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 1.85 2010/07/22 14:56:33 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
27 # Copyright (c) 2007-2010 Fabian Keil <fk@fabiankeil.de>
29 # Permission to use, copy, modify, and distribute this software for any
30 # purpose with or without fee is hereby granted, provided that the above
31 # copyright notice and this permission notice appear in all copies.
33 # THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
34 # WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
35 # MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
36 # ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
37 # WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
38 # ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
39 # OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
40 ################################################################################
47 PRIVOXY_LOG_PARSER_VERSION => '0.6',
48 # Feel free to mess with these ...
49 DEFAULT_BACKGROUND => 'black', # Choose registered colour (like 'black')
50 DEFAULT_TEXT_COLOUR => 'white', # Choose registered colour (like 'black')
51 HEADER_DEFAULT_COLOUR => 'yellow',
52 REGISTER_HEADERS_WITH_THE_SAME_COLOUR => 1,
54 CLI_OPTION_DEFAULT_TO_HTML_OUTPUT => 0,
55 CLI_OPTION_TITLE => 'Privoxy-Log-Parser in da house',
56 CLI_OPTION_NO_EMBEDDED_CSS => 0,
57 CLI_OPTION_NO_MSECS => 0,
58 CLI_OPTION_NO_SYNTAX_HIGHLIGHTING => 0,
59 CLI_OPTION_SHORTEN_THREAD_IDS => 0,
60 CLI_OPTION_SHOW_INEFFECTIVE_FILTERS => 0,
61 CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES => 0,
62 CLI_OPTION_STATISTICS => 0,
64 SUPPRESS_SUCCEEDED_FILTER_ADDITIONS => 1,
66 SHOW_FILTER_READIN_IN => 0,
67 SUPPRESS_EMPTY_LINES => 1,
68 SUPPRESS_SUCCESSFUL_CONNECTIONS => 1,
69 SUPPRESS_ACCEPTED_CONNECTIONS => 1,
70 SUPPRESS_GIF_NOT_CHANGED => 1,
71 SUPPRESS_NEED_TO_DE_CHUNK_FIRST => 1,
73 DEBUG_HEADER_REGISTERING => 0,
74 DEBUG_HEADER_HIGHLIGHTING => 0,
77 DEBUG_SUPPRESS_LOG_MESSAGES => 0,
79 PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH => 0,
80 PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH => 1,
82 LOG_UNPARSED_LINES_TO_EXTRA_FILE => 0,
83 ERROR_LOG_FILE => '/var/log/privoxy-log-parser',
85 # You better leave these alone unless you know what you're doing.
86 COLOUR_RESET => "\033[0;0m",
90 # For performance reasons, these are global.
93 my %req; # request data from previous lines
98 my $thread_colour_index = 0;
99 my $header_colour_index = 0;
100 my $time_colour_index = 0;
102 my $no_special_header_highlighting;
105 my $header_highlight_regex = '';
107 my $html_output_mode;
108 my $no_msecs_mode; # XXX: should probably be removed
109 my $shorten_thread_ids;
112 sub prepare_our_stuff () {
114 # Syntax Higlight hash
116 'red', 'green', 'brown', 'blue', 'purple', 'cyan',
117 'light_gray', 'light_red', 'light_green', 'yellow',
118 'light_blue', 'pink', 'light_cyan', 'white'
125 Filter => 'purple', # XXX: Used?
126 'Re-Filter' => 'purple',
128 Request => 'light_cyan',
129 CGI => 'light_green',
131 Error => 'light_red',
133 'Fatal error' => 'light_red',
134 'Gif-Deanimate' => 'blue',
136 Writing => 'light_green',
137 Received => 'yellow',
138 # ----------------------
141 request_ => 'brown', # host+path but no protocol
142 'ip-address' => 'yellow',
145 Truncation => 'light_red',
147 Timestamp => 'brown',
148 Crunching => 'light_red',
149 crunched => 'light_red',
150 'Request-Line' => 'pink',
152 destination => 'yellow',
153 'http-version' => 'pink',
154 'crunch-pattern' => 'pink',
159 'program-name' => 'cyan',
162 warning => 'light_red',
163 debug => 'light_red',
167 'status-message' => 'light_cyan',
168 'status-code' => 'yellow',
169 'invalid-request' => 'light_red',
171 error => 'light_red',
172 'rewritten-URL' => 'light_red',
173 'pcrs-delimiter' => 'light_red',
174 'ignored' => 'light_red',
175 'action-bits-update' => 'light_red',
176 'configuration-line' => 'red',
177 'content-type' => 'yellow',
182 # Header colours need their own hash so the keys can be accessed properly
184 # Prefilled with headers that should not appear with default header colours
185 Cookie => 'light_red',
186 'Set-Cookie' => 'light_red',
187 Warning => 'light_red',
188 Default => HEADER_DEFAULT_COLOUR,
191 # Crunch reasons need their own hash as well
193 'Unsupported HTTP feature' => 'light_red',
194 Blocked => 'light_red',
195 Untrusted => 'light_red',
196 Redirected => 'green',
197 'CGI Call' => 'white',
198 'DNS failure' => 'red',
199 'Forwarding failed' => 'light_red',
200 'Connection failure' => 'light_red',
201 'Out of memory (may mask other reasons)' => 'light_red',
202 'No reason recorded' => 'light_red',
205 @time_colours = ('white', 'light_gray');
207 # Translate highlight strings into highlight code
208 prepare_highlight_hash(\%header_colours);
209 prepare_highlight_hash(\%reason_colours);
210 prepare_highlight_hash(\%h);
211 prepare_colour_array(\@all_colours);
212 prepare_colour_array(\@time_colours);
219 ###############################################################
220 # Takes a colour string and returns an ANSI escape sequence
221 # (unless --no-syntax-highlighting is used).
222 # XXX: The Rolling Stones reference has to go.
223 ###############################################################
225 my $colour = shift @_;
227 return "" if cli_option_is_set('no-syntax-highlighting');
269 my $bg_code = get_background();
271 our $default = default_colours();
273 if (defined($text{$colour})) {
274 $colour_code = ESCAPE;
275 $colour_code .= $text{$colour};
277 $colour_code .= $light{$colour} ? "1" : "2";
279 $colour_code .= $bg_code;
281 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
283 } elsif ($colour =~ /reset/) {
285 $colour_code = default_colours();
289 die "What's $colour supposed to mean?\n";
295 sub get_semantic_html_markup ($) {
296 ###############################################################
297 # Takes a string and returns a span element
298 ###############################################################
303 if ($type =~ /Standard/) {
307 $code = '<span title="' . $type . '" class="' . $type . '">';
313 sub cli_option_is_set ($) {
316 my $cli_option = shift;
318 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
320 return $cli_options{$cli_option};
323 sub get_html_title () {
326 return $cli_options{'title'};
330 sub init_css_colours() {
338 purple => "F06", # XXX: wrong
339 cyan => "F09", # XXX: wrong
344 light_green => "33F",
353 sub get_css_colour ($) {
358 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
360 return '#' . $css_colours{$colour};
363 sub get_css_line ($) {
368 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
369 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
370 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
371 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
372 $css_line .= '}' . "\n";
377 sub get_css_line_for_colour ($) {
382 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
383 $css_line .= 'color:' . get_css_colour($colour) . ';';
384 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
385 $css_line .= '}' . "\n";
390 # XXX: Wrong solution
391 sub get_missing_css_lines () {
395 $css_line .= '.' . 'default' . ' {';
396 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
397 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
398 $css_line .= '}' . "\n";
405 our %css_colours; #XXX: Wrong solution
409 $css .= '.privoxy-log {';
410 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
411 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
414 foreach my $key (keys %h_colours) {
416 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
417 $css .= get_css_line($key);
421 foreach my $colour (keys %css_colours) {
423 $css .= get_css_line_for_colour($colour);
427 $css .= get_missing_css_lines(); #XXX: Wrong solution
436 if (cli_option_is_set('html-output')) {
438 my $title = get_html_title();
440 $intro .= '<html><head>';
441 $intro .= '<title>' . $title . '</title>';
442 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
443 $intro .= '</head><body>';
444 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
454 if (cli_option_is_set('html-output')) {
456 $outro = '</p></body></html>';
462 sub get_line_end () {
463 return cli_option_is_set('html-output') ? "<br>\n" : "\n";
466 sub get_colour_html_markup ($) {
467 ###############################################################
468 # Takes a colour string a span element. XXX: WHAT?
469 # XXX: This function shouldn't be necessary, the
470 # markup should always be semantically correct.
471 ###############################################################
476 if ($type =~ /Standard/) {
479 $code = '<span class="' . lc($type) . '">';
485 sub default_colours () {
488 return reset_colours();
491 sub show_colours () {
495 sub reset_colours () {
496 return ESCAPE . "0m";
499 sub set_background ($){
515 if (defined($backgrounds{$colour})) {
516 $bg_code = $backgrounds{$colour};
518 die "Invalid background colour: " . $colour;
522 sub get_background (){
526 sub prepare_highlight_hash ($) {
529 foreach my $key (keys %$ref) {
530 $$ref{$key} = $html_output_mode ?
531 get_semantic_html_markup($key) :
532 paint_it($$ref{$key});
536 sub prepare_colour_array ($) {
539 foreach my $i (0 ... @$ref - 1) {
540 $$ref[$i] = $html_output_mode ?
541 get_colour_html_markup($$ref[$i]) :
546 sub found_unknown_content ($) {
551 return if cli_option_is_set('accept-unknown-messages');
553 return if ($unknown =~ /\[too long, truncated\]$/);
555 $message = "found_unknown_content: Don't know how to highlight: ";
556 # Break line so the log file can later be parsed as Privoxy log file again
557 $message .= '"' . $unknown . '"' . " in:\n";
558 $message .= $req{$t}{'log-message'};
559 debug_message($message);
560 log_parse_error($req{$t}{'log-message'});
562 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
565 sub log_parse_error ($) {
569 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
570 open(ERRORLOG, ">>" . ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
571 print ERRORLOG $message;
576 sub debug_message (@) {
579 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
582 ################################################################################
583 # highlighter functions that aren't loglevel-specific
584 ################################################################################
588 # Get highlight marker
589 my $highlight = shift; # XXX: Stupid name;
593 if (defined($highlight)) {
595 $result = $h{$highlight};
599 $message = "h: Don't recognize highlighter $highlight.";
600 debug_message($message);
601 log_parser_error($message);
602 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
608 sub highlight_known_headers ($) {
612 debug_message("Searching $content for things to highlight.") if DEBUG_HEADER_HIGHLIGHTING;
614 if ($content =~ m/(?<=\s)($header_highlight_regex):/) {
616 $content =~ s@(?<=[\s|'])($header)(?=:)@$header_colours{$header}$1$h{'Standard'}@ig;
617 debug_message("Highlighted '$header' in '$content'") if DEBUG_HEADER_HIGHLIGHTING;
623 sub highlight_matched_request_line ($$) {
625 my $result = shift; # XXX: Stupid name;
627 if ($result =~ m@(.*)($regex)(.*)@) {
628 $result = $1 . highlight_request_line($2) . $3
633 sub highlight_request_line ($) {
636 my ($method, $url, $http_version);
638 #GET http://images.sourceforge.net/sfx/icon_warning.gif HTTP/1.1
639 if ($rl =~ m/Invalid request/) {
641 $rl = h('invalid-request') . $rl . h('Standard');
643 } elsif ($rl =~ m/^([-\w]+) (.*) (HTTP\/\d\.\d)/) {
645 # XXX: might not match in case of HTTP method fuzzing.
646 # XXX: save these: ($method, $path, $http_version) = ($1, $2, $3);
647 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
648 if ($rl =~ /http:\/\//) {
649 $rl = highlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
651 $rl = highlight_matched_pattern($rl, 'request_', '[^\s]*(?=\sHTTP)');
654 $rl =~ s@(HTTP\/\d\.\d)$@$h{'http-version'}$1$h{'Standard'}@;
656 } elsif ($rl =~ m/\.\.\. \[too long, truncated\]$/) {
658 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
659 $rl = highlight_matched_url($rl, '[^\s]*(?=\.\.\.)');
661 } elsif ($rl =~ m/^ $/) {
663 $rl = h('error') . "No request line specified!" . h('Standard');
667 debug_message ("Can't parse request line: $rl");
674 sub highlight_response_line ($) {
677 my ($http_version, $status_code, $status_message);
682 # TODO: Mark different status codes differently
684 if ($rl =~ m/((?:HTTP\/\d\.\d|ICY)) (\d+) (.*)/) {
685 ($http_version, $status_code, $status_message) = ($1, $2, $3);
687 debug_message ("Can't parse response line: $rl") and die 'Fix this';
690 # Rebuild highlighted
692 $rl .= h('http-version') . $http_version . h('Standard');
694 $rl .= h('status-code') . $status_code . h('Standard');
696 $rl .= h('status-message') . $status_message . h('Standard');
701 sub highlight_matched_url ($$) {
703 my $result = shift; # XXX: Stupid name;
706 #print "Got $result, regex ($regex)\n";
708 if ($result =~ m@(.*?)($regex)(.*)@) {
709 $result = $1 . highlight_url($2) . $3;
710 #print "Now the result is $result\n";
716 sub highlight_matched_host ($$) {
718 my ($result, $regex) = @_; # XXX: result ist stupid name;
720 if ($result =~ m@(.*?)($regex)(.*)@) {
721 $result = $1 . $h{host} . $2 . $h{Standard} . $3;
727 sub highlight_matched_pattern ($$$) {
729 my $result = shift; # XXX: Stupid name;
733 die "Unknown key $key" unless defined $h{$key};
735 if ($result =~ m@(.*?)($regex)(.*)@) {
736 $result = $1 . h($key) . $2 . h('Standard') . $3;
742 sub highlight_matched_path ($$) {
744 my $result = shift; # XXX: Stupid name;
747 if ($result =~ m@(.*?)($regex)(.*)@) {
748 $result = $1 . h('path') . $2 . h('Standard') . $3;
754 sub highlight_url ($) {
758 if ($html_output_mode) {
760 $url = '<a href="' . $url . '">' . $url . '</a>';
764 $url = h('URL') . $url . h('Standard');
771 sub update_header_highlight_regex ($) {
774 my $headers = join ('|', keys %header_colours);
776 $header_highlight_regex = qr/$headers/;
777 print "Registering '$header'\n" if DEBUG_HEADER_HIGHLIGHTING;
780 ################################################################################
781 # loglevel-specific highlighter functions
782 ################################################################################
784 sub handle_loglevel_header ($) {
788 if ($c =~ /^scan:/) {
790 if ($c =~ m/^scan: ([^: ]+):/) {
792 # Register new headers
793 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
795 if (!defined($header_colours{$header}) and $header =~ /^[\d\w-]*$/) {
796 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
798 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
799 $header_colours{$header} = $header_colours{'Default'};
801 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
802 $header_colour_index++;
804 update_header_highlight_regex($header);
807 } elsif ($c =~ m/^scan: ((\w+) (.+) (HTTP\/\d\.\d))/) {
809 # Client request line
810 # Save for statistics (XXX: Not implemented yet)
811 $req{$t}{'method'} = $2;
812 $req{$t}{'destination'} = $3;
813 $req{$t}{'http-version'} = $4;
815 $c = highlight_request_line($1);
817 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
819 # Server response line
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/
917 # XXX: Some of these may need highlighting
919 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
920 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
921 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
922 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
923 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
924 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
925 # Text mode is already enabled
926 # Denied request with NULL byte(s) turned into line break(s)
927 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
928 # addh-unique: Host: people.freebsd.org
929 # Suppressed offer to compress content
930 # Crunched incoming cookie -- yum!
931 # Accepted the client's request to fetch without filtering.
932 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
933 # addh-unique: Host: subkeys.pgp.net:11371
934 # Referer forged to: Referer: http://10.0.0.1/
935 # Downgraded answer to HTTP/1.0
936 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
937 # Referer overwritten with: Referer: pille-palle
938 # Referer replaced with: Referer: pille-palle
939 # crunched x-forwarded-for!
941 # modified # XXX: pretty stupid log message
942 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
943 # force-text-mode overruled the client's request to fetch without filtering!
944 # Server time in the future.
945 # content-disposition header crunched and replaced with: content-disposition: filename=baz
946 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
947 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
948 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
949 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
951 # A HTTP/1.1 response without Connection header implies keep-alive.
952 # Disabled filter mode on behalf of the client.
953 # Keeping the server header 'Connection: keep-alive' around.
954 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
955 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
956 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
957 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
958 # Removing 'Connection: close' to imply keep-alive.
959 # keep-alive support is disabled. Crunching: Keep-Alive: 300.
960 # Continue hack in da house.
962 } elsif ($c =~ m/^scanning headers for:/) {
964 return '' unless SHOW_SCAN_INTRO;
966 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
967 # crunched User-Agent!
968 # Crunching: Content-Encoding: gzip
970 $c =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
972 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
974 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
976 $c = h('warning') . $c . h('Standard');
978 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
980 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
981 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
983 $c =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
984 $c =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
986 } elsif ($c =~ m/^Removing empty header/) {
988 # Removing empty header
991 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
993 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
994 # Enable force-text-mode if you know what you're doing.
995 # XXX: Could highlight more here.
996 $c =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
998 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
1000 # Server keep-alive timeout is 5. Sticking with 10.
1001 # Client keep-alive timeout is 20. Sticking with 10.
1003 $c =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1004 $c =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1006 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1008 # Reducing keep-alive timeout from 60 to 10.
1010 $c =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1011 $c =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1015 found_unknown_content($c);
1019 unless ($c =~ m/^Transforming/) {
1020 $c = highlight_known_headers($c) unless $no_special_header_highlighting;
1026 sub handle_loglevel_re_filter ($) {
1028 my $content = shift;
1032 if ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1034 # XXX: only the second version gets highlighted properly.
1035 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1036 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1037 $req{$t}{'content_source'} = $1;
1038 $req{$t}{'content_size'} = $2;
1039 $req{$t}{'content_filter'} = $3;
1040 $req{$t}{'content_hits'} = $4;
1041 $req{$t}{'new_content_size'} = $5;
1042 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1043 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1044 if ($req{$t}{'content_hits'} == 0 and
1045 not (cli_option_is_set('show-ineffective-filters')
1046 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1050 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1051 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1052 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1054 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1055 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1058 $c .= "(" . $h{'Number'};
1059 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1060 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1063 } elsif ($c =~ /\.{3}$/
1064 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1066 # Used by Privoxy 3.0.5 and 3.0.6:
1068 # Used by Privoxy 3.0.7:
1069 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1071 $req{$t}{'filtered_header'} = $1;
1072 $req{$t}{'old_header_size'} = $2;
1073 $req{$t}{'header_filter_name'} = $3;
1075 unless (cli_option_is_set('show-ineffective-filters') or
1076 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1079 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1080 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1082 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1084 # ...produced 0 hits (new size 23).
1085 #... produced 1 hits (new size 54).
1087 $req{$t}{'header_filter_hits'} = $1;
1088 $req{$t}{'new_header_size'} = $2;
1090 unless (cli_option_is_set('show-ineffective-filters') or
1091 (defined($req{$t}{'header_filter_name'}) and
1092 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1094 if ($req{$t}{'header_filter_hits'} == 0 and
1095 not (defined($req{$t}{'header_filter_name'}) and
1096 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1099 # Reformat including information from the intro
1100 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1102 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1103 # $c .= $req{$t}{'filtered_header'} . " ";
1104 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1105 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1107 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1109 $c .= "changing size from ";
1110 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1112 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1117 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1121 # Highlight from last line (XXX: What?)
1122 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1123 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1128 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1129 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1134 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1136 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1137 # Tagger variable-test has empty joblist. Nothing to do.
1139 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1141 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1143 $req{$t}{'chunked-size'} = $1;
1144 $req{$t}{'dechunked-size'} = $2;
1145 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1147 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1149 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1150 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1152 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1154 # Decompression successful. Old size: 670, new size: 1166.
1156 $req{$t}{'size-compressed'} = $1;
1157 $req{$t}{'size-decompressed'} = $2;
1158 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1160 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1161 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1163 # XXX: Create sub get_percentage()
1164 if ($req{$t}{'size-decompressed'}) {
1165 $req{$t}{'decompression-gain-percent'} =
1166 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1168 $content .= " (saved: ";
1169 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1171 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1175 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1177 # Need to de-chunk first
1178 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1180 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1182 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1184 # Adding re_filter job ...
1185 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1186 # to filter client-ip-address succeeded.
1188 } elsif ($c =~ m/^Reading in filter/) {
1190 return '' unless SHOW_FILTER_READIN_IN;
1194 found_unknown_content($content);
1201 sub handle_loglevel_redirect ($) {
1205 if ($c =~ m/^Decoding "([^""]*)"/) {
1207 $req{$t}{'original-destination'} = $1;
1208 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1211 } elsif ($c =~ m/^Checking/) {
1213 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1214 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1216 # TODO: Change colour if really url-decoded
1217 $req{$t}{'decoded-original-destination'} = $1;
1218 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1221 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1223 # pcrs command "s@&from=rss@@" changed \
1224 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1225 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1226 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1227 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1228 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1229 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1230 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1231 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1233 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1235 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1236 # change "http://www.example.org/".
1237 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1238 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1240 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1242 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1243 # XXX: Use URL highlighter
1245 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1247 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1248 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1249 # is already properly formatted.
1250 # XXX: assume the same?
1251 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1255 found_unknown_content($c);
1262 sub handle_loglevel_gif_deanimate ($) {
1264 my $content = shift;
1266 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1268 my $bytes_from = $1;
1270 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1271 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1272 # XXX: Do we need g in case of ($1 == $2)?
1273 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1275 } elsif ($content =~ m/GIF (not) changed/) {
1277 # Gif-Deanimate: GIF not changed.
1278 return '' if SUPPRESS_GIF_NOT_CHANGED;
1279 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1281 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1283 # failed! (gif parsing)
1284 # XXX: Replace this error message with something less stupid
1285 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1287 } elsif ($content =~ m/^Need to de-chunk first/) {
1289 # Need to de-chunk first
1290 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1292 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1294 # No GIF header found (XXX: Did I ever commit this?)
1295 # failed while parsing 195 134747048 (XXX: never commited)
1297 # Ignore these for now
1301 found_unknown_content($content);
1308 sub handle_loglevel_request ($) {
1310 my $content = shift;
1312 if ($content =~ m/crunch! /) {
1314 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1316 # Highlight crunch reasons
1317 foreach my $reason (keys %reason_colours) {
1318 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1320 # Highlight request URL domain and ditch 'crunch!'
1321 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1322 $content =~ s@ crunch!@@;
1324 } elsif ($content =~ m/\[too long, truncated\]$/) {
1326 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1327 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1329 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1331 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1332 $content = h('request_') . $content . h('Standard');
1336 found_unknown_content($content);
1343 sub handle_loglevel_crunch ($) {
1345 my $content = shift;
1347 # Highlight crunch reason
1348 foreach my $reason (keys %reason_colours) {
1349 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1352 if ($content =~ m/\[too long, truncated\]$/) {
1354 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1355 # [...]&filter... [too long, truncated]
1356 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1360 # Blocked: http://ads.example.org/
1361 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1367 sub handle_loglevel_connect ($) {
1371 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1373 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1375 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1376 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1378 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1380 # connect to: www.example.org.noconnect failed: Operation not permitted
1382 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1384 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1386 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1388 # Connect: to www.nzherald.co.nz successful
1389 # Connect: to archiv.radiotux.de
1391 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1392 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1394 } elsif ($c =~ m/^to ([^\s]*)$/) {
1396 # Connect: to lists.sourceforge.net:443
1398 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1400 } elsif ($c =~ m/^accepted connection from .*/ or
1403 # accepted connection from 10.0.0.1
1404 # Privoxy 3.0.6 and earlier just say:
1406 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1407 $c = highlight_matched_host($c, '(?<=connection from ).*');
1409 } elsif ($c =~ m/^write header to: .* failed:/) {
1411 # write header to: 10.0.0.1 failed: Broken pipe
1413 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1414 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1416 } elsif ($c =~ m/^write header to client failed:/) {
1418 # write header to client failed: Broken pipe
1420 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1422 } elsif ($c =~ m/^socks4_connect:/) {
1424 # socks4_connect: SOCKS request rejected or failed.
1425 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1427 } elsif ($c =~ m/^Listening for new connections/ or
1428 $c =~ m/^accept connection/) {
1430 # Privoxy versions above 3.0.6 say:
1431 # Listening for new connections ...
1432 # earlier versions say:
1433 # accept connection ...
1436 } elsif ($c =~ m/^accept failed:/) {
1438 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1440 } elsif ($c =~ m/^Overriding forwarding settings/) {
1442 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1443 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1445 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1447 # Denying suspicious CONNECT request from 10.0.0.1
1448 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1450 } elsif ($c =~ m/^socks5_connect:/) {
1452 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1454 } elsif ($c =~ m/^Created new connection to/) {
1456 # Created new connection to www.privoxy.org:80 on socket 11.
1457 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1458 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1460 } elsif ($c =~ m/^Found reusable socket/) {
1462 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1464 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1465 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1466 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1467 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1468 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1469 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1470 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1471 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1473 } elsif ($c =~ m/^Marking open socket/) {
1475 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1476 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1477 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1478 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1480 } elsif ($c =~ m/^No reusable/) {
1482 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1483 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1485 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1487 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1488 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1490 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1491 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1492 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1494 } elsif ($c =~ m/^Socket/) {
1496 # Socket 16 already forgotten or never remembered.
1497 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1499 } elsif ($c =~ m/^The connection to/) {
1501 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1503 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1504 # Timeout is: 1. Assumed latency: 4.
1505 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1506 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1507 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1508 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1509 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1510 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1512 } elsif ($c =~ m/^Stopped waiting for the request line./) {
1514 # Stopped waiting for the request line. Timeout: 121.
1515 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1517 } elsif ($c =~ m/^Waiting for \d/) {
1519 # Waiting for 1 connections to timeout.
1520 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1522 } elsif ($c =~ m/^Initialized/) {
1524 # Initialized 20 socket slots.
1525 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1527 } elsif ($c =~ m/^Done reading from server/) {
1529 # Done reading from server. Expected content length: 24892. \
1530 # Actual content length: 24892. Most recently received: 4412.
1532 # Done reading from server. Expected content length: 24892. \
1533 # Actual content length: 24892. Bytes most recently read: 4412.
1534 # Done reading from server. Content length: 6018 as expected. \
1535 # Bytes most recently read: 294.
1536 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1537 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1538 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1540 } elsif ($c =~ m/^Continuing buffering headers/) {
1542 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1543 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1544 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1545 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1547 # Continuing buffering headers. Bytes most recently read: %d.
1548 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1550 } elsif ($c =~ m/^Received \d+ bytes while/) {
1552 # Received 206 bytes while expecting 12103.
1553 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1554 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1556 } elsif ($c =~ m/^Connection from/) {
1558 # Connection from 81.163.28.218 dropped due to ACL
1559 $c =~ s@(?<=^Connection from )((?:\d+\.?){4})@$h{'Number'}$1$h{'Standard'}@;
1561 } elsif ($c =~ m/^(?:Reusing|Closing) server socket \d./ or
1562 $c =~ m/^No additional client request/) {
1564 # Reusing server socket 4. Opened for 10.0.0.1.
1565 # Closing server socket 2. Opened for 10.0.0.1.
1566 # No additional client request received in time. \
1567 # Closing server socket 4, initially opened for 10.0.0.1.
1568 # No additional client request received in time on socket 29.
1570 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1571 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.$)');
1573 } elsif ($c =~ m/^Connected to /) {
1575 # Connected to tor-jail[10.0.0.2]:9050.
1577 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1578 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1579 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1581 } elsif ($c =~ m/^Could not connect to /) {
1583 # Could not connect to [10.0.0.1]:80.
1585 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1586 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1588 } elsif ($c =~ m/^Waiting for the next client request/ or
1589 $c =~ m/^The connection on server socket/ or
1590 $c =~ m/^Client request arrived in time or the client closed the connection/) {
1592 # Waiting for the next client request on socket 3. Keeping the server \
1593 # socket 12 to a.fsdn.com open.
1594 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1595 # Client request arrived in time or the client closed the connection on socket 12.
1597 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1598 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1599 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1601 } elsif ($c =~ m/^Marking the server socket/) {
1603 # Marking the server socket 7 tainted.
1605 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1607 } elsif ($c =~ m/^Reduced expected bytes to /) {
1609 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1610 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1611 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1613 } elsif ($c =~ m/^The client closed socket /) {
1615 # The client closed socket 2 while the server socket 4 is still open.
1616 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1617 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1619 } elsif ($c =~ m/^Expected client content length set /) {
1621 # Expected client content length set to 667325411 after reading 4999 bytes.
1622 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1623 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1625 } elsif ($c =~ m/^Waiting for up to /) {
1627 # Waiting for up to 4999 bytes from the client.
1628 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1630 } elsif ($c =~ m/^Looks like we / or
1631 $c =~ m/^Unsetting keep-alive flag/ or
1632 $c =~ m/^No connections to wait/ or
1633 $c =~ m/^Complete client request received/ or
1634 $c =~ m/^Possible pipeline attempt detected./ or
1635 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1636 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1637 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1638 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1639 $c =~ m/^Stopping to watch the client socket. There's already another request waiting./ or
1640 $c =~ m/^Done reading from the client\.$/) {
1642 # Looks like we reached the end of the last chunk. We better stop reading.
1643 # Looks like we read the end of the last chunk together with the server \
1644 # headers. We better stop reading.
1645 # Looks like we got the last chunk together with the server headers. \
1646 # We better stop reading.
1647 # Unsetting keep-alive flag.
1648 # No connections to wait for left.
1649 # Client request arrived in time or the client closed the connection.
1650 # Complete client request received
1651 # Possible pipeline attempt detected. The connection will not be \
1652 # kept alive and we will only serve the first request.
1653 # POST request detected. The connection will not be kept alive.
1654 # The server still wants to talk, but the client hung up on us.
1655 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1656 # There might be a request body. The connection will not be kept alive.
1657 # Stopping to watch the client socket. There's already another request waiting.
1658 # Done reading from the client\.
1662 found_unknown_content($c);
1670 sub handle_loglevel_info ($) {
1674 if ($c =~ m/^Rewrite detected:/) {
1676 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1677 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1679 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1680 $c =~ m/^Compressed content detected/ or
1683 # Decompressing deflated iob: 117
1684 # Decompression didn't result in any content.
1685 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1686 # with zlib support or enable the prevent-compression action.
1687 # Tagger 'complete-url' created empty tag. Ignored.
1691 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1693 # loading configuration file '/usr/local/etc/privoxy/config':
1694 # Reloading configuration file '/usr/local/etc/privoxy/config'
1695 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1697 } elsif ($c =~ m/^Loading (actions|filter) file: /) {
1699 # Loading actions file: /usr/local/etc/privoxy/default.action
1700 # Loading filter file: /usr/local/etc/privoxy/default.filter
1701 $c =~ s@(?<= file: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1703 } elsif ($c =~ m/^exiting by signal/) {
1705 # exiting by signal 15 .. bye
1706 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1708 } elsif ($c =~ m/^Privoxy version/) {
1710 # Privoxy version 3.0.7
1711 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1713 } elsif ($c =~ m/^Program name: /) {
1715 # Program name: /usr/local/sbin/privoxy
1716 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1718 } elsif ($c =~ m/^Listening on port /) {
1720 # Listening on port 8118 on IP address 10.0.0.1
1721 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1722 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1724 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1726 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1727 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1729 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1731 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1732 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1733 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1735 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1736 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1737 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1738 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1740 } elsif ($c =~ m/^Status code/) {
1742 # Status code 304 implies no body.
1743 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1745 } elsif ($c =~ m/^Method/) {
1747 # Method HEAD implies no body.
1748 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1750 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1752 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1753 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1754 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1756 } elsif ($c =~ m/^No logfile configured/ or
1757 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1758 $c =~ m/^Invalid \"chunked\" transfer/ or
1759 $c =~ m/^Support for/ or
1760 $c =~ m/^Flushing header and buffers/ or
1761 $c =~ m/^Can not resolve/
1764 # No logfile configured. Please enable it before reporting any problems.
1765 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1766 # response or even no response at all.
1767 # No logfile configured. Logging disabled.
1768 # Invalid "chunked" transfer encoding detected and ignored.
1769 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1770 # known not to work properly in some situations.
1771 # Flushing header and buffers. Stepping back from filtering.
1772 # Can not resolve doesnotexist: hostname nor servname provided, or not known
1776 found_unknown_content($c);
1783 sub handle_loglevel_cgi ($) {
1787 if ($c =~ m/^Granting access to/) {
1789 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1791 } elsif ($c =~ m/^Substituting: s(.)/) {
1793 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1794 # XXX: prone to span several lines
1797 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1798 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1799 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1805 sub handle_loglevel_force ($) {
1809 if ($c =~ m/^Ignored force prefix in request:/) {
1811 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1812 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1813 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1815 } elsif ($c =~ m/^Enforcing request:/) {
1817 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1818 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1822 found_unknown_content($c);
1829 sub handle_loglevel_error ($) {
1833 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1835 # Empty server or forwarder response received on socket 4.
1836 # Empty server or forwarder response received on socket 3. \
1837 # Closing client socket 15 without sending data.
1838 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1839 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1841 # XXX: There are probably more messages that deserve highlighting.
1847 sub handle_loglevel_ignore ($) {
1851 sub gather_loglevel_request_stats ($$) {
1859 sub gather_loglevel_crunch_stats ($$) {
1869 sub gather_loglevel_error_stats ($$) {
1876 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1878 # Empty server or forwarder response received on socket 4.
1879 $stats{'empty-responses'}++;
1880 if ($thread_data{$thread}{'new_connection'}) {
1881 $stats{'empty-responses-on-new-connections'}++;
1883 $stats{'empty-responses-on-reused-connections'}++;
1888 sub gather_loglevel_connect_stats ($$) {
1890 my ($c, $thread) = @_;
1894 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
1896 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1897 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
1899 } elsif ($c =~ m/^to ([^\s]*)$/) {
1901 # Connect: to lists.sourceforge.net:443
1903 $thread_data{$thread}{'forwarder'} = 'direct connection';
1905 } elsif ($c =~ m/^Created new connection to/) {
1907 # Created new connection to www.privoxy.org:80 on socket 11.
1909 $thread_data{$thread}{'new_connection'} = 1;
1911 } elsif ($c =~ m/^Reusing server socket \d./ or
1912 $c =~ m/^Found reusable socket/) {
1914 # Reusing server socket 4. Opened for 10.0.0.1.
1915 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1917 $thread_data{$thread}{'new_connection'} = 0;
1918 $stats{'reused-connections'}++;
1922 sub gather_loglevel_header_stats ($$) {
1924 my ($c, $thread) = @_;
1927 if ($c =~ m/^A HTTP\/1\.1 response without/ or
1928 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
1930 # A HTTP/1.1 response without Connection header implies keep-alive.
1931 # Keeping the server header 'Connection: keep-alive' around.
1932 $stats{'server-keep-alive'}++;
1940 'server-keep-alive' => 0,
1941 'reused-connections' => 0,
1942 'empty-responses' => 0,
1943 'empty-responses-on-new-connections' => 0,
1944 'empty-responses-on-reused-connections' => 0,
1948 sub get_percentage ($$) {
1951 return "NaN" if ($big eq 0);
1952 return sprintf("%.2f%%", $small / $big * 100);
1955 sub print_stats () {
1958 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
1959 my $outgoing_requests = $stats{requests} - $stats{crunches};
1961 if ($stats{requests} eq 0) {
1962 print "No requests yet.\n";
1966 print "Client requests total: " . $stats{requests} . "\n";
1967 print "Crunches: " . $stats{crunches} . " (" .
1968 get_percentage($stats{requests}, $stats{crunches}) . ")\n";
1969 print "Outgoing requests: " . $outgoing_requests . " (" .
1970 get_percentage($stats{requests}, $outgoing_requests) . ")\n";
1971 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
1972 get_percentage($stats{requests}, $stats{'server-keep-alive'}) . ")\n";
1973 print "New outgoing connections: " . $new_connections . " (" .
1974 get_percentage($stats{requests}, $new_connections) . ")\n";
1975 print "Reused connections: " . $stats{'reused-connections'} . " (" .
1976 get_percentage($stats{requests}, $stats{'reused-connections'}) . ")\n";
1977 print "Empty responses: " . $stats{'empty-responses'} . " (" .
1978 get_percentage($stats{requests}, $stats{'empty-responses'}) . ")\n";
1979 print "Empty responses on new connections: "
1980 . $stats{'empty-responses-on-new-connections'} . " (" .
1981 get_percentage($stats{requests}, $stats{'empty-responses-on-new-connections'})
1983 print "Empty responses on reused connections: " .
1984 $stats{'empty-responses-on-reused-connections'} . " (" .
1985 get_percentage($stats{requests}, $stats{'empty-responses-on-reused-connections'}) .
1990 ################################################################################
1991 # Functions that actually print stuff
1992 ################################################################################
1994 sub print_clf_message () {
1996 our ($ip, $timestamp, $request_line, $status_code, $size);
1999 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2001 # Rebuild highlighted
2002 $output .= $h{'Number'} . $ip . $h{'Standard'};
2004 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2006 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2008 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2010 $output .= $h{'Number'} . $size . $h{'Standard'};
2011 $output .= $line_end;
2016 sub print_non_clf_message ($) {
2018 my $content = shift;
2019 my $msec_string = "." . $req{$t}{'msecs'} unless $no_msecs_mode;
2020 my $line_start = $html_output_mode ? '' : $h{"Standard"};
2022 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2025 . $time_colours[$time_colour_index % 2]
2026 . $req{$t}{'time-stamp'}
2028 . $h{Standard} . " "
2029 . $thread_colours{$t}
2033 . $h{$req{$t}{'log-level'}}
2034 . $req{$t}{'log-level'}
2041 sub shorten_thread_id ($) {
2043 my $thread_id = shift;
2045 our %short_thread_ids;
2048 unless (defined $short_thread_ids{$thread_id}) {
2049 $short_thread_ids{$thread_id} = sprintf "%.3d", $max_threadid++;
2052 return $short_thread_ids{$thread_id}
2057 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2059 my $last_thread = 0;
2060 my $last_timestamp = 0;
2061 my $filters_that_did_nothing;
2064 $time_colour = paint_it('white');
2066 my %log_level_handlers = (
2067 'Re-Filter' => \&handle_loglevel_re_filter,
2068 'Header' => \&handle_loglevel_header,
2069 'Connect' => \&handle_loglevel_connect,
2070 'Redirect' => \&handle_loglevel_redirect,
2071 'Request' => \&handle_loglevel_request,
2072 'Crunch' => \&handle_loglevel_crunch,
2073 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2074 'Info' => \&handle_loglevel_info,
2075 'CGI' => \&handle_loglevel_cgi,
2076 'Force' => \&handle_loglevel_force,
2077 'Error' => \&handle_loglevel_error,
2078 'Fatal error' => \&handle_loglevel_ignore,
2079 'Writing' => \&handle_loglevel_ignore,
2080 'Received' => \&handle_loglevel_ignore,
2081 'Unknown log level' => \&handle_loglevel_ignore,
2086 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2087 $thread = $t = ($shorten_thread_ids) ? shorten_thread_id($4) : $4;
2088 $req{$t}{'day'} = $day = $1;
2089 $req{$t}{'time-stamp'} = $time_stamp = $2;
2090 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2091 $req{$t}{'log-level'} = $log_level = $5;
2092 $req{$t}{'content'} = $content = $c = $6;
2093 $req{$t}{'log-message'} = $_;
2094 $no_special_header_highlighting = 0;
2096 if (defined($log_level_handlers{$log_level})) {
2098 $content = $log_level_handlers{$log_level}($content);
2102 die "No handler found for log level \"$log_level\"\n";
2105 # Highlight Truncations
2106 if (length($_) > 4000) {
2107 $content =~ s@(too long, truncated)]$@$h{'Truncation'}$1$h{'Standard'}]@g;
2110 next unless $content;
2112 # Register threads to keep the colour constant
2113 if (!defined($thread_colours{$thread})) {
2114 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2115 $thread_colour_index++;
2118 # Switch timestamp colour if timestamps differ
2119 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2120 debug_message("Tick tack!") if DEBUG_TICKS;
2121 $time_colour = $time_colours[$time_colour_index % 2];
2122 $time_colour_index++;
2123 $last_msecs = $msecs;
2124 $last_timestamp = $time_stamp;
2127 $last_thread = $thread;
2129 print_non_clf_message($content);
2131 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2133 # LOG_LEVEL_CLF lines look like this
2134 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2135 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2136 # HTTP/1.1" 403 1730
2137 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2139 print_clf_message();
2143 # Some Privoxy log messages span more than one line,
2144 # usually to dump lots of content that doesn't need any syntax highlighting.
2145 # XXX: add mechanism to forward these lines to the right handler anyway.
2147 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2148 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2156 my ($day, $time_stamp, $msecs, $thread, $log_level, $content);
2157 my %log_level_handlers = (
2158 'Re-Filter' => \&handle_loglevel_ignore,
2159 'Header' => \&gather_loglevel_header_stats,
2160 'Connect' => \&gather_loglevel_connect_stats,
2161 'Redirect' => \&handle_loglevel_ignore,
2162 'Request' => \&gather_loglevel_request_stats,
2163 'Crunch' => \&gather_loglevel_crunch_stats,
2164 'Gif-Deanimate' => \&handle_loglevel_ignore,
2165 'Info' => \&handle_loglevel_ignore,
2166 'CGI' => \&handle_loglevel_ignore,
2167 'Force' => \&handle_loglevel_ignore,
2168 'Error' => \&gather_loglevel_error_stats,
2169 'Fatal error' => \&handle_loglevel_ignore,
2170 'Writing' => \&handle_loglevel_ignore,
2171 'Unknown log level' => \&handle_loglevel_ignore
2175 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2178 $msecs = $3 ? $3 : 0;
2183 if (defined($log_level_handlers{$log_level})) {
2185 $content = $log_level_handlers{$log_level}($content, $thread);
2189 die "No handler found for log level \"$log_level\"\n";
2199 sub VersionMessage {
2200 my $version_message;
2202 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2203 $version_message .= 'Copyright (C) 2007-2010 Fabian Keil <fk@fabiankeil.de>' . "\n";
2204 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2206 print $version_message;
2209 sub get_cli_options () {
2211 our %cli_options = (
2212 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2213 'title' => CLI_OPTION_TITLE,
2214 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2215 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2216 'no-msecs' => CLI_OPTION_NO_MSECS,
2217 'shorten-thread-ids' => CLI_OPTION_SHORTEN_THREAD_IDS,
2218 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2219 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
2220 'statistics' => CLI_OPTION_STATISTICS,
2224 'html-output' => \$cli_options{'html-output'},
2225 'title' => \$cli_options{'title'},
2226 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2227 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2228 'no-msecs' => \$cli_options{'no-msecs'},
2229 'shorten-thread-ids' => \$cli_options{'shorten-thread-ids'},
2230 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2231 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
2232 'statistics' => \$cli_options{'statistics'},
2233 'version' => sub { VersionMessage && exit(0) },
2237 $html_output_mode = cli_option_is_set('html-output');
2238 $no_msecs_mode = cli_option_is_set('no-msecs');
2239 $shorten_thread_ids = cli_option_is_set('shorten-thread-ids');
2240 $line_end = get_line_end();
2251 Options and their default values if they have any:
2252 [--accept-unknown-messages]
2256 [--no-syntax-highlighting]
2257 [--shorten-thread-ids]
2258 [--show-ineffective-filters]
2260 [--title $cli_options{'title'}]
2262 see "perldoc $0" for more information
2268 ################################################################################
2270 ################################################################################
2274 set_background(DEFAULT_BACKGROUND);
2275 prepare_our_stuff();
2279 if (cli_option_is_set('statistics')) {
2292 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2296 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
2297 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--statistics>]
2298 [B<--shorten-thread-ids>] [B<--show-ineffective-filters>] [B<--version>]
2302 B<privoxy-log-parser> reads Privoxy log messages and
2304 - syntax-highlights recognized lines,
2306 - reformats some of them for easier comprehension,
2308 - filters out less useful messages, and
2310 - (in some cases) calculates additional information,
2311 like the compression ratio or how a filter affected
2314 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2315 without getting confused by the resulting amount of output. For example for
2316 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2317 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2318 will hide the "filter foo caused 0 hits" message.
2322 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
2323 just don't highlight them.
2325 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2326 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2327 This option is only intended to make embedding log excerpts in web pages easier.
2328 It does not escape any input!
2330 [B<--no-msecs>] Don't expect milisecond resolution
2332 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2333 the filtered output is piped into less in which case the ANSI control
2334 codes don't work, or if the terminal itself doesn't support the control
2337 [B<--shorten-thread-ids>] Shorten the thread ids to a three-digit decimal number.
2338 Note that the mapping from thread ids to shortened ids is created at run-time
2339 and thus varies with the input.
2341 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2342 that didn't modify the content.
2344 [B<--statistics>] Gather various statistics instead of syntax highlighting
2345 log messages. This is an experimental feature, if the results look wrong
2346 they very well might be. Also note that the results are pretty much guaranteed
2347 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2349 [B<--version>] Print version and exit.
2353 To monitor a log file:
2355 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2357 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2358 or if the log won't get rotated anyway. The log file location depends
2359 on your system (Doh!).
2361 To monitor Privoxy without having it write to a log file:
2363 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2365 Again, the config file location depends on your system. Output redirection
2366 depends on your shell, the above works with bourne shells.
2368 To read a processed Privoxy log file from top to bottom, letting the content
2369 scroll by slightly faster than you can read:
2371 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2373 This is probably only useful to fill screens in the background of haxor movies.
2377 Syntax highlighting with ANSI escape sequences will look strange
2378 if your background color isn't black.
2380 Some messages aren't recognized yet and will not be fully highlighted.
2382 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2383 using earlier Privoxy versions will probably result in an increased amount
2384 of unrecognized log lines.
2386 Privoxy's log files tend to be rather large. If you use HTML
2387 highlighting some browsers can't handle them, get confused and
2388 will eventually crash because of segmentation faults or unexpected
2389 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2394 Many settings can't be controlled through command line options yet.
2402 Fabian Keil <fk@fabiankeil.de>