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 1.120 2011/06/29 20:29:12 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',
142 # ----------------------
145 request_ => 'brown', # host+path but no protocol
146 'ip-address' => 'yellow',
149 Truncation => 'light_red',
151 Timestamp => 'brown',
152 Crunching => 'light_red',
153 crunched => 'light_red',
154 'Request-Line' => 'pink',
156 destination => 'yellow',
157 'http-version' => 'pink',
158 'crunch-pattern' => 'pink',
163 'program-name' => 'cyan',
166 warning => 'light_red',
167 debug => 'light_red',
171 'status-message' => 'light_cyan',
172 'status-code' => 'yellow',
173 'invalid-request' => 'light_red',
175 error => 'light_red',
176 'rewritten-URL' => 'light_red',
177 'pcrs-delimiter' => 'light_red',
178 'ignored' => 'light_red',
179 'action-bits-update' => 'light_red',
180 'configuration-line' => 'red',
181 'content-type' => 'yellow',
182 'HOST' => HEADER_DEFAULT_COLOUR,
187 # Header colours need their own hash so the keys can be accessed properly
189 # Prefilled with headers that should not appear with default header colours
190 Cookie => 'light_red',
191 'Set-Cookie' => 'light_red',
192 Warning => 'light_red',
193 Default => HEADER_DEFAULT_COLOUR,
196 # Crunch reasons need their own hash as well
198 'Unsupported HTTP feature' => 'light_red',
199 Blocked => 'light_red',
200 Untrusted => 'light_red',
201 Redirected => 'green',
202 'CGI Call' => 'white',
203 'DNS failure' => 'red',
204 'Forwarding failed' => 'light_red',
205 'Connection failure' => 'light_red',
206 'Out of memory (may mask other reasons)' => 'light_red',
207 'No reason recorded' => 'light_red',
210 @time_colours = ('white', 'light_gray');
212 # Translate highlight strings into highlight code
213 prepare_highlight_hash(\%header_colours);
214 prepare_highlight_hash(\%reason_colours);
215 prepare_highlight_hash(\%h);
216 prepare_colour_array(\@all_colours);
217 prepare_colour_array(\@time_colours);
224 ###############################################################
225 # Takes a colour string and returns an ANSI escape sequence
226 # (unless --no-syntax-highlighting is used).
227 # XXX: The Rolling Stones reference has to go.
228 ###############################################################
230 my $colour = shift @_;
232 return "" if cli_option_is_set('no-syntax-highlighting');
274 my $bg_code = get_background();
276 our $default = default_colours();
278 if (defined($text{$colour})) {
279 $colour_code = ESCAPE;
280 $colour_code .= $text{$colour};
282 $colour_code .= $light{$colour} ? "1" : "2";
284 $colour_code .= $bg_code;
286 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
288 } elsif ($colour =~ /reset/) {
290 $colour_code = default_colours();
294 die "What's $colour supposed to mean?\n";
300 sub get_semantic_html_markup ($) {
301 ###############################################################
302 # Takes a string and returns a span element
303 ###############################################################
308 if ($type =~ /Standard/) {
312 $code = '<span title="' . $type . '" class="' . $type . '">';
318 sub cli_option_is_set ($) {
321 my $cli_option = shift;
323 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
325 return $cli_options{$cli_option};
328 sub get_html_title () {
331 return $cli_options{'title'};
335 sub init_css_colours() {
343 purple => "F06", # XXX: wrong
344 cyan => "F09", # XXX: wrong
349 light_green => "33F",
358 sub get_css_colour ($) {
363 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
365 return '#' . $css_colours{$colour};
368 sub get_css_line ($) {
373 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
374 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
375 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
376 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
377 $css_line .= '}' . "\n";
382 sub get_css_line_for_colour ($) {
387 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
388 $css_line .= 'color:' . get_css_colour($colour) . ';';
389 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
390 $css_line .= '}' . "\n";
395 # XXX: Wrong solution
396 sub get_missing_css_lines () {
400 $css_line .= '.' . 'default' . ' {';
401 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
402 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
403 $css_line .= '}' . "\n";
410 our %css_colours; #XXX: Wrong solution
414 $css .= '.privoxy-log {';
415 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
416 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
419 foreach my $key (keys %h_colours) {
421 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
422 $css .= get_css_line($key);
426 foreach my $colour (keys %css_colours) {
428 $css .= get_css_line_for_colour($colour);
432 $css .= get_missing_css_lines(); #XXX: Wrong solution
441 if (cli_option_is_set('html-output')) {
443 my $title = get_html_title();
445 $intro .= '<html><head>';
446 $intro .= '<title>' . $title . '</title>';
447 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
448 $intro .= '</head><body>';
449 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
459 if (cli_option_is_set('html-output')) {
461 $outro = '</p></body></html>';
467 sub get_line_end () {
468 return cli_option_is_set('html-output') ? "<br>\n" : "\n";
471 sub get_colour_html_markup ($) {
472 ###############################################################
473 # Takes a colour string a span element. XXX: WHAT?
474 # XXX: This function shouldn't be necessary, the
475 # markup should always be semantically correct.
476 ###############################################################
481 if ($type =~ /Standard/) {
484 $code = '<span class="' . lc($type) . '">';
490 sub default_colours () {
493 return reset_colours();
496 sub show_colours () {
500 sub reset_colours () {
501 return ESCAPE . "0m";
504 sub set_background ($){
520 if (defined($backgrounds{$colour})) {
521 $bg_code = $backgrounds{$colour};
523 die "Invalid background colour: " . $colour;
527 sub get_background (){
531 sub prepare_highlight_hash ($) {
534 foreach my $key (keys %$ref) {
535 $$ref{$key} = $html_output_mode ?
536 get_semantic_html_markup($key) :
537 paint_it($$ref{$key});
541 sub prepare_colour_array ($) {
544 foreach my $i (0 ... @$ref - 1) {
545 $$ref[$i] = $html_output_mode ?
546 get_colour_html_markup($$ref[$i]) :
551 sub found_unknown_content ($) {
556 return if cli_option_is_set('accept-unknown-messages');
558 return if ($unknown =~ /\[too long, truncated\]$/);
560 $message = "found_unknown_content: Don't know how to highlight: ";
561 # Break line so the log file can later be parsed as Privoxy log file again
562 $message .= '"' . $unknown . '"' . " in:\n";
563 $message .= $req{$t}{'log-message'};
564 debug_message($message);
565 log_parse_error($req{$t}{'log-message'});
567 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
570 sub log_parse_error ($) {
574 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
575 open(my $errorlog_fd, ">>" . ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
576 print $errorlog_fd $message;
581 sub debug_message (@) {
584 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
587 ################################################################################
588 # highlighter functions that aren't loglevel-specific
589 ################################################################################
593 # Get highlight marker
594 my $highlight = shift; # XXX: Stupid name;
598 if (defined($highlight)) {
600 $result = $h{$highlight};
604 $message = "h: Don't recognize highlighter $highlight.";
605 debug_message($message);
606 log_parser_error($message);
607 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
613 sub highlight_known_headers ($) {
617 debug_message("Searching $content for things to highlight.") if DEBUG_HEADER_HIGHLIGHTING;
619 if ($content =~ m/(?<=\s)($header_highlight_regex):/) {
621 $content =~ s@(?<=[\s|'])($header)(?=:)@$header_colours{$header}$1$h{'Standard'}@ig;
622 debug_message("Highlighted '$header' in '$content'") if DEBUG_HEADER_HIGHLIGHTING;
628 sub highlight_matched_request_line ($$) {
630 my $result = shift; # XXX: Stupid name;
632 if ($result =~ m@(.*)($regex)(.*)@) {
633 $result = $1 . highlight_request_line($2) . $3
638 sub highlight_request_line ($) {
641 my ($method, $url, $http_version);
643 #GET http://images.sourceforge.net/sfx/icon_warning.gif HTTP/1.1
644 if ($rl =~ m/Invalid request/) {
646 $rl = h('invalid-request') . $rl . h('Standard');
648 } elsif ($rl =~ m/^([-\w]+) (.*) (HTTP\/\d\.\d)/) {
650 # XXX: might not match in case of HTTP method fuzzing.
651 # XXX: save these: ($method, $path, $http_version) = ($1, $2, $3);
652 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
653 if ($rl =~ /http:\/\//) {
654 $rl = highlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
656 $rl = highlight_matched_pattern($rl, 'request_', '[^\s]*(?=\sHTTP)');
659 $rl =~ s@(HTTP\/\d\.\d)$@$h{'http-version'}$1$h{'Standard'}@;
661 } elsif ($rl =~ m/\.\.\. \[too long, truncated\]$/) {
663 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
664 $rl = highlight_matched_url($rl, '[^\s]*(?=\.\.\.)');
666 } elsif ($rl =~ m/^ $/) {
668 $rl = h('error') . "No request line specified!" . h('Standard');
672 debug_message ("Can't parse request line: $rl");
679 sub highlight_response_line ($) {
682 my ($http_version, $status_code, $status_message);
687 # TODO: Mark different status codes differently
689 if ($rl =~ m/((?:HTTP\/\d\.\d|ICY)) (\d+) (.*)/) {
690 ($http_version, $status_code, $status_message) = ($1, $2, $3);
692 debug_message ("Can't parse response line: $rl") and die 'Fix this';
695 # Rebuild highlighted
697 $rl .= h('http-version') . $http_version . h('Standard');
699 $rl .= h('status-code') . $status_code . h('Standard');
701 $rl .= h('status-message') . $status_message . h('Standard');
706 sub highlight_matched_url ($$) {
708 my $result = shift; # XXX: Stupid name;
711 #print "Got $result, regex ($regex)\n";
713 if ($result =~ m@(.*?)($regex)(.*)@) {
714 $result = $1 . highlight_url($2) . $3;
715 #print "Now the result is $result\n";
721 sub highlight_matched_host ($$) {
723 my ($result, $regex) = @_; # XXX: result ist stupid name;
725 if ($result =~ m@(.*?)($regex)(.*)@) {
726 $result = $1 . $h{host} . $2 . $h{Standard} . $3;
732 sub highlight_matched_pattern ($$$) {
734 my $result = shift; # XXX: Stupid name;
738 die "Unknown key $key" unless defined $h{$key};
740 if ($result =~ m@(.*?)($regex)(.*)@) {
741 $result = $1 . h($key) . $2 . h('Standard') . $3;
747 sub highlight_matched_path ($$) {
749 my $result = shift; # XXX: Stupid name;
752 if ($result =~ m@(.*?)($regex)(.*)@) {
753 $result = $1 . h('path') . $2 . h('Standard') . $3;
759 sub highlight_url ($) {
763 if ($html_output_mode) {
765 $url = '<a href="' . $url . '">' . $url . '</a>';
769 $url = h('URL') . $url . h('Standard');
776 sub update_header_highlight_regex ($) {
779 my $headers = join ('|', keys %header_colours);
781 $header_highlight_regex = qr/$headers/;
782 print "Registering '$header'\n" if DEBUG_HEADER_HIGHLIGHTING;
785 ################################################################################
786 # loglevel-specific highlighter functions
787 ################################################################################
789 sub handle_loglevel_header ($) {
793 if ($c =~ /^scan:/) {
795 if ($c =~ m/^scan: ([^: ]+):/) {
797 # Register new headers
798 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
800 if (!defined($header_colours{$header}) and $header =~ /^[\d\w-]*$/) {
801 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
803 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
804 $header_colours{$header} = $header_colours{'Default'};
806 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
807 $header_colour_index++;
809 update_header_highlight_regex($header);
812 } elsif ($c =~ m/^(scan: )(\w+ .+ HTTP\/\d\.\d)/) {
814 # scan: GET http://p.p/ HTTP/1.1
815 $c = $1 . highlight_request_line($2);
817 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
819 # scan: HTTP/1.1 200 OK
820 $req{$t}{'response_line'} = $2;
821 $req{$t}{'status_code'} = $3;
822 $req{$t}{'status_message'} = $4;
823 $c = $1 . highlight_response_line($req{$t}{'response_line'});
826 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
828 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
829 $c =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
830 $c =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
832 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
834 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/!
835 $c = highlight_matched_host($c, '(?<=New host is: )[^\s]+(?=\.)');
836 $c = highlight_matched_url($c, '(?<=Crunching Referer: )[^\s!]+');
838 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
840 # Text mode enabled by force. Take cover!
841 $c =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
843 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
845 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
846 $c = $1 . highlight_request_line($2);
848 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
850 # Adjusted Content-Length to 2132
851 # Adjust Content-Length to 33533
852 $c =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
853 $c = highlight_known_headers($c);
855 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
857 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
858 $c = highlight_matched_url($c, '(?<=New request URL: ).*');
860 } elsif ($c =~ m/^Couldn\'t parse:/) {
862 # XXX: These should probable be logged with LOG_LEVEL_ERROR
863 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
864 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
865 $c =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
867 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
868 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
870 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
871 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
872 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
874 # XXX: Save tag and tagger
876 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
877 $c =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
878 $c =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
879 $no_special_header_highlighting = 1;
881 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
883 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
884 # XXX: Save tag and tagger
886 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
887 $c =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
889 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
890 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
891 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
892 or $c =~ m/^(addh-unique|Referer forged to)/
893 or $c =~ m/^Downgraded answer to HTTP\/1.0/
894 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
895 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
896 or $c =~ m/^Referer crunched!/
897 or $c =~ m/^crunched x-forwarded-for!/
898 or $c =~ m/^crunched From!/
899 or $c =~ m/^ modified$/
900 or $c =~ m/^Content filtering is enabled. Crunching:/
901 or $c =~ m/^force-text-mode overruled the client/
902 or $c =~ m/^Server time in the future\./
903 or $c =~ m/^content-disposition header crunched and replaced with:/i
904 or $c =~ m/^Reducing white space in /
905 or $c =~ m/^Ignoring single quote in /
906 or $c =~ m/^Converting tab to space in /
907 or $c =~ m/A HTTP\/1\.1 response without/
908 or $c =~ m/Disabled filter mode on behalf of the client/
909 or $c =~ m/Keeping the (?:server|client) header /
910 or $c =~ m/Content modified with no Content-Length header set/
911 or $c =~ m/^Appended client IP address to/
912 or $c =~ m/^Removing 'Connection: close' to imply keep-alive./
913 or $c =~ m/^keep-alive support is disabled/
914 or $c =~ m/^Continue hack in da house/
915 or $c =~ m/^Merged multiple header lines to:/
916 or $c =~ m/^Added header: /
919 # XXX: Some of these may need highlighting
921 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
922 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
923 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
924 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
925 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
926 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
927 # Text mode is already enabled
928 # Denied request with NULL byte(s) turned into line break(s)
929 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
930 # addh-unique: Host: people.freebsd.org
931 # Suppressed offer to compress content
932 # Crunched incoming cookie -- yum!
933 # Accepted the client's request to fetch without filtering.
934 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
935 # addh-unique: Host: subkeys.pgp.net:11371
936 # Referer forged to: Referer: http://10.0.0.1/
937 # Downgraded answer to HTTP/1.0
938 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
939 # Referer overwritten with: Referer: pille-palle
940 # Referer replaced with: Referer: pille-palle
941 # crunched x-forwarded-for!
943 # modified # XXX: pretty stupid log message
944 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
945 # force-text-mode overruled the client's request to fetch without filtering!
946 # Server time in the future.
947 # content-disposition header crunched and replaced with: content-disposition: filename=baz
948 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
949 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
950 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
951 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
953 # A HTTP/1.1 response without Connection header implies keep-alive.
954 # Disabled filter mode on behalf of the client.
955 # Keeping the server header 'Connection: keep-alive' around.
956 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
957 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
958 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
959 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
960 # Removing 'Connection: close' to imply keep-alive.
961 # keep-alive support is disabled. Crunching: Keep-Alive: 300.
962 # Continue hack in da house.
963 # Merged multiple header lines to: 'X-FORWARDED-PROTO: http X-HOST: 127.0.0.1'
964 # Added header: Content-Encoding: deflate
966 } elsif ($c =~ m/^scanning headers for:/) {
968 return '' unless SHOW_SCAN_INTRO;
970 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
971 # crunched User-Agent!
972 # Crunching: Content-Encoding: gzip
974 $c =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
976 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
978 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
980 $c = h('warning') . $c . h('Standard');
982 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
984 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
985 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
987 $c =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
988 $c =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
990 } elsif ($c =~ m/^Removing empty header/) {
992 # Removing empty header
995 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
997 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
998 # Enable force-text-mode if you know what you're doing.
999 # XXX: Could highlight more here.
1000 $c =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
1002 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
1004 # Server keep-alive timeout is 5. Sticking with 10.
1005 # Client keep-alive timeout is 20. Sticking with 10.
1007 $c =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1008 $c =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1010 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1012 # Reducing keep-alive timeout from 60 to 10.
1014 $c =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1015 $c =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1017 } elsif ($c =~ m/^Killed all-caps Host header line: HOST:/) {
1019 # Killed all-caps Host header line: HOST: bestproxydb.com
1020 $c = highlight_matched_host($c, '(?<=HOST: )[^\s]+');
1021 $c = highlight_matched_pattern($c, 'HOST', 'HOST');
1025 found_unknown_content($c);
1029 unless ($c =~ m/^Transforming/) {
1030 $c = highlight_known_headers($c) unless $no_special_header_highlighting;
1036 sub handle_loglevel_re_filter ($) {
1038 my $content = shift;
1042 if ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1044 # XXX: only the second version gets highlighted properly.
1045 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1046 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1047 $req{$t}{'content_source'} = $1;
1048 $req{$t}{'content_size'} = $2;
1049 $req{$t}{'content_filter'} = $3;
1050 $req{$t}{'content_hits'} = $4;
1051 $req{$t}{'new_content_size'} = $5;
1052 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1053 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1054 if ($req{$t}{'content_hits'} == 0 and
1055 not (cli_option_is_set('show-ineffective-filters')
1056 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1060 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1061 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1062 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1064 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1065 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1068 $c .= "(" . $h{'Number'};
1069 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1070 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1073 } elsif ($c =~ /\.{3}$/
1074 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1076 # Used by Privoxy 3.0.5 and 3.0.6:
1078 # Used by Privoxy 3.0.7:
1079 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1081 $req{$t}{'filtered_header'} = $1;
1082 $req{$t}{'old_header_size'} = $2;
1083 $req{$t}{'header_filter_name'} = $3;
1085 unless (cli_option_is_set('show-ineffective-filters') or
1086 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1089 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1090 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1092 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1094 # ...produced 0 hits (new size 23).
1095 #... produced 1 hits (new size 54).
1097 $req{$t}{'header_filter_hits'} = $1;
1098 $req{$t}{'new_header_size'} = $2;
1100 unless (cli_option_is_set('show-ineffective-filters') or
1101 (defined($req{$t}{'header_filter_name'}) and
1102 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1104 if ($req{$t}{'header_filter_hits'} == 0 and
1105 not (defined($req{$t}{'header_filter_name'}) and
1106 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1109 # Reformat including information from the intro
1110 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1112 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1113 # $c .= $req{$t}{'filtered_header'} . " ";
1114 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1115 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1117 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1119 $c .= "changing size from ";
1120 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1122 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1127 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1131 # Highlight from last line (XXX: What?)
1132 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1133 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1138 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1139 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1144 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1146 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1147 # Tagger variable-test has empty joblist. Nothing to do.
1149 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1151 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1153 $req{$t}{'chunked-size'} = $1;
1154 $req{$t}{'dechunked-size'} = $2;
1155 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1157 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1159 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1160 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1162 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1164 # Decompression successful. Old size: 670, new size: 1166.
1166 $req{$t}{'size-compressed'} = $1;
1167 $req{$t}{'size-decompressed'} = $2;
1168 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1170 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1171 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1173 # XXX: Create sub get_percentage()
1174 if ($req{$t}{'size-decompressed'}) {
1175 $req{$t}{'decompression-gain-percent'} =
1176 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1178 $content .= " (saved: ";
1179 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1181 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1185 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1187 # Need to de-chunk first
1188 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1190 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1192 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1194 # Adding re_filter job ...
1195 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1196 # to filter client-ip-address succeeded.
1198 } elsif ($c =~ m/^Compressed content from /) {
1200 # Compressed content from 29258 to 8630 bytes. Compression level: 3
1201 $content =~ s@(?<=from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1202 $content =~ s@(?<=to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1203 $content =~ s@(?<=level: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1205 } elsif ($c =~ m/^Reading in filter/) {
1207 return '' unless SHOW_FILTER_READIN_IN;
1211 found_unknown_content($content);
1218 sub handle_loglevel_redirect ($) {
1222 if ($c =~ m/^Decoding "([^""]*)"/) {
1224 $req{$t}{'original-destination'} = $1;
1225 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1228 } elsif ($c =~ m/^Checking/) {
1230 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1231 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1233 # TODO: Change colour if really url-decoded
1234 $req{$t}{'decoded-original-destination'} = $1;
1235 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1238 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1240 # pcrs command "s@&from=rss@@" changed \
1241 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1242 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1243 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1244 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1245 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1246 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1247 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1248 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1250 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1252 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1253 # change "http://www.example.org/".
1254 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1255 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1257 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1259 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1260 # XXX: Use URL highlighter
1262 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1264 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1265 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1266 # is already properly formatted.
1267 # XXX: assume the same?
1268 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1272 found_unknown_content($c);
1279 sub handle_loglevel_gif_deanimate ($) {
1281 my $content = shift;
1283 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1285 my $bytes_from = $1;
1287 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1288 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1289 # XXX: Do we need g in case of ($1 == $2)?
1290 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1292 } elsif ($content =~ m/GIF (not) changed/) {
1294 # Gif-Deanimate: GIF not changed.
1295 return '' if SUPPRESS_GIF_NOT_CHANGED;
1296 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1298 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1300 # failed! (gif parsing)
1301 # XXX: Replace this error message with something less stupid
1302 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1304 } elsif ($content =~ m/^Need to de-chunk first/) {
1306 # Need to de-chunk first
1307 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1309 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1311 # No GIF header found (XXX: Did I ever commit this?)
1312 # failed while parsing 195 134747048 (XXX: never committed)
1314 # Ignore these for now
1318 found_unknown_content($content);
1325 sub handle_loglevel_request ($) {
1327 my $content = shift;
1329 if ($content =~ m/crunch! /) {
1331 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1333 # Highlight crunch reasons
1334 foreach my $reason (keys %reason_colours) {
1335 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1337 # Highlight request URL domain and ditch 'crunch!'
1338 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1339 $content =~ s@ crunch!@@;
1341 } elsif ($content =~ m/\[too long, truncated\]$/) {
1343 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1344 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1346 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1348 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1349 $content = h('request_') . $content . h('Standard');
1353 found_unknown_content($content);
1360 sub handle_loglevel_crunch ($) {
1362 my $content = shift;
1364 # Highlight crunch reason
1365 foreach my $reason (keys %reason_colours) {
1366 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1369 if ($content =~ m/\[too long, truncated\]$/) {
1371 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1372 # [...]&filter... [too long, truncated]
1373 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1377 # Blocked: http://ads.example.org/
1378 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1384 sub handle_loglevel_connect ($) {
1388 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1390 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1392 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1393 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1395 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1397 # connect to: www.example.org.noconnect failed: Operation not permitted
1399 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1401 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1403 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1405 # Connect: to www.nzherald.co.nz successful
1406 # Connect: to archiv.radiotux.de
1408 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1409 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1411 } elsif ($c =~ m/^to ([^\s]*)$/) {
1413 # Connect: to lists.sourceforge.net:443
1415 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1417 } elsif ($c =~ m/^accepted connection from .*/ or
1420 # accepted connection from 10.0.0.1
1421 # Privoxy 3.0.6 and earlier just say:
1423 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1424 $c = highlight_matched_host($c, '(?<=connection from ).*');
1426 } elsif ($c =~ m/^write header to: .* failed:/) {
1428 # write header to: 10.0.0.1 failed: Broken pipe
1430 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1431 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1433 } elsif ($c =~ m/^write header to client failed:/) {
1435 # write header to client failed: Broken pipe
1437 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1439 } elsif ($c =~ m/^socks4_connect:/) {
1441 # socks4_connect: SOCKS request rejected or failed.
1442 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1444 } elsif ($c =~ m/^Listening for new connections/ or
1445 $c =~ m/^accept connection/) {
1447 # Privoxy versions above 3.0.6 say:
1448 # Listening for new connections ...
1449 # earlier versions say:
1450 # accept connection ...
1453 } elsif ($c =~ m/^accept failed:/) {
1455 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1457 } elsif ($c =~ m/^Overriding forwarding settings/) {
1459 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1460 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1462 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1464 # Denying suspicious CONNECT request from 10.0.0.1
1465 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1467 } elsif ($c =~ m/^socks5_connect:/) {
1469 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1471 } elsif ($c =~ m/^Created new connection to/) {
1473 # Created new connection to www.privoxy.org:80 on socket 11.
1474 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1475 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1477 } elsif ($c =~ m/^Found reusable socket/) {
1479 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1481 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1482 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1483 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1484 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1485 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1486 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1487 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1488 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1490 } elsif ($c =~ m/^Marking open socket/) {
1492 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1493 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1494 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1495 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1497 } elsif ($c =~ m/^No reusable/) {
1499 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1500 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1502 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1504 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1505 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1507 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1508 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1509 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1511 } elsif ($c =~ m/^Socket/) {
1513 # Socket 16 already forgotten or never remembered.
1514 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1516 } elsif ($c =~ m/^The connection to/) {
1518 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1520 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1521 # Timeout is: 1. Assumed latency: 4.
1522 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1523 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1524 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1525 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1526 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1527 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1529 } elsif ($c =~ m/^Stopped waiting for the request line./) {
1531 # Stopped waiting for the request line. Timeout: 121.
1532 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1534 } elsif ($c =~ m/^Waiting for \d/) {
1536 # Waiting for 1 connections to timeout.
1537 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1539 } elsif ($c =~ m/^Initialized/) {
1541 # Initialized 20 socket slots.
1542 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1544 } elsif ($c =~ m/^Done reading from server/) {
1546 # Done reading from server. Expected content length: 24892. \
1547 # Actual content length: 24892. Most recently received: 4412.
1549 # Done reading from server. Expected content length: 24892. \
1550 # Actual content length: 24892. Bytes most recently read: 4412.
1551 # Done reading from server. Content length: 6018 as expected. \
1552 # Bytes most recently read: 294.
1553 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1554 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1555 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1557 } elsif ($c =~ m/^Continuing buffering headers/) {
1559 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1560 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1561 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1562 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1564 # Continuing buffering headers. Bytes most recently read: %d.
1565 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1567 } elsif ($c =~ m/^Received \d+ bytes while/) {
1569 # Received 206 bytes while expecting 12103.
1570 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1571 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1573 } elsif ($c =~ m/^(Rejecting c|C)onnection from/) {
1575 # Connection from 81.163.28.218 dropped due to ACL
1576 # Rejecting connection from 178.63.152.227. Maximum number of connections reached.
1577 $c =~ s@(?<=onnection from )((?:\d+\.?){3}\d+)@$h{'Number'}$1$h{'Standard'}@;
1579 } elsif ($c =~ m/^(?:Reusing|Closing) server socket \d./ or
1580 $c =~ m/^No additional client request/) {
1582 # Reusing server socket 4. Opened for 10.0.0.1.
1583 # Closing server socket 2. Opened for 10.0.0.1.
1584 # No additional client request received in time. \
1585 # Closing server socket 4, initially opened for 10.0.0.1.
1586 # No additional client request received in time on socket 29.
1588 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1589 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.$)');
1591 } elsif ($c =~ m/^Connected to /) {
1593 # Connected to tor-jail[10.0.0.2]:9050.
1595 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1596 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1597 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1599 } elsif ($c =~ m/^Could not connect to /) {
1601 # Could not connect to [10.0.0.1]:80.
1603 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1604 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1606 } elsif ($c =~ m/^Waiting for the next client request/ or
1607 $c =~ m/^The connection on server socket/ or
1608 $c =~ m/^Client request arrived in time /) {
1610 # Waiting for the next client request on socket 3. Keeping the server \
1611 # socket 12 to a.fsdn.com open.
1612 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1613 # Used by Privoxy 3.0.18 and later:
1614 # Client request arrived in time on socket 21.
1615 # Used by earlier version:
1616 # Client request arrived in time or the client closed the connection on socket 12.
1618 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1619 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1620 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1622 } elsif ($c =~ m/^Marking the server socket/) {
1624 # Marking the server socket 7 tainted.
1626 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1628 } elsif ($c =~ m/^Reduced expected bytes to /) {
1630 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1631 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1632 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1634 } elsif ($c =~ m/^The client closed socket /) {
1636 # The client closed socket 2 while the server socket 4 is still open.
1637 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1638 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1640 } elsif ($c =~ m/^Expected client content length set /) {
1642 # Expected client content length set to 667325411 after reading 4999 bytes.
1643 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1644 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1646 } elsif ($c =~ m/^Reducing expected bytes to /) {
1648 # Reducing expected bytes to 0. Marking the server socket tainted after throwing 4 bytes away.
1649 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1650 $c =~ s@(?<=after throwing )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1652 } elsif ($c =~ m/^Waiting for up to /) {
1654 # Waiting for up to 4999 bytes from the client.
1655 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1657 } elsif ($c =~ m/^Looks like we / or
1658 $c =~ m/^Unsetting keep-alive flag/ or
1659 $c =~ m/^No connections to wait/ or
1660 $c =~ m/^Complete client request received/ or
1661 $c =~ m/^Possible pipeline attempt detected./ or
1662 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1663 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1664 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1665 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1666 $c =~ m/^Stopping to watch the client socket. There's already another request waiting./ or
1667 $c =~ m/^Done reading from the client\.$/) {
1669 # Looks like we reached the end of the last chunk. We better stop reading.
1670 # Looks like we read the end of the last chunk together with the server \
1671 # headers. We better stop reading.
1672 # Looks like we got the last chunk together with the server headers. \
1673 # We better stop reading.
1674 # Unsetting keep-alive flag.
1675 # No connections to wait for left.
1676 # Client request arrived in time or the client closed the connection.
1677 # Complete client request received
1678 # Possible pipeline attempt detected. The connection will not be \
1679 # kept alive and we will only serve the first request.
1680 # POST request detected. The connection will not be kept alive.
1681 # The server still wants to talk, but the client hung up on us.
1682 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1683 # There might be a request body. The connection will not be kept alive.
1684 # Stopping to watch the client socket. There's already another request waiting.
1685 # Done reading from the client\.
1689 found_unknown_content($c);
1697 sub handle_loglevel_info ($) {
1701 if ($c =~ m/^Rewrite detected:/) {
1703 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1704 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1706 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1707 $c =~ m/^Compressed content detected/ or
1708 $c =~ m/^SDCH-compressed content detected/ or
1711 # Decompressing deflated iob: 117
1712 # Decompression didn't result in any content.
1713 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1714 # with zlib support or enable the prevent-compression action.
1715 # SDCH-compressed content detected, content filtering disabled.\
1716 # Consider suppressing SDCH offers made by the client.
1717 # Tagger 'complete-url' created empty tag. Ignored.
1721 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1723 # loading configuration file '/usr/local/etc/privoxy/config':
1724 # Reloading configuration file '/usr/local/etc/privoxy/config'
1725 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1727 } elsif ($c =~ m/^Loading (actions|filter) file: /) {
1729 # Loading actions file: /usr/local/etc/privoxy/default.action
1730 # Loading filter file: /usr/local/etc/privoxy/default.filter
1731 $c =~ s@(?<= file: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1733 } elsif ($c =~ m/^exiting by signal/) {
1735 # exiting by signal 15 .. bye
1736 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1738 } elsif ($c =~ m/^Privoxy version/) {
1740 # Privoxy version 3.0.7
1741 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1743 } elsif ($c =~ m/^Program name: /) {
1745 # Program name: /usr/local/sbin/privoxy
1746 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1748 } elsif ($c =~ m/^Listening on port /) {
1750 # Listening on port 8118 on IP address 10.0.0.1
1751 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1752 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1754 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1756 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1757 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1759 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1761 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1762 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1764 # Request from 10.0.0.1 marked for blocking. limit-connect{0} doesn't allow CONNECT requests to www.example.org:443
1765 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1767 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1768 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1769 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1770 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1771 $c =~ s@(?<=to )([^\s]+)@$h{'request_'}$1$h{'Standard'}@;
1773 } elsif ($c =~ m/^Status code/) {
1775 # Status code 304 implies no body.
1776 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1778 } elsif ($c =~ m/^Method/) {
1780 # Method HEAD implies no body.
1781 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1783 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1785 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1786 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1787 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1789 } elsif ($c =~ m/^File modification detected: /) {
1791 # File modification detected: /usr/local/etc/privoxy/user-agent.action
1792 $c =~ s@(?<= detected: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1794 } elsif ($c =~ m/^No logfile configured/ or
1795 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1796 $c =~ m/^Invalid \"chunked\" transfer/ or
1797 $c =~ m/^Support for/ or
1798 $c =~ m/^Flushing header and buffers/ or
1799 $c =~ m/^Can not resolve/
1802 # No logfile configured. Please enable it before reporting any problems.
1803 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1804 # response or even no response at all.
1805 # No logfile configured. Logging disabled.
1806 # Invalid "chunked" transfer encoding detected and ignored.
1807 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1808 # known not to work properly in some situations.
1809 # Flushing header and buffers. Stepping back from filtering.
1810 # Can not resolve doesnotexist: hostname nor servname provided, or not known
1814 found_unknown_content($c);
1821 sub handle_loglevel_cgi ($) {
1825 if ($c =~ m/^Granting access to/) {
1827 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1829 } elsif ($c =~ m/^Substituting: s(.)/) {
1831 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1832 # XXX: prone to span several lines
1835 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1836 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1837 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1843 sub handle_loglevel_force ($) {
1847 if ($c =~ m/^Ignored force prefix in request:/) {
1849 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1850 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1851 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1853 } elsif ($c =~ m/^Enforcing request:/) {
1855 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1856 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1860 found_unknown_content($c);
1867 sub handle_loglevel_error ($) {
1871 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1873 # Empty server or forwarder response received on socket 4.
1874 # Empty server or forwarder response received on socket 3. \
1875 # Closing client socket 15 without sending data.
1876 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1877 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1879 # XXX: There are probably more messages that deserve highlighting.
1885 sub handle_loglevel_ignore ($) {
1889 sub gather_loglevel_request_stats ($$) {
1897 sub gather_loglevel_crunch_stats ($$) {
1905 if ($c =~ m/^Redirected:/) {
1906 # Redirected: http://www.example.org/http://p.p/
1907 $stats{'fast-redirections'}++;
1909 } elsif ($c =~ m/^Blocked:/) {
1910 # Blocked: blogger.googleusercontent.com:443
1911 $stats{'blocked'}++;
1916 sub gather_loglevel_error_stats ($$) {
1923 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1925 # Empty server or forwarder response received on socket 4.
1926 $stats{'empty-responses'}++;
1927 if ($thread_data{$thread}{'new_connection'}) {
1928 $stats{'empty-responses-on-new-connections'}++;
1930 $stats{'empty-responses-on-reused-connections'}++;
1935 sub gather_loglevel_connect_stats ($$) {
1937 my ($c, $thread) = @_;
1941 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
1943 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1944 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
1946 } elsif ($c =~ m/^to ([^\s]*)$/) {
1948 # Connect: to lists.sourceforge.net:443
1950 $thread_data{$thread}{'forwarder'} = 'direct connection';
1952 } elsif ($c =~ m/^Created new connection to/) {
1954 # Created new connection to www.privoxy.org:80 on socket 11.
1956 $thread_data{$thread}{'new_connection'} = 1;
1958 } elsif ($c =~ m/^Reusing server socket \d./ or
1959 $c =~ m/^Found reusable socket/) {
1961 # Reusing server socket 4. Opened for 10.0.0.1.
1962 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1964 $thread_data{$thread}{'new_connection'} = 0;
1965 $stats{'reused-connections'}++;
1969 sub gather_loglevel_header_stats ($$) {
1971 my ($c, $thread) = @_;
1974 if ($c =~ m/^A HTTP\/1\.1 response without/ or
1975 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
1977 # A HTTP/1.1 response without Connection header implies keep-alive.
1978 # Keeping the server header 'Connection: keep-alive' around.
1979 $stats{'server-keep-alive'}++;
1981 } elsif ($c =~ m/^scan: ((\w+) (.+) (HTTP\/\d\.\d))/) {
1983 # scan: HTTP/1.1 200 OK
1984 $stats{'method'}{$2}++;
1985 $stats{'ressource'}{$3}++;
1986 $stats{'http-version'}{$4}++;
1988 } elsif ($c =~ m/^scan: Host: ([^\s]+)/) {
1991 $stats{'hosts'}{$1}++;
1999 'server-keep-alive' => 0,
2000 'reused-connections' => 0,
2001 'empty-responses' => 0,
2002 'empty-responses-on-new-connections' => 0,
2003 'empty-responses-on-reused-connections' => 0,
2004 'fast-redirections' => 0,
2006 'reused-connections' => 0,
2007 'server-keep-alive' => 0,
2011 sub get_percentage ($$) {
2014 return "NaN" if ($big eq 0);
2015 return sprintf("%.2f%%", $small / $big * 100);
2018 sub print_stats () {
2022 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
2023 my $outgoing_requests = $stats{requests} - $stats{crunches};
2025 if ($stats{requests} eq 0) {
2026 print "No requests yet.\n";
2030 print "Client requests total: " . $stats{requests} . "\n";
2031 print "Crunches: " . $stats{crunches} . " (" .
2032 get_percentage($stats{requests}, $stats{crunches}) . ")\n";
2033 print "Blocks: " . $stats{'blocked'} . " (" .
2034 get_percentage($stats{requests}, $stats{'blocked'}) . ")\n";
2035 print "Fast redirections: " . $stats{'fast-redirections'} . " (" .
2036 get_percentage($stats{requests}, $stats{'fast-redirections'}) . ")\n";
2037 print "Outgoing requests: " . $outgoing_requests . " (" .
2038 get_percentage($stats{requests}, $outgoing_requests) . ")\n";
2039 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
2040 get_percentage($stats{requests}, $stats{'server-keep-alive'}) . ")\n";
2041 print "New outgoing connections: " . $new_connections . " (" .
2042 get_percentage($stats{requests}, $new_connections) . ")\n";
2043 print "Reused connections: " . $stats{'reused-connections'} . " (" .
2044 get_percentage($stats{requests}, $stats{'reused-connections'}) .
2045 "; server offers accepted: " .
2046 get_percentage($stats{'server-keep-alive'}, $stats{'reused-connections'}) . ")\n";
2047 print "Empty responses: " . $stats{'empty-responses'} . " (" .
2048 get_percentage($stats{requests}, $stats{'empty-responses'}) . ")\n";
2049 print "Empty responses on new connections: "
2050 . $stats{'empty-responses-on-new-connections'} . " (" .
2051 get_percentage($stats{requests}, $stats{'empty-responses-on-new-connections'})
2053 print "Empty responses on reused connections: " .
2054 $stats{'empty-responses-on-reused-connections'} . " (" .
2055 get_percentage($stats{requests}, $stats{'empty-responses-on-reused-connections'}) .
2058 if ($stats{method} eq 0) {
2059 print "No response lines parsed yet yet.\n";
2062 print "Method distribution:\n";
2063 foreach my $method (sort {$stats{'method'}{$b} <=> $stats{'method'}{$a}} keys %{$stats{'method'}}) {
2064 printf "%8d : %-8s\n", $stats{'method'}{$method}, $method;
2066 print "Client HTTP versions:\n";
2067 foreach my $http_version (sort {$stats{'http-version'}{$b} <=> $stats{'http-version'}{$a}} keys %{$stats{'http-version'}}) {
2068 printf "%d : %s\n", $stats{'http-version'}{$http_version}, $http_version;
2071 if ($cli_options{'url-statistics-threshold'} == 0) {
2072 print "URL statistics are disabled. Increase --url-statistics-threshold to enable them.\n";
2074 print "Requested URLs:\n";
2075 foreach my $ressource (sort {$stats{'ressource'}{$b} <=> $stats{'ressource'}{$a}} keys %{$stats{'ressource'}}) {
2076 if ($stats{'ressource'}{$ressource} < $cli_options{'url-statistics-threshold'}) {
2077 print "Skipped statistics for URLs below the treshold.\n";
2080 printf "%d : %s\n", $stats{'ressource'}{$ressource}, $ressource;
2084 if ($cli_options{'host-statistics-threshold'} == 0) {
2085 print "Host statistics are disabled. Increase --host-statistics-threshold to enable them.\n";
2087 print "Requested Hosts:\n";
2088 foreach my $host (sort {$stats{'hosts'}{$b} <=> $stats{'hosts'}{$a}} keys %{$stats{'hosts'}}) {
2089 if ($stats{'hosts'}{$host} < $cli_options{'host-statistics-threshold'}) {
2090 print "Skipped statistics for Hosts below the treshold.\n";
2093 printf "%d : %s\n", $stats{'hosts'}{$host}, $host;
2099 ################################################################################
2100 # Functions that actually print stuff
2101 ################################################################################
2103 sub print_clf_message () {
2105 our ($ip, $timestamp, $request_line, $status_code, $size);
2108 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2110 # Rebuild highlighted
2111 $output .= $h{'Number'} . $ip . $h{'Standard'};
2113 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2115 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2117 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2119 $output .= $h{'Number'} . $size . $h{'Standard'};
2120 $output .= $line_end;
2125 sub print_non_clf_message ($) {
2127 my $content = shift;
2128 my $msec_string = "." . $req{$t}{'msecs'} unless $no_msecs_mode;
2129 my $line_start = $html_output_mode ? '' : $h{"Standard"};
2131 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2134 . $time_colours[$time_colour_index % 2]
2135 . $req{$t}{'time-stamp'}
2137 . $h{Standard} . " "
2138 . $thread_colours{$t}
2142 . $h{$req{$t}{'log-level'}}
2143 . $req{$t}{'log-level'}
2150 sub shorten_thread_id ($) {
2152 my $thread_id = shift;
2154 our %short_thread_ids;
2157 unless (defined $short_thread_ids{$thread_id}) {
2158 $short_thread_ids{$thread_id} = sprintf "%.3d", $max_threadid++;
2161 return $short_thread_ids{$thread_id}
2166 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2168 my $last_thread = 0;
2169 my $last_timestamp = 0;
2170 my $filters_that_did_nothing;
2173 $time_colour = paint_it('white');
2175 my %log_level_handlers = (
2176 'Re-Filter' => \&handle_loglevel_re_filter,
2177 'Header' => \&handle_loglevel_header,
2178 'Connect' => \&handle_loglevel_connect,
2179 'Redirect' => \&handle_loglevel_redirect,
2180 'Request' => \&handle_loglevel_request,
2181 'Crunch' => \&handle_loglevel_crunch,
2182 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2183 'Info' => \&handle_loglevel_info,
2184 'CGI' => \&handle_loglevel_cgi,
2185 'Force' => \&handle_loglevel_force,
2186 'Error' => \&handle_loglevel_error,
2187 'Fatal error' => \&handle_loglevel_ignore,
2188 'Writing' => \&handle_loglevel_ignore,
2189 'Received' => \&handle_loglevel_ignore,
2190 'Unknown log level' => \&handle_loglevel_ignore,
2195 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2196 $thread = $t = ($shorten_thread_ids) ? shorten_thread_id($4) : $4;
2197 $req{$t}{'day'} = $day = $1;
2198 $req{$t}{'time-stamp'} = $time_stamp = $2;
2199 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2200 $req{$t}{'log-level'} = $log_level = $5;
2201 $req{$t}{'content'} = $content = $c = $6;
2202 $req{$t}{'log-message'} = $_;
2203 $no_special_header_highlighting = 0;
2205 if (defined($log_level_handlers{$log_level})) {
2207 $content = $log_level_handlers{$log_level}($content);
2211 die "No handler found for log level \"$log_level\"\n";
2214 # Highlight Truncations
2215 if (length($_) > 4000) {
2216 $content =~ s@(too long, truncated)]$@$h{'Truncation'}$1$h{'Standard'}]@g;
2219 next unless $content;
2221 # Register threads to keep the colour constant
2222 if (!defined($thread_colours{$thread})) {
2223 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2224 $thread_colour_index++;
2227 # Switch timestamp colour if timestamps differ
2228 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2229 debug_message("Tick tack!") if DEBUG_TICKS;
2230 $time_colour = $time_colours[$time_colour_index % 2];
2231 $time_colour_index++;
2232 $last_msecs = $msecs;
2233 $last_timestamp = $time_stamp;
2236 $last_thread = $thread;
2238 print_non_clf_message($content);
2240 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2242 # LOG_LEVEL_CLF lines look like this
2243 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2244 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2245 # HTTP/1.1" 403 1730
2246 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2248 print_clf_message();
2252 # Some Privoxy log messages span more than one line,
2253 # usually to dump lots of content that doesn't need any syntax highlighting.
2254 # XXX: add mechanism to forward these lines to the right handler anyway.
2256 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2257 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2265 my ($day, $time_stamp, $msecs, $thread, $log_level, $content);
2266 my %log_level_handlers = (
2267 'Re-Filter' => \&handle_loglevel_ignore,
2268 'Header' => \&gather_loglevel_header_stats,
2269 'Connect' => \&gather_loglevel_connect_stats,
2270 'Redirect' => \&handle_loglevel_ignore,
2271 'Request' => \&gather_loglevel_request_stats,
2272 'Crunch' => \&gather_loglevel_crunch_stats,
2273 'Gif-Deanimate' => \&handle_loglevel_ignore,
2274 'Info' => \&handle_loglevel_ignore,
2275 'CGI' => \&handle_loglevel_ignore,
2276 'Force' => \&handle_loglevel_ignore,
2277 'Error' => \&gather_loglevel_error_stats,
2278 'Fatal error' => \&handle_loglevel_ignore,
2279 'Writing' => \&handle_loglevel_ignore,
2280 'Received' => \&handle_loglevel_ignore,
2281 'Unknown log level' => \&handle_loglevel_ignore
2285 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2288 $msecs = $3 ? $3 : 0;
2293 if (defined($log_level_handlers{$log_level})) {
2295 $content = $log_level_handlers{$log_level}($content, $thread);
2299 die "No handler found for log level \"$log_level\"\n";
2309 sub unbreak_lines_only_loop() {
2310 my $log_messages_reached = 0;
2314 # Log level other than LOG_LEVEL_CLF?
2315 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/ or
2317 m/^((?:\d+\.\d+\.\d+\.\d+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2318 $log_messages_reached = 1;
2323 $_ = "\n". $_ if /^(?:\d+\.\d+\.\d+\.\d+)/;
2328 print "\n" unless $log_messages_reached;
2333 sub VersionMessage {
2334 my $version_message;
2336 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2337 $version_message .= 'Copyright (C) 2007-2010 Fabian Keil <fk@fabiankeil.de>' . "\n";
2338 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2340 print $version_message;
2343 sub get_cli_options () {
2345 our %cli_options = (
2346 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2347 'title' => CLI_OPTION_TITLE,
2348 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2349 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2350 'no-msecs' => CLI_OPTION_NO_MSECS,
2351 'shorten-thread-ids' => CLI_OPTION_SHORTEN_THREAD_IDS,
2352 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2353 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
2354 'statistics' => CLI_OPTION_STATISTICS,
2355 'url-statistics-threshold' => CLI_OPTION_URL_STATISTICS_THRESHOLD,
2356 'unbreak-lines-only' => CLI_OPTION_UNBREAK_LINES_ONLY,
2357 'host-statistics-threshold'=> CLI_OPTION_HOST_STATISTICS_THRESHOLD,
2361 'html-output' => \$cli_options{'html-output'},
2362 'title' => \$cli_options{'title'},
2363 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2364 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2365 'no-msecs' => \$cli_options{'no-msecs'},
2366 'shorten-thread-ids' => \$cli_options{'shorten-thread-ids'},
2367 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2368 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
2369 'statistics' => \$cli_options{'statistics'},
2370 'unbreak-lines-only' => \$cli_options{'unbreak-lines-only'},
2371 'url-statistics-threshold=i'=> \$cli_options{'url-statistics-threshold'},
2372 'host-statistics-threshold=i'=> \$cli_options{'host-statistics-threshold'},
2373 'version' => sub { VersionMessage && exit(0) },
2377 $html_output_mode = cli_option_is_set('html-output');
2378 $no_msecs_mode = cli_option_is_set('no-msecs');
2379 $shorten_thread_ids = cli_option_is_set('shorten-thread-ids');
2380 $line_end = get_line_end();
2391 Options and their default values if they have any:
2392 [--accept-unknown-messages]
2393 [--host-statistics-threshold $cli_options{'host-statistics-threshold'}]
2397 [--no-syntax-highlighting]
2398 [--shorten-thread-ids]
2399 [--show-ineffective-filters]
2401 [--unbreak-lines-only]
2402 [--url-statistics-threshold $cli_options{'url-statistics-threshold'}]
2403 [--title $cli_options{'title'}]
2405 see "perldoc $0" for more information
2411 ################################################################################
2413 ################################################################################
2417 set_background(DEFAULT_BACKGROUND);
2418 prepare_our_stuff();
2422 # XXX: should explicitly reject incompatible argument combinations
2423 if (cli_option_is_set('unbreak-lines-only')) {
2424 unbreak_lines_only_loop();
2425 } elsif (cli_option_is_set('statistics')) {
2438 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2442 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
2443 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--statistics>]
2444 [B<--shorten-thread-ids>] [B<--show-ineffective-filters>]
2445 [B<--url-statistics-threshold>] [B<--version>]
2449 B<privoxy-log-parser> reads Privoxy log messages and
2451 - syntax-highlights recognized lines,
2453 - reformats some of them for easier comprehension,
2455 - filters out less useful messages, and
2457 - (in some cases) calculates additional information,
2458 like the compression ratio or how a filter affected
2461 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2462 without getting confused by the resulting amount of output. For example for
2463 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2464 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2465 will hide the "filter foo caused 0 hits" message.
2469 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
2470 just don't highlight them.
2472 [B<--host-statistics-threshold>] Only show the request count for a host
2473 if it's above or equal to the given threshold. If the threshold is 0, host
2474 statistics are disabled.
2476 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2477 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2478 This option is only intended to make embedding log excerpts in web pages easier.
2479 It does not escape any input!
2481 [B<--no-msecs>] Don't expect milisecond resolution
2483 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2484 the filtered output is piped into less in which case the ANSI control
2485 codes don't work, or if the terminal itself doesn't support the control
2488 [B<--shorten-thread-ids>] Shorten the thread ids to a three-digit decimal number.
2489 Note that the mapping from thread ids to shortened ids is created at run-time
2490 and thus varies with the input.
2492 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2493 that didn't modify the content.
2495 [B<--statistics>] Gather various statistics instead of syntax highlighting
2496 log messages. This is an experimental feature, if the results look wrong
2497 they very well might be. Also note that the results are pretty much guaranteed
2498 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2500 [B<--unbreak-lines-only>] Tries to fix lines that got messed up by a broken or
2501 interestingly configured mail client and thus are no longer recognized properly.
2502 Only fixes some breakage, but may be good enough or at least better than nothing.
2503 Doesn't do anything else, so you probably want to pipe the output into
2504 B<privoxy-log-parser> again.
2506 [B<--url-statistics-threshold>] Only show the request count for a ressource
2507 if it's above or equal to the given threshold. If the threshold is 0, URL
2508 statistics are disabled.
2510 [B<--version>] Print version and exit.
2514 To monitor a log file:
2516 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2518 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2519 or if the log won't get rotated anyway. The log file location depends
2520 on your system (Doh!).
2522 To monitor Privoxy without having it write to a log file:
2524 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2526 Again, the config file location depends on your system. Output redirection
2527 depends on your shell, the above works with bourne shells.
2529 To read a processed Privoxy log file from top to bottom, letting the content
2530 scroll by slightly faster than you can read:
2532 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2534 This is probably only useful to fill screens in the background of haxor movies.
2538 Syntax highlighting with ANSI escape sequences will look strange
2539 if your background color isn't black.
2541 Some messages aren't recognized yet and will not be fully highlighted.
2543 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2544 using earlier Privoxy versions will probably result in an increased amount
2545 of unrecognized log lines.
2547 Privoxy's log files tend to be rather large. If you use HTML
2548 highlighting some browsers can't handle them, get confused and
2549 will eventually crash because of segmentation faults or unexpected
2550 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2555 Many settings can't be controlled through command line options yet.
2563 Fabian Keil <fk@fabiankeil.de>