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 1.63 2009/12/30 11:14:30 fabiankeil Exp $
14 # - LOG_LEVEL_CGI, LOG_LEVEL_ERROR, LOG_LEVEL_WRITE content highlighting
15 # - create fancy statistics
16 # - grep through Privoxy sources to find unsupported log messages
17 # - hunt down substitutions that match content from variables which
18 # can contain stuff like ()?'[]
19 # - replace $h{'foo'} with h('foo') where possible
20 # - hunt down XXX comments instead of just creating them
21 # - add example log lines for every regex and mark them up for
23 # - Handle incomplete input without Perl warning about undefined variables.
24 # - Use generic highlighting function that takes a regex and the
27 # Copyright (c) 2007-2009 Fabian Keil <fk@fabiankeil.de>
29 # Permission to use, copy, modify, and distribute this software for any
30 # purpose with or without fee is hereby granted, provided that the above
31 # copyright notice and this permission notice appear in all copies.
33 # THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
34 # WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
35 # MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
36 # ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
37 # WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
38 # ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
39 # OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
40 ################################################################################
47 PRIVOXY_LOG_PARSER_VERSION => '0.5',
48 # Feel free to mess with these ...
49 DEFAULT_BACKGROUND => 'black', # Choose registered colour (like 'black')
50 DEFAULT_TEXT_COLOUR => 'white', # Choose registered colour (like 'black')
51 HEADER_DEFAULT_COLOUR => 'yellow',
52 REGISTER_HEADERS_WITH_THE_SAME_COLOUR => 1,
54 CLI_OPTION_DEFAULT_TO_HTML_OUTPUT => 0,
55 CLI_OPTION_TITLE => 'Privoxy-Log-Parser in da house',
56 CLI_OPTION_NO_EMBEDDED_CSS => 0,
57 CLI_OPTION_NO_MSECS => 0,
58 CLI_OPTION_NO_SYNTAX_HIGHLIGHTING => 0,
59 CLI_OPTION_ERROR_LOG_FILE => '/var/log/privoxy-log.log',
60 CLI_OPTION_SHOW_INEFFECTIVE_FILTERS => 0,
61 CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES => 0,
62 CLI_OPTION_STATISTICS => 0,
64 SUPPRESS_SUCCEEDED_FILTER_ADDITIONS => 1,
66 SHOW_FILTER_READIN_IN => 0,
67 SUPPRESS_EMPTY_LINES => 1,
68 SUPPRESS_SUCCESSFUL_CONNECTIONS => 1,
69 SUPPRESS_ACCEPTED_CONNECTIONS => 1,
70 SUPPRESS_GIF_NOT_CHANGED => 1,
71 SUPPRESS_NEED_TO_DE_CHUNK_FIRST => 1,
73 DEBUG_HEADER_REGISTERING => 0,
74 DEBUG_HEADER_HIGHLIGHTING => 0,
77 DEBUG_SUPPRESS_LOG_MESSAGES => 0,
79 PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH => 0,
80 PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH => 1,
82 LOG_UNPARSED_LINES_TO_EXTRA_FILE => 0,
84 # You better leave these alone unless you know what you're doing.
85 COLOUR_RESET => "\033[0;0m",
89 # For performance reasons, these are global.
92 my %req; # request data from previous lines
97 my $thread_colour_index = 0;
98 my $header_colour_index = 0;
99 my $time_colour_index = 0;
101 my $no_special_header_highlighting;
104 my $header_highlight_regex = '';
106 my $html_output_mode;
107 my $no_msecs_mode; # XXX: should probably be removed
110 sub prepare_our_stuff () {
112 # Syntax Higlight hash
114 'red', 'green', 'brown', 'blue', 'purple', 'cyan',
115 'light_gray', 'light_red', 'light_green', 'yellow',
116 'light_blue', 'pink', 'light_cyan', 'white'
123 Filter => 'purple', # XXX: Used?
124 'Re-Filter' => 'purple',
126 Request => 'light_cyan',
127 CGI => 'light_green',
129 Error => 'light_red',
131 'Fatal error' => 'light_red',
132 'Gif-Deanimate' => 'blue',
134 Writing => 'light_green',
135 # ----------------------
138 request_ => 'brown', # host+path but no protocol
139 'ip-address' => 'yellow',
142 Truncation => 'light_red',
144 Timestamp => 'brown',
145 Crunching => 'light_red',
146 crunched => 'light_red',
147 'Request-Line' => 'pink',
149 destination => 'yellow',
150 'http-version' => 'pink',
151 'crunch-pattern' => 'pink',
156 'program-name' => 'cyan',
159 warning => 'light_red',
160 debug => 'light_red',
164 'status-message' => 'light_cyan',
165 'status-code' => 'yellow',
166 'invalid-request' => 'light_red',
168 error => 'light_red',
169 'rewritten-URL' => 'light_red',
170 'pcrs-delimiter' => 'light_red',
171 'ignored' => 'light_red',
172 'action-bits-update' => 'light_red',
173 'configuration-line' => 'red',
174 'content-type' => 'yellow',
179 # Header colours need their own hash so the keys can be accessed properly
181 # Prefilled with headers that should not appear with default header colours
182 Cookie => 'light_red',
183 'Set-Cookie' => 'light_red',
184 Warning => 'light_red',
185 Default => HEADER_DEFAULT_COLOUR,
188 # Crunch reasons need their own hash as well
190 'Unsupported HTTP feature' => 'light_red',
191 Blocked => 'light_red',
192 Untrusted => 'light_red',
193 Redirected => 'green',
194 'CGI Call' => 'white',
195 'DNS failure' => 'red',
196 'Forwarding failed' => 'light_red',
197 'Connection failure' => 'light_red',
198 'Out of memory (may mask other reasons)' => 'light_red',
199 'No reason recorded' => 'light_red',
202 @time_colours = ('white', 'light_gray');
204 # Translate highlight strings into highlight code
205 prepare_highlight_hash(\%header_colours);
206 prepare_highlight_hash(\%reason_colours);
207 prepare_highlight_hash(\%h);
208 prepare_colour_array(\@all_colours);
209 prepare_colour_array(\@time_colours);
216 ###############################################################
217 # Takes a colour string and returns an ANSI escape sequence
218 # (unless --no-syntax-highlighting is used).
219 # XXX: The Rolling Stones reference has to go.
220 ###############################################################
222 my $colour = shift @_;
224 return "" if cli_option_is_set('no-syntax-highlighting');
266 my $bg_code = get_background();
268 our $default = default_colours();
270 if (defined($text{$colour})) {
271 $colour_code = ESCAPE;
272 $colour_code .= $text{$colour};
274 $colour_code .= $light{$colour} ? "1" : "2";
276 $colour_code .= $bg_code;
278 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
280 } elsif ($colour =~ /reset/) {
282 $colour_code = default_colours();
286 die "What's $colour supposed to mean?\n";
292 sub get_semantic_html_markup ($) {
293 ###############################################################
294 # Takes a string and returns a span element
295 ###############################################################
300 if ($type =~ /Standard/) {
304 $code = '<span title="' . $type . '" class="' . $type . '">';
310 sub cli_option_is_set ($) {
313 my $cli_option = shift;
315 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
317 return $cli_options{$cli_option};
320 sub get_html_title () {
323 return $cli_options{'title'};
327 sub init_css_colours() {
335 purple => "F06", # XXX: wrong
336 cyan => "F09", # XXX: wrong
341 light_green => "33F",
350 sub get_css_colour ($) {
355 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
357 return '#' . $css_colours{$colour};
360 sub get_css_line ($) {
365 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
366 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
367 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
368 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
369 $css_line .= '}' . "\n";
374 sub get_css_line_for_colour ($) {
379 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
380 $css_line .= 'color:' . get_css_colour($colour) . ';';
381 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
382 $css_line .= '}' . "\n";
387 # XXX: Wrong solution
388 sub get_missing_css_lines () {
392 $css_line .= '.' . 'default' . ' {';
393 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
394 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
395 $css_line .= '}' . "\n";
402 our %css_colours; #XXX: Wrong solution
406 $css .= '.privoxy-log {';
407 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
408 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
411 foreach my $key (keys %h_colours) {
413 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
414 $css .= get_css_line($key);
418 foreach my $colour (keys %css_colours) {
420 $css .= get_css_line_for_colour($colour);
424 $css .= get_missing_css_lines(); #XXX: Wrong solution
433 if (cli_option_is_set('html-output')) {
435 my $title = get_html_title();
437 $intro .= '<html><head>';
438 $intro .= '<title>' . $title . '</title>';
439 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
440 $intro .= '</head><body>';
441 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
451 if (cli_option_is_set('html-output')) {
453 $outro = '</p></body></html>';
459 sub get_line_end () {
460 return cli_option_is_set('html-output') ? "<br>\n" : "\n";
463 sub get_colour_html_markup ($) {
464 ###############################################################
465 # Takes a colour string a span element. XXX: WHAT?
466 # XXX: This function shouldn't be necessary, the
467 # markup should always be semantically correct.
468 ###############################################################
473 if ($type =~ /Standard/) {
476 $code = '<span class="' . lc($type) . '">';
482 sub default_colours () {
485 return reset_colours();
488 sub show_colours () {
492 sub reset_colours () {
493 return ESCAPE . "0m";
496 sub set_background ($){
512 if (defined($backgrounds{$colour})) {
513 $bg_code = $backgrounds{$colour};
515 die "Invalid background colour: " . $colour;
519 sub get_background (){
523 sub prepare_highlight_hash ($) {
526 if (!$html_output_mode) {
528 foreach my $key (keys %$ref) {
529 $$ref{$key} = paint_it($$ref{$key});
534 foreach my $key (keys %$ref) {
535 $$ref{$key} = get_semantic_html_markup($key);
541 sub prepare_colour_array ($) {
544 if (!$html_output_mode) {
546 foreach my $i (0 ... @$ref - 1) {
547 $$ref[$i] = paint_it($$ref[$i]);
552 foreach my $i (0 ... @$ref - 1) {
553 $$ref[$i] = get_colour_html_markup($$ref[$i]);
559 sub found_unknown_content ($) {
564 return if cli_option_is_set('accept-unknown-messages');
566 return if ($unknown =~ /\[too long, truncated\]$/);
568 $message = "found_unknown_content: Don't know how to highlight: ";
569 # Break line so the log file can later be parsed as Privoxy log file again
570 $message .= '"' . $unknown . '"' . " in:\n";
571 $message .= $req{$t}{'log-message'};
572 debug_message($message);
573 log_parse_error($req{$t}{'log-message'});
575 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
578 sub log_parse_error ($) {
582 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
583 open(ERRORLOG, ">>" . ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
584 print ERRORLOG $message;
589 sub debug_message (@) {
592 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
595 ################################################################################
596 # highlighter functions that aren't loglevel-specific
597 ################################################################################
601 # Get highlight marker
602 my $highlight = shift; # XXX: Stupid name;
606 if (defined($highlight)) {
608 $result = $h{$highlight};
612 $message = "h: Don't recognize highlighter $highlight.";
613 debug_message($message);
614 log_parser_error($message);
615 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
621 sub highlight_known_headers ($) {
625 debug_message("Searching $content for things to highlight.") if DEBUG_HEADER_HIGHLIGHTING;
627 if ($content =~ m/(?<=\s)($header_highlight_regex):/) {
629 $content =~ s@(?<=[\s|'])($header)(?=:)@$header_colours{$header}$1$h{'Standard'}@ig;
630 debug_message("Highlighted '$header' in '$content'") if DEBUG_HEADER_HIGHLIGHTING;
636 sub highlight_matched_request_line ($$) {
638 my $result = shift; # XXX: Stupid name;
640 if ($result =~ m@(.*)($regex)(.*)@) {
641 $result = $1 . highlight_request_line($2) . $3
646 sub highlight_request_line ($) {
649 my ($method, $url, $http_version);
651 #GET http://images.sourceforge.net/sfx/icon_warning.gif HTTP/1.1
652 if ($rl =~ m/Invalid request/) {
654 $rl = h('invalid-request') . $rl . h('Standard');
656 } elsif ($rl =~ m/^([-\w]+) (.*) (HTTP\/\d\.\d)/) {
658 # XXX: might not match in case of HTTP method fuzzing.
659 # XXX: save these: ($method, $path, $http_version) = ($1, $2, $3);
660 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
661 if ($rl =~ /http:\/\//) {
662 $rl = highlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
664 $rl = highlight_matched_pattern($rl, 'request_', '[^\s]*(?=\sHTTP)');
667 $rl =~ s@(HTTP\/\d\.\d)$@$h{'http-version'}$1$h{'Standard'}@;
669 } elsif ($rl =~ m/\.\.\. \[too long, truncated\]$/) {
671 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
672 $rl = highlight_matched_url($rl, '[^\s]*(?=\.\.\.)');
674 } elsif ($rl =~ m/^ $/) {
676 $rl = h('error') . "No request line specified!" . h('Standard');
680 debug_message ("Can't parse request line: $rl");
687 sub highlight_response_line ($) {
690 my ($http_version, $status_code, $status_message);
695 # TODO: Mark different status codes differently
697 if ($rl =~ m/((?:HTTP\/\d\.\d|ICY)) (\d+) (.*)/) {
698 ($http_version, $status_code, $status_message) = ($1, $2, $3);
700 debug_message ("Can't parse response line: $rl") and die 'Fix this';
703 # Rebuild highlighted
705 $rl .= h('http-version') . $http_version . h('Standard');
707 $rl .= h('status-code') . $status_code . h('Standard');
709 $rl .= h('status-message') . $status_message . h('Standard');
714 sub highlight_matched_url ($$) {
716 my $result = shift; # XXX: Stupid name;
719 #print "Got $result, regex ($regex)\n";
721 if ($result =~ m@(.*?)($regex)(.*)@) {
722 $result = $1 . highlight_url($2) . $3;
723 #print "Now the result is $result\n";
729 sub highlight_matched_host ($$) {
731 my ($result, $regex) = @_; # XXX: result ist stupid name;
733 if ($result =~ m@(.*?)($regex)(.*)@) {
734 $result = $1 . $h{host} . $2 . $h{Standard} . $3;
740 sub highlight_matched_pattern ($$$) {
742 my $result = shift; # XXX: Stupid name;
746 die "Unknown key $key" unless defined $h{$key};
748 if ($result =~ m@(.*?)($regex)(.*)@) {
749 $result = $1 . h($key) . $2 . h('Standard') . $3;
755 sub highlight_matched_path ($$) {
757 my $result = shift; # XXX: Stupid name;
760 if ($result =~ m@(.*?)($regex)(.*)@) {
761 $result = $1 . h('path') . $2 . h('Standard') . $3;
767 sub highlight_url ($) {
771 if ($html_output_mode) {
773 $url = '<a href="' . $url . '">' . $url . '</a>';
777 $url = h('URL') . $url . h('Standard');
784 sub update_header_highlight_regex ($) {
787 my $headers = join ('|', keys %header_colours);
789 $header_highlight_regex = qr/$headers/;
790 print "Registering '$header'\n" if DEBUG_HEADER_HIGHLIGHTING;
793 ################################################################################
794 # loglevel-specific highlighter functions
795 ################################################################################
797 sub handle_loglevel_header ($) {
802 if ($c =~ /^scan:/) {
804 if ($c =~ m/^scan: ((\w+) (.+) (HTTP\/\d\.\d))/) {
806 # Client request line
807 # Save for statistics (XXX: Not implemented yet)
808 $req{$t}{'method'} = $2;
809 $req{$t}{'destination'} = $3;
810 $req{$t}{'http-version'} = $4;
812 $content = highlight_request_line($1);
814 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
816 # Server response line
817 $req{$t}{'response_line'} = $2;
818 $req{$t}{'status_code'} = $3;
819 $req{$t}{'status_message'} = $4;
820 $content = $1 . highlight_response_line($req{$t}{'response_line'});
822 } elsif ($c =~ m/^scan: ((?>[^:]+)):/) {
824 # Register new headers
825 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
827 if (!defined($header_colours{$header}) and $header =~ /^[\d\w-]*$/) {
828 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
830 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
831 $header_colours{$header} = $header_colours{'Default'};
833 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
834 $header_colour_index++;
836 update_header_highlight_regex($header);
841 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
843 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
844 $content =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
845 $content =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
847 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
849 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/
850 $c = highlight_matched_host($c, '(?<=New host is: )[^\s]+');
851 $content = highlight_matched_url($c, '(?<=Crunching Referer: )[^\s]+');
853 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
855 # Text mode enabled by force. Take cover!
856 $content =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
858 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
860 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
861 $content = $1 . highlight_request_line($2);
863 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
865 # Adjusted Content-Length to 2132
866 # Adjust Content-Length to 33533
867 $content =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
868 $content = highlight_known_headers($content);
870 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
872 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
873 $content = highlight_matched_url($content, '(?<=New request URL: ).*');
875 } elsif ($c =~ m/^Couldn\'t parse:/) {
877 # XXX: These should probable be logged with LOG_LEVEL_ERROR
878 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
879 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
880 $content =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
882 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
883 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
885 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
886 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
887 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
889 # XXX: Save tag and tagger
891 $content =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
892 $content =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
893 $content =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
894 $no_special_header_highlighting = 1;
896 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
898 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
899 # XXX: Save tag and tagger
901 $content =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
902 $content =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
904 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
905 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
906 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
907 or $c =~ m/^(addh-unique|Referer forged to)/
908 or $c =~ m/^Downgraded answer to HTTP\/1.0/
909 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
910 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
911 or $c =~ m/^Referer crunched!/
912 or $c =~ m/^crunched x-forwarded-for!/
913 or $c =~ m/^crunched From!/
914 or $c =~ m/^ modified$/
915 or $c =~ m/^Content filtering is enabled. Crunching:/
916 or $c =~ m/^force-text-mode overruled the client/
917 or $c =~ m/^Server time in the future\./
918 or $c =~ m/^content-disposition header crunched and replaced with:/i
919 or $c =~ m/^Reducing white space in /
920 or $c =~ m/^Ignoring single quote in /
921 or $c =~ m/^Converting tab to space in /
922 or $c =~ m/A HTTP\/1\.1 response without/
923 or $c =~ m/Disabled filter mode on behalf of the client/
924 or $c =~ m/Keeping the (?:server|client) header /
925 or $c =~ m/Content modified with no Content-Length header set/
926 or $c =~ m/^Appended client IP address to/
927 or $c =~ m/^Removing 'Connection: close' to imply keep-alive./
928 or $c =~ m/^keep-alive support is disabled/
931 # XXX: Some of these may need highlighting
933 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
934 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
935 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
936 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
937 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
938 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
939 # Text mode is already enabled
940 # Denied request with NULL byte(s) turned into line break(s)
941 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
942 # addh-unique: Host: people.freebsd.org
943 # Suppressed offer to compress content
944 # Crunched incoming cookie -- yum!
945 # Accepted the client's request to fetch without filtering.
946 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
947 # addh-unique: Host: subkeys.pgp.net:11371
948 # Referer forged to: Referer: http://10.0.0.1/
949 # Downgraded answer to HTTP/1.0
950 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
951 # Referer overwritten with: Referer: pille-palle
952 # Referer replaced with: Referer: pille-palle
953 # crunched x-forwarded-for!
955 # modified # XXX: pretty stupid log message
956 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
957 # force-text-mode overruled the client's request to fetch without filtering!
958 # Server time in the future.
959 # content-disposition header crunched and replaced with: content-disposition: filename=baz
960 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
961 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
962 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
963 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
965 # A HTTP/1.1 response without Connection header implies keep-alive.
966 # Disabled filter mode on behalf of the client.
967 # Keeping the server header 'Connection: keep-alive' around.
968 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
969 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
970 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
971 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
972 # Removing 'Connection: close' to imply keep-alive.
973 # keep-alive support is disabled. Crunching: Keep-Alive: 300.
975 } elsif ($c =~ m/^scanning headers for:/) {
977 return '' unless SHOW_SCAN_INTRO;
979 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
980 # crunched User-Agent!
981 # Crunching: Content-Encoding: gzip
983 $content =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
985 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
987 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
989 $content = h('warning') . $content . h('Standard');
991 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
993 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
994 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
996 $content =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
997 $content =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
999 } elsif ($c =~ m/^Removing empty header/) {
1001 # Removing empty header
1004 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
1006 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
1007 # Enable force-text-mode if you know what you're doing.
1008 # XXX: Could highlight more here.
1009 $content =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
1011 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
1013 # Server keep-alive timeout is 5. Sticking with 10.
1014 # Client keep-alive timeout is 20. Sticking with 10.
1016 $content =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1017 $content =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1019 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1021 # Reducing keep-alive timeout from 60 to 10.
1023 $content =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1024 $content =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1028 found_unknown_content($content);
1032 unless ($c =~ m/^Transforming/) {
1033 $content = highlight_known_headers($content) unless $no_special_header_highlighting;
1039 sub handle_loglevel_re_filter ($) {
1041 my $content = shift;
1045 if ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1047 # XXX: only the second version gets highlighted properly.
1048 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1049 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1050 $req{$t}{'content_source'} = $1;
1051 $req{$t}{'content_size'} = $2;
1052 $req{$t}{'content_filter'} = $3;
1053 $req{$t}{'content_hits'} = $4;
1054 $req{$t}{'new_content_size'} = $5;
1055 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1056 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1057 if ($req{$t}{'content_hits'} == 0 and
1058 not (cli_option_is_set('show-ineffective-filters')
1059 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1063 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1064 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1065 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1067 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1068 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1071 $c .= "(" . $h{'Number'};
1072 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1073 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1076 } elsif ($c =~ /\.{3}$/
1077 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1079 # Used by Privoxy 3.0.5 and 3.0.6:
1081 # Used by Privoxy 3.0.7:
1082 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1084 $req{$t}{'filtered_header'} = $1;
1085 $req{$t}{'old_header_size'} = $2;
1086 $req{$t}{'header_filter_name'} = $3;
1088 unless (cli_option_is_set('show-ineffective-filters') or
1089 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1092 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1093 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1095 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1097 # ...produced 0 hits (new size 23).
1098 #... produced 1 hits (new size 54).
1100 $req{$t}{'header_filter_hits'} = $1;
1101 $req{$t}{'new_header_size'} = $2;
1103 unless (cli_option_is_set('show-ineffective-filters') or
1104 (defined($req{$t}{'header_filter_name'}) and
1105 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1107 if ($req{$t}{'header_filter_hits'} == 0 and
1108 not (defined($req{$t}{'header_filter_name'}) and
1109 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1112 # Reformat including information from the intro
1113 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1115 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1116 # $c .= $req{$t}{'filtered_header'} . " ";
1117 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1118 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1120 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1122 $c .= "changing size from ";
1123 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1125 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1130 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1134 # Highlight from last line (XXX: What?)
1135 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1136 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1141 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1142 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1147 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1149 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1150 # Tagger variable-test has empty joblist. Nothing to do.
1152 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1154 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1156 $req{$t}{'chunked-size'} = $1;
1157 $req{$t}{'dechunked-size'} = $2;
1158 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1160 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1162 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1163 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1165 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1167 # Decompression successful. Old size: 670, new size: 1166.
1169 $req{$t}{'size-compressed'} = $1;
1170 $req{$t}{'size-decompressed'} = $2;
1171 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1173 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1174 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1176 # XXX: Create sub get_percentage()
1177 if ($req{$t}{'size-decompressed'}) {
1178 $req{$t}{'decompression-gain-percent'} =
1179 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1181 $content .= " (saved: ";
1182 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1184 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1188 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1190 # Need to de-chunk first
1191 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1193 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1195 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1197 # Adding re_filter job ...
1198 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1199 # to filter client-ip-address succeeded.
1201 } elsif ($c =~ m/^Reading in filter/) {
1203 return '' unless SHOW_FILTER_READIN_IN;
1207 found_unknown_content($content);
1214 sub handle_loglevel_redirect ($) {
1218 if ($c =~ m/^Decoding "([^""]*)"/) {
1220 $req{$t}{'original-destination'} = $1;
1221 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1224 } elsif ($c =~ m/^Checking/) {
1226 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1227 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1229 # TODO: Change colour if really url-decoded
1230 $req{$t}{'decoded-original-destination'} = $1;
1231 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1234 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1236 # pcrs command "s@&from=rss@@" changed \
1237 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1238 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1239 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1240 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1241 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1242 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1243 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1244 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1246 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1248 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1249 # change "http://www.example.org/".
1250 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1251 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1253 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1255 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1256 # XXX: Use URL highlighter
1258 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1260 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1261 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1262 # is already properly formatted.
1263 # XXX: assume the same?
1264 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1268 found_unknown_content($c);
1275 sub handle_loglevel_gif_deanimate ($) {
1277 my $content = shift;
1279 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1281 my $bytes_from = $1;
1283 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1284 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1285 # XXX: Do we need g in case of ($1 == $2)?
1286 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1288 } elsif ($content =~ m/GIF (not) changed/) {
1290 # Gif-Deanimate: GIF not changed.
1291 return '' if SUPPRESS_GIF_NOT_CHANGED;
1292 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1294 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1296 # failed! (gif parsing)
1297 # XXX: Replace this error message with something less stupid
1298 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1300 } elsif ($content =~ m/^Need to de-chunk first/) {
1302 # Need to de-chunk first
1303 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1305 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1307 # No GIF header found (XXX: Did I ever commit this?)
1308 # failed while parsing 195 134747048 (XXX: never commited)
1310 # Ignore these for now
1314 found_unknown_content($content);
1321 sub handle_loglevel_request ($) {
1323 my $content = shift;
1325 if ($content =~ m/crunch! /) {
1327 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1329 # Highlight crunch reasons
1330 foreach my $reason (keys %reason_colours) {
1331 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1333 # Highlight request URL domain and ditch 'crunch!'
1334 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1335 $content =~ s@ crunch!@@;
1337 } elsif ($content =~ m/\[too long, truncated\]$/) {
1339 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1340 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1342 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1344 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1345 $content = h('request_') . $content . h('Standard');
1349 found_unknown_content($content);
1356 sub handle_loglevel_crunch ($) {
1358 my $content = shift;
1360 # Highlight crunch reason
1361 foreach my $reason (keys %reason_colours) {
1362 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1365 if ($content =~ m/\[too long, truncated\]$/) {
1367 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1368 # [...]&filter... [too long, truncated]
1369 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1373 # Blocked: http://ads.example.org/
1374 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1380 sub handle_loglevel_connect ($) {
1384 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1386 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1388 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1389 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1391 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1393 # connect to: www.example.org.noconnect failed: Operation not permitted
1395 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1397 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1399 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1401 # Connect: to www.nzherald.co.nz successful
1402 # Connect: to archiv.radiotux.de
1404 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1405 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1407 } elsif ($c =~ m/^to ([^\s]*)$/) {
1409 # Connect: to lists.sourceforge.net:443
1411 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1413 } elsif ($c =~ m/^accepted connection from .*/ or
1416 # accepted connection from 10.0.0.1
1417 # Privoxy 3.0.6 and earlier just say:
1419 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1420 $c = highlight_matched_host($c, '(?<=connection from ).*');
1422 } elsif ($c =~ m/^write header to: .* failed:/) {
1424 # write header to: 10.0.0.1 failed: Broken pipe
1426 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1427 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1429 } elsif ($c =~ m/^write header to client failed:/) {
1431 # write header to client failed: Broken pipe
1433 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1435 } elsif ($c =~ m/^socks4_connect:/) {
1437 # socks4_connect: SOCKS request rejected or failed.
1438 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1440 } elsif ($c =~ m/^Listening for new connections/ or
1441 $c =~ m/^accept connection/) {
1443 # Privoxy versions above 3.0.6 say:
1444 # Listening for new connections ...
1445 # earlier versions say:
1446 # accept connection ...
1449 } elsif ($c =~ m/^accept failed:/) {
1451 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1453 } elsif ($c =~ m/^Overriding forwarding settings/) {
1455 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1456 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1458 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1460 # Denying suspicious CONNECT request from 10.0.0.1
1461 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1463 } elsif ($c =~ m/^socks5_connect:/) {
1465 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1467 } elsif ($c =~ m/^Created new connection to/) {
1469 # Created new connection to www.privoxy.org:80 on socket 11.
1470 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1471 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1473 } elsif ($c =~ m/^Found reusable socket/) {
1475 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1477 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1478 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1479 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1480 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1481 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1482 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1483 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1484 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1486 } elsif ($c =~ m/^Marking open socket/) {
1488 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1489 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1490 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1491 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1493 } elsif ($c =~ m/^No reusable/) {
1495 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1496 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1498 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1500 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1501 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1503 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1504 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1505 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1507 } elsif ($c =~ m/^Socket/) {
1509 # Socket 16 already forgotten or never remembered.
1510 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1512 } elsif ($c =~ m/^The connection to/) {
1514 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1516 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1517 # Timeout is: 1. Assumed latency: 4.
1518 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1519 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1520 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1521 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1522 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1523 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1525 } elsif ($c =~ m/^Stopped waiting for the request line./) {
1527 # Stopped waiting for the request line. Timeout: 121.
1528 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1530 } elsif ($c =~ m/^Waiting for \d/) {
1532 # Waiting for 1 connections to timeout.
1533 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1535 } elsif ($c =~ m/^Initialized/) {
1537 # Initialized 20 socket slots.
1538 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1540 } elsif ($c =~ m/^Done reading from server/) {
1542 # Done reading from server. Expected content length: 24892. \
1543 # Actual content length: 24892. Most recently received: 4412.
1545 # Done reading from server. Expected content length: 24892. \
1546 # Actual content length: 24892. Bytes most recently read: 4412.
1547 # Done reading from server. Content length: 6018 as expected. \
1548 # Bytes most recently read: 294.
1549 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1550 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1551 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1553 } elsif ($c =~ m/^Continuing buffering headers/) {
1555 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1556 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1557 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1558 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1560 # Continuing buffering headers. Bytes most recently read: %d.
1561 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1563 } elsif ($c =~ m/^Received \d+ bytes while/) {
1565 # Received 206 bytes while expecting 12103.
1566 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1567 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1569 } elsif ($c =~ m/^Connection from/) {
1571 # Connection from 81.163.28.218 dropped due to ACL
1572 $c =~ s@(?<=^Connection from )((?:\d+\.?){4})@$h{'Number'}$1$h{'Standard'}@;
1574 } elsif ($c =~ m/^(?:Reusing|Closing) server socket \d./ or
1575 $c =~ m/^No additional client request/) {
1577 # Reusing server socket 4. Opened for 10.0.0.1.
1578 # Closing server socket 2. Opened for 10.0.0.1.
1579 # No additional client request received in time. \
1580 # Closing server socket 4, initially opened for 10.0.0.1.
1581 # No additional client request received in time on socket 29.
1583 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1584 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.$)');
1586 } elsif ($c =~ m/^Connected to /) {
1588 # Connected to tor-jail[10.0.0.2]:9050.
1590 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1591 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1592 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1594 } elsif ($c =~ m/^Could not connect to /) {
1596 # Could not connect to [10.0.0.1]:80.
1598 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1599 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1601 } elsif ($c =~ m/^Waiting for the next client request/ or
1602 $c =~ m/^The connection on server socket/ or
1603 $c =~ m/^Client request arrived in time or the client closed the connection/) {
1605 # Waiting for the next client request on socket 3. Keeping the server \
1606 # socket 12 to a.fsdn.com open.
1607 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1608 # Client request arrived in time or the client closed the connection on socket 12.
1610 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1611 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1612 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1614 } elsif ($c =~ m/^Marking the server socket/) {
1616 # Marking the server socket 7 tainted.
1618 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1620 } elsif ($c =~ m/^Reduced expected bytes to /) {
1622 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1623 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1624 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1626 } elsif ($c =~ m/^The client closed socket /) {
1628 # The client closed socket 2 while the server socket 4 is still open.
1629 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1630 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1632 } elsif ($c =~ m/^Expected client content length set /) {
1634 # Expected client content length set to 667325411 after reading 4999 bytes.
1635 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1636 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1638 } elsif ($c =~ m/^Waiting for up to /) {
1640 # Waiting for up to 4999 bytes from the client.
1641 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1643 } elsif ($c =~ m/^Looks like we rea/ or
1644 $c =~ m/^Unsetting keep-alive flag/ or
1645 $c =~ m/^No connections to wait/ or
1646 $c =~ m/^Complete client request received/ or
1647 $c =~ m/^Possible pipeline attempt detected./ or
1648 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1649 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1650 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1651 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1652 $c =~ m/^Stopping to watch the client socket. There's already another request waiting./ or
1653 $c =~ m/^Done reading from the client\.$/) {
1655 # Looks like we reached the end of the last chunk. We better stop reading.
1656 # Looks like we read the end of the last chunk together with the server \
1657 # headers. We better stop reading.
1658 # Unsetting keep-alive flag.
1659 # No connections to wait for left.
1660 # Client request arrived in time or the client closed the connection.
1661 # Complete client request received
1662 # Possible pipeline attempt detected. The connection will not be \
1663 # kept alive and we will only serve the first request.
1664 # POST request detected. The connection will not be kept alive.
1665 # The server still wants to talk, but the client hung up on us.
1666 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1667 # There might be a request body. The connection will not be kept alive.
1668 # Stopping to watch the client socket. There's already another request waiting.
1669 # Done reading from the client\.
1673 found_unknown_content($c);
1681 sub handle_loglevel_info ($) {
1685 if ($c =~ m/^Rewrite detected:/) {
1687 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1688 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1690 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1691 $c =~ m/^Compressed content detected/ or
1694 # Decompressing deflated iob: 117
1695 # Decompression didn't result in any content.
1696 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1697 # with zlib support or enable the prevent-compression action.
1698 # Tagger 'complete-url' created empty tag. Ignored.
1702 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1704 # loading configuration file '/usr/local/etc/privoxy/config':
1705 # Reloading configuration file '/usr/local/etc/privoxy/config'
1706 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1708 } elsif ($c =~ m/^exiting by signal/) {
1710 # exiting by signal 15 .. bye
1711 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1713 } elsif ($c =~ m/^Privoxy version/) {
1715 # Privoxy version 3.0.7
1716 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1718 } elsif ($c =~ m/^Program name: /) {
1720 # Program name: /usr/local/sbin/privoxy
1721 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1723 } elsif ($c =~ m/^Listening on port /) {
1725 # Listening on port 8118 on IP address 10.0.0.1
1726 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1727 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1729 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1731 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1732 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1734 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1736 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1737 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1738 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1740 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1741 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1742 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1743 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1745 } elsif ($c =~ m/^Status code/) {
1747 # Status code 304 implies no body.
1748 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1750 } elsif ($c =~ m/^Method/) {
1752 # Method HEAD implies no body.
1753 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1755 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1757 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1758 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1759 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1761 } elsif ($c =~ m/^No logfile configured/ or
1762 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1763 $c =~ m/^Invalid \"chunked\" transfer/ or
1764 $c =~ m/^Support for/ or
1765 $c =~ m/^Flushing header and buffers/
1768 # No logfile configured. Please enable it before reporting any problems.
1769 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1770 # response or even no response at all.
1771 # No logfile configured. Logging disabled.
1772 # Invalid "chunked" transfer encoding detected and ignored.
1773 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1774 # known not to work properly in some situations.
1775 # Flushing header and buffers. Stepping back from filtering.
1779 found_unknown_content($c);
1786 sub handle_loglevel_cgi ($) {
1790 if ($c =~ m/^Granting access to/) {
1792 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1794 } elsif ($c =~ m/^Substituting: s(.)/) {
1796 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1797 # XXX: prone to span several lines
1800 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1801 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1802 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1808 sub handle_loglevel_force ($) {
1812 if ($c =~ m/^Ignored force prefix in request:/) {
1814 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1815 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1816 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1818 } elsif ($c =~ m/^Enforcing request:/) {
1820 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1821 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1825 found_unknown_content($c);
1832 sub handle_loglevel_error ($) {
1836 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1838 # Empty server or forwarder response received on socket 4.
1839 # Empty server or forwarder response received on socket 3. \
1840 # Closing client socket 15 without sending data.
1841 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1842 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1844 # XXX: There are probably more messages that deserve highlighting.
1850 sub handle_loglevel_ignore ($) {
1854 sub gather_loglevel_request_stats ($$) {
1862 sub gather_loglevel_crunch_stats ($$) {
1872 sub gather_loglevel_error_stats ($$) {
1879 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1881 # Empty server or forwarder response received on socket 4.
1882 $stats{'empty-responses'}++;
1883 if ($thread_data{$thread}{'new_connection'}) {
1884 $stats{'empty-responses-on-new-connections'}++;
1886 $stats{'empty-responses-on-reused-connections'}++;
1891 sub gather_loglevel_connect_stats ($$) {
1898 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
1900 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1901 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
1903 } elsif ($c =~ m/^to ([^\s]*)$/) {
1905 # Connect: to lists.sourceforge.net:443
1907 $thread_data{$thread}{'forwarder'} = 'direct connection';
1909 } elsif ($c =~ m/^Created new connection to/) {
1911 # Created new connection to www.privoxy.org:80 on socket 11.
1913 $thread_data{$thread}{'new_connection'} = 1;
1915 } elsif ($c =~ m/^Reusing server socket \d./ or
1916 $c =~ m/^Found reusable socket/) {
1918 # Reusing server socket 4. Opened for 10.0.0.1.
1919 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1921 $thread_data{$thread}{'new_connection'} = 0;
1922 $stats{'reused-connections'}++;
1926 sub gather_loglevel_header_stats ($) {
1932 if ($c =~ m/^A HTTP\/1\.1 response without/ or
1933 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
1935 # A HTTP/1.1 response without Connection header implies keep-alive.
1936 # Keeping the server header 'Connection: keep-alive' around.
1937 $stats{'server-keep-alive'}++;
1945 'server-keep-alive' => 0,
1946 'reused-connections' => 0,
1947 'empty-responses' => 0,
1948 'empty-responses-on-new-connections' => 0,
1949 'empty-responses-on-reused-connections' => 0,
1953 sub get_percentage ($$) {
1956 return "NaN" if ($big eq 0);
1957 return sprintf("%.2f%%", $small / $big * 100);
1960 sub print_stats () {
1963 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
1964 my $outgoing_requests = $stats{requests} - $stats{crunches};
1966 if ($stats{requests} eq 0) {
1967 print "No requests yet.\n";
1971 print "Client requests total: " . $stats{requests} . "\n";
1972 print "Crunches: " . $stats{crunches} . " (" .
1973 get_percentage($stats{requests}, $stats{crunches}) . ")\n";
1974 print "Outgoing requests: " . $outgoing_requests . " (" .
1975 get_percentage($stats{requests}, $outgoing_requests) . ")\n";
1976 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
1977 get_percentage($stats{requests}, $stats{'server-keep-alive'}) . ")\n";
1978 print "New outgoing connections: " . $new_connections . " (" .
1979 get_percentage($stats{requests}, $new_connections) . ")\n";
1980 print "Reused connections: " . $stats{'reused-connections'} . " (" .
1981 get_percentage($stats{requests}, $stats{'reused-connections'}) . ")\n";
1982 print "Empty responses: " . $stats{'empty-responses'} . " (" .
1983 get_percentage($stats{requests}, $stats{'empty-responses'}) . ")\n";
1984 print "Empty responses on new connections: "
1985 . $stats{'empty-responses-on-new-connections'} . " (" .
1986 get_percentage($stats{requests}, $stats{'empty-responses-on-new-connections'})
1988 print "Empty responses on reused connections: " .
1989 $stats{'empty-responses-on-reused-connections'} . " (" .
1990 get_percentage($stats{requests}, $stats{'empty-responses-on-reused-connections'}) .
1995 ################################################################################
1996 # Functions that actually print stuff
1997 ################################################################################
1999 sub print_clf_message () {
2001 our ($ip, $timestamp, $request_line, $status_code, $size);
2004 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2006 # Rebuild highlighted
2007 $output .= $h{'Number'} . $ip . $h{'Standard'};
2009 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2011 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2013 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2015 $output .= $h{'Number'} . $size . $h{'Standard'};
2016 $output .= $line_end;
2021 sub print_non_clf_message ($) {
2023 my $content = shift;
2024 my $log_level = $req{$t}{'log-level'};
2025 my $msec_string = "." . $req{$t}{'msecs'} unless $no_msecs_mode;
2026 my $thread_color_string = $thread_colours{$t} if defined($thread_colours{$t});
2027 my $log_level_string = $h{$log_level} if defined($h{$log_level});
2029 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2032 . $time_colours[$time_colour_index % 2]
2033 . $req{$t}{'time-stamp'}
2035 . $h{Standard} . " "
2036 . $thread_color_string
2050 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2052 my $last_thread = 0;
2053 my $last_timestamp = 0;
2054 my $filters_that_did_nothing;
2057 $time_colour = paint_it('white');
2059 my %log_level_handlers = (
2060 'Re-Filter' => \&handle_loglevel_re_filter,
2061 'Header' => \&handle_loglevel_header,
2062 'Connect' => \&handle_loglevel_connect,
2063 'Redirect' => \&handle_loglevel_redirect,
2064 'Request' => \&handle_loglevel_request,
2065 'Crunch' => \&handle_loglevel_crunch,
2066 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2067 'Info' => \&handle_loglevel_info,
2068 'CGI' => \&handle_loglevel_cgi,
2069 'Force' => \&handle_loglevel_force,
2070 'Error' => \&handle_loglevel_error,
2071 'Fatal error' => \&handle_loglevel_ignore,
2072 'Writing' => \&handle_loglevel_ignore,
2073 'Unknown log level' => \&handle_loglevel_ignore,
2078 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2080 $req{$t}{'day'} = $day = $1;
2081 $req{$t}{'time-stamp'} = $time_stamp = $2;
2082 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2083 $req{$t}{'log-level'} = $log_level = $5;
2084 $req{$t}{'content'} = $content = $c = $6;
2085 $req{$t}{'log-message'} = $_;
2086 $no_special_header_highlighting = 0;
2088 if (defined($log_level_handlers{$log_level})) {
2090 $content = $log_level_handlers{$log_level}($content);
2094 die "No handler found for log level \"$log_level\"\n";
2098 # Highlight Truncations
2099 if (m/\.\.\. \[(too long, truncated)/) {
2100 $content =~ s@($1)@$h{'Truncation'}$1$h{'Standard'}@g;
2103 next unless $content;
2105 # Register threads to keep the colour constant
2106 if (!defined($thread_colours{$thread})) {
2107 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2108 $thread_colour_index++;
2111 # Switch timestamp colour if timestamps differ
2112 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2113 debug_message("Tick tack!") if DEBUG_TICKS;
2114 $time_colour = $time_colours[$time_colour_index % 2];
2115 $time_colour_index++;
2116 $last_msecs = $msecs;
2117 $last_timestamp = $time_stamp;
2120 $last_thread = $thread;
2122 print_non_clf_message($content);
2124 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2126 # LOG_LEVEL_CLF lines look like this
2127 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2128 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2129 # HTTP/1.1" 403 1730
2130 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2132 print_clf_message();
2136 # Some Privoxy log messages span more than one line,
2137 # usually to dump lots of content that doesn't need any syntax highlighting.
2138 # XXX: add mechanism to forward these lines to the right handler anyway.
2140 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2141 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2149 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2150 my %log_level_handlers = (
2151 'Re-Filter' => \&handle_loglevel_ignore,
2152 'Header' => \&gather_loglevel_header_stats,
2153 'Connect' => \&gather_loglevel_connect_stats,
2154 'Redirect' => \&handle_loglevel_ignore,
2155 'Request' => \&gather_loglevel_request_stats,
2156 'Crunch' => \&gather_loglevel_crunch_stats,
2157 'Gif-Deanimate' => \&handle_loglevel_ignore,
2158 'Info' => \&handle_loglevel_ignore,
2159 'CGI' => \&handle_loglevel_ignore,
2160 'Force' => \&handle_loglevel_ignore,
2161 'Error' => \&gather_loglevel_error_stats,
2162 'Fatal error' => \&handle_loglevel_ignore,
2163 'Writing' => \&handle_loglevel_ignore,
2164 'Unknown log level' => \&handle_loglevel_ignore
2168 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2171 $msecs = $3 ? $3 : 0;
2176 if (defined($log_level_handlers{$log_level})) {
2178 $content = $log_level_handlers{$log_level}($content, $thread);
2182 die "No handler found for log level \"$log_level\"\n";
2192 sub VersionMessage {
2193 my $version_message;
2195 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2196 $version_message .= 'Copyright (C) 2007-2009 Fabian Keil <fk@fabiankeil.de>' . "\n";
2197 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2199 print $version_message;
2202 sub get_cli_options () {
2204 our %cli_options = (
2205 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2206 'title' => CLI_OPTION_TITLE,
2207 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2208 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2209 'no-msecs' => CLI_OPTION_NO_MSECS,
2210 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2211 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
2212 'statistics' => CLI_OPTION_STATISTICS,
2216 'html-output' => \$cli_options{'html-output'},
2217 'title' => \$cli_options{'title'},
2218 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2219 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2220 'no-msecs' => \$cli_options{'no-msecs'},
2221 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2222 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
2223 'statistics' => \$cli_options{'statistics'},
2224 'version' => sub { VersionMessage && exit(0) },
2228 $html_output_mode = cli_option_is_set('html-output');
2229 $no_msecs_mode = cli_option_is_set('no-msecs');
2230 $line_end = get_line_end();
2241 Options and their default values if they have any:
2242 [--accept-unknown-messages]
2246 [--no-syntax-highlighting]
2247 [--show-ineffective-filters]
2249 [--title $cli_options{'title'}]
2251 see "perldoc $0" for more information
2257 ################################################################################
2259 ################################################################################
2263 set_background(DEFAULT_BACKGROUND);
2264 prepare_our_stuff();
2268 if (cli_option_is_set('statistics')) {
2281 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2285 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
2286 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--show-ineffective-filters>]
2291 B<privoxy-log-parser> reads Privoxy log messages and
2293 - syntax-highlights recognized lines,
2295 - reformats some of them for easier comprehension,
2297 - filters out less useful messages, and
2299 - (in some cases) calculates additional information,
2300 like the compression ratio or how a filter affected
2303 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2304 without getting confused by the resulting amount of output. For example for
2305 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2306 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2307 will hide the "filter foo caused 0 hits" message.
2311 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
2312 just don't highlight them.
2314 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2315 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2316 This option is only intended to make embedding log excerpts in web pages easier.
2317 It does not escape any input!
2319 [B<--no-msecs>] Don't expect milisecond resolution
2321 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2322 the filtered output is piped into less in which case the ANSI control
2323 codes don't work, or if the terminal itself doesn't support the control
2326 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2327 that didn't modify the content.
2329 [B<--statistics>] Gather various statistics instead of syntax highlighting
2330 log messages. This is an experimental feature, if the results look wrong
2331 they very well might be. Also note that the results a pretty much guaranteed
2332 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2334 [B<--version>] Print version and exit.
2338 To monitor a log file:
2340 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2342 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2343 or if the log won't get rotated anyway. The log file location depends
2344 on your system (Doh!).
2346 To monitor Privoxy without having it write to a log file:
2348 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2350 Again, the config file location depends on your system. Output redirection
2351 depends on your shell, the above works with bourne shells.
2353 To read a processed Privoxy log file from top to bottom, letting the content
2354 scroll by slightly faster than you can read:
2356 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2358 This is probably only useful to fill screens in the background of haxor movies.
2362 Syntax highlighting with ANSI escape sequences will look strange
2363 if your background color isn't black.
2365 Some messages aren't recognized yet and will not be fully highlighted.
2367 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2368 using earlier Privoxy versions will probably result in an increased amount
2369 of unrecognized log lines.
2371 Privoxy's log files tend to be rather large. If you use HTML
2372 highlighting some browsers can't handle them, get confused and
2373 will eventually crash because of segmentation faults or unexpected
2374 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2379 Many settings can't be controlled through command line options yet.
2387 Fabian Keil <fk@fabiankeil.de>