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 1.111 2008/08/02 07:31:54 fk Exp $
14 # - LOG_LEVEL_CGI, LOG_LEVEL_ERROR, LOG_LEVEL_WRITE content highlighting
15 # - create fancy statistics
16 # - grep through Privoxy sources to find unsupported log messages
17 # - hunt down substitutions that match content from variables which
18 # can contain stuff like ()?'[]
19 # - replace $h{'foo'} with h('foo') where possible
20 # - hunt down XXX comments instead of just creating them
21 # - add example log lines for every regex and mark them up for
23 # - Handle incomplete input without Perl warning about undefined variables.
24 # - Use generic highlighting function that takes a regex and the
27 # Copyright (c) 2007-2008 Fabian Keil <fk@fabiankeil.de>
29 # Permission to use, copy, modify, and distribute this software for any
30 # purpose with or without fee is hereby granted, provided that the above
31 # copyright notice and this permission notice appear in all copies.
33 # THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
34 # WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
35 # MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
36 # ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
37 # WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
38 # ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
39 # OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
40 ################################################################################
47 PRIVOXY_LOG_PARSER_VERSION => '0.4',
48 # Feel free to mess with these ...
49 DEFAULT_BACKGROUND => 'black', # Choose registered colour (like 'black')
50 DEFAULT_TEXT_COLOUR => 'white', # Choose registered colour (like 'black')
51 HEADER_DEFAULT_COLOUR => 'yellow',
52 REGISTER_HEADERS_WITH_THE_SAME_COLOUR => 1,
54 CLI_OPTION_DEFAULT_TO_HTML_OUTPUT => 0,
55 CLI_OPTION_TITLE => 'Privoxy-Log-Parser in da house',
56 CLI_OPTION_NO_EMBEDDED_CSS => 0,
57 CLI_OPTION_NO_MSECS => 0,
58 CLI_OPTION_NO_SYNTAX_HIGHLIGHTING => 0,
59 CLI_OPTION_ERROR_LOG_FILE => '/var/log/privoxy-log.log',
60 CLI_OPTION_SHOW_INEFFECTIVE_FILTERS => 0,
61 CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES => 0,
62 CLI_OPTION_STATISTIC => 0,
64 SUPPRESS_SUCCEEDED_FILTER_ADDITIONS => 1,
66 SHOW_FILTER_READIN_IN => 0,
67 SUPPRESS_EMPTY_LINES => 1,
68 SUPPRESS_SUCCESSFUL_CONNECTIONS => 1,
69 SUPPRESS_ACCEPTED_CONNECTIONS => 1,
70 SUPPRESS_GIF_NOT_CHANGED => 1,
71 SUPPRESS_NEED_TO_DE_CHUNK_FIRST => 1,
73 DEBUG_HEADER_REGISTERING => 0,
74 DEBUG_HEADER_HIGHLIGHTING => 0,
77 DEBUG_SUPPRESS_LOG_MESSAGES => 0,
79 PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH => 0,
80 PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH => 1,
82 LOG_UNPARSED_LINES_TO_EXTRA_FILE => 0,
84 # You better leave these alone unless you know what you're doing.
85 COLOUR_RESET => "\033[0;0m",
89 sub prepare_our_stuff () {
91 # Syntax Higlight hash
93 'red', 'green', 'brown', 'blue', 'purple', 'cyan',
94 'light_gray', 'light_red', 'light_green', 'yellow',
95 'light_blue', 'pink', 'light_cyan', 'white'
102 Filter => 'purple', # XXX: Used?
103 'Re-Filter' => 'purple',
105 Request => 'light_cyan',
106 CGI => 'light_green',
108 Error => 'light_red',
109 'Fatal error' => 'light_red',
110 'Gif-Deanimate' => 'blue',
112 Writing => 'light_green',
113 # ----------------------
116 request_ => 'brown', # host+path but no protocol
117 'ip-address' => 'yellow',
120 Truncation => 'light_red',
122 Timestamp => 'brown',
123 Crunching => 'light_red',
124 crunched => 'light_red',
125 'Request-Line' => 'pink',
127 destination => 'yellow',
128 'http-version' => 'pink',
129 'crunch-pattern' => 'pink',
134 'program-name' => 'cyan',
137 warning => 'light_red',
138 debug => 'light_red',
142 'status-message' => 'light_cyan',
143 'status-code' => 'yellow',
144 'invalid-request' => 'light_red',
146 error => 'light_red',
147 'rewritten-URL' => 'light_red',
148 'pcrs-delimiter' => 'light_red',
149 'ignored' => 'light_red',
150 'action-bits-update' => 'light_red',
151 'configuration-line' => 'red',
152 'content-type' => 'yellow',
157 # Header colours need their own hash so the keys can be accessed properly
158 our %header_colours = (
159 # Prefilled with headers that should not appear with default header colours
160 Cookie => 'light_red',
161 'Set-Cookie' => 'light_red',
162 Warning => 'light_red',
163 Default => HEADER_DEFAULT_COLOUR,
166 # Crunch reasons need their own hash as well
167 our %reason_colours = (
168 'Unsupported HTTP feature' => 'light_red',
169 Blocked => 'light_red',
170 Untrusted => 'light_red',
171 Redirected => 'green',
172 'CGI Call' => 'white',
173 'DNS failure' => 'red',
174 'Forwarding failed' => 'light_red',
175 'Connection failure' => 'light_red',
176 'Out of memory (may mask other reasons)' => 'light_red',
177 'No reason recorded' => 'light_red',
180 our @time_colours = ('white', 'light_gray');
182 # Translate highlight strings into highlight code
183 prepare_highlight_hash(\%header_colours);
184 prepare_highlight_hash(\%reason_colours);
185 prepare_highlight_hash(\%h);
186 prepare_colour_array(\@all_colours);
187 prepare_colour_array(\@time_colours);
192 ###############################################################
193 # Takes a colour string and returns an ANSI escape sequence
194 # (unless --no-syntax-highlighting is used).
195 # XXX: The Rolling Stones reference has to go.
196 ###############################################################
198 my $colour = shift @_;
200 return "" if cli_option_is_set('no-syntax-highlighting');
242 my $bg_code = get_background();
244 our $default = default_colours();
246 if (defined($text{$colour})) {
247 $colour_code = ESCAPE;
248 $colour_code .= $text{$colour};
250 $colour_code .= $light{$colour} ? "1" : "2";
252 $colour_code .= $bg_code;
254 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
256 } elsif ($colour =~ /reset/) {
258 $colour_code = default_colours();
262 die "What's $colour supposed to mean?\n";
269 sub get_semantic_html_markup ($) {
270 ###############################################################
271 # Takes a string and returns a span element
272 ###############################################################
277 if ($type =~ /Standard/) {
281 $code = '<span title="' . $type . '" class="' . $type . '">';
287 sub cli_option_is_set ($) {
290 my $cli_option = shift;
292 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
294 return $cli_options{$cli_option};
297 sub get_html_title () {
300 return $cli_options{'title'};
304 sub init_css_colours() {
312 purple => "F06", # XXX: wrong
313 cyan => "F09", # XXX: wrong
318 light_green => "33F",
327 sub get_css_colour ($) {
332 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
334 return '#' . $css_colours{$colour};
337 sub get_css_line ($) {
344 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
345 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
346 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
347 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
348 $css_line .= '}' . "\n";
353 sub get_css_line_for_colour ($) {
360 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
361 $css_line .= 'color:' . get_css_colour($colour) . ';';
362 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
363 $css_line .= '}' . "\n";
368 # XXX: Wrong solution
369 sub get_missing_css_lines () {
373 $css_line .= '.' . 'default' . ' {'; # XXX: lc() shouldn't be necessary
374 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
375 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
376 $css_line .= '}' . "\n";
384 our %css_colours; #XXX: Wrong solution
388 $css .= '.privoxy-log {';
389 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
390 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
393 foreach my $key (keys %h_colours) {
395 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
396 $css .= get_css_line($key);
400 foreach my $colour (keys %css_colours) {
402 $css .= get_css_line_for_colour($colour);
406 $css .= get_missing_css_lines(); #XXX: Wrong solution
415 if (cli_option_is_set('html-output')) {
417 my $title = get_html_title();
419 $intro .= '<html><head>';
420 $intro .= '<title>' . $title . '</title>';
421 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
422 $intro .= '</head><body>';
423 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
433 if (cli_option_is_set('html-output')) {
435 $outro = '</p></body></html>';
442 sub get_line_end () {
446 $line_end = '<br>' . $line_end if cli_option_is_set('html-output');
451 sub get_colour_html_markup ($) {
452 ###############################################################
453 # Takes a colour string a span element. XXX: WHAT?
454 # XXX: This function shouldn't be necessary, the
455 # markup should always be semantically correct.
456 ###############################################################
461 if ($type =~ /Standard/) {
464 $code = '<span class="' . lc($type) . '">';
471 sub default_colours () {
474 return reset_colours();
477 sub show_colours () {
481 sub reset_colours () {
482 return ESCAPE . "0m";
485 sub set_background ($){
501 if (defined($backgrounds{$colour})) {
502 $bg_code = $backgrounds{$colour};
504 die "Invalid background colour: " . $colour;
508 sub get_background (){
512 sub prepare_highlight_hash ($) {
515 if (!cli_option_is_set('html-output')) {
517 foreach my $key (keys %$ref) {
518 $$ref{$key} = paint_it($$ref{$key});
523 foreach my $key (keys %$ref) {
524 $$ref{$key} = get_semantic_html_markup($key);
530 sub prepare_colour_array ($) {
533 if (!cli_option_is_set('html-output')) {
535 foreach my $i (0 ... @$ref - 1) {
536 $$ref[$i] = paint_it($$ref[$i]);
541 foreach my $i (0 ... @$ref - 1) {
542 $$ref[$i] = get_colour_html_markup($$ref[$i]);
548 sub found_unknown_content ($) {
556 return if cli_option_is_set('accept-unknown-messages');
558 return if ($unknown =~ /\[too long, truncated\]$/);
560 $message = "found_unknown_content: Don't know how to highlight: ";
561 # Break line so the log file can later be parsed as Privoxy log file again
562 $message .= '"' . $unknown . '"' . " in:\n";
563 $message .= $req{$t}{'log-message'};
564 debug_message($message);
565 log_parse_error($req{$t}{'log-message'});
567 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
570 sub log_parse_error ($) {
574 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
575 open(ERRORLOG, ">>" . ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
576 print ERRORLOG $message;
581 sub debug_message (@) {
585 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
588 ################################################################################
589 # highlighter functions that aren't loglevel-specific
590 ################################################################################
594 # Get highlight marker
596 my $highlight = shift; # XXX: Stupid name;
600 if (defined($highlight)) {
602 $result = $h{$highlight};
606 $message = "h: Don't recognize highlighter $highlight.";
607 debug_message($message);
608 log_parser_error($message);
609 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
615 sub higlight_known_headers ($) {
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 higlight_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 $rl = higlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
662 $rl = higlight_matched_pattern($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 $rl = higlight_matched_url($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, $status_message);
692 # TODO: Mark different status codes differently
694 if ($rl =~ m/(HTTP\/\d\.\d) (\d+) (.*)/) {
695 ($http_version, $status_code, $status_message) = ($1, $2, $3);
697 debug_message ("Can't parse response line: $rl") and die 'Fix this';
700 # Rebuild highlighted
702 $rl .= h('http-version') . $http_version . h('Standard');
704 $rl .= h('status-code') . $status_code . h('Standard');
706 $rl .= h('status-message') . $status_message . h('Standard');
711 sub higlight_matched_url ($$) {
713 my $result = shift; # XXX: Stupid name;
716 #print "Got $result, regex ($regex)\n";
718 if ($result =~ m@(.*?)($regex)(.*)@) {
719 $result = $1 . highlight_url($2) . $3;
720 #print "Now the result is $result\n";
726 sub higlight_matched_host ($$) {
728 my $result = shift; # XXX: Stupid name;
731 if ($result =~ m@(.*?)($regex)(.*)@) {
732 $result = $1 . h('host') . $2 . h('Standard') . $3;
738 sub higlight_matched_pattern ($$$) {
741 my $result = shift; # XXX: Stupid name;
745 die "Unknown key $key" unless defined $h{$key};
747 if ($result =~ m@(.*?)($regex)(.*)@) {
748 $result = $1 . h($key) . $2 . h('Standard') . $3;
755 sub higlight_matched_path ($$) {
757 my $result = shift; # XXX: Stupid name;
760 if ($result =~ m@(.*?)($regex)(.*)@) {
761 $result = $1 . h('path') . $2 . h('Standard') . $3;
768 sub highlight_url ($) {
772 if (cli_option_is_set('html-output')) {
774 $url = '<a href="' . $url . '">' . $url . '</a>';
778 $url = h('URL') . $url . h('Standard');
785 ################################################################################
786 # loglevel-specific highlighter functions
787 ################################################################################
789 sub handle_loglevel_header ($) {
798 our $header_colour_index;
799 our $no_special_header_highlighting;
801 # Register new headers
802 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
803 if ($c =~ m/^scan: ((?>[^:]+)):/) {
805 if (!defined($header_colours{$header})) {
806 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
808 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
809 $header_colours{$header} = $header_colours{'Default'};
811 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
812 $header_colour_index++;
817 if ($c =~ m/^scan: ((\w*) (.*) (HTTP\/\d\.\d))/) {
819 # Client request line
820 # Save for statistics (XXX: Not implemented yet)
821 $req{$t}{'method'} = $2;
822 $req{$t}{'destination'} = $3;
823 $req{$t}{'http-version'} = $4;
825 $content = highlight_request_line($1);
827 } elsif ($c =~ m/^(scan: )(HTTP\/\d\.\d (\d+) (.*))/) {
829 # Server response line
830 $req{$t}{'response_line'} = $2;
831 $req{$t}{'status_code'} = $3;
832 $req{$t}{'status_message'} = $4;
833 $content = $1 . highlight_response_line($req{$t}{'response_line'});
835 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
837 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
838 $content =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
839 $content =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
841 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
843 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/
844 $c = higlight_matched_host($c, '(?<=New host is: )[^\s]+');
845 $content = higlight_matched_url($c, '(?<=Crunching Referer: )[^\s]+');
847 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
849 # Text mode enabled by force. Take cover!
850 $content =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
852 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
854 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
855 $content = $1 . highlight_request_line($2);
857 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
859 # Adjusted Content-Length to 2132
860 # Adjust Content-Length to 33533
861 $content =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
862 $content = higlight_known_headers($content);
864 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
866 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
867 $content = higlight_matched_url($content, '(?<=New request URL: ).*');
869 } elsif ($c =~ m/^Couldn\'t parse:/) {
871 # XXX: These should probable be logged with LOG_LEVEL_ERROR
872 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
873 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
874 $content =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
876 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
877 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
879 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
880 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
881 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
883 # XXX: Save tag and tagger
885 $content =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
886 $content =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
887 $content =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
888 $no_special_header_highlighting = 1;
890 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
892 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
893 # XXX: Save tag and tagger
895 $content =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
896 $content =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
898 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
899 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
900 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
901 or $c =~ m/^(addh-unique|Referer forged to)/
902 or $c =~ m/^Downgraded answer to HTTP\/1.0/
903 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
904 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
905 or $c =~ m/^Referer crunched!/
906 or $c =~ m/^crunched x-forwarded-for!/
907 or $c =~ m/^crunched From!/
908 or $c =~ m/^ modified$/
909 or $c =~ m/^Content filtering is enabled. Crunching:/
910 or $c =~ m/^force-text-mode overruled the client/
911 or $c =~ m/^Server time in the future\./
912 or $c =~ m/^content-disposition header crunched and replaced with:/i
913 or $c =~ m/^Reducing white space in /
914 or $c =~ m/^Ignoring single quote in /
915 or $c =~ m/^Converting tab to space in /
918 # XXX: Some of these may need highlighting
920 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
921 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
922 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
923 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
924 # Text mode is already enabled
925 # Denied request with NULL byte(s) turned into line break(s)
926 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
927 # addh-unique: Host: people.freebsd.org
928 # Suppressed offer to compress content
929 # Crunched incoming cookie -- yum!
930 # Accepted the client's request to fetch without filtering.
931 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
932 # addh-unique: Host: subkeys.pgp.net:11371
933 # Referer forged to: Referer: http://10.0.0.1/
934 # Downgraded answer to HTTP/1.0
935 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
936 # Referer overwritten with: Referer: pille-palle
937 # Referer replaced with: Referer: pille-palle
938 # crunched x-forwarded-for!
940 # modified # XXX: pretty stupid log message
941 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
942 # force-text-mode overruled the client's request to fetch without filtering!
943 # Server time in the future.
944 # content-disposition header crunched and replaced with: content-disposition: filename=baz
945 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
946 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
947 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
948 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
950 } elsif ($c =~ m/^scanning headers for:/) {
952 return '' unless SHOW_SCAN_INTRO;
954 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
955 # crunched User-Agent!
956 # Crunching: Content-Encoding: gzip
958 $content =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
960 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
962 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
964 $content = h('warning') . $content . h('Standard');
966 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
968 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to "Proxy-Authenticate: Basic realm="Correos Proxy Server""
970 $content =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
971 $content =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
973 } elsif ($c =~ m/^Removing empty header/) {
975 # Removing empty header
978 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
980 # Content-Type: application/octet-stream not replaced. It doesn't look like text. Enable force-text-mode if you know what you're doing.
981 # XXX: Could highlight more here.
982 $content =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
986 found_unknown_content($content);
990 unless ($c =~ m/^Transforming/) {
991 $content = higlight_known_headers($content) unless $no_special_header_highlighting;
997 sub handle_loglevel_re_filter ($) {
1005 our %header_colours;
1007 our $header_colour_index;
1010 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1012 # Used by Privoxy 3.0.5 and 3.0.6:
1014 # Used by Privoxy 3.0.7:
1015 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1017 $req{$t}{'filtered_header'} = $1;
1018 $req{$t}{'old_header_size'} = $2;
1019 $req{$t}{'header_filter_name'} = $3;
1021 unless (cli_option_is_set('show-ineffective-filters') or
1022 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1025 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1026 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1028 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1030 # ...produced 0 hits (new size 23).
1031 #... produced 1 hits (new size 54).
1033 $req{$t}{'header_filter_hits'} = $1;
1034 $req{$t}{'new_header_size'} = $2;
1036 unless (cli_option_is_set('show-ineffective-filters') or
1037 (defined($req{$t}{'header_filter_name'}) and
1038 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1040 if ($req{$t}{'header_filter_hits'} == 0 and
1041 not (defined($req{$t}{'header_filter_name'}) and
1042 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1045 # Reformat including information from the intro
1046 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1048 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1049 # $c .= $req{$t}{'filtered_header'} . " ";
1050 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1051 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1053 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1055 $c .= "changing size from ";
1056 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1058 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1063 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1067 # Highlight from last line (XXX: What?)
1068 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1069 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1074 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1075 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1080 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1082 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1083 # Tagger variable-test has empty joblist. Nothing to do.
1085 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1087 } elsif ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1089 # XXX: only the second version gets highlighted properly.
1090 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1091 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1092 $req{$t}{'content_source'} = $1;
1093 $req{$t}{'content_size'} = $2;
1094 $req{$t}{'content_filter'} = $3;
1095 $req{$t}{'content_hits'} = $4;
1096 $req{$t}{'new_content_size'} = $5;
1097 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1098 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1099 if ($req{$t}{'content_hits'} == 0 and
1100 not (cli_option_is_set('show-ineffective-filters')
1101 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1105 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1106 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1107 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1109 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1110 $c = higlight_matched_host($c, '(?<=filtering )[^\s]+');
1113 $c .= "(" . $h{'Number'};
1114 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1115 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1118 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1120 $req{$t}{'chunked-size'} = $1;
1121 $req{$t}{'dechunked-size'} = $2;
1122 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1124 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1126 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1127 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1129 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1131 # Decompression successful. Old size: 670, new size: 1166.
1133 $req{$t}{'size-compressed'} = $1;
1134 $req{$t}{'size-decompressed'} = $2;
1135 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1137 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1138 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1140 # XXX: Create sub get_percentage()
1141 if ($req{$t}{'size-decompressed'}) {
1142 $req{$t}{'decompression-gain-percent'} =
1143 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1145 $content .= " (saved: ";
1146 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1148 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1152 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1154 # Need to de-chunk first
1155 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1157 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1159 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1161 # Adding re_filter job ...
1162 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D to filter client-ip-address succeeded.
1164 } elsif ($c =~ m/^Reading in filter/) {
1166 return '' unless SHOW_FILTER_READIN_IN;
1170 found_unknown_content($content);
1178 sub handle_loglevel_redirect ($) {
1185 if ($c =~ m/^Decoding "([^""]*)"/) {
1187 $req{$t}{'original-destination'} = $1;
1188 $c = higlight_matched_path($c, '(?<=Decoding ")[^"]*');
1191 } elsif ($c =~ m/^Checking/) {
1193 # Checking /_ylt=A0geu.Z76BRGR9kAH2RXNyoA/SIG=14gqhtscv/EXP=1175861755/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd&ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1195 # TODO: Change colour if really url-decoded
1196 $req{$t}{'decoded-original-destination'} = $1;
1197 $c = higlight_matched_path($c, '(?<=Checking ")[^"]*');
1200 } elsif ($c =~ m/^pcrs command "([^""]*)" changed "([^""]*)" to "([^""]*)" \((\d+) hits?\)/) {
1202 # pcrs command "s@&from=rss@@" changed "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss" to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1204 my ($pcrs_command, $url_before, $url_after, $hits) = ($1, $2, $3, $4); # XXX: save these?
1206 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1207 $c = higlight_matched_url($c, '(?<=changed ")[^""]*');
1208 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1209 $c = higlight_matched_url($c, '(?<=to ")[^""]*');
1210 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1211 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1213 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1215 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1216 # XXX: Use URL highlighter
1218 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1220 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1221 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png" is already properly formatted.
1222 # XXX: assume the same?
1223 $c = higlight_matched_url($c, '(?<=assuming that \")[^"]*');
1227 found_unknown_content($c);
1234 sub handle_loglevel_gif_deanimate ($) {
1236 my $content = shift;
1241 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1243 my $bytes_from = $1;
1245 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1246 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1247 # XXX: Do we need g in case of ($1 == $2)?
1248 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1250 } elsif ($content =~ m/GIF (not) changed/) {
1252 # Gif-Deanimate: GIF not changed.
1253 return '' if SUPPRESS_GIF_NOT_CHANGED;
1254 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1256 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1258 # failed! (gif parsing)
1259 # XXX: Replace this error message with something less stupid
1260 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1262 } elsif ($content =~ m/^Need to de-chunk first/) {
1264 # Need to de-chunk first
1265 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1267 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1269 # No GIF header found (XXX: Did I ever commit this?)
1270 # failed while parsing 195 134747048 (XXX: never commited)
1272 # Ignore these for now
1276 found_unknown_content($content);
1284 sub handle_loglevel_request ($) {
1286 my $content = shift;
1290 our %reason_colours;
1292 if ($content =~ m/crunch! /) {
1294 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1296 # Highlight crunch reasons
1297 foreach my $reason (keys %reason_colours) {
1298 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1300 # Highlight request URL domain and ditch 'crunch!'
1301 $content = higlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1302 $content =~ s@ crunch!@@;
1304 } elsif ($content =~ m/\[too long, truncated\]$/) {
1306 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1307 $content = higlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1309 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1311 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1312 $content = h('request_') . $content . h('Standard');
1316 found_unknown_content($content);
1323 sub handle_loglevel_connect ($) {
1330 if ($c =~ m/via [^\s]+ to: [^\s]+/) {
1332 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1334 $c = higlight_matched_host($c, '(?<=via )[^\s]+');
1335 $c = higlight_matched_host($c, '(?<=to: )[^\s]+');
1337 } elsif ($c =~ m/connect to: .* failed: .*/) {
1339 # connect to: www.example.org.noconnect failed: Operation not permitted
1341 $c = higlight_matched_host($c, '(?<=connect to: )[^\s]+');
1343 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1345 } elsif ($c =~ m/to ([^\s]*) successful$/) {
1347 # Connect: to www.nzherald.co.nz successful
1349 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1350 $c = higlight_matched_host($c, '(?<=to )[^\s]+');
1352 } elsif ($c =~ m/to ([^\s]*)$/) {
1354 # Connect: to lists.sourceforge.net:443
1356 $c = higlight_matched_host($c, '(?<=to )[^\s]+');
1358 } elsif ($c =~ m/^accepted connection from .*/ or
1361 # accepted connection from 10.0.0.1
1362 # Privoxy 3.0.6 and earlier just say:
1364 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1365 $c = higlight_matched_host($c, '(?<=connection from ).*');
1367 } elsif ($c =~ m/^write header to: .* failed:/) {
1369 # write header to: 10.0.0.1 failed: Broken pipe
1371 $c = higlight_matched_host($c, '(?<=write header to: )[^\s]*');
1372 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1374 } elsif ($c =~ m/^write header to client failed:/) {
1376 # write header to client failed: Broken pipe
1378 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1380 } elsif ($c =~ m/^socks4_connect:/) {
1382 # socks4_connect: SOCKS request rejected or failed.
1383 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1385 } elsif ($c =~ m/^Listening for new connections/ or
1386 $c =~ m/^accept connection/) {
1388 # Privoxy versions above 3.0.6 say:
1389 # Listening for new connections ...
1390 # earlier versions say:
1391 # accept connection ...
1394 } elsif ($c =~ m/^accept failed:/) {
1396 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1398 } elsif ($c =~ m/^Overriding forwarding settings/) {
1400 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1401 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1403 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1405 # Denying suspicious CONNECT request from 10.0.0.1
1406 $c = higlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1408 } elsif ($c =~ m/^socks5_connect:/) {
1410 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1414 found_unknown_content($c);
1422 sub handle_loglevel_info ($) {
1429 if ($c =~ m/^Rewrite detected:/) {
1431 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1432 $c = higlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1434 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1435 $c =~ m/^Compressed content detected/ or
1438 # Decompressing deflated iob: 117
1439 # Decompression didn't result in any content.
1440 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy with zlib support or enable the prevent-compression action.
1441 # Tagger 'complete-url' created empty tag. Ignored.
1445 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1447 # loading configuration file '/usr/local/etc/privoxy/config':
1448 # Reloading configuration file '/usr/local/etc/privoxy/config'
1449 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1451 } elsif ($c =~ m/^exiting by signal/) {
1453 # exiting by signal 15 .. bye
1454 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1456 } elsif ($c =~ m/^Privoxy version/) {
1458 # Privoxy version 3.0.7
1459 $c =~ s@(?<=Privoxy version )(\d.\d.\d)@$h{'version'}$1$h{'Standard'}@;
1461 } elsif ($c =~ m/^Program name: /) {
1463 # Program name: /usr/local/sbin/privoxy
1464 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1466 } elsif ($c =~ m/^Listening on port /) {
1468 # Listening on port 8118 on IP address 10.0.0.1
1469 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1470 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1472 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1474 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1475 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1477 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1479 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1480 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1481 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1483 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1484 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1485 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1486 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1488 } elsif ($c =~ m/^No logfile configured/ or
1489 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1490 $c =~ m/^Invalid \"chunked\" transfer/
1493 # No logfile configured. Please enable it before reporting any problems.
1494 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid response or even no response at all.
1495 # No logfile configured. Logging disabled.
1496 # Invalid "chunked" transfer encoding detected and ignored.
1500 found_unknown_content($c);
1507 sub handle_loglevel_cgi ($) {
1514 if ($c =~ m/^Granting access to/) {
1516 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1518 } elsif ($c =~ m/^Substituting: s(.)/) {
1520 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1521 # XXX: prone to span several lines
1524 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1525 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1526 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1532 sub handle_loglevel_force ($) {
1539 if ($c =~ m/^Ignored force prefix in request:/) {
1541 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1542 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1543 $c = higlight_matched_request_line($c, '(?<=request: ")[^"]*');
1545 } elsif ($c =~ m/^Enforcing request:/) {
1547 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1548 $c = higlight_matched_request_line($c, '(?<=request: ")[^"]*');
1552 found_unknown_content($c);
1559 sub handle_loglevel_ignore ($) {
1563 ################################################################################
1564 # Functions that actually print stuff
1565 ################################################################################
1567 sub print_clf_message () {
1569 our ($ip, $timestamp, $request_line, $status_code, $size);
1573 return if DEBUG_SUPPRESS_LOG_MESSAGES;
1575 # Rebuild highlighted
1576 $output .= $h{'Number'} . $ip . $h{'Standard'};
1578 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
1580 $output .= "\"" . highlight_request_line("$request_line") . "\"";
1582 $output .= $h{'Status'} . $status_code . $h{'Standard'};
1584 $output .= $h{'Number'} . $size . $h{'Standard'};
1585 $output .= get_line_end();
1590 sub print_non_clf_message ($) {
1593 our %thread_colours;
1596 our $time_colour_index;
1599 my $content = shift;
1600 my ($day, $time_stamp, $msecs, $thread, $log_level)
1601 = ($req{$t}{'day'}, $req{$t}{'time-stamp'}, $req{$t}{'msecs'}, $t, $req{$t}{'log-level'} );
1603 return if DEBUG_SUPPRESS_LOG_MESSAGES;
1605 $output .= $h{"Standard"} unless cli_option_is_set('html-output');
1606 # $output .= "$day ";
1607 $output .= $time_colours[$time_colour_index % 2];
1609 $output .= $time_stamp;
1610 $output .= ".$msecs" unless cli_option_is_set('no-msecs');
1611 $output .= $h{"Standard"};
1613 $output .= $thread_colours{$thread} if (defined($thread_colours{$thread}));
1615 $output .= $h{"Standard"} . " ";
1616 $output .= $h{$log_level} if (defined($h{$log_level}));
1617 $output .= $log_level;
1618 $output .= $h{"Standard"} . ": ";
1619 $output .= "$content";
1620 $output .= get_line_end();
1628 our %req; # request data from previous lines
1630 our %thread_colours;
1633 our $thread_colour_index = 0;
1634 our $header_colour_index = 0;
1635 our $time_colour_index = 0;
1637 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
1639 my $last_thread = 0;
1640 my $last_timestamp = 0;
1642 my $filters_that_did_nothing;
1645 our $no_special_header_highlighting;
1646 $time_colour = paint_it('white');
1647 my %log_level_count;
1649 my %log_level_handlers = (
1650 'Re-Filter' => \&handle_loglevel_re_filter,
1651 'Header' => \&handle_loglevel_header,
1652 'Connect' => \&handle_loglevel_connect,
1653 'Redirect' => \&handle_loglevel_redirect,
1654 'Request' => \&handle_loglevel_request,
1655 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
1656 'Info' => \&handle_loglevel_info,
1657 'CGI' => \&handle_loglevel_cgi,
1658 'Force' => \&handle_loglevel_force,
1659 'Error' => \&handle_loglevel_ignore,
1660 'Fatal error' => \&handle_loglevel_ignore,
1661 'Writing' => \&handle_loglevel_ignore,
1668 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\(([^\)]*)\)) ([\w -]*): (.*)$/) {
1669 # XXX: Put in req hash?
1672 $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution
1677 $req{$t}{'day'} = $day;
1678 $req{$t}{'time-stamp'} = $time_stamp;
1679 $req{$t}{'msecs'} = $msecs; # Only the cool kids have micro second resolution;
1680 $req{$t}{'log-level'} = $log_level;
1681 $req{$t}{'content'} = $content;
1682 $req{$t}{'log-message'} = $_;
1683 $no_special_header_highlighting = 0;
1685 $log_level_count{$log_level}++;
1687 if (defined($log_level_handlers{$log_level})) {
1689 $content = $log_level_handlers{$log_level}($content);
1693 die "No handler found for log level \"$log_level\"\n";
1697 # Highlight Truncations
1698 if (m/\.\.\. \[(too long, truncated)/) {
1699 $content =~ s@($1)@$h{'Truncation'}$1$h{'Standard'}@g;
1702 next unless $content;
1704 # Register threads to keep the colour constant
1705 if (!defined($thread_colours{$thread})) {
1706 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
1707 $thread_colour_index++;
1710 # Switch timestamp colour if timestamps differ
1711 if ($msecs != $last_msecs || !($time_stamp =~ m/$last_timestamp/)) {
1712 debug_message("Tick tack!") if DEBUG_TICKS;
1713 $time_colour = $time_colours[$time_colour_index % 2];
1714 $time_colour_index++
1717 $last_msecs = $msecs;
1718 $last_thread = $thread;
1719 $last_timestamp = $time_stamp;
1721 print_non_clf_message($content);
1723 } elsif (m/^(\d+\.\d+\.\d+\.\d+) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
1725 # LOG_LEVEL_CLF lines look like this
1726 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1 HTTP/1.1" 403 1730
1727 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
1729 print_clf_message();
1733 # Some Privoxy log messages span more than one line,
1734 # usually to dump lots of content that doesn't need any syntax highlighting.
1735 # XXX: add mechanism to forward these lines to the right handler anyway.
1737 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
1738 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
1743 if (cli_option_is_set('statistic')) {
1744 foreach (keys %log_level_count) {
1745 print $_ . ": " . $log_level_count{$_} . " ";
1750 sub VersionMessage {
1751 my $version_message;
1753 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
1754 $version_message .= 'Copyright (C) 2007-2008 Fabian Keil <fk@fabiankeil.de>' . "\n";
1755 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
1757 print $version_message;
1760 sub get_cli_options () {
1762 our %cli_options = (
1763 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
1764 'title' => CLI_OPTION_TITLE,
1765 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
1766 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
1767 'no-msecs' => CLI_OPTION_NO_MSECS,
1768 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
1769 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
1770 'statistic' => CLI_OPTION_STATISTIC,
1774 'html-output' => \$cli_options{'html-output'},
1775 'title' => \$cli_options{'title'},
1776 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
1777 'no-embedded-css' => \$cli_options{'no-embedded-css'},
1778 'no-msecs' => \$cli_options{'no-msecs'},
1779 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
1780 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
1781 'statistic' => \$cli_options{'statistic'},
1782 'version' => sub { VersionMessage && exit(0) }
1788 ################################################################################
1790 ################################################################################
1794 set_background(DEFAULT_BACKGROUND);
1795 prepare_our_stuff();
1808 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
1812 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
1813 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--show-ineffective-filters>]
1818 B<privoxy-log-parser> reads Privoxy log messages and
1820 - syntax-highlights recognized lines,
1822 - reformats some of them for easier comprehension,
1824 - filters out less useful messages, and
1826 - (in some cases) calculates additional information,
1827 like the compression ratio or how a filter affected
1830 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
1831 without getting confused by the resulting amount of output. For example for
1832 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
1833 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
1834 will hide the "filter foo caused 0 hits" message.
1838 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
1839 just don't highlight them.
1841 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
1842 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
1843 This option is only intended to make embedding log excerpts in web pages easier.
1844 It does not excape any input!
1846 [B<--no-msecs>] Don't expect milisecond resolution
1848 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
1849 the filtered output is piped into less in which case the ANSI control
1850 codes don't work, or if the terminal itself doesn't support the control
1853 [B<--show-ineffective-filters>] Don't suppress log lines for filters
1854 that didn't modify the content.
1856 [B<--version>] Print version and exit.
1860 To monitor a log file:
1862 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
1864 Replace '-F' with '-f' if your tail implementation lacks '-F' support
1865 or if the log won't get rotated anyway. The log file location depends
1866 on your system (Doh!).
1868 To monitor Privoxy without having it write to a log file:
1870 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
1872 Again, the config file location depends on your system. Output redirection
1873 depends on your shell, the above works with bourne shells.
1875 To read a processed Privoxy log file from top to bottom, letting the content
1876 scroll by slightly faster than you can read:
1878 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
1880 This is probably only useful to fill screens in the background of haxor movies.
1884 Syntax highlighting with ANSI escape sequences will look strange
1885 if your background color isn't black.
1887 Some messages aren't recognized yet and will not be fully highlighted.
1889 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
1890 using earlier Privoxy versions will probably result in an increased amount
1891 of unrecognized log lines.
1893 Privoxy's log files tend to be rather large. If you use HTML
1894 highlighting some browsers can't handle them, get confused and
1895 will eventually crash because of segmentation faults or unexpected
1896 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
1901 Many settings can't be controlled through command line options yet.
1909 Fabian Keil <fk@fabiankeil.de>