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 1.55 2009/10/08 11:47:12 fabiankeil Exp $
14 # - LOG_LEVEL_CGI, LOG_LEVEL_ERROR, LOG_LEVEL_WRITE content highlighting
15 # - create fancy statistics
16 # - grep through Privoxy sources to find unsupported log messages
17 # - hunt down substitutions that match content from variables which
18 # can contain stuff like ()?'[]
19 # - replace $h{'foo'} with h('foo') where possible
20 # - hunt down XXX comments instead of just creating them
21 # - add example log lines for every regex and mark them up for
23 # - Handle incomplete input without Perl warning about undefined variables.
24 # - Use generic highlighting function that takes a regex and the
27 # Copyright (c) 2007-2009 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 => '0.5',
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_STATISTICS => 0,
64 SUPPRESS_SUCCEEDED_FILTER_ADDITIONS => 1,
66 SHOW_FILTER_READIN_IN => 0,
67 SUPPRESS_EMPTY_LINES => 1,
68 SUPPRESS_SUCCESSFUL_CONNECTIONS => 1,
69 SUPPRESS_ACCEPTED_CONNECTIONS => 1,
70 SUPPRESS_GIF_NOT_CHANGED => 1,
71 SUPPRESS_NEED_TO_DE_CHUNK_FIRST => 1,
73 DEBUG_HEADER_REGISTERING => 0,
74 DEBUG_HEADER_HIGHLIGHTING => 0,
77 DEBUG_SUPPRESS_LOG_MESSAGES => 0,
79 PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH => 0,
80 PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH => 1,
82 LOG_UNPARSED_LINES_TO_EXTRA_FILE => 0,
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',
110 'Fatal error' => 'light_red',
111 'Gif-Deanimate' => 'blue',
113 Writing => 'light_green',
114 # ----------------------
117 request_ => 'brown', # host+path but no protocol
118 'ip-address' => 'yellow',
121 Truncation => 'light_red',
123 Timestamp => 'brown',
124 Crunching => 'light_red',
125 crunched => 'light_red',
126 'Request-Line' => 'pink',
128 destination => 'yellow',
129 'http-version' => 'pink',
130 'crunch-pattern' => 'pink',
135 'program-name' => 'cyan',
138 warning => 'light_red',
139 debug => 'light_red',
143 'status-message' => 'light_cyan',
144 'status-code' => 'yellow',
145 'invalid-request' => 'light_red',
147 error => 'light_red',
148 'rewritten-URL' => 'light_red',
149 'pcrs-delimiter' => 'light_red',
150 'ignored' => 'light_red',
151 'action-bits-update' => 'light_red',
152 'configuration-line' => 'red',
153 'content-type' => 'yellow',
158 # Header colours need their own hash so the keys can be accessed properly
159 our %header_colours = (
160 # Prefilled with headers that should not appear with default header colours
161 Cookie => 'light_red',
162 'Set-Cookie' => 'light_red',
163 Warning => 'light_red',
164 Default => HEADER_DEFAULT_COLOUR,
167 # Crunch reasons need their own hash as well
168 our %reason_colours = (
169 'Unsupported HTTP feature' => 'light_red',
170 Blocked => 'light_red',
171 Untrusted => 'light_red',
172 Redirected => 'green',
173 'CGI Call' => 'white',
174 'DNS failure' => 'red',
175 'Forwarding failed' => 'light_red',
176 'Connection failure' => 'light_red',
177 'Out of memory (may mask other reasons)' => 'light_red',
178 'No reason recorded' => 'light_red',
181 our @time_colours = ('white', 'light_gray');
183 # Translate highlight strings into highlight code
184 prepare_highlight_hash(\%header_colours);
185 prepare_highlight_hash(\%reason_colours);
186 prepare_highlight_hash(\%h);
187 prepare_colour_array(\@all_colours);
188 prepare_colour_array(\@time_colours);
195 ###############################################################
196 # Takes a colour string and returns an ANSI escape sequence
197 # (unless --no-syntax-highlighting is used).
198 # XXX: The Rolling Stones reference has to go.
199 ###############################################################
201 my $colour = shift @_;
203 return "" if cli_option_is_set('no-syntax-highlighting');
245 my $bg_code = get_background();
247 our $default = default_colours();
249 if (defined($text{$colour})) {
250 $colour_code = ESCAPE;
251 $colour_code .= $text{$colour};
253 $colour_code .= $light{$colour} ? "1" : "2";
255 $colour_code .= $bg_code;
257 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
259 } elsif ($colour =~ /reset/) {
261 $colour_code = default_colours();
265 die "What's $colour supposed to mean?\n";
271 sub get_semantic_html_markup ($) {
272 ###############################################################
273 # Takes a string and returns a span element
274 ###############################################################
279 if ($type =~ /Standard/) {
283 $code = '<span title="' . $type . '" class="' . $type . '">';
289 sub cli_option_is_set ($) {
292 my $cli_option = shift;
294 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
296 return $cli_options{$cli_option};
299 sub get_html_title () {
302 return $cli_options{'title'};
306 sub init_css_colours() {
314 purple => "F06", # XXX: wrong
315 cyan => "F09", # XXX: wrong
320 light_green => "33F",
329 sub get_css_colour ($) {
334 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
336 return '#' . $css_colours{$colour};
339 sub get_css_line ($) {
346 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
347 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
348 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
349 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
350 $css_line .= '}' . "\n";
355 sub get_css_line_for_colour ($) {
362 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
363 $css_line .= 'color:' . get_css_colour($colour) . ';';
364 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
365 $css_line .= '}' . "\n";
370 # XXX: Wrong solution
371 sub get_missing_css_lines () {
375 $css_line .= '.' . 'default' . ' {';
376 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
377 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
378 $css_line .= '}' . "\n";
386 our %css_colours; #XXX: Wrong solution
390 $css .= '.privoxy-log {';
391 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
392 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
395 foreach my $key (keys %h_colours) {
397 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
398 $css .= get_css_line($key);
402 foreach my $colour (keys %css_colours) {
404 $css .= get_css_line_for_colour($colour);
408 $css .= get_missing_css_lines(); #XXX: Wrong solution
417 if (cli_option_is_set('html-output')) {
419 my $title = get_html_title();
421 $intro .= '<html><head>';
422 $intro .= '<title>' . $title . '</title>';
423 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
424 $intro .= '</head><body>';
425 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
435 if (cli_option_is_set('html-output')) {
437 $outro = '</p></body></html>';
443 sub get_line_end () {
447 $line_end = '<br>' . $line_end if cli_option_is_set('html-output');
452 sub get_colour_html_markup ($) {
453 ###############################################################
454 # Takes a colour string a span element. XXX: WHAT?
455 # XXX: This function shouldn't be necessary, the
456 # markup should always be semantically correct.
457 ###############################################################
462 if ($type =~ /Standard/) {
465 $code = '<span class="' . lc($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 ($$$) {
742 my $result = shift; # XXX: Stupid name;
746 die "Unknown key $key" unless defined $h{$key};
748 if ($result =~ m@(.*?)($regex)(.*)@) {
749 $result = $1 . h($key) . $2 . h('Standard') . $3;
755 sub highlight_matched_path ($$) {
757 my $result = shift; # XXX: Stupid name;
760 if ($result =~ m@(.*?)($regex)(.*)@) {
761 $result = $1 . h('path') . $2 . h('Standard') . $3;
767 sub highlight_url ($) {
771 if (cli_option_is_set('html-output')) {
773 $url = '<a href="' . $url . '">' . $url . '</a>';
777 $url = h('URL') . $url . h('Standard');
784 ################################################################################
785 # loglevel-specific highlighter functions
786 ################################################################################
788 sub handle_loglevel_header ($) {
797 our $header_colour_index;
798 our $no_special_header_highlighting;
800 # Register new headers
801 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
802 if ($c =~ m/^scan: ((?>[^:]+)):/) {
804 if (!defined($header_colours{$header}) and $header =~ /^[\d\w-]*$/) {
805 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
807 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
808 $header_colours{$header} = $header_colours{'Default'};
810 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
811 $header_colour_index++;
816 if ($c =~ m/^scan: ((\w*) (.*) (HTTP\/\d\.\d))/) {
818 # Client request line
819 # Save for statistics (XXX: Not implemented yet)
820 $req{$t}{'method'} = $2;
821 $req{$t}{'destination'} = $3;
822 $req{$t}{'http-version'} = $4;
824 $content = highlight_request_line($1);
826 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
828 # Server response line
829 $req{$t}{'response_line'} = $2;
830 $req{$t}{'status_code'} = $3;
831 $req{$t}{'status_message'} = $4;
832 $content = $1 . highlight_response_line($req{$t}{'response_line'});
834 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
836 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
837 $content =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
838 $content =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
840 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
842 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/
843 $c = highlight_matched_host($c, '(?<=New host is: )[^\s]+');
844 $content = highlight_matched_url($c, '(?<=Crunching Referer: )[^\s]+');
846 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
848 # Text mode enabled by force. Take cover!
849 $content =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
851 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
853 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
854 $content = $1 . highlight_request_line($2);
856 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
858 # Adjusted Content-Length to 2132
859 # Adjust Content-Length to 33533
860 $content =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
861 $content = highlight_known_headers($content);
863 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
865 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
866 $content = highlight_matched_url($content, '(?<=New request URL: ).*');
868 } elsif ($c =~ m/^Couldn\'t parse:/) {
870 # XXX: These should probable be logged with LOG_LEVEL_ERROR
871 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
872 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
873 $content =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
875 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
876 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
878 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
879 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
880 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
882 # XXX: Save tag and tagger
884 $content =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
885 $content =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
886 $content =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
887 $no_special_header_highlighting = 1;
889 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
891 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
892 # XXX: Save tag and tagger
894 $content =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
895 $content =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
897 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
898 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
899 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
900 or $c =~ m/^(addh-unique|Referer forged to)/
901 or $c =~ m/^Downgraded answer to HTTP\/1.0/
902 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
903 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
904 or $c =~ m/^Referer crunched!/
905 or $c =~ m/^crunched x-forwarded-for!/
906 or $c =~ m/^crunched From!/
907 or $c =~ m/^ modified$/
908 or $c =~ m/^Content filtering is enabled. Crunching:/
909 or $c =~ m/^force-text-mode overruled the client/
910 or $c =~ m/^Server time in the future\./
911 or $c =~ m/^content-disposition header crunched and replaced with:/i
912 or $c =~ m/^Reducing white space in /
913 or $c =~ m/^Ignoring single quote in /
914 or $c =~ m/^Converting tab to space in /
915 or $c =~ m/A HTTP\/1\.1 response without/
916 or $c =~ m/Disabled filter mode on behalf of the client/
917 or $c =~ m/Keeping the (?:server|client) header /
918 or $c =~ m/Content modified with no Content-Length header set/
919 or $c =~ m/^Appended client IP address to/
920 or $c =~ m/^Removing 'Connection: close' to imply keep-alive./
923 # XXX: Some of these may need highlighting
925 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
926 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
927 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
928 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
929 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
930 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
931 # Text mode is already enabled
932 # Denied request with NULL byte(s) turned into line break(s)
933 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
934 # addh-unique: Host: people.freebsd.org
935 # Suppressed offer to compress content
936 # Crunched incoming cookie -- yum!
937 # Accepted the client's request to fetch without filtering.
938 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
939 # addh-unique: Host: subkeys.pgp.net:11371
940 # Referer forged to: Referer: http://10.0.0.1/
941 # Downgraded answer to HTTP/1.0
942 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
943 # Referer overwritten with: Referer: pille-palle
944 # Referer replaced with: Referer: pille-palle
945 # crunched x-forwarded-for!
947 # modified # XXX: pretty stupid log message
948 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
949 # force-text-mode overruled the client's request to fetch without filtering!
950 # Server time in the future.
951 # content-disposition header crunched and replaced with: content-disposition: filename=baz
952 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
953 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
954 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
955 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
957 # A HTTP/1.1 response without Connection header implies keep-alive.
958 # Disabled filter mode on behalf of the client.
959 # Keeping the server header 'Connection: keep-alive' around.
960 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
961 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
962 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
963 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
964 # Removing 'Connection: close' to imply keep-alive.
966 } elsif ($c =~ m/^scanning headers for:/) {
968 return '' unless SHOW_SCAN_INTRO;
970 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
971 # crunched User-Agent!
972 # Crunching: Content-Encoding: gzip
974 $content =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
976 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
978 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
980 $content = h('warning') . $content . h('Standard');
982 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
984 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
985 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
987 $content =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
988 $content =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
990 } elsif ($c =~ m/^Removing empty header/) {
992 # Removing empty header
995 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
997 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
998 # Enable force-text-mode if you know what you're doing.
999 # XXX: Could highlight more here.
1000 $content =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
1002 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
1004 # Server keep-alive timeout is 5. Sticking with 10.
1005 # Client keep-alive timeout is 20. Sticking with 10.
1007 $content =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1008 $content =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1010 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1012 # Reducing keep-alive timeout from 60 to 10.
1014 $content =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1015 $content =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1019 found_unknown_content($content);
1023 unless ($c =~ m/^Transforming/) {
1024 $content = highlight_known_headers($content) unless $no_special_header_highlighting;
1030 sub handle_loglevel_re_filter ($) {
1032 my $content = shift;
1038 our %header_colours;
1040 our $header_colour_index;
1043 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1045 # Used by Privoxy 3.0.5 and 3.0.6:
1047 # Used by Privoxy 3.0.7:
1048 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1050 $req{$t}{'filtered_header'} = $1;
1051 $req{$t}{'old_header_size'} = $2;
1052 $req{$t}{'header_filter_name'} = $3;
1054 unless (cli_option_is_set('show-ineffective-filters') or
1055 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1058 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1059 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1061 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1063 # ...produced 0 hits (new size 23).
1064 #... produced 1 hits (new size 54).
1066 $req{$t}{'header_filter_hits'} = $1;
1067 $req{$t}{'new_header_size'} = $2;
1069 unless (cli_option_is_set('show-ineffective-filters') or
1070 (defined($req{$t}{'header_filter_name'}) and
1071 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1073 if ($req{$t}{'header_filter_hits'} == 0 and
1074 not (defined($req{$t}{'header_filter_name'}) and
1075 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1078 # Reformat including information from the intro
1079 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1081 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1082 # $c .= $req{$t}{'filtered_header'} . " ";
1083 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1084 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1086 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1088 $c .= "changing size from ";
1089 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1091 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1096 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1100 # Highlight from last line (XXX: What?)
1101 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1102 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1107 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1108 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1113 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1115 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1116 # Tagger variable-test has empty joblist. Nothing to do.
1118 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1120 } elsif ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1122 # XXX: only the second version gets highlighted properly.
1123 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1124 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1125 $req{$t}{'content_source'} = $1;
1126 $req{$t}{'content_size'} = $2;
1127 $req{$t}{'content_filter'} = $3;
1128 $req{$t}{'content_hits'} = $4;
1129 $req{$t}{'new_content_size'} = $5;
1130 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1131 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1132 if ($req{$t}{'content_hits'} == 0 and
1133 not (cli_option_is_set('show-ineffective-filters')
1134 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1138 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1139 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1140 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1142 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1143 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1146 $c .= "(" . $h{'Number'};
1147 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1148 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1151 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1153 $req{$t}{'chunked-size'} = $1;
1154 $req{$t}{'dechunked-size'} = $2;
1155 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1157 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1159 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1160 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1162 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1164 # Decompression successful. Old size: 670, new size: 1166.
1166 $req{$t}{'size-compressed'} = $1;
1167 $req{$t}{'size-decompressed'} = $2;
1168 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1170 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1171 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1173 # XXX: Create sub get_percentage()
1174 if ($req{$t}{'size-decompressed'}) {
1175 $req{$t}{'decompression-gain-percent'} =
1176 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1178 $content .= " (saved: ";
1179 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1181 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1185 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1187 # Need to de-chunk first
1188 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1190 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1192 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1194 # Adding re_filter job ...
1195 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1196 # to filter client-ip-address succeeded.
1198 } elsif ($c =~ m/^Reading in filter/) {
1200 return '' unless SHOW_FILTER_READIN_IN;
1204 found_unknown_content($content);
1211 sub handle_loglevel_redirect ($) {
1218 if ($c =~ m/^Decoding "([^""]*)"/) {
1220 $req{$t}{'original-destination'} = $1;
1221 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1224 } elsif ($c =~ m/^Checking/) {
1226 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1227 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1229 # TODO: Change colour if really url-decoded
1230 $req{$t}{'decoded-original-destination'} = $1;
1231 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1234 } elsif ($c =~ m/^pcrs command "([^""]*)" changed "([^""]*)" to "([^""]*)" \((\d+) hits?\)/) {
1236 # pcrs command "s@&from=rss@@" changed "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1237 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1239 my ($pcrs_command, $url_before, $url_after, $hits) = ($1, $2, $3, $4); # XXX: save these?
1241 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1242 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1243 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1244 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1245 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1246 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1248 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1250 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1251 # XXX: Use URL highlighter
1253 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1255 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1256 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1257 # is already properly formatted.
1258 # XXX: assume the same?
1259 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1263 found_unknown_content($c);
1270 sub handle_loglevel_gif_deanimate ($) {
1272 my $content = shift;
1277 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1279 my $bytes_from = $1;
1281 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1282 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1283 # XXX: Do we need g in case of ($1 == $2)?
1284 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1286 } elsif ($content =~ m/GIF (not) changed/) {
1288 # Gif-Deanimate: GIF not changed.
1289 return '' if SUPPRESS_GIF_NOT_CHANGED;
1290 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1292 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1294 # failed! (gif parsing)
1295 # XXX: Replace this error message with something less stupid
1296 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1298 } elsif ($content =~ m/^Need to de-chunk first/) {
1300 # Need to de-chunk first
1301 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1303 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1305 # No GIF header found (XXX: Did I ever commit this?)
1306 # failed while parsing 195 134747048 (XXX: never commited)
1308 # Ignore these for now
1312 found_unknown_content($content);
1319 sub handle_loglevel_request ($) {
1321 my $content = shift;
1325 our %reason_colours;
1327 if ($content =~ m/crunch! /) {
1329 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1331 # Highlight crunch reasons
1332 foreach my $reason (keys %reason_colours) {
1333 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1335 # Highlight request URL domain and ditch 'crunch!'
1336 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1337 $content =~ s@ crunch!@@;
1339 } elsif ($content =~ m/\[too long, truncated\]$/) {
1341 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1342 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1344 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1346 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1347 $content = h('request_') . $content . h('Standard');
1351 found_unknown_content($content);
1358 sub handle_loglevel_crunch ($) {
1360 my $content = shift;
1362 our %reason_colours;
1364 # Highlight crunch reason
1365 foreach my $reason (keys %reason_colours) {
1366 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1369 if ($content =~ m/\[too long, truncated\]$/) {
1371 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1372 # [...]&filter... [too long, truncated]
1373 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1377 # Blocked: http://ads.example.org/
1378 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1384 sub handle_loglevel_connect ($) {
1391 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1393 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1395 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1396 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1398 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1400 # connect to: www.example.org.noconnect failed: Operation not permitted
1402 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1404 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1406 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1408 # Connect: to www.nzherald.co.nz successful
1409 # Connect: to archiv.radiotux.de
1411 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1412 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1414 } elsif ($c =~ m/^to ([^\s]*)$/) {
1416 # Connect: to lists.sourceforge.net:443
1418 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1420 } elsif ($c =~ m/^accepted connection from .*/ or
1423 # accepted connection from 10.0.0.1
1424 # Privoxy 3.0.6 and earlier just say:
1426 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1427 $c = highlight_matched_host($c, '(?<=connection from ).*');
1429 } elsif ($c =~ m/^write header to: .* failed:/) {
1431 # write header to: 10.0.0.1 failed: Broken pipe
1433 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1434 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1436 } elsif ($c =~ m/^write header to client failed:/) {
1438 # write header to client failed: Broken pipe
1440 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1442 } elsif ($c =~ m/^socks4_connect:/) {
1444 # socks4_connect: SOCKS request rejected or failed.
1445 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1447 } elsif ($c =~ m/^Listening for new connections/ or
1448 $c =~ m/^accept connection/) {
1450 # Privoxy versions above 3.0.6 say:
1451 # Listening for new connections ...
1452 # earlier versions say:
1453 # accept connection ...
1456 } elsif ($c =~ m/^accept failed:/) {
1458 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1460 } elsif ($c =~ m/^Overriding forwarding settings/) {
1462 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1463 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1465 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1467 # Denying suspicious CONNECT request from 10.0.0.1
1468 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1470 } elsif ($c =~ m/^socks5_connect:/) {
1472 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1474 } elsif ($c =~ m/^Created new connection to/) {
1476 # Created new connection to www.privoxy.org:80 on socket 11.
1477 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1478 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1480 } elsif ($c =~ m/^Found reusable socket/) {
1482 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1484 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1485 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1486 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1487 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1488 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1489 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1490 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1491 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1493 } elsif ($c =~ m/^Marking open socket/) {
1495 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1496 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1497 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1498 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1500 } elsif ($c =~ m/^No reusable/) {
1502 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1503 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1505 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1507 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1508 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1510 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1511 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1512 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1514 } elsif ($c =~ m/^Socket/) {
1516 # Socket 16 already forgotten or never remembered.
1517 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1519 } elsif ($c =~ m/^The connection to/) {
1521 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1523 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1524 # Timeout is: 1. Assumed latency: 4.
1525 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1526 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1527 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1528 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1529 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1530 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1532 } elsif ($c =~ m/^Stopped waiting for the request line./) {
1534 # Stopped waiting for the request line. Timeout: 121.
1535 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1537 } elsif ($c =~ m/^Waiting for \d/) {
1539 # Waiting for 1 connections to timeout.
1540 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1542 } elsif ($c =~ m/^Initialized/) {
1544 # Initialized 20 socket slots.
1545 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1547 } elsif ($c =~ m/^Done reading from server/) {
1549 # Done reading from server. Expected content length: 24892. \
1550 # Actual content length: 24892. Most recently received: 4412.
1552 # Done reading from server. Expected content length: 24892. \
1553 # Actual content length: 24892. Bytes most recently read: 4412.
1554 # Done reading from server. Content length: 6018 as expected. \
1555 # Bytes most recently read: 294.
1556 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1557 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1558 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1560 } elsif ($c =~ m/^Continuing buffering headers/) {
1562 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1563 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1564 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1565 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1567 # Continuing buffering headers. Bytes most recently read: %d.
1568 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1570 } elsif ($c =~ m/^Received \d+ bytes while/) {
1572 # Received 206 bytes while expecting 12103.
1573 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1574 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1576 } elsif ($c =~ m/^Connection from/) {
1578 # Connection from 81.163.28.218 dropped due to ACL
1579 $c =~ s@(?<=^Connection from )((?:\d+\.?){4})@$h{'Number'}$1$h{'Standard'}@;
1581 } elsif ($c =~ m/^(?:Reusing|Closing) server socket \d./ or
1582 $c =~ m/^No additional client request/) {
1584 # Reusing server socket 4. Opened for 10.0.0.1.
1585 # Closing server socket 2. Opened for 10.0.0.1.
1586 # No additional client request received in time. \
1587 # Closing server socket 4, initially opened for 10.0.0.1.
1589 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1590 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.$)');
1592 } elsif ($c =~ m/^Connected to /) {
1594 # Connected to tor-jail[10.0.0.2]:9050.
1596 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1597 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1598 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1600 } elsif ($c =~ m/^Could not connect to /) {
1602 # Could not connect to [10.0.0.1]:80.
1604 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1605 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1607 } elsif ($c =~ m/^Waiting for the next client request/ or
1608 $c =~ m/^The connection on server socket/ ) {
1610 # Waiting for the next client request. Keeping the server socket 5 to 10.0.0.1 open.
1611 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1613 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1614 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1616 } elsif ($c =~ m/^Marking the server socket/) {
1618 # Marking the server socket 7 tainted.
1620 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1622 } elsif ($c =~ m/^Reduced expected bytes to /) {
1624 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1625 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1626 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1628 } elsif ($c =~ m/^The client closed socket /) {
1630 # The client closed socket 2 while the server socket 4 is still open.
1631 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1632 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1634 } elsif ($c =~ m/^Looks like we rea/ or
1635 $c =~ m/^Unsetting keep-alive flag/ or
1636 $c =~ m/^No connections to wait/ or
1637 $c =~ m/^Client request arrived in time or the client closed the connection/ or
1638 $c =~ m/^Complete client request received/ or
1639 $c =~ m/^Possible pipeline attempt detected./ or
1640 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1641 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1642 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1643 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1644 $c =~ m/^Stopping to watch the client socket. There's already another request waiting./) {
1646 # Looks like we reached the end of the last chunk. We better stop reading.
1647 # Looks like we read the end of the last chunk together with the server \
1648 # headers. We better stop reading.
1649 # Unsetting keep-alive flag.
1650 # No connections to wait for left.
1651 # Client request arrived in time or the client closed the connection.
1652 # Complete client request received
1653 # Possible pipeline attempt detected. The connection will not be \
1654 # kept alive and we will only serve the first request.
1655 # POST request detected. The connection will not be kept alive.
1656 # The server still wants to talk, but the client hung up on us.
1657 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1658 # There might be a request body. The connection will not be kept alive.
1659 # Stopping to watch the client socket. There's already another request waiting.
1663 found_unknown_content($c);
1671 sub handle_loglevel_info ($) {
1678 if ($c =~ m/^Rewrite detected:/) {
1680 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1681 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1683 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1684 $c =~ m/^Compressed content detected/ or
1687 # Decompressing deflated iob: 117
1688 # Decompression didn't result in any content.
1689 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1690 # with zlib support or enable the prevent-compression action.
1691 # Tagger 'complete-url' created empty tag. Ignored.
1695 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1697 # loading configuration file '/usr/local/etc/privoxy/config':
1698 # Reloading configuration file '/usr/local/etc/privoxy/config'
1699 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1701 } elsif ($c =~ m/^exiting by signal/) {
1703 # exiting by signal 15 .. bye
1704 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1706 } elsif ($c =~ m/^Privoxy version/) {
1708 # Privoxy version 3.0.7
1709 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1711 } elsif ($c =~ m/^Program name: /) {
1713 # Program name: /usr/local/sbin/privoxy
1714 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1716 } elsif ($c =~ m/^Listening on port /) {
1718 # Listening on port 8118 on IP address 10.0.0.1
1719 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1720 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1722 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1724 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1725 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1727 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1729 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1730 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1731 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1733 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1734 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1735 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1736 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1738 } elsif ($c =~ m/^Status code/) {
1740 # Status code 304 implies no body.
1741 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1743 } elsif ($c =~ m/^Method/) {
1745 # Method HEAD implies no body.
1746 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1748 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1750 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1751 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1752 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1754 } elsif ($c =~ m/^No logfile configured/ or
1755 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1756 $c =~ m/^Invalid \"chunked\" transfer/ or
1757 $c =~ m/^Support for/ or
1758 $c =~ m/^Flushing header and buffers/
1761 # No logfile configured. Please enable it before reporting any problems.
1762 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1763 # response or even no response at all.
1764 # No logfile configured. Logging disabled.
1765 # Invalid "chunked" transfer encoding detected and ignored.
1766 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1767 # known not to work properly in some situations.
1768 # Flushing header and buffers. Stepping back from filtering.
1772 found_unknown_content($c);
1779 sub handle_loglevel_cgi ($) {
1786 if ($c =~ m/^Granting access to/) {
1788 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1790 } elsif ($c =~ m/^Substituting: s(.)/) {
1792 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1793 # XXX: prone to span several lines
1796 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1797 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1798 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1804 sub handle_loglevel_force ($) {
1811 if ($c =~ m/^Ignored force prefix in request:/) {
1813 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1814 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1815 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1817 } elsif ($c =~ m/^Enforcing request:/) {
1819 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1820 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1824 found_unknown_content($c);
1831 sub handle_loglevel_error ($) {
1836 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1838 # Empty server or forwarder response received on socket 4.
1839 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1841 # XXX: There are probably more messages that deserve highlighting.
1847 sub handle_loglevel_ignore ($) {
1851 sub gather_loglevel_request_stats ($$) {
1859 sub gather_loglevel_crunch_stats ($$) {
1869 sub gather_loglevel_error_stats ($$) {
1876 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1878 # Empty server or forwarder response received on socket 4.
1879 $stats{'empty-responses'}++;
1880 if ($thread_data{$thread}{'new_connection'}) {
1881 $stats{'empty-responses-on-new-connections'}++;
1883 $stats{'empty-responses-on-reused-connections'}++;
1888 sub gather_loglevel_connect_stats ($$) {
1895 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
1897 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1898 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
1900 } elsif ($c =~ m/^to ([^\s]*)$/) {
1902 # Connect: to lists.sourceforge.net:443
1904 $thread_data{$thread}{'forwarder'} = 'direct connection';
1906 } elsif ($c =~ m/^Created new connection to/) {
1908 # Created new connection to www.privoxy.org:80 on socket 11.
1910 $thread_data{$thread}{'new_connection'} = 1;
1912 } elsif ($c =~ m/^Reusing server socket \d./ or
1913 $c =~ m/^Found reusable socket/) {
1915 # Reusing server socket 4. Opened for 10.0.0.1.
1916 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1918 $thread_data{$thread}{'new_connection'} = 0;
1919 $stats{'reused-connections'}++;
1923 sub gather_loglevel_header_stats ($) {
1929 if ($c =~ m/^A HTTP\/1\.1 response without/ or
1930 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
1932 # A HTTP/1.1 response without Connection header implies keep-alive.
1933 # Keeping the server header 'Connection: keep-alive' around.
1934 $stats{'server-keep-alive'}++;
1942 'server-keep-alive' => 0,
1943 'reused-connections' => 0,
1944 'empty-responses' => 0,
1945 'empty-responses-on-new-connections' => 0,
1946 'empty-responses-on-reused-connections' => 0,
1950 sub get_percentage ($$) {
1953 return "NaN" if ($big eq 0);
1954 return sprintf("%.2f%%", $small / $big * 100);
1957 sub print_stats () {
1960 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
1961 my $outgoing_requests = $stats{requests} - $stats{crunches};
1963 if ($stats{requests} eq 0) {
1964 print "No requests yet.\n";
1968 print "Client requests total: " . $stats{requests} . "\n";
1969 print "Crunches: " . $stats{crunches} . " (" .
1970 get_percentage($stats{requests}, $stats{crunches}) . ")\n";
1971 print "Outgoing requests: " . $outgoing_requests . " (" .
1972 get_percentage($stats{requests}, $outgoing_requests) . ")\n";
1973 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
1974 get_percentage($stats{requests}, $stats{'server-keep-alive'}) . ")\n";
1975 print "New outgoing connections: " . $new_connections . " (" .
1976 get_percentage($stats{requests}, $new_connections) . ")\n";
1977 print "Reused connections: " . $stats{'reused-connections'} . " (" .
1978 get_percentage($stats{requests}, $stats{'reused-connections'}) . ")\n";
1979 print "Empty responses: " . $stats{'empty-responses'} . " (" .
1980 get_percentage($stats{requests}, $stats{'empty-responses'}) . ")\n";
1981 print "Empty responses on new connections: "
1982 . $stats{'empty-responses-on-new-connections'} . " (" .
1983 get_percentage($stats{requests}, $stats{'empty-responses-on-new-connections'})
1985 print "Empty responses on reused connections: " .
1986 $stats{'empty-responses-on-reused-connections'} . " (" .
1987 get_percentage($stats{requests}, $stats{'empty-responses-on-reused-connections'}) .
1992 ################################################################################
1993 # Functions that actually print stuff
1994 ################################################################################
1996 sub print_clf_message () {
1998 our ($ip, $timestamp, $request_line, $status_code, $size);
2002 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2004 # Rebuild highlighted
2005 $output .= $h{'Number'} . $ip . $h{'Standard'};
2007 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2009 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2011 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2013 $output .= $h{'Number'} . $size . $h{'Standard'};
2014 $output .= get_line_end();
2019 sub print_non_clf_message ($) {
2022 our %thread_colours;
2025 our $time_colour_index;
2028 my $content = shift;
2029 my ($day, $time_stamp, $msecs, $thread, $log_level)
2030 = ($req{$t}{'day'}, $req{$t}{'time-stamp'}, $req{$t}{'msecs'}, $t, $req{$t}{'log-level'} );
2032 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2034 $output .= $h{"Standard"} unless cli_option_is_set('html-output');
2035 # $output .= "$day ";
2036 $output .= $time_colours[$time_colour_index % 2];
2038 $output .= $time_stamp;
2039 $output .= ".$msecs" unless cli_option_is_set('no-msecs');
2040 $output .= $h{"Standard"};
2042 $output .= $thread_colours{$thread} if (defined($thread_colours{$thread}));
2044 $output .= $h{"Standard"} . " ";
2045 $output .= $h{$log_level} if (defined($h{$log_level}));
2046 $output .= $log_level;
2047 $output .= $h{"Standard"} . ": ";
2048 $output .= "$content";
2049 $output .= get_line_end();
2057 our %req; # request data from previous lines
2059 our %thread_colours;
2062 our $thread_colour_index = 0;
2063 our $header_colour_index = 0;
2064 our $time_colour_index = 0;
2066 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2068 my $last_thread = 0;
2069 my $last_timestamp = 0;
2071 my $filters_that_did_nothing;
2074 our $no_special_header_highlighting;
2075 $time_colour = paint_it('white');
2077 my %log_level_handlers = (
2078 'Re-Filter' => \&handle_loglevel_re_filter,
2079 'Header' => \&handle_loglevel_header,
2080 'Connect' => \&handle_loglevel_connect,
2081 'Redirect' => \&handle_loglevel_redirect,
2082 'Request' => \&handle_loglevel_request,
2083 'Crunch' => \&handle_loglevel_crunch,
2084 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2085 'Info' => \&handle_loglevel_info,
2086 'CGI' => \&handle_loglevel_cgi,
2087 'Force' => \&handle_loglevel_force,
2088 'Error' => \&handle_loglevel_error,
2089 'Fatal error' => \&handle_loglevel_ignore,
2090 'Writing' => \&handle_loglevel_ignore,
2091 'Unknown log level' => \&handle_loglevel_ignore,
2098 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2099 # XXX: Put in req hash?
2102 $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution
2107 $req{$t}{'day'} = $day;
2108 $req{$t}{'time-stamp'} = $time_stamp;
2109 $req{$t}{'msecs'} = $msecs; # Only the cool kids have micro second resolution;
2110 $req{$t}{'log-level'} = $log_level;
2111 $req{$t}{'content'} = $content;
2112 $req{$t}{'log-message'} = $_;
2113 $no_special_header_highlighting = 0;
2115 if (defined($log_level_handlers{$log_level})) {
2117 $content = $log_level_handlers{$log_level}($content);
2121 die "No handler found for log level \"$log_level\"\n";
2125 # Highlight Truncations
2126 if (m/\.\.\. \[(too long, truncated)/) {
2127 $content =~ s@($1)@$h{'Truncation'}$1$h{'Standard'}@g;
2130 next unless $content;
2132 # Register threads to keep the colour constant
2133 if (!defined($thread_colours{$thread})) {
2134 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2135 $thread_colour_index++;
2138 # Switch timestamp colour if timestamps differ
2139 if ($msecs != $last_msecs || !($time_stamp =~ m/$last_timestamp/)) {
2140 debug_message("Tick tack!") if DEBUG_TICKS;
2141 $time_colour = $time_colours[$time_colour_index % 2];
2142 $time_colour_index++
2145 $last_msecs = $msecs;
2146 $last_thread = $thread;
2147 $last_timestamp = $time_stamp;
2149 print_non_clf_message($content);
2151 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2153 # LOG_LEVEL_CLF lines look like this
2154 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2155 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2156 # HTTP/1.1" 403 1730
2157 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2159 print_clf_message();
2163 # Some Privoxy log messages span more than one line,
2164 # usually to dump lots of content that doesn't need any syntax highlighting.
2165 # XXX: add mechanism to forward these lines to the right handler anyway.
2167 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2168 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2176 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2177 my %log_level_handlers = (
2178 'Re-Filter' => \&handle_loglevel_ignore,
2179 'Header' => \&gather_loglevel_header_stats,
2180 'Connect' => \&gather_loglevel_connect_stats,
2181 'Redirect' => \&handle_loglevel_ignore,
2182 'Request' => \&gather_loglevel_request_stats,
2183 'Crunch' => \&gather_loglevel_crunch_stats,
2184 'Gif-Deanimate' => \&handle_loglevel_ignore,
2185 'Info' => \&handle_loglevel_ignore,
2186 'CGI' => \&handle_loglevel_ignore,
2187 'Force' => \&handle_loglevel_ignore,
2188 'Error' => \&gather_loglevel_error_stats,
2189 'Fatal error' => \&handle_loglevel_ignore,
2190 'Writing' => \&handle_loglevel_ignore,
2191 'Unknown log level' => \&handle_loglevel_ignore
2195 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2198 $msecs = $3 ? $3 : 0;
2203 if (defined($log_level_handlers{$log_level})) {
2205 $content = $log_level_handlers{$log_level}($content, $thread);
2209 die "No handler found for log level \"$log_level\"\n";
2219 sub VersionMessage {
2220 my $version_message;
2222 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2223 $version_message .= 'Copyright (C) 2007-2009 Fabian Keil <fk@fabiankeil.de>' . "\n";
2224 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2226 print $version_message;
2229 sub get_cli_options () {
2231 our %cli_options = (
2232 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2233 'title' => CLI_OPTION_TITLE,
2234 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2235 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2236 'no-msecs' => CLI_OPTION_NO_MSECS,
2237 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2238 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
2239 'statistics' => CLI_OPTION_STATISTICS,
2243 'html-output' => \$cli_options{'html-output'},
2244 'title' => \$cli_options{'title'},
2245 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2246 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2247 'no-msecs' => \$cli_options{'no-msecs'},
2248 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2249 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
2250 'statistics' => \$cli_options{'statistics'},
2251 'version' => sub { VersionMessage && exit(0) },
2264 Options and their default values if they have any:
2265 [--accept-unknown-messages]
2269 [--no-syntax-highlighting]
2270 [--show-ineffective-filters]
2272 [--title $cli_options{'title'}]
2274 see "perldoc $0" for more information
2280 ################################################################################
2282 ################################################################################
2286 set_background(DEFAULT_BACKGROUND);
2287 prepare_our_stuff();
2291 if (cli_option_is_set('statistics')) {
2304 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2308 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
2309 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--show-ineffective-filters>]
2314 B<privoxy-log-parser> reads Privoxy log messages and
2316 - syntax-highlights recognized lines,
2318 - reformats some of them for easier comprehension,
2320 - filters out less useful messages, and
2322 - (in some cases) calculates additional information,
2323 like the compression ratio or how a filter affected
2326 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2327 without getting confused by the resulting amount of output. For example for
2328 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2329 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2330 will hide the "filter foo caused 0 hits" message.
2334 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
2335 just don't highlight them.
2337 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2338 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2339 This option is only intended to make embedding log excerpts in web pages easier.
2340 It does not escape any input!
2342 [B<--no-msecs>] Don't expect milisecond resolution
2344 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2345 the filtered output is piped into less in which case the ANSI control
2346 codes don't work, or if the terminal itself doesn't support the control
2349 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2350 that didn't modify the content.
2352 [B<--statistics>] Gather various statistics instead of syntax highlighting
2353 log messages. This is an experimental feature, if the results look wrong
2354 they very well might be. Also note that the results a pretty much guaranteed
2355 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2357 [B<--version>] Print version and exit.
2361 To monitor a log file:
2363 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2365 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2366 or if the log won't get rotated anyway. The log file location depends
2367 on your system (Doh!).
2369 To monitor Privoxy without having it write to a log file:
2371 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2373 Again, the config file location depends on your system. Output redirection
2374 depends on your shell, the above works with bourne shells.
2376 To read a processed Privoxy log file from top to bottom, letting the content
2377 scroll by slightly faster than you can read:
2379 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2381 This is probably only useful to fill screens in the background of haxor movies.
2385 Syntax highlighting with ANSI escape sequences will look strange
2386 if your background color isn't black.
2388 Some messages aren't recognized yet and will not be fully highlighted.
2390 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2391 using earlier Privoxy versions will probably result in an increased amount
2392 of unrecognized log lines.
2394 Privoxy's log files tend to be rather large. If you use HTML
2395 highlighting some browsers can't handle them, get confused and
2396 will eventually crash because of segmentation faults or unexpected
2397 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2402 Many settings can't be controlled through command line options yet.
2410 Fabian Keil <fk@fabiankeil.de>