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 1.23 2009/03/14 15:31:58 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_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',
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);
193 ###############################################################
194 # Takes a colour string and returns an ANSI escape sequence
195 # (unless --no-syntax-highlighting is used).
196 # XXX: The Rolling Stones reference has to go.
197 ###############################################################
199 my $colour = shift @_;
201 return "" if cli_option_is_set('no-syntax-highlighting');
243 my $bg_code = get_background();
245 our $default = default_colours();
247 if (defined($text{$colour})) {
248 $colour_code = ESCAPE;
249 $colour_code .= $text{$colour};
251 $colour_code .= $light{$colour} ? "1" : "2";
253 $colour_code .= $bg_code;
255 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
257 } elsif ($colour =~ /reset/) {
259 $colour_code = default_colours();
263 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>';
441 sub get_line_end () {
445 $line_end = '<br>' . $line_end if cli_option_is_set('html-output');
450 sub get_colour_html_markup ($) {
451 ###############################################################
452 # Takes a colour string a span element. XXX: WHAT?
453 # XXX: This function shouldn't be necessary, the
454 # markup should always be semantically correct.
455 ###############################################################
460 if ($type =~ /Standard/) {
463 $code = '<span class="' . lc($type) . '">';
469 sub default_colours () {
472 return reset_colours();
475 sub show_colours () {
479 sub reset_colours () {
480 return ESCAPE . "0m";
483 sub set_background ($){
499 if (defined($backgrounds{$colour})) {
500 $bg_code = $backgrounds{$colour};
502 die "Invalid background colour: " . $colour;
506 sub get_background (){
510 sub prepare_highlight_hash ($) {
513 if (!cli_option_is_set('html-output')) {
515 foreach my $key (keys %$ref) {
516 $$ref{$key} = paint_it($$ref{$key});
521 foreach my $key (keys %$ref) {
522 $$ref{$key} = get_semantic_html_markup($key);
528 sub prepare_colour_array ($) {
531 if (!cli_option_is_set('html-output')) {
533 foreach my $i (0 ... @$ref - 1) {
534 $$ref[$i] = paint_it($$ref[$i]);
539 foreach my $i (0 ... @$ref - 1) {
540 $$ref[$i] = get_colour_html_markup($$ref[$i]);
546 sub found_unknown_content ($) {
554 return if cli_option_is_set('accept-unknown-messages');
556 return if ($unknown =~ /\[too long, truncated\]$/);
558 $message = "found_unknown_content: Don't know how to highlight: ";
559 # Break line so the log file can later be parsed as Privoxy log file again
560 $message .= '"' . $unknown . '"' . " in:\n";
561 $message .= $req{$t}{'log-message'};
562 debug_message($message);
563 log_parse_error($req{$t}{'log-message'});
565 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
568 sub log_parse_error ($) {
572 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
573 open(ERRORLOG, ">>" . ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
574 print ERRORLOG $message;
579 sub debug_message (@) {
583 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
586 ################################################################################
587 # highlighter functions that aren't loglevel-specific
588 ################################################################################
592 # Get highlight marker
594 my $highlight = shift; # XXX: Stupid name;
598 if (defined($highlight)) {
600 $result = $h{$highlight};
604 $message = "h: Don't recognize highlighter $highlight.";
605 debug_message($message);
606 log_parser_error($message);
607 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
613 sub highlight_known_headers ($) {
618 my $headers = join ('|', keys %header_colours);
620 debug_message("Searching $content for things to highlight.") if DEBUG_HEADER_HIGHLIGHTING;
622 if ($content =~ m/(?<=\s)($headers):/) {
624 $content =~ s@(?<=[\s|'])($header)(?=:)@$header_colours{$header}$1$h{'Standard'}@ig;
625 debug_message("Highlighted $content") if DEBUG_HEADER_HIGHLIGHTING;
631 sub highlight_matched_request_line ($$) {
633 my $result = shift; # XXX: Stupid name;
635 if ($result =~ m@(.*)($regex)(.*)@) {
636 $result = $1 . highlight_request_line($2) . $3
641 sub highlight_request_line ($) {
644 my ($method, $url, $http_version);
647 #GET http://images.sourceforge.net/sfx/icon_warning.gif HTTP/1.1
648 if ($rl =~ m/Invalid request/) {
650 $rl = h('invalid-request') . $rl . h('Standard');
652 } elsif ($rl =~ m/^([-\w]+) (.*) (HTTP\/\d\.\d)/) {
654 # XXX: might not match in case of HTTP method fuzzing.
655 # XXX: save these: ($method, $path, $http_version) = ($1, $2, $3);
656 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
657 if ($rl =~ /http:\/\//) {
658 $rl = highlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
660 $rl = highlight_matched_pattern($rl, 'request_', '[^\s]*(?=\sHTTP)');
663 $rl =~ s@(HTTP\/\d\.\d)$@$h{'http-version'}$1$h{'Standard'}@;
665 } elsif ($rl =~ m/\.\.\. \[too long, truncated\]$/) {
667 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
668 $rl = highlight_matched_url($rl, '[^\s]*(?=\.\.\.)');
670 } elsif ($rl =~ m/^ $/) {
672 $rl = h('error') . "No request line specified!" . h('Standard');
676 debug_message ("Can't parse request line: $rl");
683 sub highlight_response_line ($) {
686 my ($http_version, $status_code, $status_message);
691 # TODO: Mark different status codes differently
693 if ($rl =~ m/((?:HTTP\/\d\.\d|ICY)) (\d+) (.*)/) {
694 ($http_version, $status_code, $status_message) = ($1, $2, $3);
696 debug_message ("Can't parse response line: $rl") and die 'Fix this';
699 # Rebuild highlighted
701 $rl .= h('http-version') . $http_version . h('Standard');
703 $rl .= h('status-code') . $status_code . h('Standard');
705 $rl .= h('status-message') . $status_message . h('Standard');
710 sub highlight_matched_url ($$) {
712 my $result = shift; # XXX: Stupid name;
715 #print "Got $result, regex ($regex)\n";
717 if ($result =~ m@(.*?)($regex)(.*)@) {
718 $result = $1 . highlight_url($2) . $3;
719 #print "Now the result is $result\n";
725 sub highlight_matched_host ($$) {
727 my $result = shift; # XXX: Stupid name;
730 if ($result =~ m@(.*?)($regex)(.*)@) {
731 $result = $1 . h('host') . $2 . h('Standard') . $3;
737 sub highlight_matched_pattern ($$$) {
740 my $result = shift; # XXX: Stupid name;
744 die "Unknown key $key" unless defined $h{$key};
746 if ($result =~ m@(.*?)($regex)(.*)@) {
747 $result = $1 . h($key) . $2 . h('Standard') . $3;
753 sub highlight_matched_path ($$) {
755 my $result = shift; # XXX: Stupid name;
758 if ($result =~ m@(.*?)($regex)(.*)@) {
759 $result = $1 . h('path') . $2 . h('Standard') . $3;
765 sub highlight_url ($) {
769 if (cli_option_is_set('html-output')) {
771 $url = '<a href="' . $url . '">' . $url . '</a>';
775 $url = h('URL') . $url . h('Standard');
782 ################################################################################
783 # loglevel-specific highlighter functions
784 ################################################################################
786 sub handle_loglevel_header ($) {
795 our $header_colour_index;
796 our $no_special_header_highlighting;
798 # Register new headers
799 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
800 if ($c =~ m/^scan: ((?>[^:]+)):/) {
802 if (!defined($header_colours{$header}) and $header =~ /^[\d\w-]*$/) {
803 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
805 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
806 $header_colours{$header} = $header_colours{'Default'};
808 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
809 $header_colour_index++;
814 if ($c =~ m/^scan: ((\w*) (.*) (HTTP\/\d\.\d))/) {
816 # Client request line
817 # Save for statistics (XXX: Not implemented yet)
818 $req{$t}{'method'} = $2;
819 $req{$t}{'destination'} = $3;
820 $req{$t}{'http-version'} = $4;
822 $content = highlight_request_line($1);
824 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
826 # Server response line
827 $req{$t}{'response_line'} = $2;
828 $req{$t}{'status_code'} = $3;
829 $req{$t}{'status_message'} = $4;
830 $content = $1 . highlight_response_line($req{$t}{'response_line'});
832 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
834 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
835 $content =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
836 $content =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
838 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
840 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/
841 $c = highlight_matched_host($c, '(?<=New host is: )[^\s]+');
842 $content = highlight_matched_url($c, '(?<=Crunching Referer: )[^\s]+');
844 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
846 # Text mode enabled by force. Take cover!
847 $content =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
849 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
851 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
852 $content = $1 . highlight_request_line($2);
854 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
856 # Adjusted Content-Length to 2132
857 # Adjust Content-Length to 33533
858 $content =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
859 $content = highlight_known_headers($content);
861 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
863 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
864 $content = highlight_matched_url($content, '(?<=New request URL: ).*');
866 } elsif ($c =~ m/^Couldn\'t parse:/) {
868 # XXX: These should probable be logged with LOG_LEVEL_ERROR
869 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
870 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
871 $content =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
873 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
874 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
876 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
877 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
878 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
880 # XXX: Save tag and tagger
882 $content =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
883 $content =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
884 $content =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
885 $no_special_header_highlighting = 1;
887 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
889 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
890 # XXX: Save tag and tagger
892 $content =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
893 $content =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
895 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
896 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
897 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
898 or $c =~ m/^(addh-unique|Referer forged to)/
899 or $c =~ m/^Downgraded answer to HTTP\/1.0/
900 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
901 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
902 or $c =~ m/^Referer crunched!/
903 or $c =~ m/^crunched x-forwarded-for!/
904 or $c =~ m/^crunched From!/
905 or $c =~ m/^ modified$/
906 or $c =~ m/^Content filtering is enabled. Crunching:/
907 or $c =~ m/^force-text-mode overruled the client/
908 or $c =~ m/^Server time in the future\./
909 or $c =~ m/^content-disposition header crunched and replaced with:/i
910 or $c =~ m/^Reducing white space in /
911 or $c =~ m/^Ignoring single quote in /
912 or $c =~ m/^Converting tab to space in /
913 or $c =~ m/A HTTP\/1\.1 response without/
914 or $c =~ m/Disabled filter mode on behalf of the client/
917 # XXX: Some of these may need highlighting
919 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
920 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
921 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
922 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
923 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
924 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
925 # Text mode is already enabled
926 # Denied request with NULL byte(s) turned into line break(s)
927 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
928 # addh-unique: Host: people.freebsd.org
929 # Suppressed offer to compress content
930 # Crunched incoming cookie -- yum!
931 # Accepted the client's request to fetch without filtering.
932 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
933 # addh-unique: Host: subkeys.pgp.net:11371
934 # Referer forged to: Referer: http://10.0.0.1/
935 # Downgraded answer to HTTP/1.0
936 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
937 # Referer overwritten with: Referer: pille-palle
938 # Referer replaced with: Referer: pille-palle
939 # crunched x-forwarded-for!
941 # modified # XXX: pretty stupid log message
942 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
943 # force-text-mode overruled the client's request to fetch without filtering!
944 # Server time in the future.
945 # content-disposition header crunched and replaced with: content-disposition: filename=baz
946 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
947 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
948 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
949 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
951 # A HTTP/1.1 response without Connection header implies keep-alive.
952 # Disabled filter mode on behalf of the client.
954 } elsif ($c =~ m/^scanning headers for:/) {
956 return '' unless SHOW_SCAN_INTRO;
958 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
959 # crunched User-Agent!
960 # Crunching: Content-Encoding: gzip
962 $content =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
964 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
966 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
968 $content = h('warning') . $content . h('Standard');
970 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
972 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
973 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
975 $content =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
976 $content =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
978 } elsif ($c =~ m/^Removing empty header/) {
980 # Removing empty header
983 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
985 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
986 # Enable force-text-mode if you know what you're doing.
987 # XXX: Could highlight more here.
988 $content =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
992 found_unknown_content($content);
996 unless ($c =~ m/^Transforming/) {
997 $content = highlight_known_headers($content) unless $no_special_header_highlighting;
1003 sub handle_loglevel_re_filter ($) {
1005 my $content = shift;
1011 our %header_colours;
1013 our $header_colour_index;
1016 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1018 # Used by Privoxy 3.0.5 and 3.0.6:
1020 # Used by Privoxy 3.0.7:
1021 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1023 $req{$t}{'filtered_header'} = $1;
1024 $req{$t}{'old_header_size'} = $2;
1025 $req{$t}{'header_filter_name'} = $3;
1027 unless (cli_option_is_set('show-ineffective-filters') or
1028 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1031 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1032 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1034 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1036 # ...produced 0 hits (new size 23).
1037 #... produced 1 hits (new size 54).
1039 $req{$t}{'header_filter_hits'} = $1;
1040 $req{$t}{'new_header_size'} = $2;
1042 unless (cli_option_is_set('show-ineffective-filters') or
1043 (defined($req{$t}{'header_filter_name'}) and
1044 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1046 if ($req{$t}{'header_filter_hits'} == 0 and
1047 not (defined($req{$t}{'header_filter_name'}) and
1048 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1051 # Reformat including information from the intro
1052 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1054 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1055 # $c .= $req{$t}{'filtered_header'} . " ";
1056 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1057 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1059 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1061 $c .= "changing size from ";
1062 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1064 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1069 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1073 # Highlight from last line (XXX: What?)
1074 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1075 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1080 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1081 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1086 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1088 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1089 # Tagger variable-test has empty joblist. Nothing to do.
1091 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1093 } elsif ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1095 # XXX: only the second version gets highlighted properly.
1096 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1097 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1098 $req{$t}{'content_source'} = $1;
1099 $req{$t}{'content_size'} = $2;
1100 $req{$t}{'content_filter'} = $3;
1101 $req{$t}{'content_hits'} = $4;
1102 $req{$t}{'new_content_size'} = $5;
1103 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1104 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1105 if ($req{$t}{'content_hits'} == 0 and
1106 not (cli_option_is_set('show-ineffective-filters')
1107 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1111 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1112 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1113 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1115 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1116 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1119 $c .= "(" . $h{'Number'};
1120 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1121 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1124 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1126 $req{$t}{'chunked-size'} = $1;
1127 $req{$t}{'dechunked-size'} = $2;
1128 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1130 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1132 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1133 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1135 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1137 # Decompression successful. Old size: 670, new size: 1166.
1139 $req{$t}{'size-compressed'} = $1;
1140 $req{$t}{'size-decompressed'} = $2;
1141 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1143 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1144 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1146 # XXX: Create sub get_percentage()
1147 if ($req{$t}{'size-decompressed'}) {
1148 $req{$t}{'decompression-gain-percent'} =
1149 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1151 $content .= " (saved: ";
1152 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1154 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1158 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1160 # Need to de-chunk first
1161 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1163 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1165 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1167 # Adding re_filter job ...
1168 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1169 # to filter client-ip-address succeeded.
1171 } elsif ($c =~ m/^Reading in filter/) {
1173 return '' unless SHOW_FILTER_READIN_IN;
1177 found_unknown_content($content);
1184 sub handle_loglevel_redirect ($) {
1191 if ($c =~ m/^Decoding "([^""]*)"/) {
1193 $req{$t}{'original-destination'} = $1;
1194 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1197 } elsif ($c =~ m/^Checking/) {
1199 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1200 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1202 # TODO: Change colour if really url-decoded
1203 $req{$t}{'decoded-original-destination'} = $1;
1204 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1207 } elsif ($c =~ m/^pcrs command "([^""]*)" changed "([^""]*)" to "([^""]*)" \((\d+) hits?\)/) {
1209 # pcrs command "s@&from=rss@@" changed "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1210 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1212 my ($pcrs_command, $url_before, $url_after, $hits) = ($1, $2, $3, $4); # XXX: save these?
1214 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1215 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1216 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1217 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1218 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1219 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1221 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1223 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1224 # XXX: Use URL highlighter
1226 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1228 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1229 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1230 # is already properly formatted.
1231 # XXX: assume the same?
1232 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1236 found_unknown_content($c);
1243 sub handle_loglevel_gif_deanimate ($) {
1245 my $content = shift;
1250 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1252 my $bytes_from = $1;
1254 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1255 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1256 # XXX: Do we need g in case of ($1 == $2)?
1257 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1259 } elsif ($content =~ m/GIF (not) changed/) {
1261 # Gif-Deanimate: GIF not changed.
1262 return '' if SUPPRESS_GIF_NOT_CHANGED;
1263 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1265 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1267 # failed! (gif parsing)
1268 # XXX: Replace this error message with something less stupid
1269 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1271 } elsif ($content =~ m/^Need to de-chunk first/) {
1273 # Need to de-chunk first
1274 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1276 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1278 # No GIF header found (XXX: Did I ever commit this?)
1279 # failed while parsing 195 134747048 (XXX: never commited)
1281 # Ignore these for now
1285 found_unknown_content($content);
1292 sub handle_loglevel_request ($) {
1294 my $content = shift;
1298 our %reason_colours;
1300 if ($content =~ m/crunch! /) {
1302 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1304 # Highlight crunch reasons
1305 foreach my $reason (keys %reason_colours) {
1306 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1308 # Highlight request URL domain and ditch 'crunch!'
1309 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1310 $content =~ s@ crunch!@@;
1312 } elsif ($content =~ m/\[too long, truncated\]$/) {
1314 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1315 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1317 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1319 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1320 $content = h('request_') . $content . h('Standard');
1324 found_unknown_content($content);
1331 sub handle_loglevel_crunch ($) {
1333 my $content = shift;
1335 our %reason_colours;
1337 # Highlight crunch reason
1338 foreach my $reason (keys %reason_colours) {
1339 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1342 if ($content =~ m/\[too long, truncated\]$/) {
1344 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1345 # [...]&filter... [too long, truncated]
1346 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1350 # Blocked: http://ads.example.org/
1351 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1357 sub handle_loglevel_connect ($) {
1364 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1366 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1368 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1369 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1371 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1373 # connect to: www.example.org.noconnect failed: Operation not permitted
1375 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1377 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1379 } elsif ($c =~ m/^to ([^\s]*) successful$/) {
1381 # Connect: to www.nzherald.co.nz successful
1383 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1384 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1386 } elsif ($c =~ m/^to ([^\s]*)$/) {
1388 # Connect: to lists.sourceforge.net:443
1390 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1392 } elsif ($c =~ m/^accepted connection from .*/ or
1395 # accepted connection from 10.0.0.1
1396 # Privoxy 3.0.6 and earlier just say:
1398 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1399 $c = highlight_matched_host($c, '(?<=connection from ).*');
1401 } elsif ($c =~ m/^write header to: .* failed:/) {
1403 # write header to: 10.0.0.1 failed: Broken pipe
1405 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1406 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1408 } elsif ($c =~ m/^write header to client failed:/) {
1410 # write header to client failed: Broken pipe
1412 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1414 } elsif ($c =~ m/^socks4_connect:/) {
1416 # socks4_connect: SOCKS request rejected or failed.
1417 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1419 } elsif ($c =~ m/^Listening for new connections/ or
1420 $c =~ m/^accept connection/) {
1422 # Privoxy versions above 3.0.6 say:
1423 # Listening for new connections ...
1424 # earlier versions say:
1425 # accept connection ...
1428 } elsif ($c =~ m/^accept failed:/) {
1430 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1432 } elsif ($c =~ m/^Overriding forwarding settings/) {
1434 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1435 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1437 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1439 # Denying suspicious CONNECT request from 10.0.0.1
1440 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1442 } elsif ($c =~ m/^socks5_connect:/) {
1444 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1446 } elsif ($c =~ m/^Created new connection to/) {
1448 # Created new connection to www.privoxy.org:80 on socket 11.
1449 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1450 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1452 } elsif ($c =~ m/^Found reusable socket/) {
1454 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1455 $c =~ s@(?<=Found reusable socket )(\d+)@$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/^Marking open socket/) {
1461 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1462 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1463 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1464 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1466 } elsif ($c =~ m/^No reusable/) {
1468 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1469 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1471 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1473 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1474 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1475 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1476 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1477 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1479 } elsif ($c =~ m/^Socket/) {
1481 # Socket 16 already forgotten or never remembered.
1482 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1484 } elsif ($c =~ m/^The connection to/) {
1486 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1487 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1488 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1489 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1490 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1491 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1493 } elsif ($c =~ m/^Waiting for/) {
1495 # Waiting for 1 connections to timeout.
1496 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1498 } elsif ($c =~ m/^Initialized/) {
1500 # Initialized 20 socket slots.
1501 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1503 } elsif ($c =~ m/^Done reading from server/) {
1505 # Done reading from server. Expected content length: 24892. \
1506 # Actual content length: 24892. Most recently received: 4412.
1507 $c =~ s@(?<=Expected content length: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1508 $c =~ s@(?<=Actual content length: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1509 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1511 } elsif ($c =~ m/^Continuing buffering headers/) {
1513 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1514 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1515 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1516 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1518 } elsif ($c =~ m/^Received \d+ bytes while/) {
1520 # Received 206 bytes while expecting 12103.
1521 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1522 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1524 } elsif ($c =~ m/^Connection from/) {
1526 # Connection from 81.163.28.218 dropped due to ACL
1527 $c =~ s@(?<=^Connection from )((?:\d+\.?){4})@$h{'Number'}$1$h{'Standard'}@;
1529 } elsif ($c =~ m/^Looks like we rea/ or
1530 $c =~ m/^Unsetting keep-alive flag/ or
1531 $c =~ m/^No connections to wait/) {
1533 # Looks like we reached the end of the last chunk. We better stop reading.
1534 # Looks like we read the end of the last chunk together with the server \
1535 # headers. We better stop reading.
1536 # Unsetting keep-alive flag.
1537 # No connections to wait for left.
1541 found_unknown_content($c);
1549 sub handle_loglevel_info ($) {
1556 if ($c =~ m/^Rewrite detected:/) {
1558 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1559 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1561 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1562 $c =~ m/^Compressed content detected/ or
1565 # Decompressing deflated iob: 117
1566 # Decompression didn't result in any content.
1567 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1568 # with zlib support or enable the prevent-compression action.
1569 # Tagger 'complete-url' created empty tag. Ignored.
1573 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1575 # loading configuration file '/usr/local/etc/privoxy/config':
1576 # Reloading configuration file '/usr/local/etc/privoxy/config'
1577 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1579 } elsif ($c =~ m/^exiting by signal/) {
1581 # exiting by signal 15 .. bye
1582 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1584 } elsif ($c =~ m/^Privoxy version/) {
1586 # Privoxy version 3.0.7
1587 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1589 } elsif ($c =~ m/^Program name: /) {
1591 # Program name: /usr/local/sbin/privoxy
1592 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1594 } elsif ($c =~ m/^Listening on port /) {
1596 # Listening on port 8118 on IP address 10.0.0.1
1597 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1598 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1600 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1602 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1603 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1605 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1607 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1608 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1609 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1611 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1612 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1613 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1614 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1616 } elsif ($c =~ m/^Status code/) {
1618 # Status code 304 implies no body.
1619 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1621 } elsif ($c =~ m/^Method/) {
1623 # Method HEAD implies no body.
1624 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1626 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1628 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1629 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1630 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1632 } elsif ($c =~ m/^No logfile configured/ or
1633 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1634 $c =~ m/^Invalid \"chunked\" transfer/ or
1635 $c =~ m/^Support for/ or
1636 $c =~ m/^Flushing header and buffers/
1639 # No logfile configured. Please enable it before reporting any problems.
1640 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1641 # response or even no response at all.
1642 # No logfile configured. Logging disabled.
1643 # Invalid "chunked" transfer encoding detected and ignored.
1644 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1645 # known not to work properly in some situations.
1646 # Flushing header and buffers. Stepping back from filtering.
1650 found_unknown_content($c);
1657 sub handle_loglevel_cgi ($) {
1664 if ($c =~ m/^Granting access to/) {
1666 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1668 } elsif ($c =~ m/^Substituting: s(.)/) {
1670 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1671 # XXX: prone to span several lines
1674 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1675 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1676 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1682 sub handle_loglevel_force ($) {
1689 if ($c =~ m/^Ignored force prefix in request:/) {
1691 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1692 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1693 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1695 } elsif ($c =~ m/^Enforcing request:/) {
1697 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1698 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1702 found_unknown_content($c);
1709 sub handle_loglevel_ignore ($) {
1713 ################################################################################
1714 # Functions that actually print stuff
1715 ################################################################################
1717 sub print_clf_message () {
1719 our ($ip, $timestamp, $request_line, $status_code, $size);
1723 return if DEBUG_SUPPRESS_LOG_MESSAGES;
1725 # Rebuild highlighted
1726 $output .= $h{'Number'} . $ip . $h{'Standard'};
1728 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
1730 $output .= "\"" . highlight_request_line("$request_line") . "\"";
1732 $output .= $h{'Status'} . $status_code . $h{'Standard'};
1734 $output .= $h{'Number'} . $size . $h{'Standard'};
1735 $output .= get_line_end();
1740 sub print_non_clf_message ($) {
1743 our %thread_colours;
1746 our $time_colour_index;
1749 my $content = shift;
1750 my ($day, $time_stamp, $msecs, $thread, $log_level)
1751 = ($req{$t}{'day'}, $req{$t}{'time-stamp'}, $req{$t}{'msecs'}, $t, $req{$t}{'log-level'} );
1753 return if DEBUG_SUPPRESS_LOG_MESSAGES;
1755 $output .= $h{"Standard"} unless cli_option_is_set('html-output');
1756 # $output .= "$day ";
1757 $output .= $time_colours[$time_colour_index % 2];
1759 $output .= $time_stamp;
1760 $output .= ".$msecs" unless cli_option_is_set('no-msecs');
1761 $output .= $h{"Standard"};
1763 $output .= $thread_colours{$thread} if (defined($thread_colours{$thread}));
1765 $output .= $h{"Standard"} . " ";
1766 $output .= $h{$log_level} if (defined($h{$log_level}));
1767 $output .= $log_level;
1768 $output .= $h{"Standard"} . ": ";
1769 $output .= "$content";
1770 $output .= get_line_end();
1778 our %req; # request data from previous lines
1780 our %thread_colours;
1783 our $thread_colour_index = 0;
1784 our $header_colour_index = 0;
1785 our $time_colour_index = 0;
1787 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
1789 my $last_thread = 0;
1790 my $last_timestamp = 0;
1792 my $filters_that_did_nothing;
1795 our $no_special_header_highlighting;
1796 $time_colour = paint_it('white');
1797 my %log_level_count;
1799 my %log_level_handlers = (
1800 'Re-Filter' => \&handle_loglevel_re_filter,
1801 'Header' => \&handle_loglevel_header,
1802 'Connect' => \&handle_loglevel_connect,
1803 'Redirect' => \&handle_loglevel_redirect,
1804 'Request' => \&handle_loglevel_request,
1805 'Crunch' => \&handle_loglevel_crunch,
1806 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
1807 'Info' => \&handle_loglevel_info,
1808 'CGI' => \&handle_loglevel_cgi,
1809 'Force' => \&handle_loglevel_force,
1810 'Error' => \&handle_loglevel_ignore,
1811 'Fatal error' => \&handle_loglevel_ignore,
1812 'Writing' => \&handle_loglevel_ignore,
1819 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*)$/) {
1820 # XXX: Put in req hash?
1823 $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution
1828 $req{$t}{'day'} = $day;
1829 $req{$t}{'time-stamp'} = $time_stamp;
1830 $req{$t}{'msecs'} = $msecs; # Only the cool kids have micro second resolution;
1831 $req{$t}{'log-level'} = $log_level;
1832 $req{$t}{'content'} = $content;
1833 $req{$t}{'log-message'} = $_;
1834 $no_special_header_highlighting = 0;
1836 $log_level_count{$log_level}++;
1838 if (defined($log_level_handlers{$log_level})) {
1840 $content = $log_level_handlers{$log_level}($content);
1844 die "No handler found for log level \"$log_level\"\n";
1848 # Highlight Truncations
1849 if (m/\.\.\. \[(too long, truncated)/) {
1850 $content =~ s@($1)@$h{'Truncation'}$1$h{'Standard'}@g;
1853 next unless $content;
1855 # Register threads to keep the colour constant
1856 if (!defined($thread_colours{$thread})) {
1857 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
1858 $thread_colour_index++;
1861 # Switch timestamp colour if timestamps differ
1862 if ($msecs != $last_msecs || !($time_stamp =~ m/$last_timestamp/)) {
1863 debug_message("Tick tack!") if DEBUG_TICKS;
1864 $time_colour = $time_colours[$time_colour_index % 2];
1865 $time_colour_index++
1868 $last_msecs = $msecs;
1869 $last_thread = $thread;
1870 $last_timestamp = $time_stamp;
1872 print_non_clf_message($content);
1874 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
1876 # LOG_LEVEL_CLF lines look like this
1877 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
1878 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
1879 # HTTP/1.1" 403 1730
1880 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
1882 print_clf_message();
1886 # Some Privoxy log messages span more than one line,
1887 # usually to dump lots of content that doesn't need any syntax highlighting.
1888 # XXX: add mechanism to forward these lines to the right handler anyway.
1890 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
1891 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
1896 if (cli_option_is_set('statistic')) {
1897 foreach (keys %log_level_count) {
1898 print $_ . ": " . $log_level_count{$_} . " ";
1903 sub VersionMessage {
1904 my $version_message;
1906 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
1907 $version_message .= 'Copyright (C) 2007-2009 Fabian Keil <fk@fabiankeil.de>' . "\n";
1908 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
1910 print $version_message;
1913 sub get_cli_options () {
1915 our %cli_options = (
1916 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
1917 'title' => CLI_OPTION_TITLE,
1918 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
1919 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
1920 'no-msecs' => CLI_OPTION_NO_MSECS,
1921 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
1922 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
1923 'statistic' => CLI_OPTION_STATISTIC,
1927 'html-output' => \$cli_options{'html-output'},
1928 'title' => \$cli_options{'title'},
1929 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
1930 'no-embedded-css' => \$cli_options{'no-embedded-css'},
1931 'no-msecs' => \$cli_options{'no-msecs'},
1932 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
1933 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
1934 'statistic' => \$cli_options{'statistic'},
1935 'version' => sub { VersionMessage && exit(0) }
1941 ################################################################################
1943 ################################################################################
1947 set_background(DEFAULT_BACKGROUND);
1948 prepare_our_stuff();
1961 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
1965 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
1966 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--show-ineffective-filters>]
1971 B<privoxy-log-parser> reads Privoxy log messages and
1973 - syntax-highlights recognized lines,
1975 - reformats some of them for easier comprehension,
1977 - filters out less useful messages, and
1979 - (in some cases) calculates additional information,
1980 like the compression ratio or how a filter affected
1983 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
1984 without getting confused by the resulting amount of output. For example for
1985 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
1986 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
1987 will hide the "filter foo caused 0 hits" message.
1991 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
1992 just don't highlight them.
1994 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
1995 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
1996 This option is only intended to make embedding log excerpts in web pages easier.
1997 It does not excape any input!
1999 [B<--no-msecs>] Don't expect milisecond resolution
2001 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2002 the filtered output is piped into less in which case the ANSI control
2003 codes don't work, or if the terminal itself doesn't support the control
2006 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2007 that didn't modify the content.
2009 [B<--version>] Print version and exit.
2013 To monitor a log file:
2015 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2017 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2018 or if the log won't get rotated anyway. The log file location depends
2019 on your system (Doh!).
2021 To monitor Privoxy without having it write to a log file:
2023 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2025 Again, the config file location depends on your system. Output redirection
2026 depends on your shell, the above works with bourne shells.
2028 To read a processed Privoxy log file from top to bottom, letting the content
2029 scroll by slightly faster than you can read:
2031 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2033 This is probably only useful to fill screens in the background of haxor movies.
2037 Syntax highlighting with ANSI escape sequences will look strange
2038 if your background color isn't black.
2040 Some messages aren't recognized yet and will not be fully highlighted.
2042 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2043 using earlier Privoxy versions will probably result in an increased amount
2044 of unrecognized log lines.
2046 Privoxy's log files tend to be rather large. If you use HTML
2047 highlighting some browsers can't handle them, get confused and
2048 will eventually crash because of segmentation faults or unexpected
2049 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2054 Many settings can't be controlled through command line options yet.
2062 Fabian Keil <fk@fabiankeil.de>