configured. Logging disabled.
1774 # Invalid "chunked" transfer encoding detected and ignored.
1775 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1776 # known not to work properly in some situations.
1777 # Flushing header and buffers. Stepping back from filtering.
1778 # Can not resolve doesnotexist: hostname nor servname provided, or not known
1782 found_unknown_content($c);
1789 sub handle_loglevel_cgi ($) {
1793 if ($c =~ m/^Granting access to/) {
1795 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1797 } elsif ($c =~ m/^Substituting: s(.)/) {
1799 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1800 # XXX: prone to span several lines
1803 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1804 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1805 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1811 sub handle_loglevel_force ($) {
1815 if ($c =~ m/^Ignored force prefix in request:/) {
1817 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1818 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1819 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1821 } elsif ($c =~ m/^Enforcing request:/) {
1823 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1824 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1828 found_unknown_content($c);
1835 sub handle_loglevel_error ($) {
1839 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1841 # Empty server or forwarder response received on socket 4.
1842 # Empty server or forwarder response received on socket 3. \
1843 # Closing client socket 15 without sending data.
1844 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1845 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1847 # XXX: There are probably more messages that deserve highlighting.
1853 sub handle_loglevel_ignore ($) {
1857 sub gather_loglevel_request_stats ($$) {
1865 sub gather_loglevel_crunch_stats ($$) {
1873 if ($c =~ m/^Redirected:/) {
1874 # Redirected: http://www.example.org/http://p.p/
1875 $stats{'fast-redirections'}++;
1877 } elsif ($c =~ m/^Blocked:/) {
1878 # Blocked: blogger.googleusercontent.com:443
1879 $stats{'blocked'}++;
1884 sub gather_loglevel_error_stats ($$) {
1891 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1893 # Empty server or forwarder response received on socket 4.
1894 $stats{'empty-responses'}++;
1895 if ($thread_data{$thread}{'new_connection'}) {
1896 $stats{'empty-responses-on-new-connections'}++;
1898 $stats{'empty-responses-on-reused-connections'}++;
1903 sub gather_loglevel_connect_stats ($$) {
1905 my ($c, $thread) = @_;
1909 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
1911 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1912 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
1914 } elsif ($c =~ m/^to ([^\s]*)$/) {
1916 # Connect: to lists.sourceforge.net:443
1918 $thread_data{$thread}{'forwarder'} = 'direct connection';
1920 } elsif ($c =~ m/^Created new connection to/) {
1922 # Created new connection to www.privoxy.org:80 on socket 11.
1924 $thread_data{$thread}{'new_connection'} = 1;
1926 } elsif ($c =~ m/^Reusing server socket \d./ or
1927 $c =~ m/^Found reusable socket/) {
1929 # Reusing server socket 4. Opened for 10.0.0.1.
1930 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1932 $thread_data{$thread}{'new_connection'} = 0;
1933 $stats{'reused-connections'}++;
1937 sub gather_loglevel_header_stats ($$) {
1939 my ($c, $thread) = @_;
1942 if ($c =~ m/^A HTTP\/1\.1 response without/ or
1943 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
1945 # A HTTP/1.1 response without Connection header implies keep-alive.
1946 # Keeping the server header 'Connection: keep-alive' around.
1947 $stats{'server-keep-alive'}++;
1949 } elsif ($c =~ m/^scan: ((\w+) (.+) (HTTP\/\d\.\d))/) {
1951 # scan: HTTP/1.1 200 OK
1952 $stats{'method'}{$2}++;
1953 $stats{'ressource'}{$3}++;
1954 $stats{'http-version'}{$4}++;
1956 } elsif ($c =~ m/^scan: Host: ([^\s]+)/) {
1959 $stats{'hosts'}{$1}++;
1967 'server-keep-alive' => 0,
1968 'reused-connections' => 0,
1969 'empty-responses' => 0,
1970 'empty-responses-on-new-connections' => 0,
1971 'empty-responses-on-reused-connections' => 0,
1972 'fast-redirections' => 0,
1974 'reused-connections' => 0,
1975 'server-keep-alive' => 0,
1979 sub get_percentage ($$) {
1982 return "NaN" if ($big eq 0);
1983 return sprintf("%.2f%%", $small / $big * 100);
1986 sub print_stats () {
1990 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
1991 my $outgoing_requests = $stats{requests} - $stats{crunches};
1993 if ($stats{requests} eq 0) {
1994 print "No requests yet.\n";
1998 print "Client requests total: " . $stats{requests} . "\n";
1999 print "Crunches: " . $stats{crunches} . " (" .
2000 get_percentage($stats{requests}, $stats{crunches}) . ")\n";
2001 print "Blocks: " . $stats{'blocked'} . " (" .
2002 get_percentage($stats{requests}, $stats{'blocked'}) . ")\n";
2003 print "Fast redirections: " . $stats{'fast-redirections'} . " (" .
2004 get_percentage($stats{requests}, $stats{'fast-redirections'}) . ")\n";
2005 print "Outgoing requests: " . $outgoing_requests . " (" .
2006 get_percentage($stats{requests}, $outgoing_requests) . ")\n";
2007 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
2008 get_percentage($stats{requests}, $stats{'server-keep-alive'}) . ")\n";
2009 print "New outgoing connections: " . $new_connections . " (" .
2010 get_percentage($stats{requests}, $new_connections) . ")\n";
2011 print "Reused connections: " . $stats{'reused-connections'} . " (" .
2012 get_percentage($stats{requests}, $stats{'reused-connections'}) .
2013 "; server offers accepted: " .
2014 get_percentage($stats{'server-keep-alive'}, $stats{'reused-connections'}) . ")\n";
2015 print "Empty responses: " . $stats{'empty-responses'} . " (" .
2016 get_percentage($stats{requests}, $stats{'empty-responses'}) . ")\n";
2017 print "Empty responses on new connections: "
2018 . $stats{'empty-responses-on-new-connections'} . " (" .
2019 get_percentage($stats{requests}, $stats{'empty-responses-on-new-connections'})
2021 print "Empty responses on reused connections: " .
2022 $stats{'empty-responses-on-reused-connections'} . " (" .
2023 get_percentage($stats{requests}, $stats{'empty-responses-on-reused-connections'}) .
2026 if ($stats{method} eq 0) {
2027 print "No response lines parsed yet yet.\n";
2030 print "Method distribution:\n";
2031 foreach my $method (sort {$stats{'method'}{$b} <=> $stats{'method'}{$a}} keys %{$stats{'method'}}) {
2032 printf "%8d : %-8s\n", $stats{'method'}{$method}, $method;
2034 print "Client HTTP versions:\n";
2035 foreach my $http_version (sort {$stats{'http-version'}{$b} <=> $stats{'http-version'}{$a}} keys %{$stats{'http-version'}}) {
2036 printf "%d : %s\n", $stats{'http-version'}{$http_version}, $http_version;
2039 if ($cli_options{'url-statistics-threshold'} == 0) {
2040 print "URL statistics are disabled. Increase --url-statistics-threshold to enable them.\n";
2042 print "Requested URLs:\n";
2043 foreach my $ressource (sort {$stats{'ressource'}{$b} <=> $stats{'ressource'}{$a}} keys %{$stats{'ressource'}}) {
2044 if ($stats{'ressource'}{$ressource} < $cli_options{'url-statistics-threshold'}) {
2045 print "Skipped statistics for URLs below the treshold.\n";
2048 printf "%d : %s\n", $stats{'ressource'}{$ressource}, $ressource;
2052 if ($cli_options{'host-statistics-threshold'} == 0) {
2053 print "Host statistics are disabled. Increase --host-statistics-threshold to enable them.\n";
2055 print "Requested Hosts:\n";
2056 foreach my $host (sort {$stats{'hosts'}{$b} <=> $stats{'hosts'}{$a}} keys %{$stats{'hosts'}}) {
2057 if ($stats{'hosts'}{$host} < $cli_options{'host-statistics-threshold'}) {
2058 print "Skipped statistics for Hosts below the treshold.\n";
2061 printf "%d : %s\n", $stats{'hosts'}{$host}, $host;
2067 ################################################################################
2068 # Functions that actually print stuff
2069 ################################################################################
2071 sub print_clf_message () {
2073 our ($ip, $timestamp, $request_line, $status_code, $size);
2076 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2078 # Rebuild highlighted
2079 $output .= $h{'Number'} . $ip . $h{'Standard'};
2081 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2083 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2085 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2087 $output .= $h{'Number'} . $size . $h{'Standard'};
2088 $output .= $line_end;
2093 sub print_non_clf_message ($) {
2095 my $content = shift;
2096 my $msec_string = "." . $req{$t}{'msecs'} unless $no_msecs_mode;
2097 my $line_start = $html_output_mode ? '' : $h{"Standard"};
2099 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2102 . $time_colours[$time_colour_index % 2]
2103 . $req{$t}{'time-stamp'}
2105 . $h{Standard} . " "
2106 . $thread_colours{$t}
2110 . $h{$req{$t}{'log-level'}}
2111 . $req{$t}{'log-level'}
2118 sub shorten_thread_id ($) {
2120 my $thread_id = shift;
2122 our %short_thread_ids;
2125 unless (defined $short_thread_ids{$thread_id}) {
2126 $short_thread_ids{$thread_id} = sprintf "%.3d", $max_threadid++;
2129 return $short_thread_ids{$thread_id}
2134 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2136 my $last_thread = 0;
2137 my $last_timestamp = 0;
2138 my $filters_that_did_nothing;
2141 $time_colour = paint_it('white');
2143 my %log_level_handlers = (
2144 'Re-Filter' => \&handle_loglevel_re_filter,
2145 'Header' => \&handle_loglevel_header,
2146 'Connect' => \&handle_loglevel_connect,
2147 'Redirect' => \&handle_loglevel_redirect,
2148 'Request' => \&handle_loglevel_request,
2149 'Crunch' => \&handle_loglevel_crunch,
2150 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2151 'Info' => \&handle_loglevel_info,
2152 'CGI' => \&handle_loglevel_cgi,
2153 'Force' => \&handle_loglevel_force,
2154 'Error' => \&handle_loglevel_error,
2155 'Fatal error' => \&handle_loglevel_ignore,
2156 'Writing' => \&handle_loglevel_ignore,
2157 'Received' => \&handle_loglevel_ignore,
2158 'Unknown log level' => \&handle_loglevel_ignore,
2163 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2164 $thread = $t = ($shorten_thread_ids) ? shorten_thread_id($4) : $4;
2165 $req{$t}{'day'} = $day = $1;
2166 $req{$t}{'time-stamp'} = $time_stamp = $2;
2167 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2168 $req{$t}{'log-level'} = $log_level = $5;
2169 $req{$t}{'content'} = $content = $c = $6;
2170 $req{$t}{'log-message'} = $_;
2171 $no_special_header_highlighting = 0;
2173 if (defined($log_level_handlers{$log_level})) {
2175 $content = $log_level_handlers{$log_level}($content);
2179 die "No handler found for log level \"$log_level\"\n";
2182 # Highlight Truncations
2183 if (length($_) > 4000) {
2184 $content =~ s@(too long, truncated)]$@$h{'Truncation'}$1$h{'Standard'}]@g;
2187 next unless $content;
2189 # Register threads to keep the colour constant
2190 if (!defined($thread_colours{$thread})) {
2191 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2192 $thread_colour_index++;
2195 # Switch timestamp colour if timestamps differ
2196 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2197 debug_message("Tick tack!") if DEBUG_TICKS;
2198 $time_colour = $time_colours[$time_colour_index % 2];
2199 $time_colour_index++;
2200 $last_msecs = $msecs;
2201 $last_timestamp = $time_stamp;
2204 $last_thread = $thread;
2206 print_non_clf_message($content);
2208 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2210 # LOG_LEVEL_CLF lines look like this
2211 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2212 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2213 # HTTP/1.1" 403 1730
2214 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2216 print_clf_message();
2220 # Some Privoxy log messages span more than one line,
2221 # usually to dump lots of content that doesn't need any syntax highlighting.
2222 # XXX: add mechanism to forward these lines to the right handler anyway.
2224 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2225 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2233 my ($day, $time_stamp, $msecs, $thread, $log_level, $content);
2234 my %log_level_handlers = (
2235 'Re-Filter' => \&handle_loglevel_ignore,
2236 'Header' => \&gather_loglevel_header_stats,
2237 'Connect' => \&gather_loglevel_connect_stats,
2238 'Redirect' => \&handle_loglevel_ignore,
2239 'Request' => \&gather_loglevel_request_stats,
2240 'Crunch' => \&gather_loglevel_crunch_stats,
2241 'Gif-Deanimate' => \&handle_loglevel_ignore,
2242 'Info' => \&handle_loglevel_ignore,
2243 'CGI' => \&handle_loglevel_ignore,
2244 'Force' => \&handle_loglevel_ignore,
2245 'Error' => \&gather_loglevel_error_stats,
2246 'Fatal error' => \&handle_loglevel_ignore,
2247 'Writing' => \&handle_loglevel_ignore,
2248 'Received' => \&handle_loglevel_ignore,
2249 'Unknown log level' => \&handle_loglevel_ignore
2253 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2256 $msecs = $3 ? $3 : 0;
2261 if (defined($log_level_handlers{$log_level})) {
2263 $content = $log_level_handlers{$log_level}($content, $thread);
2267 die "No handler found for log level \"$log_level\"\n";
2277 sub VersionMessage {
2278 my $version_message;
2280 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2281 $version_message .= 'Copyright (C) 2007-2010 Fabian Keil <fk@fabiankeil.de>' . "\n";
2282 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2284 print $version_message;
2287 sub get_cli_options () {
2289 our %cli_options = (
2290 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2291 'title' => CLI_OPTION_TITLE,
2292 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2293 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2294 'no-msecs' => CLI_OPTION_NO_MSECS,
2295 'shorten-thread-ids' => CLI_OPTION_SHORTEN_THREAD_IDS,
2296 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2297 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
2298 'statistics' => CLI_OPTION_STATISTICS,
2299 'url-statistics-threshold' => CLI_OPTION_URL_STATISTICS_THRESHOLD,
2300 'host-statistics-threshold'=> CLI_OPTION_HOST_STATISTICS_THRESHOLD,
2304 'html-output' => \$cli_options{'html-output'},
2305 'title' => \$cli_options{'title'},
2306 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2307 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2308 'no-msecs' => \$cli_options{'no-msecs'},
2309 'shorten-thread-ids' => \$cli_options{'shorten-thread-ids'},
2310 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2311 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
2312 'statistics' => \$cli_options{'statistics'},
2313 'url-statistics-threshold=s'=> \$cli_options{'url-statistics-threshold'},
2314 'host-statistics-threshold=s'=> \$cli_options{'host-statistics-threshold'},
2315 'version' => sub { VersionMessage && exit(0) },
2319 $html_output_mode = cli_option_is_set('html-output');
2320 $no_msecs_mode = cli_option_is_set('no-msecs');
2321 $shorten_thread_ids = cli_option_is_set('shorten-thread-ids');
2322 $line_end = get_line_end();
2333 Options and their default values if they have any:
2334 [--accept-unknown-messages]
2335 [--host-statistics-threshold $cli_options{'host-statistics-threshold'}]
2339 [--no-syntax-highlighting]
2340 [--shorten-thread-ids]
2341 [--show-ineffective-filters]
2343 [--url-statistics-threshold $cli_options{'url-statistics-threshold'}]
2344 [--title $cli_options{'title'}]
2346 see "perldoc $0" for more information
2352 ################################################################################
2354 ################################################################################
2358 set_background(DEFAULT_BACKGROUND);
2359 prepare_our_stuff();
2363 if (cli_option_is_set('statistics')) {
2376 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2380 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
2381 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--statistics>]
2382 [B<--shorten-thread-ids>] [B<--show-ineffective-filters>]
2383 [B<--url-statistics-threshold>] [B<--version>]
2387 B<privoxy-log-parser> reads Privoxy log messages and
2389 - syntax-highlights recognized lines,
2391 - reformats some of them for easier comprehension,
2393 - filters out less useful messages, and
2395 - (in some cases) calculates additional information,
2396 like the compression ratio or how a filter affected
2399 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2400 without getting confused by the resulting amount of output. For example for
2401 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2402 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2403 will hide the "filter foo caused 0 hits" message.
2407 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
2408 just don't highlight them.
2410 [B<--host-statistics-threshold>] Only show the request count for a host
2411 if it's above or equal to the given threshold. If the threshold is 0, host
2412 statistics are disabled.
2414 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2415 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2416 This option is only intended to make embedding log excerpts in web pages easier.
2417 It does not escape any input!
2419 [B<--no-msecs>] Don't expect milisecond resolution
2421 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2422 the filtered output is piped into less in which case the ANSI control
2423 codes don't work, or if the terminal itself doesn't support the control
2426 [B<--shorten-thread-ids>] Shorten the thread ids to a three-digit decimal number.
2427 Note that the mapping from thread ids to shortened ids is created at run-time
2428 and thus varies with the input.
2430 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2431 that didn't modify the content.
2433 [B<--statistics>] Gather various statistics instead of syntax highlighting
2434 log messages. This is an experimental feature, if the results look wrong
2435 they very well might be. Also note that the results are pretty much guaranteed
2436 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2438 [B<--url-statistics-threshold>] Only show the request count for a ressource
2439 if it's above or equal to the given threshold. If the threshold is 0, URL
2440 statistics are disabled.
2442 [B<--version>] Print version and exit.
2446 To monitor a log file:
2448 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2450 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2451 or if the log won't get rotated anyway. The log file location depends
2452 on your system (Doh!).
2454 To monitor Privoxy without having it write to a log file:
2456 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2458 Again, the config file location depends on your system. Output redirection
2459 depends on your shell, the above works with bourne shells.
2461 To read a processed Privoxy log file from top to bottom, letting the content
2462 scroll by slightly faster than you can read:
2464 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2466 This is probably only useful to fill screens in the background of haxor movies.
2470 Syntax highlighting with ANSI escape sequences will look strange
2471 if your background color isn't black.
2473 Some messages aren't recognized yet and will not be fully highlighted.
2475 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2476 using earlier Privoxy versions will probably result in an increased amount
2477 of unrecognized log lines.
2479 Privoxy's log files tend to be rather large. If you use HTML
2480 highlighting some browsers can't handle them, get confused and
2481 will eventually crash because of segmentation faults or unexpected
2482 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2487 Many settings can't be controlled through command line options yet.
2495 Fabian Keil <fk@fabiankeil.de>