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 1.110 2011/01/14 19:47:53 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:/
918 # XXX: Some of these may need highlighting
920 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
921 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
922 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
923 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
924 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
925 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
926 # Text mode is already enabled
927 # Denied request with NULL byte(s) turned into line break(s)
928 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
929 # addh-unique: Host: people.freebsd.org
930 # Suppressed offer to compress content
931 # Crunched incoming cookie -- yum!
932 # Accepted the client's request to fetch without filtering.
933 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
934 # addh-unique: Host: subkeys.pgp.net:11371
935 # Referer forged to: Referer: http://10.0.0.1/
936 # Downgraded answer to HTTP/1.0
937 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
938 # Referer overwritten with: Referer: pille-palle
939 # Referer replaced with: Referer: pille-palle
940 # crunched x-forwarded-for!
942 # modified # XXX: pretty stupid log message
943 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
944 # force-text-mode overruled the client's request to fetch without filtering!
945 # Server time in the future.
946 # content-disposition header crunched and replaced with: content-disposition: filename=baz
947 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
948 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
949 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
950 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
952 # A HTTP/1.1 response without Connection header implies keep-alive.
953 # Disabled filter mode on behalf of the client.
954 # Keeping the server header 'Connection: keep-alive' around.
955 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
956 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
957 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
958 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
959 # Removing 'Connection: close' to imply keep-alive.
960 # keep-alive support is disabled. Crunching: Keep-Alive: 300.
961 # Continue hack in da house.
962 # Merged multiple header lines to: 'X-FORWARDED-PROTO: http X-HOST: 127.0.0.1'
964 } elsif ($c =~ m/^scanning headers for:/) {
966 return '' unless SHOW_SCAN_INTRO;
968 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
969 # crunched User-Agent!
970 # Crunching: Content-Encoding: gzip
972 $c =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
974 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
976 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
978 $c = h('warning') . $c . h('Standard');
980 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
982 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
983 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
985 $c =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
986 $c =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
988 } elsif ($c =~ m/^Removing empty header/) {
990 # Removing empty header
993 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
995 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
996 # Enable force-text-mode if you know what you're doing.
997 # XXX: Could highlight more here.
998 $c =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
1000 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
1002 # Server keep-alive timeout is 5. Sticking with 10.
1003 # Client keep-alive timeout is 20. Sticking with 10.
1005 $c =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1006 $c =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1008 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1010 # Reducing keep-alive timeout from 60 to 10.
1012 $c =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1013 $c =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1015 } elsif ($c =~ m/^Killed all-caps Host header line: HOST:/) {
1017 # Killed all-caps Host header line: HOST: bestproxydb.com
1018 $c = highlight_matched_host($c, '(?<=HOST: )[^\s]+');
1019 $c = highlight_matched_pattern($c, 'HOST', 'HOST');
1023 found_unknown_content($c);
1027 unless ($c =~ m/^Transforming/) {
1028 $c = highlight_known_headers($c) unless $no_special_header_highlighting;
1034 sub handle_loglevel_re_filter ($) {
1036 my $content = shift;
1040 if ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1042 # XXX: only the second version gets highlighted properly.
1043 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1044 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1045 $req{$t}{'content_source'} = $1;
1046 $req{$t}{'content_size'} = $2;
1047 $req{$t}{'content_filter'} = $3;
1048 $req{$t}{'content_hits'} = $4;
1049 $req{$t}{'new_content_size'} = $5;
1050 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1051 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1052 if ($req{$t}{'content_hits'} == 0 and
1053 not (cli_option_is_set('show-ineffective-filters')
1054 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1058 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1059 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1060 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1062 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1063 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1066 $c .= "(" . $h{'Number'};
1067 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1068 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1071 } elsif ($c =~ /\.{3}$/
1072 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1074 # Used by Privoxy 3.0.5 and 3.0.6:
1076 # Used by Privoxy 3.0.7:
1077 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1079 $req{$t}{'filtered_header'} = $1;
1080 $req{$t}{'old_header_size'} = $2;
1081 $req{$t}{'header_filter_name'} = $3;
1083 unless (cli_option_is_set('show-ineffective-filters') or
1084 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1087 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1088 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1090 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1092 # ...produced 0 hits (new size 23).
1093 #... produced 1 hits (new size 54).
1095 $req{$t}{'header_filter_hits'} = $1;
1096 $req{$t}{'new_header_size'} = $2;
1098 unless (cli_option_is_set('show-ineffective-filters') or
1099 (defined($req{$t}{'header_filter_name'}) and
1100 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1102 if ($req{$t}{'header_filter_hits'} == 0 and
1103 not (defined($req{$t}{'header_filter_name'}) and
1104 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1107 # Reformat including information from the intro
1108 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1110 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1111 # $c .= $req{$t}{'filtered_header'} . " ";
1112 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1113 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1115 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1117 $c .= "changing size from ";
1118 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1120 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1125 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1129 # Highlight from last line (XXX: What?)
1130 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1131 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1136 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1137 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1142 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1144 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1145 # Tagger variable-test has empty joblist. Nothing to do.
1147 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1149 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1151 $req{$t}{'chunked-size'} = $1;
1152 $req{$t}{'dechunked-size'} = $2;
1153 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1155 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1157 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1158 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1160 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1162 # Decompression successful. Old size: 670, new size: 1166.
1164 $req{$t}{'size-compressed'} = $1;
1165 $req{$t}{'size-decompressed'} = $2;
1166 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1168 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1169 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1171 # XXX: Create sub get_percentage()
1172 if ($req{$t}{'size-decompressed'}) {
1173 $req{$t}{'decompression-gain-percent'} =
1174 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1176 $content .= " (saved: ";
1177 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1179 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1183 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1185 # Need to de-chunk first
1186 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1188 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1190 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1192 # Adding re_filter job ...
1193 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1194 # to filter client-ip-address succeeded.
1196 } elsif ($c =~ m/^Reading in filter/) {
1198 return '' unless SHOW_FILTER_READIN_IN;
1202 found_unknown_content($content);
1209 sub handle_loglevel_redirect ($) {
1213 if ($c =~ m/^Decoding "([^""]*)"/) {
1215 $req{$t}{'original-destination'} = $1;
1216 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1219 } elsif ($c =~ m/^Checking/) {
1221 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1222 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1224 # TODO: Change colour if really url-decoded
1225 $req{$t}{'decoded-original-destination'} = $1;
1226 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1229 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1231 # pcrs command "s@&from=rss@@" changed \
1232 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1233 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1234 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1235 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1236 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1237 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1238 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1239 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1241 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1243 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1244 # change "http://www.example.org/".
1245 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1246 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1248 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1250 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1251 # XXX: Use URL highlighter
1253 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1255 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1256 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1257 # is already properly formatted.
1258 # XXX: assume the same?
1259 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1263 found_unknown_content($c);
1270 sub handle_loglevel_gif_deanimate ($) {
1272 my $content = shift;
1274 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1276 my $bytes_from = $1;
1278 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1279 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1280 # XXX: Do we need g in case of ($1 == $2)?
1281 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1283 } elsif ($content =~ m/GIF (not) changed/) {
1285 # Gif-Deanimate: GIF not changed.
1286 return '' if SUPPRESS_GIF_NOT_CHANGED;
1287 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1289 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1291 # failed! (gif parsing)
1292 # XXX: Replace this error message with something less stupid
1293 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1295 } elsif ($content =~ m/^Need to de-chunk first/) {
1297 # Need to de-chunk first
1298 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1300 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1302 # No GIF header found (XXX: Did I ever commit this?)
1303 # failed while parsing 195 134747048 (XXX: never commited)
1305 # Ignore these for now
1309 found_unknown_content($content);
1316 sub handle_loglevel_request ($) {
1318 my $content = shift;
1320 if ($content =~ m/crunch! /) {
1322 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1324 # Highlight crunch reasons
1325 foreach my $reason (keys %reason_colours) {
1326 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1328 # Highlight request URL domain and ditch 'crunch!'
1329 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1330 $content =~ s@ crunch!@@;
1332 } elsif ($content =~ m/\[too long, truncated\]$/) {
1334 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1335 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1337 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1339 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1340 $content = h('request_') . $content . h('Standard');
1344 found_unknown_content($content);
1351 sub handle_loglevel_crunch ($) {
1353 my $content = shift;
1355 # Highlight crunch reason
1356 foreach my $reason (keys %reason_colours) {
1357 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1360 if ($content =~ m/\[too long, truncated\]$/) {
1362 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1363 # [...]&filter... [too long, truncated]
1364 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1368 # Blocked: http://ads.example.org/
1369 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1375 sub handle_loglevel_connect ($) {
1379 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1381 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1383 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1384 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1386 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1388 # connect to: www.example.org.noconnect failed: Operation not permitted
1390 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1392 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1394 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1396 # Connect: to www.nzherald.co.nz successful
1397 # Connect: to archiv.radiotux.de
1399 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1400 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1402 } elsif ($c =~ m/^to ([^\s]*)$/) {
1404 # Connect: to lists.sourceforge.net:443
1406 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1408 } elsif ($c =~ m/^accepted connection from .*/ or
1411 # accepted connection from 10.0.0.1
1412 # Privoxy 3.0.6 and earlier just say:
1414 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1415 $c = highlight_matched_host($c, '(?<=connection from ).*');
1417 } elsif ($c =~ m/^write header to: .* failed:/) {
1419 # write header to: 10.0.0.1 failed: Broken pipe
1421 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1422 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1424 } elsif ($c =~ m/^write header to client failed:/) {
1426 # write header to client failed: Broken pipe
1428 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1430 } elsif ($c =~ m/^socks4_connect:/) {
1432 # socks4_connect: SOCKS request rejected or failed.
1433 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1435 } elsif ($c =~ m/^Listening for new connections/ or
1436 $c =~ m/^accept connection/) {
1438 # Privoxy versions above 3.0.6 say:
1439 # Listening for new connections ...
1440 # earlier versions say:
1441 # accept connection ...
1444 } elsif ($c =~ m/^accept failed:/) {
1446 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1448 } elsif ($c =~ m/^Overriding forwarding settings/) {
1450 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1451 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1453 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1455 # Denying suspicious CONNECT request from 10.0.0.1
1456 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1458 } elsif ($c =~ m/^socks5_connect:/) {
1460 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1462 } elsif ($c =~ m/^Created new connection to/) {
1464 # Created new connection to www.privoxy.org:80 on socket 11.
1465 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1466 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1468 } elsif ($c =~ m/^Found reusable socket/) {
1470 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1472 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1473 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1474 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1475 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1476 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1477 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1478 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1479 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1481 } elsif ($c =~ m/^Marking open socket/) {
1483 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1484 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1485 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1486 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1488 } elsif ($c =~ m/^No reusable/) {
1490 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1491 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1493 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1495 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1496 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1498 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1499 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1500 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1502 } elsif ($c =~ m/^Socket/) {
1504 # Socket 16 already forgotten or never remembered.
1505 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1507 } elsif ($c =~ m/^The connection to/) {
1509 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1511 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1512 # Timeout is: 1. Assumed latency: 4.
1513 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1514 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1515 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1516 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1517 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1518 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1520 } elsif ($c =~ m/^Stopped waiting for the request line./) {
1522 # Stopped waiting for the request line. Timeout: 121.
1523 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1525 } elsif ($c =~ m/^Waiting for \d/) {
1527 # Waiting for 1 connections to timeout.
1528 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1530 } elsif ($c =~ m/^Initialized/) {
1532 # Initialized 20 socket slots.
1533 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1535 } elsif ($c =~ m/^Done reading from server/) {
1537 # Done reading from server. Expected content length: 24892. \
1538 # Actual content length: 24892. Most recently received: 4412.
1540 # Done reading from server. Expected content length: 24892. \
1541 # Actual content length: 24892. Bytes most recently read: 4412.
1542 # Done reading from server. Content length: 6018 as expected. \
1543 # Bytes most recently read: 294.
1544 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1545 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1546 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1548 } elsif ($c =~ m/^Continuing buffering headers/) {
1550 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1551 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1552 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1553 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1555 # Continuing buffering headers. Bytes most recently read: %d.
1556 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1558 } elsif ($c =~ m/^Received \d+ bytes while/) {
1560 # Received 206 bytes while expecting 12103.
1561 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1562 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1564 } elsif ($c =~ m/^(Rejecting c|C)onnection from/) {
1566 # Connection from 81.163.28.218 dropped due to ACL
1567 # Rejecting connection from 178.63.152.227. Maximum number of connections reached.
1568 $c =~ s@(?<=onnection from )((?:\d+\.?){3}\d+)@$h{'Number'}$1$h{'Standard'}@;
1570 } elsif ($c =~ m/^(?:Reusing|Closing) server socket \d./ or
1571 $c =~ m/^No additional client request/) {
1573 # Reusing server socket 4. Opened for 10.0.0.1.
1574 # Closing server socket 2. Opened for 10.0.0.1.
1575 # No additional client request received in time. \
1576 # Closing server socket 4, initially opened for 10.0.0.1.
1577 # No additional client request received in time on socket 29.
1579 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1580 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.$)');
1582 } elsif ($c =~ m/^Connected to /) {
1584 # Connected to tor-jail[10.0.0.2]:9050.
1586 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1587 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1588 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1590 } elsif ($c =~ m/^Could not connect to /) {
1592 # Could not connect to [10.0.0.1]:80.
1594 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1595 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1597 } elsif ($c =~ m/^Waiting for the next client request/ or
1598 $c =~ m/^The connection on server socket/ or
1599 $c =~ m/^Client request arrived in time /) {
1601 # Waiting for the next client request on socket 3. Keeping the server \
1602 # socket 12 to a.fsdn.com open.
1603 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1604 # Used by Privoxy 3.0.18 and later:
1605 # Client request arrived in time on socket 21.
1606 # Used by earlier version:
1607 # Client request arrived in time or the client closed the connection on socket 12.
1609 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1610 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1611 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1613 } elsif ($c =~ m/^Marking the server socket/) {
1615 # Marking the server socket 7 tainted.
1617 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1619 } elsif ($c =~ m/^Reduced expected bytes to /) {
1621 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1622 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1623 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1625 } elsif ($c =~ m/^The client closed socket /) {
1627 # The client closed socket 2 while the server socket 4 is still open.
1628 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1629 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1631 } elsif ($c =~ m/^Expected client content length set /) {
1633 # Expected client content length set to 667325411 after reading 4999 bytes.
1634 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1635 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1637 } elsif ($c =~ m/^Reducing expected bytes to /) {
1639 # Reducing expected bytes to 0. Marking the server socket tainted after throwing 4 bytes away.
1640 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1641 $c =~ s@(?<=after throwing )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1643 } elsif ($c =~ m/^Waiting for up to /) {
1645 # Waiting for up to 4999 bytes from the client.
1646 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1648 } elsif ($c =~ m/^Looks like we / or
1649 $c =~ m/^Unsetting keep-alive flag/ or
1650 $c =~ m/^No connections to wait/ or
1651 $c =~ m/^Complete client request received/ or
1652 $c =~ m/^Possible pipeline attempt detected./ or
1653 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1654 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1655 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1656 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1657 $c =~ m/^Stopping to watch the client socket. There's already another request waiting./ or
1658 $c =~ m/^Done reading from the client\.$/) {
1660 # Looks like we reached the end of the last chunk. We better stop reading.
1661 # Looks like we read the end of the last chunk together with the server \
1662 # headers. We better stop reading.
1663 # Looks like we got the last chunk together with the server headers. \
1664 # We better stop reading.
1665 # Unsetting keep-alive flag.
1666 # No connections to wait for left.
1667 # Client request arrived in time or the client closed the connection.
1668 # Complete client request received
1669 # Possible pipeline attempt detected. The connection will not be \
1670 # kept alive and we will only serve the first request.
1671 # POST request detected. The connection will not be kept alive.
1672 # The server still wants to talk, but the client hung up on us.
1673 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1674 # There might be a request body. The connection will not be kept alive.
1675 # Stopping to watch the client socket. There's already another request waiting.
1676 # Done reading from the client\.
1680 found_unknown_content($c);
1688 sub handle_loglevel_info ($) {
1692 if ($c =~ m/^Rewrite detected:/) {
1694 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1695 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1697 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1698 $c =~ m/^Compressed content detected/ or
1701 # Decompressing deflated iob: 117
1702 # Decompression didn't result in any content.
1703 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1704 # with zlib support or enable the prevent-compression action.
1705 # Tagger 'complete-url' created empty tag. Ignored.
1709 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1711 # loading configuration file '/usr/local/etc/privoxy/config':
1712 # Reloading configuration file '/usr/local/etc/privoxy/config'
1713 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1715 } elsif ($c =~ m/^Loading (actions|filter) file: /) {
1717 # Loading actions file: /usr/local/etc/privoxy/default.action
1718 # Loading filter file: /usr/local/etc/privoxy/default.filter
1719 $c =~ s@(?<= file: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1721 } elsif ($c =~ m/^exiting by signal/) {
1723 # exiting by signal 15 .. bye
1724 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1726 } elsif ($c =~ m/^Privoxy version/) {
1728 # Privoxy version 3.0.7
1729 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1731 } elsif ($c =~ m/^Program name: /) {
1733 # Program name: /usr/local/sbin/privoxy
1734 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1736 } elsif ($c =~ m/^Listening on port /) {
1738 # Listening on port 8118 on IP address 10.0.0.1
1739 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1740 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1742 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1744 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1745 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1747 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1749 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1750 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1751 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1753 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1754 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1755 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1756 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1758 } elsif ($c =~ m/^Status code/) {
1760 # Status code 304 implies no body.
1761 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1763 } elsif ($c =~ m/^Method/) {
1765 # Method HEAD implies no body.
1766 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1768 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1770 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1771 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1772 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1774 } elsif ($c =~ m/^File modification detected: /) {
1776 # File modification detected: /usr/local/etc/privoxy/user-agent.action
1777 $c =~ s@(?<= detected: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1779 } elsif ($c =~ m/^No logfile configured/ or
1780 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1781 $c =~ m/^Invalid \"chunked\" transfer/ or
1782 $c =~ m/^Support for/ or
1783 $c =~ m/^Flushing header and buffers/ or
1784 $c =~ m/^Can not resolve/
1787 # No logfile configured. Please enable it before reporting any problems.
1788 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1789 # response or even no response at all.
1790 # No logfile configured. Logging disabled.
1791 # Invalid "chunked" transfer encoding detected and ignored.
1792 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1793 # known not to work properly in some situations.
1794 # Flushing header and buffers. Stepping back from filtering.
1795 # Can not resolve doesnotexist: hostname nor servname provided, or not known
1799 found_unknown_content($c);
1806 sub handle_loglevel_cgi ($) {
1810 if ($c =~ m/^Granting access to/) {
1812 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1814 } elsif ($c =~ m/^Substituting: s(.)/) {
1816 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1817 # XXX: prone to span several lines
1820 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1821 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1822 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1828 sub handle_loglevel_force ($) {
1832 if ($c =~ m/^Ignored force prefix in request:/) {
1834 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1835 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1836 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1838 } elsif ($c =~ m/^Enforcing request:/) {
1840 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1841 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1845 found_unknown_content($c);
1852 sub handle_loglevel_error ($) {
1856 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1858 # Empty server or forwarder response received on socket 4.
1859 # Empty server or forwarder response received on socket 3. \
1860 # Closing client socket 15 without sending data.
1861 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1862 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1864 # XXX: There are probably more messages that deserve highlighting.
1870 sub handle_loglevel_ignore ($) {
1874 sub gather_loglevel_request_stats ($$) {
1882 sub gather_loglevel_crunch_stats ($$) {
1890 if ($c =~ m/^Redirected:/) {
1891 # Redirected: http://www.example.org/http://p.p/
1892 $stats{'fast-redirections'}++;
1894 } elsif ($c =~ m/^Blocked:/) {
1895 # Blocked: blogger.googleusercontent.com:443
1896 $stats{'blocked'}++;
1901 sub gather_loglevel_error_stats ($$) {
1908 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1910 # Empty server or forwarder response received on socket 4.
1911 $stats{'empty-responses'}++;
1912 if ($thread_data{$thread}{'new_connection'}) {
1913 $stats{'empty-responses-on-new-connections'}++;
1915 $stats{'empty-responses-on-reused-connections'}++;
1920 sub gather_loglevel_connect_stats ($$) {
1922 my ($c, $thread) = @_;
1926 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
1928 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1929 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
1931 } elsif ($c =~ m/^to ([^\s]*)$/) {
1933 # Connect: to lists.sourceforge.net:443
1935 $thread_data{$thread}{'forwarder'} = 'direct connection';
1937 } elsif ($c =~ m/^Created new connection to/) {
1939 # Created new connection to www.privoxy.org:80 on socket 11.
1941 $thread_data{$thread}{'new_connection'} = 1;
1943 } elsif ($c =~ m/^Reusing server socket \d./ or
1944 $c =~ m/^Found reusable socket/) {
1946 # Reusing server socket 4. Opened for 10.0.0.1.
1947 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1949 $thread_data{$thread}{'new_connection'} = 0;
1950 $stats{'reused-connections'}++;
1954 sub gather_loglevel_header_stats ($$) {
1956 my ($c, $thread) = @_;
1959 if ($c =~ m/^A HTTP\/1\.1 response without/ or
1960 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
1962 # A HTTP/1.1 response without Connection header implies keep-alive.
1963 # Keeping the server header 'Connection: keep-alive' around.
1964 $stats{'server-keep-alive'}++;
1966 } elsif ($c =~ m/^scan: ((\w+) (.+) (HTTP\/\d\.\d))/) {
1968 # scan: HTTP/1.1 200 OK
1969 $stats{'method'}{$2}++;
1970 $stats{'ressource'}{$3}++;
1971 $stats{'http-version'}{$4}++;
1973 } elsif ($c =~ m/^scan: Host: ([^\s]+)/) {
1976 $stats{'hosts'}{$1}++;
1984 'server-keep-alive' => 0,
1985 'reused-connections' => 0,
1986 'empty-responses' => 0,
1987 'empty-responses-on-new-connections' => 0,
1988 'empty-responses-on-reused-connections' => 0,
1989 'fast-redirections' => 0,
1991 'reused-connections' => 0,
1992 'server-keep-alive' => 0,
1996 sub get_percentage ($$) {
1999 return "NaN" if ($big eq 0);
2000 return sprintf("%.2f%%", $small / $big * 100);
2003 sub print_stats () {
2007 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
2008 my $outgoing_requests = $stats{requests} - $stats{crunches};
2010 if ($stats{requests} eq 0) {
2011 print "No requests yet.\n";
2015 print "Client requests total: " . $stats{requests} . "\n";
2016 print "Crunches: " . $stats{crunches} . " (" .
2017 get_percentage($stats{requests}, $stats{crunches}) . ")\n";
2018 print "Blocks: " . $stats{'blocked'} . " (" .
2019 get_percentage($stats{requests}, $stats{'blocked'}) . ")\n";
2020 print "Fast redirections: " . $stats{'fast-redirections'} . " (" .
2021 get_percentage($stats{requests}, $stats{'fast-redirections'}) . ")\n";
2022 print "Outgoing requests: " . $outgoing_requests . " (" .
2023 get_percentage($stats{requests}, $outgoing_requests) . ")\n";
2024 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
2025 get_percentage($stats{requests}, $stats{'server-keep-alive'}) . ")\n";
2026 print "New outgoing connections: " . $new_connections . " (" .
2027 get_percentage($stats{requests}, $new_connections) . ")\n";
2028 print "Reused connections: " . $stats{'reused-connections'} . " (" .
2029 get_percentage($stats{requests}, $stats{'reused-connections'}) .
2030 "; server offers accepted: " .
2031 get_percentage($stats{'server-keep-alive'}, $stats{'reused-connections'}) . ")\n";
2032 print "Empty responses: " . $stats{'empty-responses'} . " (" .
2033 get_percentage($stats{requests}, $stats{'empty-responses'}) . ")\n";
2034 print "Empty responses on new connections: "
2035 . $stats{'empty-responses-on-new-connections'} . " (" .
2036 get_percentage($stats{requests}, $stats{'empty-responses-on-new-connections'})
2038 print "Empty responses on reused connections: " .
2039 $stats{'empty-responses-on-reused-connections'} . " (" .
2040 get_percentage($stats{requests}, $stats{'empty-responses-on-reused-connections'}) .
2043 if ($stats{method} eq 0) {
2044 print "No response lines parsed yet yet.\n";
2047 print "Method distribution:\n";
2048 foreach my $method (sort {$stats{'method'}{$b} <=> $stats{'method'}{$a}} keys %{$stats{'method'}}) {
2049 printf "%8d : %-8s\n", $stats{'method'}{$method}, $method;
2051 print "Client HTTP versions:\n";
2052 foreach my $http_version (sort {$stats{'http-version'}{$b} <=> $stats{'http-version'}{$a}} keys %{$stats{'http-version'}}) {
2053 printf "%d : %s\n", $stats{'http-version'}{$http_version}, $http_version;
2056 if ($cli_options{'url-statistics-threshold'} == 0) {
2057 print "URL statistics are disabled. Increase --url-statistics-threshold to enable them.\n";
2059 print "Requested URLs:\n";
2060 foreach my $ressource (sort {$stats{'ressource'}{$b} <=> $stats{'ressource'}{$a}} keys %{$stats{'ressource'}}) {
2061 if ($stats{'ressource'}{$ressource} < $cli_options{'url-statistics-threshold'}) {
2062 print "Skipped statistics for URLs below the treshold.\n";
2065 printf "%d : %s\n", $stats{'ressource'}{$ressource}, $ressource;
2069 if ($cli_options{'host-statistics-threshold'} == 0) {
2070 print "Host statistics are disabled. Increase --host-statistics-threshold to enable them.\n";
2072 print "Requested Hosts:\n";
2073 foreach my $host (sort {$stats{'hosts'}{$b} <=> $stats{'hosts'}{$a}} keys %{$stats{'hosts'}}) {
2074 if ($stats{'hosts'}{$host} < $cli_options{'host-statistics-threshold'}) {
2075 print "Skipped statistics for Hosts below the treshold.\n";
2078 printf "%d : %s\n", $stats{'hosts'}{$host}, $host;
2084 ################################################################################
2085 # Functions that actually print stuff
2086 ################################################################################
2088 sub print_clf_message () {
2090 our ($ip, $timestamp, $request_line, $status_code, $size);
2093 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2095 # Rebuild highlighted
2096 $output .= $h{'Number'} . $ip . $h{'Standard'};
2098 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2100 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2102 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2104 $output .= $h{'Number'} . $size . $h{'Standard'};
2105 $output .= $line_end;
2110 sub print_non_clf_message ($) {
2112 my $content = shift;
2113 my $msec_string = "." . $req{$t}{'msecs'} unless $no_msecs_mode;
2114 my $line_start = $html_output_mode ? '' : $h{"Standard"};
2116 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2119 . $time_colours[$time_colour_index % 2]
2120 . $req{$t}{'time-stamp'}
2122 . $h{Standard} . " "
2123 . $thread_colours{$t}
2127 . $h{$req{$t}{'log-level'}}
2128 . $req{$t}{'log-level'}
2135 sub shorten_thread_id ($) {
2137 my $thread_id = shift;
2139 our %short_thread_ids;
2142 unless (defined $short_thread_ids{$thread_id}) {
2143 $short_thread_ids{$thread_id} = sprintf "%.3d", $max_threadid++;
2146 return $short_thread_ids{$thread_id}
2151 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2153 my $last_thread = 0;
2154 my $last_timestamp = 0;
2155 my $filters_that_did_nothing;
2158 $time_colour = paint_it('white');
2160 my %log_level_handlers = (
2161 'Re-Filter' => \&handle_loglevel_re_filter,
2162 'Header' => \&handle_loglevel_header,
2163 'Connect' => \&handle_loglevel_connect,
2164 'Redirect' => \&handle_loglevel_redirect,
2165 'Request' => \&handle_loglevel_request,
2166 'Crunch' => \&handle_loglevel_crunch,
2167 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2168 'Info' => \&handle_loglevel_info,
2169 'CGI' => \&handle_loglevel_cgi,
2170 'Force' => \&handle_loglevel_force,
2171 'Error' => \&handle_loglevel_error,
2172 'Fatal error' => \&handle_loglevel_ignore,
2173 'Writing' => \&handle_loglevel_ignore,
2174 'Received' => \&handle_loglevel_ignore,
2175 'Unknown log level' => \&handle_loglevel_ignore,
2180 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2181 $thread = $t = ($shorten_thread_ids) ? shorten_thread_id($4) : $4;
2182 $req{$t}{'day'} = $day = $1;
2183 $req{$t}{'time-stamp'} = $time_stamp = $2;
2184 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2185 $req{$t}{'log-level'} = $log_level = $5;
2186 $req{$t}{'content'} = $content = $c = $6;
2187 $req{$t}{'log-message'} = $_;
2188 $no_special_header_highlighting = 0;
2190 if (defined($log_level_handlers{$log_level})) {
2192 $content = $log_level_handlers{$log_level}($content);
2196 die "No handler found for log level \"$log_level\"\n";
2199 # Highlight Truncations
2200 if (length($_) > 4000) {
2201 $content =~ s@(too long, truncated)]$@$h{'Truncation'}$1$h{'Standard'}]@g;
2204 next unless $content;
2206 # Register threads to keep the colour constant
2207 if (!defined($thread_colours{$thread})) {
2208 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2209 $thread_colour_index++;
2212 # Switch timestamp colour if timestamps differ
2213 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2214 debug_message("Tick tack!") if DEBUG_TICKS;
2215 $time_colour = $time_colours[$time_colour_index % 2];
2216 $time_colour_index++;
2217 $last_msecs = $msecs;
2218 $last_timestamp = $time_stamp;
2221 $last_thread = $thread;
2223 print_non_clf_message($content);
2225 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2227 # LOG_LEVEL_CLF lines look like this
2228 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2229 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2230 # HTTP/1.1" 403 1730
2231 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2233 print_clf_message();
2237 # Some Privoxy log messages span more than one line,
2238 # usually to dump lots of content that doesn't need any syntax highlighting.
2239 # XXX: add mechanism to forward these lines to the right handler anyway.
2241 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2242 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2250 my ($day, $time_stamp, $msecs, $thread, $log_level, $content);
2251 my %log_level_handlers = (
2252 'Re-Filter' => \&handle_loglevel_ignore,
2253 'Header' => \&gather_loglevel_header_stats,
2254 'Connect' => \&gather_loglevel_connect_stats,
2255 'Redirect' => \&handle_loglevel_ignore,
2256 'Request' => \&gather_loglevel_request_stats,
2257 'Crunch' => \&gather_loglevel_crunch_stats,
2258 'Gif-Deanimate' => \&handle_loglevel_ignore,
2259 'Info' => \&handle_loglevel_ignore,
2260 'CGI' => \&handle_loglevel_ignore,
2261 'Force' => \&handle_loglevel_ignore,
2262 'Error' => \&gather_loglevel_error_stats,
2263 'Fatal error' => \&handle_loglevel_ignore,
2264 'Writing' => \&handle_loglevel_ignore,
2265 'Received' => \&handle_loglevel_ignore,
2266 'Unknown log level' => \&handle_loglevel_ignore
2270 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2273 $msecs = $3 ? $3 : 0;
2278 if (defined($log_level_handlers{$log_level})) {
2280 $content = $log_level_handlers{$log_level}($content, $thread);
2284 die "No handler found for log level \"$log_level\"\n";
2294 sub unbreak_lines_only_loop() {
2295 my $log_messages_reached = 0;
2299 # Log level other than LOG_LEVEL_CLF?
2300 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/ or
2302 m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2303 $log_messages_reached = 1;
2311 print "\n" unless $log_messages_reached;
2315 sub VersionMessage {
2316 my $version_message;
2318 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2319 $version_message .= 'Copyright (C) 2007-2010 Fabian Keil <fk@fabiankeil.de>' . "\n";
2320 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2322 print $version_message;
2325 sub get_cli_options () {
2327 our %cli_options = (
2328 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2329 'title' => CLI_OPTION_TITLE,
2330 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2331 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2332 'no-msecs' => CLI_OPTION_NO_MSECS,
2333 'shorten-thread-ids' => CLI_OPTION_SHORTEN_THREAD_IDS,
2334 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2335 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
2336 'statistics' => CLI_OPTION_STATISTICS,
2337 'url-statistics-threshold' => CLI_OPTION_URL_STATISTICS_THRESHOLD,
2338 'unbreak-lines-only' => CLI_OPTION_UNBREAK_LINES_ONLY,
2339 'host-statistics-threshold'=> CLI_OPTION_HOST_STATISTICS_THRESHOLD,
2343 'html-output' => \$cli_options{'html-output'},
2344 'title' => \$cli_options{'title'},
2345 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2346 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2347 'no-msecs' => \$cli_options{'no-msecs'},
2348 'shorten-thread-ids' => \$cli_options{'shorten-thread-ids'},
2349 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2350 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
2351 'statistics' => \$cli_options{'statistics'},
2352 'unbreak-lines-only' => \$cli_options{'unbreak-lines-only'},
2353 'url-statistics-threshold=s'=> \$cli_options{'url-statistics-threshold'},
2354 'host-statistics-threshold=s'=> \$cli_options{'host-statistics-threshold'},
2355 'version' => sub { VersionMessage && exit(0) },
2359 $html_output_mode = cli_option_is_set('html-output');
2360 $no_msecs_mode = cli_option_is_set('no-msecs');
2361 $shorten_thread_ids = cli_option_is_set('shorten-thread-ids');
2362 $line_end = get_line_end();
2373 Options and their default values if they have any:
2374 [--accept-unknown-messages]
2375 [--host-statistics-threshold $cli_options{'host-statistics-threshold'}]
2379 [--no-syntax-highlighting]
2380 [--shorten-thread-ids]
2381 [--show-ineffective-filters]
2383 [--unbreak-lines-only]
2384 [--url-statistics-threshold $cli_options{'url-statistics-threshold'}]
2385 [--title $cli_options{'title'}]
2387 see "perldoc $0" for more information
2393 ################################################################################
2395 ################################################################################
2399 set_background(DEFAULT_BACKGROUND);
2400 prepare_our_stuff();
2404 # XXX: should explicitly reject incompatible argument combinations
2405 if (cli_option_is_set('unbreak-lines-only')) {
2406 unbreak_lines_only_loop();
2407 } elsif (cli_option_is_set('statistics')) {
2420 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2424 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
2425 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--statistics>]
2426 [B<--shorten-thread-ids>] [B<--show-ineffective-filters>]
2427 [B<--url-statistics-threshold>] [B<--version>]
2431 B<privoxy-log-parser> reads Privoxy log messages and
2433 - syntax-highlights recognized lines,
2435 - reformats some of them for easier comprehension,
2437 - filters out less useful messages, and
2439 - (in some cases) calculates additional information,
2440 like the compression ratio or how a filter affected
2443 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2444 without getting confused by the resulting amount of output. For example for
2445 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2446 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2447 will hide the "filter foo caused 0 hits" message.
2451 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
2452 just don't highlight them.
2454 [B<--host-statistics-threshold>] Only show the request count for a host
2455 if it's above or equal to the given threshold. If the threshold is 0, host
2456 statistics are disabled.
2458 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2459 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2460 This option is only intended to make embedding log excerpts in web pages easier.
2461 It does not escape any input!
2463 [B<--no-msecs>] Don't expect milisecond resolution
2465 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2466 the filtered output is piped into less in which case the ANSI control
2467 codes don't work, or if the terminal itself doesn't support the control
2470 [B<--shorten-thread-ids>] Shorten the thread ids to a three-digit decimal number.
2471 Note that the mapping from thread ids to shortened ids is created at run-time
2472 and thus varies with the input.
2474 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2475 that didn't modify the content.
2477 [B<--statistics>] Gather various statistics instead of syntax highlighting
2478 log messages. This is an experimental feature, if the results look wrong
2479 they very well might be. Also note that the results are pretty much guaranteed
2480 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2482 [B<--unbreak-lines-only] Tries to fix lines that got messed up by a broken or
2483 interestingly configured mail client and thus are no longer recognized properly.
2484 Only fixes some breakage, but may be good enough or at least better than nothing.
2485 Doesn't do anything else, so you probably want to pipe the output into
2486 B<privoxy-log-parser> again.
2488 [B<--url-statistics-threshold>] Only show the request count for a ressource
2489 if it's above or equal to the given threshold. If the threshold is 0, URL
2490 statistics are disabled.
2492 [B<--version>] Print version and exit.
2496 To monitor a log file:
2498 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2500 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2501 or if the log won't get rotated anyway. The log file location depends
2502 on your system (Doh!).
2504 To monitor Privoxy without having it write to a log file:
2506 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2508 Again, the config file location depends on your system. Output redirection
2509 depends on your shell, the above works with bourne shells.
2511 To read a processed Privoxy log file from top to bottom, letting the content
2512 scroll by slightly faster than you can read:
2514 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2516 This is probably only useful to fill screens in the background of haxor movies.
2520 Syntax highlighting with ANSI escape sequences will look strange
2521 if your background color isn't black.
2523 Some messages aren't recognized yet and will not be fully highlighted.
2525 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2526 using earlier Privoxy versions will probably result in an increased amount
2527 of unrecognized log lines.
2529 Privoxy's log files tend to be rather large. If you use HTML
2530 highlighting some browsers can't handle them, get confused and
2531 will eventually crash because of segmentation faults or unexpected
2532 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2537 Many settings can't be controlled through command line options yet.
2545 Fabian Keil <fk@fabiankeil.de>