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 1.124 2008/11/22 10:02:52 fk 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) 2007-2008 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 => '0.4',
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_ERROR_LOG_FILE => '/var/log/privoxy-log.log',
60 CLI_OPTION_SHOW_INEFFECTIVE_FILTERS => 0,
61 CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES => 0,
62 CLI_OPTION_STATISTIC => 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,
84 # You better leave these alone unless you know what you're doing.
85 COLOUR_RESET => "\033[0;0m",
89 sub prepare_our_stuff () {
91 # Syntax Higlight hash
93 'red', 'green', 'brown', 'blue', 'purple', 'cyan',
94 'light_gray', 'light_red', 'light_green', 'yellow',
95 'light_blue', 'pink', 'light_cyan', 'white'
102 Filter => 'purple', # XXX: Used?
103 'Re-Filter' => 'purple',
105 Request => 'light_cyan',
106 CGI => 'light_green',
108 Error => 'light_red',
109 'Fatal error' => 'light_red',
110 'Gif-Deanimate' => 'blue',
112 Writing => 'light_green',
113 # ----------------------
116 request_ => 'brown', # host+path but no protocol
117 'ip-address' => 'yellow',
120 Truncation => 'light_red',
122 Timestamp => 'brown',
123 Crunching => 'light_red',
124 crunched => 'light_red',
125 'Request-Line' => 'pink',
127 destination => 'yellow',
128 'http-version' => 'pink',
129 'crunch-pattern' => 'pink',
134 'program-name' => 'cyan',
137 warning => 'light_red',
138 debug => 'light_red',
142 'status-message' => 'light_cyan',
143 'status-code' => 'yellow',
144 'invalid-request' => 'light_red',
146 error => 'light_red',
147 'rewritten-URL' => 'light_red',
148 'pcrs-delimiter' => 'light_red',
149 'ignored' => 'light_red',
150 'action-bits-update' => 'light_red',
151 'configuration-line' => 'red',
152 'content-type' => 'yellow',
157 # Header colours need their own hash so the keys can be accessed properly
158 our %header_colours = (
159 # Prefilled with headers that should not appear with default header colours
160 Cookie => 'light_red',
161 'Set-Cookie' => 'light_red',
162 Warning => 'light_red',
163 Default => HEADER_DEFAULT_COLOUR,
166 # Crunch reasons need their own hash as well
167 our %reason_colours = (
168 'Unsupported HTTP feature' => 'light_red',
169 Blocked => 'light_red',
170 Untrusted => 'light_red',
171 Redirected => 'green',
172 'CGI Call' => 'white',
173 'DNS failure' => 'red',
174 'Forwarding failed' => 'light_red',
175 'Connection failure' => 'light_red',
176 'Out of memory (may mask other reasons)' => 'light_red',
177 'No reason recorded' => 'light_red',
180 our @time_colours = ('white', 'light_gray');
182 # Translate highlight strings into highlight code
183 prepare_highlight_hash(\%header_colours);
184 prepare_highlight_hash(\%reason_colours);
185 prepare_highlight_hash(\%h);
186 prepare_colour_array(\@all_colours);
187 prepare_colour_array(\@time_colours);
192 ###############################################################
193 # Takes a colour string and returns an ANSI escape sequence
194 # (unless --no-syntax-highlighting is used).
195 # XXX: The Rolling Stones reference has to go.
196 ###############################################################
198 my $colour = shift @_;
200 return "" if cli_option_is_set('no-syntax-highlighting');
242 my $bg_code = get_background();
244 our $default = default_colours();
246 if (defined($text{$colour})) {
247 $colour_code = ESCAPE;
248 $colour_code .= $text{$colour};
250 $colour_code .= $light{$colour} ? "1" : "2";
252 $colour_code .= $bg_code;
254 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
256 } elsif ($colour =~ /reset/) {
258 $colour_code = default_colours();
262 die "What's $colour supposed to mean?\n";
269 sub get_semantic_html_markup ($) {
270 ###############################################################
271 # Takes a string and returns a span element
272 ###############################################################
277 if ($type =~ /Standard/) {
281 $code = '<span title="' . $type . '" class="' . $type . '">';
287 sub cli_option_is_set ($) {
290 my $cli_option = shift;
292 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
294 return $cli_options{$cli_option};
297 sub get_html_title () {
300 return $cli_options{'title'};
304 sub init_css_colours() {
312 purple => "F06", # XXX: wrong
313 cyan => "F09", # XXX: wrong
318 light_green => "33F",
327 sub get_css_colour ($) {
332 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
334 return '#' . $css_colours{$colour};
337 sub get_css_line ($) {
344 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
345 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
346 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
347 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
348 $css_line .= '}' . "\n";
353 sub get_css_line_for_colour ($) {
360 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
361 $css_line .= 'color:' . get_css_colour($colour) . ';';
362 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
363 $css_line .= '}' . "\n";
368 # XXX: Wrong solution
369 sub get_missing_css_lines () {
373 $css_line .= '.' . 'default' . ' {'; # XXX: lc() shouldn't be necessary
374 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
375 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
376 $css_line .= '}' . "\n";
384 our %css_colours; #XXX: Wrong solution
388 $css .= '.privoxy-log {';
389 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
390 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
393 foreach my $key (keys %h_colours) {
395 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
396 $css .= get_css_line($key);
400 foreach my $colour (keys %css_colours) {
402 $css .= get_css_line_for_colour($colour);
406 $css .= get_missing_css_lines(); #XXX: Wrong solution
415 if (cli_option_is_set('html-output')) {
417 my $title = get_html_title();
419 $intro .= '<html><head>';
420 $intro .= '<title>' . $title . '</title>';
421 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
422 $intro .= '</head><body>';
423 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
433 if (cli_option_is_set('html-output')) {
435 $outro = '</p></body></html>';
442 sub get_line_end () {
446 $line_end = '<br>' . $line_end if cli_option_is_set('html-output');
451 sub get_colour_html_markup ($) {
452 ###############################################################
453 # Takes a colour string a span element. XXX: WHAT?
454 # XXX: This function shouldn't be necessary, the
455 # markup should always be semantically correct.
456 ###############################################################
461 if ($type =~ /Standard/) {
464 $code = '<span class="' . lc($type) . '">';
471 sub default_colours () {
474 return reset_colours();
477 sub show_colours () {
481 sub reset_colours () {
482 return ESCAPE . "0m";
485 sub set_background ($){
501 if (defined($backgrounds{$colour})) {
502 $bg_code = $backgrounds{$colour};
504 die "Invalid background colour: " . $colour;
508 sub get_background (){
512 sub prepare_highlight_hash ($) {
515 if (!cli_option_is_set('html-output')) {
517 foreach my $key (keys %$ref) {
518 $$ref{$key} = paint_it($$ref{$key});
523 foreach my $key (keys %$ref) {
524 $$ref{$key} = get_semantic_html_markup($key);
530 sub prepare_colour_array ($) {
533 if (!cli_option_is_set('html-output')) {
535 foreach my $i (0 ... @$ref - 1) {
536 $$ref[$i] = paint_it($$ref[$i]);
541 foreach my $i (0 ... @$ref - 1) {
542 $$ref[$i] = get_colour_html_markup($$ref[$i]);
548 sub found_unknown_content ($) {
556 return if cli_option_is_set('accept-unknown-messages');
558 return if ($unknown =~ /\[too long, truncated\]$/);
560 $message = "found_unknown_content: Don't know how to highlight: ";
561 # Break line so the log file can later be parsed as Privoxy log file again
562 $message .= '"' . $unknown . '"' . " in:\n";
563 $message .= $req{$t}{'log-message'};
564 debug_message($message);
565 log_parse_error($req{$t}{'log-message'});
567 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
570 sub log_parse_error ($) {
574 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
575 open(ERRORLOG, ">>" . ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
576 print ERRORLOG $message;
581 sub debug_message (@) {
585 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
588 ################################################################################
589 # highlighter functions that aren't loglevel-specific
590 ################################################################################
594 # Get highlight marker
596 my $highlight = shift; # XXX: Stupid name;
600 if (defined($highlight)) {
602 $result = $h{$highlight};
606 $message = "h: Don't recognize highlighter $highlight.";
607 debug_message($message);
608 log_parser_error($message);
609 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
615 sub highlight_known_headers ($) {
620 my $headers = join ('|', keys %header_colours);
622 debug_message("Searching $content for things to highlight.") if DEBUG_HEADER_HIGHLIGHTING;
624 if ($content =~ m/(?<=\s)($headers):/) {
626 $content =~ s@(?<=[\s|'])($header)(?=:)@$header_colours{$header}$1$h{'Standard'}@ig;
627 debug_message("Highlighted $content") if DEBUG_HEADER_HIGHLIGHTING;
633 sub highlight_matched_request_line ($$) {
635 my $result = shift; # XXX: Stupid name;
637 if ($result =~ m@(.*)($regex)(.*)@) {
638 $result = $1 . highlight_request_line($2) . $3
643 sub highlight_request_line ($) {
646 my ($method, $url, $http_version);
649 #GET http://images.sourceforge.net/sfx/icon_warning.gif HTTP/1.1
650 if ($rl =~ m/Invalid request/) {
652 $rl = h('invalid-request') . $rl . h('Standard');
654 } elsif ($rl =~ m/^([-\w]+) (.*) (HTTP\/\d\.\d)/) {
656 # XXX: might not match in case of HTTP method fuzzing.
657 # XXX: save these: ($method, $path, $http_version) = ($1, $2, $3);
658 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
659 if ($rl =~ /http:\/\//) {
660 $rl = highlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
662 $rl = highlight_matched_pattern($rl, 'request_', '[^\s]*(?=\sHTTP)');
665 $rl =~ s@(HTTP\/\d\.\d)$@$h{'http-version'}$1$h{'Standard'}@;
667 } elsif ($rl =~ m/\.\.\. \[too long, truncated\]$/) {
669 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
670 $rl = highlight_matched_url($rl, '[^\s]*(?=\.\.\.)');
672 } elsif ($rl =~ m/^ $/) {
674 $rl = h('error') . "No request line specified!" . h('Standard');
678 debug_message ("Can't parse request line: $rl");
685 sub highlight_response_line ($) {
688 my ($http_version, $status_code, $status_message);
693 # TODO: Mark different status codes differently
695 if ($rl =~ m/((?:HTTP\/\d\.\d|ICY)) (\d+) (.*)/) {
696 ($http_version, $status_code, $status_message) = ($1, $2, $3);
698 debug_message ("Can't parse response line: $rl") and die 'Fix this';
701 # Rebuild highlighted
703 $rl .= h('http-version') . $http_version . h('Standard');
705 $rl .= h('status-code') . $status_code . h('Standard');
707 $rl .= h('status-message') . $status_message . h('Standard');
712 sub highlight_matched_url ($$) {
714 my $result = shift; # XXX: Stupid name;
717 #print "Got $result, regex ($regex)\n";
719 if ($result =~ m@(.*?)($regex)(.*)@) {
720 $result = $1 . highlight_url($2) . $3;
721 #print "Now the result is $result\n";
727 sub highlight_matched_host ($$) {
729 my $result = shift; # XXX: Stupid name;
732 if ($result =~ m@(.*?)($regex)(.*)@) {
733 $result = $1 . h('host') . $2 . h('Standard') . $3;
739 sub highlight_matched_pattern ($$$) {
742 my $result = shift; # XXX: Stupid name;
746 die "Unknown key $key" unless defined $h{$key};
748 if ($result =~ m@(.*?)($regex)(.*)@) {
749 $result = $1 . h($key) . $2 . h('Standard') . $3;
756 sub highlight_matched_path ($$) {
758 my $result = shift; # XXX: Stupid name;
761 if ($result =~ m@(.*?)($regex)(.*)@) {
762 $result = $1 . h('path') . $2 . h('Standard') . $3;
769 sub highlight_url ($) {
773 if (cli_option_is_set('html-output')) {
775 $url = '<a href="' . $url . '">' . $url . '</a>';
779 $url = h('URL') . $url . h('Standard');
786 ################################################################################
787 # loglevel-specific highlighter functions
788 ################################################################################
790 sub handle_loglevel_header ($) {
799 our $header_colour_index;
800 our $no_special_header_highlighting;
802 # Register new headers
803 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
804 if ($c =~ m/^scan: ((?>[^:]+)):/) {
806 if (!defined($header_colours{$header})) {
807 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
809 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
810 $header_colours{$header} = $header_colours{'Default'};
812 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
813 $header_colour_index++;
818 if ($c =~ m/^scan: ((\w*) (.*) (HTTP\/\d\.\d))/) {
820 # Client request line
821 # Save for statistics (XXX: Not implemented yet)
822 $req{$t}{'method'} = $2;
823 $req{$t}{'destination'} = $3;
824 $req{$t}{'http-version'} = $4;
826 $content = highlight_request_line($1);
828 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
830 # Server response line
831 $req{$t}{'response_line'} = $2;
832 $req{$t}{'status_code'} = $3;
833 $req{$t}{'status_message'} = $4;
834 $content = $1 . highlight_response_line($req{$t}{'response_line'});
836 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
838 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
839 $content =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
840 $content =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
842 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
844 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/
845 $c = highlight_matched_host($c, '(?<=New host is: )[^\s]+');
846 $content = highlight_matched_url($c, '(?<=Crunching Referer: )[^\s]+');
848 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
850 # Text mode enabled by force. Take cover!
851 $content =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
853 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
855 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
856 $content = $1 . highlight_request_line($2);
858 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
860 # Adjusted Content-Length to 2132
861 # Adjust Content-Length to 33533
862 $content =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
863 $content = highlight_known_headers($content);
865 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
867 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
868 $content = highlight_matched_url($content, '(?<=New request URL: ).*');
870 } elsif ($c =~ m/^Couldn\'t parse:/) {
872 # XXX: These should probable be logged with LOG_LEVEL_ERROR
873 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
874 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
875 $content =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
877 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
878 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
880 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
881 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
882 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
884 # XXX: Save tag and tagger
886 $content =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
887 $content =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
888 $content =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
889 $no_special_header_highlighting = 1;
891 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
893 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
894 # XXX: Save tag and tagger
896 $content =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
897 $content =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
899 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
900 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
901 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
902 or $c =~ m/^(addh-unique|Referer forged to)/
903 or $c =~ m/^Downgraded answer to HTTP\/1.0/
904 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
905 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
906 or $c =~ m/^Referer crunched!/
907 or $c =~ m/^crunched x-forwarded-for!/
908 or $c =~ m/^crunched From!/
909 or $c =~ m/^ modified$/
910 or $c =~ m/^Content filtering is enabled. Crunching:/
911 or $c =~ m/^force-text-mode overruled the client/
912 or $c =~ m/^Server time in the future\./
913 or $c =~ m/^content-disposition header crunched and replaced with:/i
914 or $c =~ m/^Reducing white space in /
915 or $c =~ m/^Ignoring single quote in /
916 or $c =~ m/^Converting tab to space in /
917 or $c =~ m/A HTTP\/1\.1 response without/
918 or $c =~ m/Disabled filter mode on behalf of the client/
921 # XXX: Some of these may need highlighting
923 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
924 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
925 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
926 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
927 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
928 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
929 # Text mode is already enabled
930 # Denied request with NULL byte(s) turned into line break(s)
931 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
932 # addh-unique: Host: people.freebsd.org
933 # Suppressed offer to compress content
934 # Crunched incoming cookie -- yum!
935 # Accepted the client's request to fetch without filtering.
936 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
937 # addh-unique: Host: subkeys.pgp.net:11371
938 # Referer forged to: Referer: http://10.0.0.1/
939 # Downgraded answer to HTTP/1.0
940 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
941 # Referer overwritten with: Referer: pille-palle
942 # Referer replaced with: Referer: pille-palle
943 # crunched x-forwarded-for!
945 # modified # XXX: pretty stupid log message
946 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
947 # force-text-mode overruled the client's request to fetch without filtering!
948 # Server time in the future.
949 # content-disposition header crunched and replaced with: content-disposition: filename=baz
950 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
951 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
952 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
953 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
955 # A HTTP/1.1 response without Connection header implies keep-alive.
956 # Disabled filter mode on behalf of the client.
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 $content =~ s@(Crunching|crunched)@$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 $content = h('warning') . $content . h('Standard');
974 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
976 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
977 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
979 $content =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
980 $content =~ s@(?<=to \")(.*)(?=\")@$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 $content =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
996 found_unknown_content($content);
1000 unless ($c =~ m/^Transforming/) {
1001 $content = highlight_known_headers($content) unless $no_special_header_highlighting;
1007 sub handle_loglevel_re_filter ($) {
1009 my $content = shift;
1015 our %header_colours;
1017 our $header_colour_index;
1020 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1022 # Used by Privoxy 3.0.5 and 3.0.6:
1024 # Used by Privoxy 3.0.7:
1025 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1027 $req{$t}{'filtered_header'} = $1;
1028 $req{$t}{'old_header_size'} = $2;
1029 $req{$t}{'header_filter_name'} = $3;
1031 unless (cli_option_is_set('show-ineffective-filters') or
1032 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1035 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1036 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1038 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1040 # ...produced 0 hits (new size 23).
1041 #... produced 1 hits (new size 54).
1043 $req{$t}{'header_filter_hits'} = $1;
1044 $req{$t}{'new_header_size'} = $2;
1046 unless (cli_option_is_set('show-ineffective-filters') or
1047 (defined($req{$t}{'header_filter_name'}) and
1048 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1050 if ($req{$t}{'header_filter_hits'} == 0 and
1051 not (defined($req{$t}{'header_filter_name'}) and
1052 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1055 # Reformat including information from the intro
1056 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1058 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1059 # $c .= $req{$t}{'filtered_header'} . " ";
1060 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1061 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1063 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1065 $c .= "changing size from ";
1066 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1068 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1073 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1077 # Highlight from last line (XXX: What?)
1078 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1079 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1084 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1085 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1090 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1092 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1093 # Tagger variable-test has empty joblist. Nothing to do.
1095 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1097 } elsif ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1099 # XXX: only the second version gets highlighted properly.
1100 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1101 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1102 $req{$t}{'content_source'} = $1;
1103 $req{$t}{'content_size'} = $2;
1104 $req{$t}{'content_filter'} = $3;
1105 $req{$t}{'content_hits'} = $4;
1106 $req{$t}{'new_content_size'} = $5;
1107 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1108 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1109 if ($req{$t}{'content_hits'} == 0 and
1110 not (cli_option_is_set('show-ineffective-filters')
1111 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1115 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1116 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1117 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1119 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1120 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1123 $c .= "(" . $h{'Number'};
1124 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1125 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1128 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1130 $req{$t}{'chunked-size'} = $1;
1131 $req{$t}{'dechunked-size'} = $2;
1132 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1134 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1136 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1137 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1139 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1141 # Decompression successful. Old size: 670, new size: 1166.
1143 $req{$t}{'size-compressed'} = $1;
1144 $req{$t}{'size-decompressed'} = $2;
1145 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1147 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1148 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1150 # XXX: Create sub get_percentage()
1151 if ($req{$t}{'size-decompressed'}) {
1152 $req{$t}{'decompression-gain-percent'} =
1153 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1155 $content .= " (saved: ";
1156 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1158 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1162 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1164 # Need to de-chunk first
1165 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1167 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1169 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1171 # Adding re_filter job ...
1172 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1173 # to filter client-ip-address succeeded.
1175 } elsif ($c =~ m/^Reading in filter/) {
1177 return '' unless SHOW_FILTER_READIN_IN;
1181 found_unknown_content($content);
1189 sub handle_loglevel_redirect ($) {
1196 if ($c =~ m/^Decoding "([^""]*)"/) {
1198 $req{$t}{'original-destination'} = $1;
1199 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1202 } elsif ($c =~ m/^Checking/) {
1204 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1205 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1207 # TODO: Change colour if really url-decoded
1208 $req{$t}{'decoded-original-destination'} = $1;
1209 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1212 } elsif ($c =~ m/^pcrs command "([^""]*)" changed "([^""]*)" to "([^""]*)" \((\d+) hits?\)/) {
1214 # pcrs command "s@&from=rss@@" changed "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1215 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1217 my ($pcrs_command, $url_before, $url_after, $hits) = ($1, $2, $3, $4); # XXX: save these?
1219 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1220 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1221 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1222 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1223 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1224 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1226 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1228 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1229 # XXX: Use URL highlighter
1231 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1233 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1234 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1235 # is already properly formatted.
1236 # XXX: assume the same?
1237 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1241 found_unknown_content($c);
1248 sub handle_loglevel_gif_deanimate ($) {
1250 my $content = shift;
1255 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1257 my $bytes_from = $1;
1259 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1260 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1261 # XXX: Do we need g in case of ($1 == $2)?
1262 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1264 } elsif ($content =~ m/GIF (not) changed/) {
1266 # Gif-Deanimate: GIF not changed.
1267 return '' if SUPPRESS_GIF_NOT_CHANGED;
1268 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1270 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1272 # failed! (gif parsing)
1273 # XXX: Replace this error message with something less stupid
1274 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1276 } elsif ($content =~ m/^Need to de-chunk first/) {
1278 # Need to de-chunk first
1279 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1281 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1283 # No GIF header found (XXX: Did I ever commit this?)
1284 # failed while parsing 195 134747048 (XXX: never commited)
1286 # Ignore these for now
1290 found_unknown_content($content);
1298 sub handle_loglevel_request ($) {
1300 my $content = shift;
1304 our %reason_colours;
1306 if ($content =~ m/crunch! /) {
1308 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1310 # Highlight crunch reasons
1311 foreach my $reason (keys %reason_colours) {
1312 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1314 # Highlight request URL domain and ditch 'crunch!'
1315 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1316 $content =~ s@ crunch!@@;
1318 } elsif ($content =~ m/\[too long, truncated\]$/) {
1320 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1321 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1323 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1325 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1326 $content = h('request_') . $content . h('Standard');
1330 found_unknown_content($content);
1337 sub handle_loglevel_connect ($) {
1344 if ($c =~ m/via [^\s]+ to: [^\s]+/) {
1346 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1348 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1349 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1351 } elsif ($c =~ m/connect to: .* failed: .*/) {
1353 # connect to: www.example.org.noconnect failed: Operation not permitted
1355 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1357 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1359 } elsif ($c =~ m/to ([^\s]*) successful$/) {
1361 # Connect: to www.nzherald.co.nz successful
1363 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1364 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1366 } elsif ($c =~ m/to ([^\s]*)$/) {
1368 # Connect: to lists.sourceforge.net:443
1370 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1372 } elsif ($c =~ m/^accepted connection from .*/ or
1375 # accepted connection from 10.0.0.1
1376 # Privoxy 3.0.6 and earlier just say:
1378 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1379 $c = highlight_matched_host($c, '(?<=connection from ).*');
1381 } elsif ($c =~ m/^write header to: .* failed:/) {
1383 # write header to: 10.0.0.1 failed: Broken pipe
1385 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1386 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1388 } elsif ($c =~ m/^write header to client failed:/) {
1390 # write header to client failed: Broken pipe
1392 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1394 } elsif ($c =~ m/^socks4_connect:/) {
1396 # socks4_connect: SOCKS request rejected or failed.
1397 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1399 } elsif ($c =~ m/^Listening for new connections/ or
1400 $c =~ m/^accept connection/) {
1402 # Privoxy versions above 3.0.6 say:
1403 # Listening for new connections ...
1404 # earlier versions say:
1405 # accept connection ...
1408 } elsif ($c =~ m/^accept failed:/) {
1410 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1412 } elsif ($c =~ m/^Overriding forwarding settings/) {
1414 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1415 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1417 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1419 # Denying suspicious CONNECT request from 10.0.0.1
1420 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1422 } elsif ($c =~ m/^socks5_connect:/) {
1424 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1426 } elsif ($c =~ m/^Created new connection to/) {
1428 # Created new connection to www.privoxy.org:80 on socket 11.
1429 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1430 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1432 } elsif ($c =~ m/^Found reusable socket/) {
1434 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1435 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1436 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1437 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1439 } elsif ($c =~ m/^Marking open socket/) {
1441 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1442 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1443 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1444 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1446 } elsif ($c =~ m/^No reusable/) {
1448 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1449 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1451 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1453 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1454 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1455 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1456 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1457 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1459 } elsif ($c =~ m/^Socket/) {
1461 # Socket 16 already forgotten or never remembered.
1462 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1464 } elsif ($c =~ m/^The connection to/) {
1466 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19.
1467 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1468 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1469 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1470 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1472 } elsif ($c =~ m/^Initialized/) {
1474 # Initialized 20 socket slots.
1475 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1477 } elsif ($c =~ m/^Done reading from server/) {
1479 # Done reading from server. Expected content length: 24892. \
1480 # Actual content length: 24892. Most recently received: 4412.
1481 $c =~ s@(?<=Expected content length: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1482 $c =~ s@(?<=Actual content length: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1483 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1485 } elsif ($c =~ m/^Looks like we reached/ or
1486 $c =~ m/^Unsetting keep-alive flag/) {
1488 # Looks like we reached the end of the last chunk. We better stop reading.
1489 # Unsetting keep-alive flag.
1493 found_unknown_content($c);
1501 sub handle_loglevel_info ($) {
1508 if ($c =~ m/^Rewrite detected:/) {
1510 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1511 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1513 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1514 $c =~ m/^Compressed content detected/ or
1517 # Decompressing deflated iob: 117
1518 # Decompression didn't result in any content.
1519 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1520 # with zlib support or enable the prevent-compression action.
1521 # Tagger 'complete-url' created empty tag. Ignored.
1525 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1527 # loading configuration file '/usr/local/etc/privoxy/config':
1528 # Reloading configuration file '/usr/local/etc/privoxy/config'
1529 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1531 } elsif ($c =~ m/^exiting by signal/) {
1533 # exiting by signal 15 .. bye
1534 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1536 } elsif ($c =~ m/^Privoxy version/) {
1538 # Privoxy version 3.0.7
1539 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1541 } elsif ($c =~ m/^Program name: /) {
1543 # Program name: /usr/local/sbin/privoxy
1544 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1546 } elsif ($c =~ m/^Listening on port /) {
1548 # Listening on port 8118 on IP address 10.0.0.1
1549 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1550 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1552 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1554 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1555 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1557 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1559 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1560 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1561 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1563 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1564 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1565 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1566 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1568 } elsif ($c =~ m/^Status code/) {
1570 # Status code 304 implies no body.
1571 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1573 } elsif ($c =~ m/^Method/) {
1575 # Method HEAD implies no body.
1576 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1578 } elsif ($c =~ m/^No logfile configured/ or
1579 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1580 $c =~ m/^Invalid \"chunked\" transfer/ or
1581 $c =~ m/^Support for/
1584 # No logfile configured. Please enable it before reporting any problems.
1585 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1586 # response or even no response at all.
1587 # No logfile configured. Logging disabled.
1588 # Invalid "chunked" transfer encoding detected and ignored.
1589 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1590 # known not to work properly in some situations.
1594 found_unknown_content($c);
1601 sub handle_loglevel_cgi ($) {
1608 if ($c =~ m/^Granting access to/) {
1610 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1612 } elsif ($c =~ m/^Substituting: s(.)/) {
1614 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1615 # XXX: prone to span several lines
1618 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1619 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1620 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1626 sub handle_loglevel_force ($) {
1633 if ($c =~ m/^Ignored force prefix in request:/) {
1635 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1636 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1637 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1639 } elsif ($c =~ m/^Enforcing request:/) {
1641 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1642 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1646 found_unknown_content($c);
1653 sub handle_loglevel_ignore ($) {
1657 ################################################################################
1658 # Functions that actually print stuff
1659 ################################################################################
1661 sub print_clf_message () {
1663 our ($ip, $timestamp, $request_line, $status_code, $size);
1667 return if DEBUG_SUPPRESS_LOG_MESSAGES;
1669 # Rebuild highlighted
1670 $output .= $h{'Number'} . $ip . $h{'Standard'};
1672 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
1674 $output .= "\"" . highlight_request_line("$request_line") . "\"";
1676 $output .= $h{'Status'} . $status_code . $h{'Standard'};
1678 $output .= $h{'Number'} . $size . $h{'Standard'};
1679 $output .= get_line_end();
1684 sub print_non_clf_message ($) {
1687 our %thread_colours;
1690 our $time_colour_index;
1693 my $content = shift;
1694 my ($day, $time_stamp, $msecs, $thread, $log_level)
1695 = ($req{$t}{'day'}, $req{$t}{'time-stamp'}, $req{$t}{'msecs'}, $t, $req{$t}{'log-level'} );
1697 return if DEBUG_SUPPRESS_LOG_MESSAGES;
1699 $output .= $h{"Standard"} unless cli_option_is_set('html-output');
1700 # $output .= "$day ";
1701 $output .= $time_colours[$time_colour_index % 2];
1703 $output .= $time_stamp;
1704 $output .= ".$msecs" unless cli_option_is_set('no-msecs');
1705 $output .= $h{"Standard"};
1707 $output .= $thread_colours{$thread} if (defined($thread_colours{$thread}));
1709 $output .= $h{"Standard"} . " ";
1710 $output .= $h{$log_level} if (defined($h{$log_level}));
1711 $output .= $log_level;
1712 $output .= $h{"Standard"} . ": ";
1713 $output .= "$content";
1714 $output .= get_line_end();
1722 our %req; # request data from previous lines
1724 our %thread_colours;
1727 our $thread_colour_index = 0;
1728 our $header_colour_index = 0;
1729 our $time_colour_index = 0;
1731 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
1733 my $last_thread = 0;
1734 my $last_timestamp = 0;
1736 my $filters_that_did_nothing;
1739 our $no_special_header_highlighting;
1740 $time_colour = paint_it('white');
1741 my %log_level_count;
1743 my %log_level_handlers = (
1744 'Re-Filter' => \&handle_loglevel_re_filter,
1745 'Header' => \&handle_loglevel_header,
1746 'Connect' => \&handle_loglevel_connect,
1747 'Redirect' => \&handle_loglevel_redirect,
1748 'Request' => \&handle_loglevel_request,
1749 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
1750 'Info' => \&handle_loglevel_info,
1751 'CGI' => \&handle_loglevel_cgi,
1752 'Force' => \&handle_loglevel_force,
1753 'Error' => \&handle_loglevel_ignore,
1754 'Fatal error' => \&handle_loglevel_ignore,
1755 'Writing' => \&handle_loglevel_ignore,
1762 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\(([^\)]*)\)) ([\w -]*): (.*)$/) {
1763 # XXX: Put in req hash?
1766 $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution
1771 $req{$t}{'day'} = $day;
1772 $req{$t}{'time-stamp'} = $time_stamp;
1773 $req{$t}{'msecs'} = $msecs; # Only the cool kids have micro second resolution;
1774 $req{$t}{'log-level'} = $log_level;
1775 $req{$t}{'content'} = $content;
1776 $req{$t}{'log-message'} = $_;
1777 $no_special_header_highlighting = 0;
1779 $log_level_count{$log_level}++;
1781 if (defined($log_level_handlers{$log_level})) {
1783 $content = $log_level_handlers{$log_level}($content);
1787 die "No handler found for log level \"$log_level\"\n";
1791 # Highlight Truncations
1792 if (m/\.\.\. \[(too long, truncated)/) {
1793 $content =~ s@($1)@$h{'Truncation'}$1$h{'Standard'}@g;
1796 next unless $content;
1798 # Register threads to keep the colour constant
1799 if (!defined($thread_colours{$thread})) {
1800 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
1801 $thread_colour_index++;
1804 # Switch timestamp colour if timestamps differ
1805 if ($msecs != $last_msecs || !($time_stamp =~ m/$last_timestamp/)) {
1806 debug_message("Tick tack!") if DEBUG_TICKS;
1807 $time_colour = $time_colours[$time_colour_index % 2];
1808 $time_colour_index++
1811 $last_msecs = $msecs;
1812 $last_thread = $thread;
1813 $last_timestamp = $time_stamp;
1815 print_non_clf_message($content);
1817 } elsif (m/^(\d+\.\d+\.\d+\.\d+) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
1819 # LOG_LEVEL_CLF lines look like this
1820 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
1821 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
1822 # HTTP/1.1" 403 1730
1823 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
1825 print_clf_message();
1829 # Some Privoxy log messages span more than one line,
1830 # usually to dump lots of content that doesn't need any syntax highlighting.
1831 # XXX: add mechanism to forward these lines to the right handler anyway.
1833 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
1834 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
1839 if (cli_option_is_set('statistic')) {
1840 foreach (keys %log_level_count) {
1841 print $_ . ": " . $log_level_count{$_} . " ";
1846 sub VersionMessage {
1847 my $version_message;
1849 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
1850 $version_message .= 'Copyright (C) 2007-2008 Fabian Keil <fk@fabiankeil.de>' . "\n";
1851 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
1853 print $version_message;
1856 sub get_cli_options () {
1858 our %cli_options = (
1859 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
1860 'title' => CLI_OPTION_TITLE,
1861 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
1862 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
1863 'no-msecs' => CLI_OPTION_NO_MSECS,
1864 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
1865 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
1866 'statistic' => CLI_OPTION_STATISTIC,
1870 'html-output' => \$cli_options{'html-output'},
1871 'title' => \$cli_options{'title'},
1872 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
1873 'no-embedded-css' => \$cli_options{'no-embedded-css'},
1874 'no-msecs' => \$cli_options{'no-msecs'},
1875 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
1876 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
1877 'statistic' => \$cli_options{'statistic'},
1878 'version' => sub { VersionMessage && exit(0) }
1884 ################################################################################
1886 ################################################################################
1890 set_background(DEFAULT_BACKGROUND);
1891 prepare_our_stuff();
1904 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
1908 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
1909 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--show-ineffective-filters>]
1914 B<privoxy-log-parser> reads Privoxy log messages and
1916 - syntax-highlights recognized lines,
1918 - reformats some of them for easier comprehension,
1920 - filters out less useful messages, and
1922 - (in some cases) calculates additional information,
1923 like the compression ratio or how a filter affected
1926 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
1927 without getting confused by the resulting amount of output. For example for
1928 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
1929 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
1930 will hide the "filter foo caused 0 hits" message.
1934 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
1935 just don't highlight them.
1937 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
1938 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
1939 This option is only intended to make embedding log excerpts in web pages easier.
1940 It does not excape any input!
1942 [B<--no-msecs>] Don't expect milisecond resolution
1944 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
1945 the filtered output is piped into less in which case the ANSI control
1946 codes don't work, or if the terminal itself doesn't support the control
1949 [B<--show-ineffective-filters>] Don't suppress log lines for filters
1950 that didn't modify the content.
1952 [B<--version>] Print version and exit.
1956 To monitor a log file:
1958 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
1960 Replace '-F' with '-f' if your tail implementation lacks '-F' support
1961 or if the log won't get rotated anyway. The log file location depends
1962 on your system (Doh!).
1964 To monitor Privoxy without having it write to a log file:
1966 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
1968 Again, the config file location depends on your system. Output redirection
1969 depends on your shell, the above works with bourne shells.
1971 To read a processed Privoxy log file from top to bottom, letting the content
1972 scroll by slightly faster than you can read:
1974 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
1976 This is probably only useful to fill screens in the background of haxor movies.
1980 Syntax highlighting with ANSI escape sequences will look strange
1981 if your background color isn't black.
1983 Some messages aren't recognized yet and will not be fully highlighted.
1985 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
1986 using earlier Privoxy versions will probably result in an increased amount
1987 of unrecognized log lines.
1989 Privoxy's log files tend to be rather large. If you use HTML
1990 highlighting some browsers can't handle them, get confused and
1991 will eventually crash because of segmentation faults or unexpected
1992 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
1997 Many settings can't be controlled through command line options yet.
2005 Fabian Keil <fk@fabiankeil.de>