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 1.89 2010/08/28 13:20:23 fabiankeil Exp $
14 # - LOG_LEVEL_CGI, LOG_LEVEL_ERROR, LOG_LEVEL_WRITE content highlighting
15 # - create fancy statistics
16 # - grep through Privoxy sources to find unsupported log messages
17 # - hunt down substitutions that match content from variables which
18 # can contain stuff like ()?'[]
19 # - replace $h{'foo'} with h('foo') where possible
20 # - hunt down XXX comments instead of just creating them
21 # - add example log lines for every regex and mark them up for
23 # - Handle incomplete input without Perl warning about undefined variables.
24 # - Use generic highlighting function that takes a regex and the
26 # - Add --compress and --decompress options.
28 # Copyright (c) 2007-2010 Fabian Keil <fk@fabiankeil.de>
30 # Permission to use, copy, modify, and distribute this software for any
31 # purpose with or without fee is hereby granted, provided that the above
32 # copyright notice and this permission notice appear in all copies.
34 # THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
35 # WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
36 # MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
37 # ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
38 # WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
39 # ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
40 # OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
41 ################################################################################
48 PRIVOXY_LOG_PARSER_VERSION => '0.6',
49 # Feel free to mess with these ...
50 DEFAULT_BACKGROUND => 'black', # Choose registered colour (like 'black')
51 DEFAULT_TEXT_COLOUR => 'white', # Choose registered colour (like 'black')
52 HEADER_DEFAULT_COLOUR => 'yellow',
53 REGISTER_HEADERS_WITH_THE_SAME_COLOUR => 1,
55 CLI_OPTION_DEFAULT_TO_HTML_OUTPUT => 0,
56 CLI_OPTION_TITLE => 'Privoxy-Log-Parser in da house',
57 CLI_OPTION_NO_EMBEDDED_CSS => 0,
58 CLI_OPTION_NO_MSECS => 0,
59 CLI_OPTION_NO_SYNTAX_HIGHLIGHTING => 0,
60 CLI_OPTION_SHORTEN_THREAD_IDS => 0,
61 CLI_OPTION_SHOW_INEFFECTIVE_FILTERS => 0,
62 CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES => 0,
63 CLI_OPTION_STATISTICS => 0,
65 SUPPRESS_SUCCEEDED_FILTER_ADDITIONS => 1,
67 SHOW_FILTER_READIN_IN => 0,
68 SUPPRESS_EMPTY_LINES => 1,
69 SUPPRESS_SUCCESSFUL_CONNECTIONS => 1,
70 SUPPRESS_ACCEPTED_CONNECTIONS => 1,
71 SUPPRESS_GIF_NOT_CHANGED => 1,
72 SUPPRESS_NEED_TO_DE_CHUNK_FIRST => 1,
74 DEBUG_HEADER_REGISTERING => 0,
75 DEBUG_HEADER_HIGHLIGHTING => 0,
78 DEBUG_SUPPRESS_LOG_MESSAGES => 0,
80 PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH => 0,
81 PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH => 1,
83 LOG_UNPARSED_LINES_TO_EXTRA_FILE => 0,
84 ERROR_LOG_FILE => '/var/log/privoxy-log-parser',
86 # You better leave these alone unless you know what you're doing.
87 COLOUR_RESET => "\033[0;0m",
91 # For performance reasons, these are global.
94 my %req; # request data from previous lines
99 my $thread_colour_index = 0;
100 my $header_colour_index = 0;
101 my $time_colour_index = 0;
103 my $no_special_header_highlighting;
106 my $header_highlight_regex = '';
108 my $html_output_mode;
109 my $no_msecs_mode; # XXX: should probably be removed
110 my $shorten_thread_ids;
113 sub prepare_our_stuff () {
115 # Syntax Higlight hash
117 'red', 'green', 'brown', 'blue', 'purple', 'cyan',
118 'light_gray', 'light_red', 'light_green', 'yellow',
119 'light_blue', 'pink', 'light_cyan', 'white'
126 Filter => 'purple', # XXX: Used?
127 'Re-Filter' => 'purple',
129 Request => 'light_cyan',
130 CGI => 'light_green',
132 Error => 'light_red',
134 'Fatal error' => 'light_red',
135 'Gif-Deanimate' => 'blue',
137 Writing => 'light_green',
138 Received => 'yellow',
139 # ----------------------
142 request_ => 'brown', # host+path but no protocol
143 'ip-address' => 'yellow',
146 Truncation => 'light_red',
148 Timestamp => 'brown',
149 Crunching => 'light_red',
150 crunched => 'light_red',
151 'Request-Line' => 'pink',
153 destination => 'yellow',
154 'http-version' => 'pink',
155 'crunch-pattern' => 'pink',
160 'program-name' => 'cyan',
163 warning => 'light_red',
164 debug => 'light_red',
168 'status-message' => 'light_cyan',
169 'status-code' => 'yellow',
170 'invalid-request' => 'light_red',
172 error => 'light_red',
173 'rewritten-URL' => 'light_red',
174 'pcrs-delimiter' => 'light_red',
175 'ignored' => 'light_red',
176 'action-bits-update' => 'light_red',
177 'configuration-line' => 'red',
178 'content-type' => 'yellow',
183 # Header colours need their own hash so the keys can be accessed properly
185 # Prefilled with headers that should not appear with default header colours
186 Cookie => 'light_red',
187 'Set-Cookie' => 'light_red',
188 Warning => 'light_red',
189 Default => HEADER_DEFAULT_COLOUR,
192 # Crunch reasons need their own hash as well
194 'Unsupported HTTP feature' => 'light_red',
195 Blocked => 'light_red',
196 Untrusted => 'light_red',
197 Redirected => 'green',
198 'CGI Call' => 'white',
199 'DNS failure' => 'red',
200 'Forwarding failed' => 'light_red',
201 'Connection failure' => 'light_red',
202 'Out of memory (may mask other reasons)' => 'light_red',
203 'No reason recorded' => 'light_red',
206 @time_colours = ('white', 'light_gray');
208 # Translate highlight strings into highlight code
209 prepare_highlight_hash(\%header_colours);
210 prepare_highlight_hash(\%reason_colours);
211 prepare_highlight_hash(\%h);
212 prepare_colour_array(\@all_colours);
213 prepare_colour_array(\@time_colours);
220 ###############################################################
221 # Takes a colour string and returns an ANSI escape sequence
222 # (unless --no-syntax-highlighting is used).
223 # XXX: The Rolling Stones reference has to go.
224 ###############################################################
226 my $colour = shift @_;
228 return "" if cli_option_is_set('no-syntax-highlighting');
270 my $bg_code = get_background();
272 our $default = default_colours();
274 if (defined($text{$colour})) {
275 $colour_code = ESCAPE;
276 $colour_code .= $text{$colour};
278 $colour_code .= $light{$colour} ? "1" : "2";
280 $colour_code .= $bg_code;
282 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
284 } elsif ($colour =~ /reset/) {
286 $colour_code = default_colours();
290 die "What's $colour supposed to mean?\n";
296 sub get_semantic_html_markup ($) {
297 ###############################################################
298 # Takes a string and returns a span element
299 ###############################################################
304 if ($type =~ /Standard/) {
308 $code = '<span title="' . $type . '" class="' . $type . '">';
314 sub cli_option_is_set ($) {
317 my $cli_option = shift;
319 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
321 return $cli_options{$cli_option};
324 sub get_html_title () {
327 return $cli_options{'title'};
331 sub init_css_colours() {
339 purple => "F06", # XXX: wrong
340 cyan => "F09", # XXX: wrong
345 light_green => "33F",
354 sub get_css_colour ($) {
359 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
361 return '#' . $css_colours{$colour};
364 sub get_css_line ($) {
369 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
370 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
371 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
372 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
373 $css_line .= '}' . "\n";
378 sub get_css_line_for_colour ($) {
383 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
384 $css_line .= 'color:' . get_css_colour($colour) . ';';
385 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
386 $css_line .= '}' . "\n";
391 # XXX: Wrong solution
392 sub get_missing_css_lines () {
396 $css_line .= '.' . 'default' . ' {';
397 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
398 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
399 $css_line .= '}' . "\n";
406 our %css_colours; #XXX: Wrong solution
410 $css .= '.privoxy-log {';
411 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
412 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
415 foreach my $key (keys %h_colours) {
417 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
418 $css .= get_css_line($key);
422 foreach my $colour (keys %css_colours) {
424 $css .= get_css_line_for_colour($colour);
428 $css .= get_missing_css_lines(); #XXX: Wrong solution
437 if (cli_option_is_set('html-output')) {
439 my $title = get_html_title();
441 $intro .= '<html><head>';
442 $intro .= '<title>' . $title . '</title>';
443 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
444 $intro .= '</head><body>';
445 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
455 if (cli_option_is_set('html-output')) {
457 $outro = '</p></body></html>';
463 sub get_line_end () {
464 return cli_option_is_set('html-output') ? "<br>\n" : "\n";
467 sub get_colour_html_markup ($) {
468 ###############################################################
469 # Takes a colour string a span element. XXX: WHAT?
470 # XXX: This function shouldn't be necessary, the
471 # markup should always be semantically correct.
472 ###############################################################
477 if ($type =~ /Standard/) {
480 $code = '<span class="' . lc($type) . '">';
486 sub default_colours () {
489 return reset_colours();
492 sub show_colours () {
496 sub reset_colours () {
497 return ESCAPE . "0m";
500 sub set_background ($){
516 if (defined($backgrounds{$colour})) {
517 $bg_code = $backgrounds{$colour};
519 die "Invalid background colour: " . $colour;
523 sub get_background (){
527 sub prepare_highlight_hash ($) {
530 foreach my $key (keys %$ref) {
531 $$ref{$key} = $html_output_mode ?
532 get_semantic_html_markup($key) :
533 paint_it($$ref{$key});
537 sub prepare_colour_array ($) {
540 foreach my $i (0 ... @$ref - 1) {
541 $$ref[$i] = $html_output_mode ?
542 get_colour_html_markup($$ref[$i]) :
547 sub found_unknown_content ($) {
552 return if cli_option_is_set('accept-unknown-messages');
554 return if ($unknown =~ /\[too long, truncated\]$/);
556 $message = "found_unknown_content: Don't know how to highlight: ";
557 # Break line so the log file can later be parsed as Privoxy log file again
558 $message .= '"' . $unknown . '"' . " in:\n";
559 $message .= $req{$t}{'log-message'};
560 debug_message($message);
561 log_parse_error($req{$t}{'log-message'});
563 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
566 sub log_parse_error ($) {
570 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
571 open(ERRORLOG, ">>" . ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
572 print ERRORLOG $message;
577 sub debug_message (@) {
580 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
583 ################################################################################
584 # highlighter functions that aren't loglevel-specific
585 ################################################################################
589 # Get highlight marker
590 my $highlight = shift; # XXX: Stupid name;
594 if (defined($highlight)) {
596 $result = $h{$highlight};
600 $message = "h: Don't recognize highlighter $highlight.";
601 debug_message($message);
602 log_parser_error($message);
603 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
609 sub highlight_known_headers ($) {
613 debug_message("Searching $content for things to highlight.") if DEBUG_HEADER_HIGHLIGHTING;
615 if ($content =~ m/(?<=\s)($header_highlight_regex):/) {
617 $content =~ s@(?<=[\s|'])($header)(?=:)@$header_colours{$header}$1$h{'Standard'}@ig;
618 debug_message("Highlighted '$header' in '$content'") if DEBUG_HEADER_HIGHLIGHTING;
624 sub highlight_matched_request_line ($$) {
626 my $result = shift; # XXX: Stupid name;
628 if ($result =~ m@(.*)($regex)(.*)@) {
629 $result = $1 . highlight_request_line($2) . $3
634 sub highlight_request_line ($) {
637 my ($method, $url, $http_version);
639 #GET http://images.sourceforge.net/sfx/icon_warning.gif HTTP/1.1
640 if ($rl =~ m/Invalid request/) {
642 $rl = h('invalid-request') . $rl . h('Standard');
644 } elsif ($rl =~ m/^([-\w]+) (.*) (HTTP\/\d\.\d)/) {
646 # XXX: might not match in case of HTTP method fuzzing.
647 # XXX: save these: ($method, $path, $http_version) = ($1, $2, $3);
648 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
649 if ($rl =~ /http:\/\//) {
650 $rl = highlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
652 $rl = highlight_matched_pattern($rl, 'request_', '[^\s]*(?=\sHTTP)');
655 $rl =~ s@(HTTP\/\d\.\d)$@$h{'http-version'}$1$h{'Standard'}@;
657 } elsif ($rl =~ m/\.\.\. \[too long, truncated\]$/) {
659 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
660 $rl = highlight_matched_url($rl, '[^\s]*(?=\.\.\.)');
662 } elsif ($rl =~ m/^ $/) {
664 $rl = h('error') . "No request line specified!" . h('Standard');
668 debug_message ("Can't parse request line: $rl");
675 sub highlight_response_line ($) {
678 my ($http_version, $status_code, $status_message);
683 # TODO: Mark different status codes differently
685 if ($rl =~ m/((?:HTTP\/\d\.\d|ICY)) (\d+) (.*)/) {
686 ($http_version, $status_code, $status_message) = ($1, $2, $3);
688 debug_message ("Can't parse response line: $rl") and die 'Fix this';
691 # Rebuild highlighted
693 $rl .= h('http-version') . $http_version . h('Standard');
695 $rl .= h('status-code') . $status_code . h('Standard');
697 $rl .= h('status-message') . $status_message . h('Standard');
702 sub highlight_matched_url ($$) {
704 my $result = shift; # XXX: Stupid name;
707 #print "Got $result, regex ($regex)\n";
709 if ($result =~ m@(.*?)($regex)(.*)@) {
710 $result = $1 . highlight_url($2) . $3;
711 #print "Now the result is $result\n";
717 sub highlight_matched_host ($$) {
719 my ($result, $regex) = @_; # XXX: result ist stupid name;
721 if ($result =~ m@(.*?)($regex)(.*)@) {
722 $result = $1 . $h{host} . $2 . $h{Standard} . $3;
728 sub highlight_matched_pattern ($$$) {
730 my $result = shift; # XXX: Stupid name;
734 die "Unknown key $key" unless defined $h{$key};
736 if ($result =~ m@(.*?)($regex)(.*)@) {
737 $result = $1 . h($key) . $2 . h('Standard') . $3;
743 sub highlight_matched_path ($$) {
745 my $result = shift; # XXX: Stupid name;
748 if ($result =~ m@(.*?)($regex)(.*)@) {
749 $result = $1 . h('path') . $2 . h('Standard') . $3;
755 sub highlight_url ($) {
759 if ($html_output_mode) {
761 $url = '<a href="' . $url . '">' . $url . '</a>';
765 $url = h('URL') . $url . h('Standard');
772 sub update_header_highlight_regex ($) {
775 my $headers = join ('|', keys %header_colours);
777 $header_highlight_regex = qr/$headers/;
778 print "Registering '$header'\n" if DEBUG_HEADER_HIGHLIGHTING;
781 ################################################################################
782 # loglevel-specific highlighter functions
783 ################################################################################
785 sub handle_loglevel_header ($) {
789 if ($c =~ /^scan:/) {
791 if ($c =~ m/^scan: ([^: ]+):/) {
793 # Register new headers
794 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
796 if (!defined($header_colours{$header}) and $header =~ /^[\d\w-]*$/) {
797 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
799 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
800 $header_colours{$header} = $header_colours{'Default'};
802 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
803 $header_colour_index++;
805 update_header_highlight_regex($header);
808 } elsif ($c =~ m/^(scan: )(\w+ .+ HTTP\/\d\.\d)/) {
810 # scan: HTTP/1.1 200 OK
811 $c = $1 . highlight_request_line($2);
813 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
815 # Server response line
816 $req{$t}{'response_line'} = $2;
817 $req{$t}{'status_code'} = $3;
818 $req{$t}{'status_message'} = $4;
819 $c = $1 . highlight_response_line($req{$t}{'response_line'});
822 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
824 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
825 $c =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
826 $c =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
828 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
830 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/!
831 $c = highlight_matched_host($c, '(?<=New host is: )[^\s]+(?=\.)');
832 $c = highlight_matched_url($c, '(?<=Crunching Referer: )[^\s!]+');
834 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
836 # Text mode enabled by force. Take cover!
837 $c =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
839 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
841 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
842 $c = $1 . highlight_request_line($2);
844 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
846 # Adjusted Content-Length to 2132
847 # Adjust Content-Length to 33533
848 $c =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
849 $c = highlight_known_headers($c);
851 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
853 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
854 $c = highlight_matched_url($c, '(?<=New request URL: ).*');
856 } elsif ($c =~ m/^Couldn\'t parse:/) {
858 # XXX: These should probable be logged with LOG_LEVEL_ERROR
859 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
860 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
861 $c =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
863 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
864 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
866 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
867 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
868 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
870 # XXX: Save tag and tagger
872 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
873 $c =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
874 $c =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
875 $no_special_header_highlighting = 1;
877 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
879 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
880 # XXX: Save tag and tagger
882 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
883 $c =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
885 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
886 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
887 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
888 or $c =~ m/^(addh-unique|Referer forged to)/
889 or $c =~ m/^Downgraded answer to HTTP\/1.0/
890 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
891 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
892 or $c =~ m/^Referer crunched!/
893 or $c =~ m/^crunched x-forwarded-for!/
894 or $c =~ m/^crunched From!/
895 or $c =~ m/^ modified$/
896 or $c =~ m/^Content filtering is enabled. Crunching:/
897 or $c =~ m/^force-text-mode overruled the client/
898 or $c =~ m/^Server time in the future\./
899 or $c =~ m/^content-disposition header crunched and replaced with:/i
900 or $c =~ m/^Reducing white space in /
901 or $c =~ m/^Ignoring single quote in /
902 or $c =~ m/^Converting tab to space in /
903 or $c =~ m/A HTTP\/1\.1 response without/
904 or $c =~ m/Disabled filter mode on behalf of the client/
905 or $c =~ m/Keeping the (?:server|client) header /
906 or $c =~ m/Content modified with no Content-Length header set/
907 or $c =~ m/^Appended client IP address to/
908 or $c =~ m/^Removing 'Connection: close' to imply keep-alive./
909 or $c =~ m/^keep-alive support is disabled/
910 or $c =~ m/^Continue hack in da house/
913 # XXX: Some of these may need highlighting
915 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
916 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
917 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
918 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
919 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
920 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
921 # Text mode is already enabled
922 # Denied request with NULL byte(s) turned into line break(s)
923 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
924 # addh-unique: Host: people.freebsd.org
925 # Suppressed offer to compress content
926 # Crunched incoming cookie -- yum!
927 # Accepted the client's request to fetch without filtering.
928 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
929 # addh-unique: Host: subkeys.pgp.net:11371
930 # Referer forged to: Referer: http://10.0.0.1/
931 # Downgraded answer to HTTP/1.0
932 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
933 # Referer overwritten with: Referer: pille-palle
934 # Referer replaced with: Referer: pille-palle
935 # crunched x-forwarded-for!
937 # modified # XXX: pretty stupid log message
938 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
939 # force-text-mode overruled the client's request to fetch without filtering!
940 # Server time in the future.
941 # content-disposition header crunched and replaced with: content-disposition: filename=baz
942 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
943 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
944 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
945 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
947 # A HTTP/1.1 response without Connection header implies keep-alive.
948 # Disabled filter mode on behalf of the client.
949 # Keeping the server header 'Connection: keep-alive' around.
950 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
951 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
952 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
953 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
954 # Removing 'Connection: close' to imply keep-alive.
955 # keep-alive support is disabled. Crunching: Keep-Alive: 300.
956 # Continue hack in da house.
958 } elsif ($c =~ m/^scanning headers for:/) {
960 return '' unless SHOW_SCAN_INTRO;
962 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
963 # crunched User-Agent!
964 # Crunching: Content-Encoding: gzip
966 $c =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
968 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
970 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
972 $c = h('warning') . $c . h('Standard');
974 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
976 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
977 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
979 $c =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
980 $c =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
982 } elsif ($c =~ m/^Removing empty header/) {
984 # Removing empty header
987 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
989 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
990 # Enable force-text-mode if you know what you're doing.
991 # XXX: Could highlight more here.
992 $c =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
994 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
996 # Server keep-alive timeout is 5. Sticking with 10.
997 # Client keep-alive timeout is 20. Sticking with 10.
999 $c =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1000 $c =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1002 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1004 # Reducing keep-alive timeout from 60 to 10.
1006 $c =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1007 $c =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1011 found_unknown_content($c);
1015 unless ($c =~ m/^Transforming/) {
1016 $c = highlight_known_headers($c) unless $no_special_header_highlighting;
1022 sub handle_loglevel_re_filter ($) {
1024 my $content = shift;
1028 if ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1030 # XXX: only the second version gets highlighted properly.
1031 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1032 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1033 $req{$t}{'content_source'} = $1;
1034 $req{$t}{'content_size'} = $2;
1035 $req{$t}{'content_filter'} = $3;
1036 $req{$t}{'content_hits'} = $4;
1037 $req{$t}{'new_content_size'} = $5;
1038 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1039 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1040 if ($req{$t}{'content_hits'} == 0 and
1041 not (cli_option_is_set('show-ineffective-filters')
1042 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1046 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1047 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1048 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1050 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1051 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1054 $c .= "(" . $h{'Number'};
1055 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1056 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1059 } elsif ($c =~ /\.{3}$/
1060 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1062 # Used by Privoxy 3.0.5 and 3.0.6:
1064 # Used by Privoxy 3.0.7:
1065 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1067 $req{$t}{'filtered_header'} = $1;
1068 $req{$t}{'old_header_size'} = $2;
1069 $req{$t}{'header_filter_name'} = $3;
1071 unless (cli_option_is_set('show-ineffective-filters') or
1072 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1075 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1076 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1078 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1080 # ...produced 0 hits (new size 23).
1081 #... produced 1 hits (new size 54).
1083 $req{$t}{'header_filter_hits'} = $1;
1084 $req{$t}{'new_header_size'} = $2;
1086 unless (cli_option_is_set('show-ineffective-filters') or
1087 (defined($req{$t}{'header_filter_name'}) and
1088 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1090 if ($req{$t}{'header_filter_hits'} == 0 and
1091 not (defined($req{$t}{'header_filter_name'}) and
1092 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1095 # Reformat including information from the intro
1096 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1098 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1099 # $c .= $req{$t}{'filtered_header'} . " ";
1100 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1101 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1103 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1105 $c .= "changing size from ";
1106 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1108 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1113 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1117 # Highlight from last line (XXX: What?)
1118 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1119 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1124 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1125 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1130 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1132 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1133 # Tagger variable-test has empty joblist. Nothing to do.
1135 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1137 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1139 $req{$t}{'chunked-size'} = $1;
1140 $req{$t}{'dechunked-size'} = $2;
1141 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1143 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1145 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1146 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1148 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1150 # Decompression successful. Old size: 670, new size: 1166.
1152 $req{$t}{'size-compressed'} = $1;
1153 $req{$t}{'size-decompressed'} = $2;
1154 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1156 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1157 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1159 # XXX: Create sub get_percentage()
1160 if ($req{$t}{'size-decompressed'}) {
1161 $req{$t}{'decompression-gain-percent'} =
1162 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1164 $content .= " (saved: ";
1165 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1167 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1171 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1173 # Need to de-chunk first
1174 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1176 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1178 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1180 # Adding re_filter job ...
1181 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1182 # to filter client-ip-address succeeded.
1184 } elsif ($c =~ m/^Reading in filter/) {
1186 return '' unless SHOW_FILTER_READIN_IN;
1190 found_unknown_content($content);
1197 sub handle_loglevel_redirect ($) {
1201 if ($c =~ m/^Decoding "([^""]*)"/) {
1203 $req{$t}{'original-destination'} = $1;
1204 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1207 } elsif ($c =~ m/^Checking/) {
1209 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1210 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1212 # TODO: Change colour if really url-decoded
1213 $req{$t}{'decoded-original-destination'} = $1;
1214 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1217 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1219 # pcrs command "s@&from=rss@@" changed \
1220 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1221 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1222 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1223 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1224 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1225 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1226 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1227 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1229 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1231 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1232 # change "http://www.example.org/".
1233 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1234 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1236 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1238 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1239 # XXX: Use URL highlighter
1241 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1243 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1244 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1245 # is already properly formatted.
1246 # XXX: assume the same?
1247 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1251 found_unknown_content($c);
1258 sub handle_loglevel_gif_deanimate ($) {
1260 my $content = shift;
1262 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1264 my $bytes_from = $1;
1266 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1267 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1268 # XXX: Do we need g in case of ($1 == $2)?
1269 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1271 } elsif ($content =~ m/GIF (not) changed/) {
1273 # Gif-Deanimate: GIF not changed.
1274 return '' if SUPPRESS_GIF_NOT_CHANGED;
1275 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1277 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1279 # failed! (gif parsing)
1280 # XXX: Replace this error message with something less stupid
1281 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1283 } elsif ($content =~ m/^Need to de-chunk first/) {
1285 # Need to de-chunk first
1286 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1288 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1290 # No GIF header found (XXX: Did I ever commit this?)
1291 # failed while parsing 195 134747048 (XXX: never commited)
1293 # Ignore these for now
1297 found_unknown_content($content);
1304 sub handle_loglevel_request ($) {
1306 my $content = shift;
1308 if ($content =~ m/crunch! /) {
1310 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1312 # Highlight crunch reasons
1313 foreach my $reason (keys %reason_colours) {
1314 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1316 # Highlight request URL domain and ditch 'crunch!'
1317 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1318 $content =~ s@ crunch!@@;
1320 } elsif ($content =~ m/\[too long, truncated\]$/) {
1322 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1323 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1325 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1327 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1328 $content = h('request_') . $content . h('Standard');
1332 found_unknown_content($content);
1339 sub handle_loglevel_crunch ($) {
1341 my $content = shift;
1343 # Highlight crunch reason
1344 foreach my $reason (keys %reason_colours) {
1345 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1348 if ($content =~ m/\[too long, truncated\]$/) {
1350 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1351 # [...]&filter... [too long, truncated]
1352 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1356 # Blocked: http://ads.example.org/
1357 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1363 sub handle_loglevel_connect ($) {
1367 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1369 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1371 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1372 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1374 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1376 # connect to: www.example.org.noconnect failed: Operation not permitted
1378 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1380 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1382 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1384 # Connect: to www.nzherald.co.nz successful
1385 # Connect: to archiv.radiotux.de
1387 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1388 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1390 } elsif ($c =~ m/^to ([^\s]*)$/) {
1392 # Connect: to lists.sourceforge.net:443
1394 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1396 } elsif ($c =~ m/^accepted connection from .*/ or
1399 # accepted connection from 10.0.0.1
1400 # Privoxy 3.0.6 and earlier just say:
1402 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1403 $c = highlight_matched_host($c, '(?<=connection from ).*');
1405 } elsif ($c =~ m/^write header to: .* failed:/) {
1407 # write header to: 10.0.0.1 failed: Broken pipe
1409 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1410 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1412 } elsif ($c =~ m/^write header to client failed:/) {
1414 # write header to client failed: Broken pipe
1416 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1418 } elsif ($c =~ m/^socks4_connect:/) {
1420 # socks4_connect: SOCKS request rejected or failed.
1421 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1423 } elsif ($c =~ m/^Listening for new connections/ or
1424 $c =~ m/^accept connection/) {
1426 # Privoxy versions above 3.0.6 say:
1427 # Listening for new connections ...
1428 # earlier versions say:
1429 # accept connection ...
1432 } elsif ($c =~ m/^accept failed:/) {
1434 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1436 } elsif ($c =~ m/^Overriding forwarding settings/) {
1438 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1439 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1441 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1443 # Denying suspicious CONNECT request from 10.0.0.1
1444 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1446 } elsif ($c =~ m/^socks5_connect:/) {
1448 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1450 } elsif ($c =~ m/^Created new connection to/) {
1452 # Created new connection to www.privoxy.org:80 on socket 11.
1453 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1454 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1456 } elsif ($c =~ m/^Found reusable socket/) {
1458 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1460 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1461 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1462 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1463 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1464 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1465 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1466 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1467 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1469 } elsif ($c =~ m/^Marking open socket/) {
1471 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1472 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1473 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1474 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1476 } elsif ($c =~ m/^No reusable/) {
1478 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1479 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1481 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1483 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1484 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1486 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1487 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1488 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1490 } elsif ($c =~ m/^Socket/) {
1492 # Socket 16 already forgotten or never remembered.
1493 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1495 } elsif ($c =~ m/^The connection to/) {
1497 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1499 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1500 # Timeout is: 1. Assumed latency: 4.
1501 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1502 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1503 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1504 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1505 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1506 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1508 } elsif ($c =~ m/^Stopped waiting for the request line./) {
1510 # Stopped waiting for the request line. Timeout: 121.
1511 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1513 } elsif ($c =~ m/^Waiting for \d/) {
1515 # Waiting for 1 connections to timeout.
1516 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1518 } elsif ($c =~ m/^Initialized/) {
1520 # Initialized 20 socket slots.
1521 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1523 } elsif ($c =~ m/^Done reading from server/) {
1525 # Done reading from server. Expected content length: 24892. \
1526 # Actual content length: 24892. Most recently received: 4412.
1528 # Done reading from server. Expected content length: 24892. \
1529 # Actual content length: 24892. Bytes most recently read: 4412.
1530 # Done reading from server. Content length: 6018 as expected. \
1531 # Bytes most recently read: 294.
1532 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1533 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1534 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1536 } elsif ($c =~ m/^Continuing buffering headers/) {
1538 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1539 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1540 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1541 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1543 # Continuing buffering headers. Bytes most recently read: %d.
1544 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1546 } elsif ($c =~ m/^Received \d+ bytes while/) {
1548 # Received 206 bytes while expecting 12103.
1549 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1550 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1552 } elsif ($c =~ m/^(Rejecting c|C)onnection from/) {
1554 # Connection from 81.163.28.218 dropped due to ACL
1555 # Rejecting connection from 178.63.152.227. Maximum number of connections reached.
1556 $c =~ s@(?<=onnection from )((?:\d+\.?){3}\d+)@$h{'Number'}$1$h{'Standard'}@;
1558 } elsif ($c =~ m/^(?:Reusing|Closing) server socket \d./ or
1559 $c =~ m/^No additional client request/) {
1561 # Reusing server socket 4. Opened for 10.0.0.1.
1562 # Closing server socket 2. Opened for 10.0.0.1.
1563 # No additional client request received in time. \
1564 # Closing server socket 4, initially opened for 10.0.0.1.
1565 # No additional client request received in time on socket 29.
1567 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1568 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.$)');
1570 } elsif ($c =~ m/^Connected to /) {
1572 # Connected to tor-jail[10.0.0.2]:9050.
1574 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1575 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1576 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1578 } elsif ($c =~ m/^Could not connect to /) {
1580 # Could not connect to [10.0.0.1]:80.
1582 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1583 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1585 } elsif ($c =~ m/^Waiting for the next client request/ or
1586 $c =~ m/^The connection on server socket/ or
1587 $c =~ m/^Client request arrived in time or the client closed the connection/) {
1589 # Waiting for the next client request on socket 3. Keeping the server \
1590 # socket 12 to a.fsdn.com open.
1591 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1592 # Client request arrived in time or the client closed the connection on socket 12.
1594 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1595 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1596 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1598 } elsif ($c =~ m/^Marking the server socket/) {
1600 # Marking the server socket 7 tainted.
1602 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1604 } elsif ($c =~ m/^Reduced expected bytes to /) {
1606 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1607 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1608 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1610 } elsif ($c =~ m/^The client closed socket /) {
1612 # The client closed socket 2 while the server socket 4 is still open.
1613 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1614 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1616 } elsif ($c =~ m/^Expected client content length set /) {
1618 # Expected client content length set to 667325411 after reading 4999 bytes.
1619 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1620 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1622 } elsif ($c =~ m/^Waiting for up to /) {
1624 # Waiting for up to 4999 bytes from the client.
1625 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1627 } elsif ($c =~ m/^Looks like we / or
1628 $c =~ m/^Unsetting keep-alive flag/ or
1629 $c =~ m/^No connections to wait/ or
1630 $c =~ m/^Complete client request received/ or
1631 $c =~ m/^Possible pipeline attempt detected./ or
1632 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1633 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1634 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1635 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1636 $c =~ m/^Stopping to watch the client socket. There's already another request waiting./ or
1637 $c =~ m/^Done reading from the client\.$/) {
1639 # Looks like we reached the end of the last chunk. We better stop reading.
1640 # Looks like we read the end of the last chunk together with the server \
1641 # headers. We better stop reading.
1642 # Looks like we got the last chunk together with the server headers. \
1643 # We better stop reading.
1644 # Unsetting keep-alive flag.
1645 # No connections to wait for left.
1646 # Client request arrived in time or the client closed the connection.
1647 # Complete client request received
1648 # Possible pipeline attempt detected. The connection will not be \
1649 # kept alive and we will only serve the first request.
1650 # POST request detected. The connection will not be kept alive.
1651 # The server still wants to talk, but the client hung up on us.
1652 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1653 # There might be a request body. The connection will not be kept alive.
1654 # Stopping to watch the client socket. There's already another request waiting.
1655 # Done reading from the client\.
1659 found_unknown_content($c);
1667 sub handle_loglevel_info ($) {
1671 if ($c =~ m/^Rewrite detected:/) {
1673 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1674 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1676 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1677 $c =~ m/^Compressed content detected/ or
1680 # Decompressing deflated iob: 117
1681 # Decompression didn't result in any content.
1682 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1683 # with zlib support or enable the prevent-compression action.
1684 # Tagger 'complete-url' created empty tag. Ignored.
1688 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1690 # loading configuration file '/usr/local/etc/privoxy/config':
1691 # Reloading configuration file '/usr/local/etc/privoxy/config'
1692 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1694 } elsif ($c =~ m/^Loading (actions|filter) file: /) {
1696 # Loading actions file: /usr/local/etc/privoxy/default.action
1697 # Loading filter file: /usr/local/etc/privoxy/default.filter
1698 $c =~ s@(?<= file: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1700 } elsif ($c =~ m/^exiting by signal/) {
1702 # exiting by signal 15 .. bye
1703 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1705 } elsif ($c =~ m/^Privoxy version/) {
1707 # Privoxy version 3.0.7
1708 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1710 } elsif ($c =~ m/^Program name: /) {
1712 # Program name: /usr/local/sbin/privoxy
1713 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1715 } elsif ($c =~ m/^Listening on port /) {
1717 # Listening on port 8118 on IP address 10.0.0.1
1718 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1719 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1721 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1723 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1724 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1726 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1728 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1729 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1730 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1732 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1733 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1734 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1735 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1737 } elsif ($c =~ m/^Status code/) {
1739 # Status code 304 implies no body.
1740 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1742 } elsif ($c =~ m/^Method/) {
1744 # Method HEAD implies no body.
1745 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1747 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1749 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1750 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1751 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1753 } elsif ($c =~ m/^No logfile configured/ or
1754 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1755 $c =~ m/^Invalid \"chunked\" transfer/ or
1756 $c =~ m/^Support for/ or
1757 $c =~ m/^Flushing header and buffers/ or
1758 $c =~ m/^Can not resolve/
1761 # No logfile configured. Please enable it before reporting any problems.
1762 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1763 # response or even no response at all.
1764 # No logfile configured. Logging disabled.
1765 # Invalid "chunked" transfer encoding detected and ignored.
1766 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1767 # known not to work properly in some situations.
1768 # Flushing header and buffers. Stepping back from filtering.
1769 # Can not resolve doesnotexist: hostname nor servname provided, or not known
1773 found_unknown_content($c);
1780 sub handle_loglevel_cgi ($) {
1784 if ($c =~ m/^Granting access to/) {
1786 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1788 } elsif ($c =~ m/^Substituting: s(.)/) {
1790 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1791 # XXX: prone to span several lines
1794 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1795 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1796 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1802 sub handle_loglevel_force ($) {
1806 if ($c =~ m/^Ignored force prefix in request:/) {
1808 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1809 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1810 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1812 } elsif ($c =~ m/^Enforcing request:/) {
1814 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1815 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1819 found_unknown_content($c);
1826 sub handle_loglevel_error ($) {
1830 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1832 # Empty server or forwarder response received on socket 4.
1833 # Empty server or forwarder response received on socket 3. \
1834 # Closing client socket 15 without sending data.
1835 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1836 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1838 # XXX: There are probably more messages that deserve highlighting.
1844 sub handle_loglevel_ignore ($) {
1848 sub gather_loglevel_request_stats ($$) {
1856 sub gather_loglevel_crunch_stats ($$) {
1866 sub gather_loglevel_error_stats ($$) {
1873 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1875 # Empty server or forwarder response received on socket 4.
1876 $stats{'empty-responses'}++;
1877 if ($thread_data{$thread}{'new_connection'}) {
1878 $stats{'empty-responses-on-new-connections'}++;
1880 $stats{'empty-responses-on-reused-connections'}++;
1885 sub gather_loglevel_connect_stats ($$) {
1887 my ($c, $thread) = @_;
1891 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
1893 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1894 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
1896 } elsif ($c =~ m/^to ([^\s]*)$/) {
1898 # Connect: to lists.sourceforge.net:443
1900 $thread_data{$thread}{'forwarder'} = 'direct connection';
1902 } elsif ($c =~ m/^Created new connection to/) {
1904 # Created new connection to www.privoxy.org:80 on socket 11.
1906 $thread_data{$thread}{'new_connection'} = 1;
1908 } elsif ($c =~ m/^Reusing server socket \d./ or
1909 $c =~ m/^Found reusable socket/) {
1911 # Reusing server socket 4. Opened for 10.0.0.1.
1912 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1914 $thread_data{$thread}{'new_connection'} = 0;
1915 $stats{'reused-connections'}++;
1919 sub gather_loglevel_header_stats ($$) {
1921 my ($c, $thread) = @_;
1924 if ($c =~ m/^A HTTP\/1\.1 response without/ or
1925 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
1927 # A HTTP/1.1 response without Connection header implies keep-alive.
1928 # Keeping the server header 'Connection: keep-alive' around.
1929 $stats{'server-keep-alive'}++;
1931 } elsif ($c =~ m/^scan: ((\w+) (.+) (HTTP\/\d\.\d))/) {
1933 # scan: HTTP/1.1 200 OK
1934 $stats{'method'}{$2}++;
1935 $stats{'ressource'}{$3}++;
1936 $stats{'http-version'}{$4}++;
1944 'server-keep-alive' => 0,
1945 'reused-connections' => 0,
1946 'empty-responses' => 0,
1947 'empty-responses-on-new-connections' => 0,
1948 'empty-responses-on-reused-connections' => 0,
1952 sub get_percentage ($$) {
1955 return "NaN" if ($big eq 0);
1956 return sprintf("%.2f%%", $small / $big * 100);
1959 sub print_stats () {
1962 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
1963 my $outgoing_requests = $stats{requests} - $stats{crunches};
1965 if ($stats{requests} eq 0) {
1966 print "No requests yet.\n";
1970 print "Client requests total: " . $stats{requests} . "\n";
1971 print "Crunches: " . $stats{crunches} . " (" .
1972 get_percentage($stats{requests}, $stats{crunches}) . ")\n";
1973 print "Outgoing requests: " . $outgoing_requests . " (" .
1974 get_percentage($stats{requests}, $outgoing_requests) . ")\n";
1975 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
1976 get_percentage($stats{requests}, $stats{'server-keep-alive'}) . ")\n";
1977 print "New outgoing connections: " . $new_connections . " (" .
1978 get_percentage($stats{requests}, $new_connections) . ")\n";
1979 print "Reused connections: " . $stats{'reused-connections'} . " (" .
1980 get_percentage($stats{requests}, $stats{'reused-connections'}) . ")\n";
1981 print "Empty responses: " . $stats{'empty-responses'} . " (" .
1982 get_percentage($stats{requests}, $stats{'empty-responses'}) . ")\n";
1983 print "Empty responses on new connections: "
1984 . $stats{'empty-responses-on-new-connections'} . " (" .
1985 get_percentage($stats{requests}, $stats{'empty-responses-on-new-connections'})
1987 print "Empty responses on reused connections: " .
1988 $stats{'empty-responses-on-reused-connections'} . " (" .
1989 get_percentage($stats{requests}, $stats{'empty-responses-on-reused-connections'}) .
1992 if ($stats{method} eq 0) {
1993 print "No response lines parsed yet yet.\n";
1996 print "Method distribution:\n";
1997 foreach my $method (sort {$stats{'method'}{$b} <=> $stats{'method'}{$a}} keys %{$stats{'method'}}) {
1998 printf "%8d : %-8s\n", $stats{'method'}{$method}, $method;
2000 print "Client HTTP versions:\n";
2001 foreach my $http_version (sort {$stats{'http-version'}{$b} <=> $stats{'http-version'}{$a}} keys %{$stats{'http-version'}}) {
2002 printf "%d : %s\n", $stats{'http-version'}{$http_version}, $http_version;
2004 print "Requested ressources:\n";
2005 foreach my $ressource (sort {$stats{'ressource'}{$b} <=> $stats{'ressource'}{$a}} keys %{$stats{'ressource'}}) {
2006 printf "%d : %s\n", $stats{'ressource'}{$ressource}, $ressource;
2011 ################################################################################
2012 # Functions that actually print stuff
2013 ################################################################################
2015 sub print_clf_message () {
2017 our ($ip, $timestamp, $request_line, $status_code, $size);
2020 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2022 # Rebuild highlighted
2023 $output .= $h{'Number'} . $ip . $h{'Standard'};
2025 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2027 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2029 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2031 $output .= $h{'Number'} . $size . $h{'Standard'};
2032 $output .= $line_end;
2037 sub print_non_clf_message ($) {
2039 my $content = shift;
2040 my $msec_string = "." . $req{$t}{'msecs'} unless $no_msecs_mode;
2041 my $line_start = $html_output_mode ? '' : $h{"Standard"};
2043 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2046 . $time_colours[$time_colour_index % 2]
2047 . $req{$t}{'time-stamp'}
2049 . $h{Standard} . " "
2050 . $thread_colours{$t}
2054 . $h{$req{$t}{'log-level'}}
2055 . $req{$t}{'log-level'}
2062 sub shorten_thread_id ($) {
2064 my $thread_id = shift;
2066 our %short_thread_ids;
2069 unless (defined $short_thread_ids{$thread_id}) {
2070 $short_thread_ids{$thread_id} = sprintf "%.3d", $max_threadid++;
2073 return $short_thread_ids{$thread_id}
2078 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2080 my $last_thread = 0;
2081 my $last_timestamp = 0;
2082 my $filters_that_did_nothing;
2085 $time_colour = paint_it('white');
2087 my %log_level_handlers = (
2088 'Re-Filter' => \&handle_loglevel_re_filter,
2089 'Header' => \&handle_loglevel_header,
2090 'Connect' => \&handle_loglevel_connect,
2091 'Redirect' => \&handle_loglevel_redirect,
2092 'Request' => \&handle_loglevel_request,
2093 'Crunch' => \&handle_loglevel_crunch,
2094 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2095 'Info' => \&handle_loglevel_info,
2096 'CGI' => \&handle_loglevel_cgi,
2097 'Force' => \&handle_loglevel_force,
2098 'Error' => \&handle_loglevel_error,
2099 'Fatal error' => \&handle_loglevel_ignore,
2100 'Writing' => \&handle_loglevel_ignore,
2101 'Received' => \&handle_loglevel_ignore,
2102 'Unknown log level' => \&handle_loglevel_ignore,
2107 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2108 $thread = $t = ($shorten_thread_ids) ? shorten_thread_id($4) : $4;
2109 $req{$t}{'day'} = $day = $1;
2110 $req{$t}{'time-stamp'} = $time_stamp = $2;
2111 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2112 $req{$t}{'log-level'} = $log_level = $5;
2113 $req{$t}{'content'} = $content = $c = $6;
2114 $req{$t}{'log-message'} = $_;
2115 $no_special_header_highlighting = 0;
2117 if (defined($log_level_handlers{$log_level})) {
2119 $content = $log_level_handlers{$log_level}($content);
2123 die "No handler found for log level \"$log_level\"\n";
2126 # Highlight Truncations
2127 if (length($_) > 4000) {
2128 $content =~ s@(too long, truncated)]$@$h{'Truncation'}$1$h{'Standard'}]@g;
2131 next unless $content;
2133 # Register threads to keep the colour constant
2134 if (!defined($thread_colours{$thread})) {
2135 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2136 $thread_colour_index++;
2139 # Switch timestamp colour if timestamps differ
2140 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2141 debug_message("Tick tack!") if DEBUG_TICKS;
2142 $time_colour = $time_colours[$time_colour_index % 2];
2143 $time_colour_index++;
2144 $last_msecs = $msecs;
2145 $last_timestamp = $time_stamp;
2148 $last_thread = $thread;
2150 print_non_clf_message($content);
2152 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2154 # LOG_LEVEL_CLF lines look like this
2155 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2156 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2157 # HTTP/1.1" 403 1730
2158 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2160 print_clf_message();
2164 # Some Privoxy log messages span more than one line,
2165 # usually to dump lots of content that doesn't need any syntax highlighting.
2166 # XXX: add mechanism to forward these lines to the right handler anyway.
2168 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2169 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2177 my ($day, $time_stamp, $msecs, $thread, $log_level, $content);
2178 my %log_level_handlers = (
2179 'Re-Filter' => \&handle_loglevel_ignore,
2180 'Header' => \&gather_loglevel_header_stats,
2181 'Connect' => \&gather_loglevel_connect_stats,
2182 'Redirect' => \&handle_loglevel_ignore,
2183 'Request' => \&gather_loglevel_request_stats,
2184 'Crunch' => \&gather_loglevel_crunch_stats,
2185 'Gif-Deanimate' => \&handle_loglevel_ignore,
2186 'Info' => \&handle_loglevel_ignore,
2187 'CGI' => \&handle_loglevel_ignore,
2188 'Force' => \&handle_loglevel_ignore,
2189 'Error' => \&gather_loglevel_error_stats,
2190 'Fatal error' => \&handle_loglevel_ignore,
2191 'Writing' => \&handle_loglevel_ignore,
2192 'Unknown log level' => \&handle_loglevel_ignore
2196 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2199 $msecs = $3 ? $3 : 0;
2204 if (defined($log_level_handlers{$log_level})) {
2206 $content = $log_level_handlers{$log_level}($content, $thread);
2210 die "No handler found for log level \"$log_level\"\n";
2220 sub VersionMessage {
2221 my $version_message;
2223 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2224 $version_message .= 'Copyright (C) 2007-2010 Fabian Keil <fk@fabiankeil.de>' . "\n";
2225 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2227 print $version_message;
2230 sub get_cli_options () {
2232 our %cli_options = (
2233 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2234 'title' => CLI_OPTION_TITLE,
2235 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2236 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2237 'no-msecs' => CLI_OPTION_NO_MSECS,
2238 'shorten-thread-ids' => CLI_OPTION_SHORTEN_THREAD_IDS,
2239 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2240 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
2241 'statistics' => CLI_OPTION_STATISTICS,
2245 'html-output' => \$cli_options{'html-output'},
2246 'title' => \$cli_options{'title'},
2247 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2248 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2249 'no-msecs' => \$cli_options{'no-msecs'},
2250 'shorten-thread-ids' => \$cli_options{'shorten-thread-ids'},
2251 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2252 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
2253 'statistics' => \$cli_options{'statistics'},
2254 'version' => sub { VersionMessage && exit(0) },
2258 $html_output_mode = cli_option_is_set('html-output');
2259 $no_msecs_mode = cli_option_is_set('no-msecs');
2260 $shorten_thread_ids = cli_option_is_set('shorten-thread-ids');
2261 $line_end = get_line_end();
2272 Options and their default values if they have any:
2273 [--accept-unknown-messages]
2277 [--no-syntax-highlighting]
2278 [--shorten-thread-ids]
2279 [--show-ineffective-filters]
2281 [--title $cli_options{'title'}]
2283 see "perldoc $0" for more information
2289 ################################################################################
2291 ################################################################################
2295 set_background(DEFAULT_BACKGROUND);
2296 prepare_our_stuff();
2300 if (cli_option_is_set('statistics')) {
2313 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2317 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
2318 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--statistics>]
2319 [B<--shorten-thread-ids>] [B<--show-ineffective-filters>] [B<--version>]
2323 B<privoxy-log-parser> reads Privoxy log messages and
2325 - syntax-highlights recognized lines,
2327 - reformats some of them for easier comprehension,
2329 - filters out less useful messages, and
2331 - (in some cases) calculates additional information,
2332 like the compression ratio or how a filter affected
2335 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2336 without getting confused by the resulting amount of output. For example for
2337 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2338 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2339 will hide the "filter foo caused 0 hits" message.
2343 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
2344 just don't highlight them.
2346 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2347 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2348 This option is only intended to make embedding log excerpts in web pages easier.
2349 It does not escape any input!
2351 [B<--no-msecs>] Don't expect milisecond resolution
2353 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2354 the filtered output is piped into less in which case the ANSI control
2355 codes don't work, or if the terminal itself doesn't support the control
2358 [B<--shorten-thread-ids>] Shorten the thread ids to a three-digit decimal number.
2359 Note that the mapping from thread ids to shortened ids is created at run-time
2360 and thus varies with the input.
2362 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2363 that didn't modify the content.
2365 [B<--statistics>] Gather various statistics instead of syntax highlighting
2366 log messages. This is an experimental feature, if the results look wrong
2367 they very well might be. Also note that the results are pretty much guaranteed
2368 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2370 [B<--version>] Print version and exit.
2374 To monitor a log file:
2376 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2378 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2379 or if the log won't get rotated anyway. The log file location depends
2380 on your system (Doh!).
2382 To monitor Privoxy without having it write to a log file:
2384 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2386 Again, the config file location depends on your system. Output redirection
2387 depends on your shell, the above works with bourne shells.
2389 To read a processed Privoxy log file from top to bottom, letting the content
2390 scroll by slightly faster than you can read:
2392 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2394 This is probably only useful to fill screens in the background of haxor movies.
2398 Syntax highlighting with ANSI escape sequences will look strange
2399 if your background color isn't black.
2401 Some messages aren't recognized yet and will not be fully highlighted.
2403 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2404 using earlier Privoxy versions will probably result in an increased amount
2405 of unrecognized log lines.
2407 Privoxy's log files tend to be rather large. If you use HTML
2408 highlighting some browsers can't handle them, get confused and
2409 will eventually crash because of segmentation faults or unexpected
2410 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2415 Many settings can't be controlled through command line options yet.
2423 Fabian Keil <fk@fabiankeil.de>