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 1.102 2010/11/08 17:53:14 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
26 # - Add --compress and --decompress options.
28 # Copyright (c) 2007-2010 Fabian Keil <fk@fabiankeil.de>
30 # Permission to use, copy, modify, and distribute this software for any
31 # purpose with or without fee is hereby granted, provided that the above
32 # copyright notice and this permission notice appear in all copies.
34 # THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
35 # WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
36 # MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
37 # ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
38 # WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
39 # ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
40 # OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
41 ################################################################################
48 PRIVOXY_LOG_PARSER_VERSION => '0.6',
49 # Feel free to mess with these ...
50 DEFAULT_BACKGROUND => 'black', # Choose registered colour (like 'black')
51 DEFAULT_TEXT_COLOUR => 'white', # Choose registered colour (like 'black')
52 HEADER_DEFAULT_COLOUR => 'yellow',
53 REGISTER_HEADERS_WITH_THE_SAME_COLOUR => 1,
55 CLI_OPTION_DEFAULT_TO_HTML_OUTPUT => 0,
56 CLI_OPTION_TITLE => 'Privoxy-Log-Parser in da house',
57 CLI_OPTION_NO_EMBEDDED_CSS => 0,
58 CLI_OPTION_NO_MSECS => 0,
59 CLI_OPTION_NO_SYNTAX_HIGHLIGHTING => 0,
60 CLI_OPTION_SHORTEN_THREAD_IDS => 0,
61 CLI_OPTION_SHOW_INEFFECTIVE_FILTERS => 0,
62 CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES => 0,
63 CLI_OPTION_STATISTICS => 0,
64 CLI_OPTION_URL_STATISTICS_THRESHOLD => 0,
65 CLI_OPTION_HOST_STATISTICS_THRESHOLD => 0,
67 SUPPRESS_SUCCEEDED_FILTER_ADDITIONS => 1,
69 SHOW_FILTER_READIN_IN => 0,
70 SUPPRESS_EMPTY_LINES => 1,
71 SUPPRESS_SUCCESSFUL_CONNECTIONS => 1,
72 SUPPRESS_ACCEPTED_CONNECTIONS => 1,
73 SUPPRESS_GIF_NOT_CHANGED => 1,
74 SUPPRESS_NEED_TO_DE_CHUNK_FIRST => 1,
76 DEBUG_HEADER_REGISTERING => 0,
77 DEBUG_HEADER_HIGHLIGHTING => 0,
80 DEBUG_SUPPRESS_LOG_MESSAGES => 0,
82 PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH => 0,
83 PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH => 1,
85 LOG_UNPARSED_LINES_TO_EXTRA_FILE => 0,
86 ERROR_LOG_FILE => '/var/log/privoxy-log-parser',
88 # You better leave these alone unless you know what you're doing.
89 COLOUR_RESET => "\033[0;0m",
93 # For performance reasons, these are global.
96 my %req; # request data from previous lines
101 my $thread_colour_index = 0;
102 my $header_colour_index = 0;
103 my $time_colour_index = 0;
105 my $no_special_header_highlighting;
108 my $header_highlight_regex = '';
110 my $html_output_mode;
111 my $no_msecs_mode; # XXX: should probably be removed
112 my $shorten_thread_ids;
115 sub prepare_our_stuff () {
117 # Syntax Higlight hash
119 'red', 'green', 'brown', 'blue', 'purple', 'cyan',
120 'light_gray', 'light_red', 'light_green', 'yellow',
121 'light_blue', 'pink', 'light_cyan', 'white'
128 Filter => 'purple', # XXX: Used?
129 'Re-Filter' => 'purple',
131 Request => 'light_cyan',
132 CGI => 'light_green',
134 Error => 'light_red',
136 'Fatal error' => 'light_red',
137 'Gif-Deanimate' => 'blue',
139 Writing => 'light_green',
140 Received => 'yellow',
141 # ----------------------
144 request_ => 'brown', # host+path but no protocol
145 'ip-address' => 'yellow',
148 Truncation => 'light_red',
150 Timestamp => 'brown',
151 Crunching => 'light_red',
152 crunched => 'light_red',
153 'Request-Line' => 'pink',
155 destination => 'yellow',
156 'http-version' => 'pink',
157 'crunch-pattern' => 'pink',
162 'program-name' => 'cyan',
165 warning => 'light_red',
166 debug => 'light_red',
170 'status-message' => 'light_cyan',
171 'status-code' => 'yellow',
172 'invalid-request' => 'light_red',
174 error => 'light_red',
175 'rewritten-URL' => 'light_red',
176 'pcrs-delimiter' => 'light_red',
177 'ignored' => 'light_red',
178 'action-bits-update' => 'light_red',
179 'configuration-line' => 'red',
180 'content-type' => 'yellow',
181 'HOST' => HEADER_DEFAULT_COLOUR,
186 # Header colours need their own hash so the keys can be accessed properly
188 # Prefilled with headers that should not appear with default header colours
189 Cookie => 'light_red',
190 'Set-Cookie' => 'light_red',
191 Warning => 'light_red',
192 Default => HEADER_DEFAULT_COLOUR,
195 # Crunch reasons need their own hash as well
197 'Unsupported HTTP feature' => 'light_red',
198 Blocked => 'light_red',
199 Untrusted => 'light_red',
200 Redirected => 'green',
201 'CGI Call' => 'white',
202 'DNS failure' => 'red',
203 'Forwarding failed' => 'light_red',
204 'Connection failure' => 'light_red',
205 'Out of memory (may mask other reasons)' => 'light_red',
206 'No reason recorded' => 'light_red',
209 @time_colours = ('white', 'light_gray');
211 # Translate highlight strings into highlight code
212 prepare_highlight_hash(\%header_colours);
213 prepare_highlight_hash(\%reason_colours);
214 prepare_highlight_hash(\%h);
215 prepare_colour_array(\@all_colours);
216 prepare_colour_array(\@time_colours);
223 ###############################################################
224 # Takes a colour string and returns an ANSI escape sequence
225 # (unless --no-syntax-highlighting is used).
226 # XXX: The Rolling Stones reference has to go.
227 ###############################################################
229 my $colour = shift @_;
231 return "" if cli_option_is_set('no-syntax-highlighting');
273 my $bg_code = get_background();
275 our $default = default_colours();
277 if (defined($text{$colour})) {
278 $colour_code = ESCAPE;
279 $colour_code .= $text{$colour};
281 $colour_code .= $light{$colour} ? "1" : "2";
283 $colour_code .= $bg_code;
285 debug_message $colour . " is \'" . $colour_code . $colour . $default . "\'" if DEBUG_PAINT_IT;
287 } elsif ($colour =~ /reset/) {
289 $colour_code = default_colours();
293 die "What's $colour supposed to mean?\n";
299 sub get_semantic_html_markup ($) {
300 ###############################################################
301 # Takes a string and returns a span element
302 ###############################################################
307 if ($type =~ /Standard/) {
311 $code = '<span title="' . $type . '" class="' . $type . '">';
317 sub cli_option_is_set ($) {
320 my $cli_option = shift;
322 die "Unknown CLI option: $cli_option" unless defined $cli_options{$cli_option};
324 return $cli_options{$cli_option};
327 sub get_html_title () {
330 return $cli_options{'title'};
334 sub init_css_colours() {
342 purple => "F06", # XXX: wrong
343 cyan => "F09", # XXX: wrong
348 light_green => "33F",
357 sub get_css_colour ($) {
362 die "What's $colour supposed to mean?\n" unless defined($css_colours{$colour});
364 return '#' . $css_colours{$colour};
367 sub get_css_line ($) {
372 $css_line .= '.' . lc($class) . ' {'; # XXX: lc() shouldn't be necessary
373 die "What's $class supposed to mean?\n" unless defined($h_colours{$class});
374 $css_line .= 'color:' . get_css_colour($h_colours{$class}) . ';';
375 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
376 $css_line .= '}' . "\n";
381 sub get_css_line_for_colour ($) {
386 $css_line .= '.' . lc($colour) . ' {'; # XXX: lc() shouldn't be necessary
387 $css_line .= 'color:' . get_css_colour($colour) . ';';
388 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
389 $css_line .= '}' . "\n";
394 # XXX: Wrong solution
395 sub get_missing_css_lines () {
399 $css_line .= '.' . 'default' . ' {';
400 $css_line .= 'color:' . HEADER_DEFAULT_COLOUR . ';';
401 $css_line .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
402 $css_line .= '}' . "\n";
409 our %css_colours; #XXX: Wrong solution
413 $css .= '.privoxy-log {';
414 $css .= 'color:' . get_css_colour(DEFAULT_TEXT_COLOUR) . ';';
415 $css .= 'background-color:' . get_css_colour(DEFAULT_BACKGROUND) . ';';
418 foreach my $key (keys %h_colours) {
420 next if ($h_colours{$key} =~ m/reset/); #XXX: Wrong solution.
421 $css .= get_css_line($key);
425 foreach my $colour (keys %css_colours) {
427 $css .= get_css_line_for_colour($colour);
431 $css .= get_missing_css_lines(); #XXX: Wrong solution
440 if (cli_option_is_set('html-output')) {
442 my $title = get_html_title();
444 $intro .= '<html><head>';
445 $intro .= '<title>' . $title . '</title>';
446 $intro .= '<style>' . get_css() . '</style>' unless cli_option_is_set('no-embedded-css');
447 $intro .= '</head><body>';
448 $intro .= '<h1>' . $title . '</h1><p class="privoxy-log">';
458 if (cli_option_is_set('html-output')) {
460 $outro = '</p></body></html>';
466 sub get_line_end () {
467 return cli_option_is_set('html-output') ? "<br>\n" : "\n";
470 sub get_colour_html_markup ($) {
471 ###############################################################
472 # Takes a colour string a span element. XXX: WHAT?
473 # XXX: This function shouldn't be necessary, the
474 # markup should always be semantically correct.
475 ###############################################################
480 if ($type =~ /Standard/) {
483 $code = '<span class="' . lc($type) . '">';
489 sub default_colours () {
492 return reset_colours();
495 sub show_colours () {
499 sub reset_colours () {
500 return ESCAPE . "0m";
503 sub set_background ($){
519 if (defined($backgrounds{$colour})) {
520 $bg_code = $backgrounds{$colour};
522 die "Invalid background colour: " . $colour;
526 sub get_background (){
530 sub prepare_highlight_hash ($) {
533 foreach my $key (keys %$ref) {
534 $$ref{$key} = $html_output_mode ?
535 get_semantic_html_markup($key) :
536 paint_it($$ref{$key});
540 sub prepare_colour_array ($) {
543 foreach my $i (0 ... @$ref - 1) {
544 $$ref[$i] = $html_output_mode ?
545 get_colour_html_markup($$ref[$i]) :
550 sub found_unknown_content ($) {
555 return if cli_option_is_set('accept-unknown-messages');
557 return if ($unknown =~ /\[too long, truncated\]$/);
559 $message = "found_unknown_content: Don't know how to highlight: ";
560 # Break line so the log file can later be parsed as Privoxy log file again
561 $message .= '"' . $unknown . '"' . " in:\n";
562 $message .= $req{$t}{'log-message'};
563 debug_message($message);
564 log_parse_error($req{$t}{'log-message'});
566 die "Unworthy content parser" if PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH;
569 sub log_parse_error ($) {
573 if (LOG_UNPARSED_LINES_TO_EXTRA_FILE) {
574 open(ERRORLOG, ">>" . ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
575 print ERRORLOG $message;
580 sub debug_message (@) {
583 print $h{'debug'} . "@message" . $h{'Standard'} . "\n";
586 ################################################################################
587 # highlighter functions that aren't loglevel-specific
588 ################################################################################
592 # Get highlight marker
593 my $highlight = shift; # XXX: Stupid name;
597 if (defined($highlight)) {
599 $result = $h{$highlight};
603 $message = "h: Don't recognize highlighter $highlight.";
604 debug_message($message);
605 log_parser_error($message);
606 die "Unworthy highlighter function" if PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH;
612 sub highlight_known_headers ($) {
616 debug_message("Searching $content for things to highlight.") if DEBUG_HEADER_HIGHLIGHTING;
618 if ($content =~ m/(?<=\s)($header_highlight_regex):/) {
620 $content =~ s@(?<=[\s|'])($header)(?=:)@$header_colours{$header}$1$h{'Standard'}@ig;
621 debug_message("Highlighted '$header' in '$content'") if DEBUG_HEADER_HIGHLIGHTING;
627 sub highlight_matched_request_line ($$) {
629 my $result = shift; # XXX: Stupid name;
631 if ($result =~ m@(.*)($regex)(.*)@) {
632 $result = $1 . highlight_request_line($2) . $3
637 sub highlight_request_line ($) {
640 my ($method, $url, $http_version);
642 #GET http://images.sourceforge.net/sfx/icon_warning.gif HTTP/1.1
643 if ($rl =~ m/Invalid request/) {
645 $rl = h('invalid-request') . $rl . h('Standard');
647 } elsif ($rl =~ m/^([-\w]+) (.*) (HTTP\/\d\.\d)/) {
649 # XXX: might not match in case of HTTP method fuzzing.
650 # XXX: save these: ($method, $path, $http_version) = ($1, $2, $3);
651 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
652 if ($rl =~ /http:\/\//) {
653 $rl = highlight_matched_url($rl, '[^\s]*(?=\sHTTP)');
655 $rl = highlight_matched_pattern($rl, 'request_', '[^\s]*(?=\sHTTP)');
658 $rl =~ s@(HTTP\/\d\.\d)$@$h{'http-version'}$1$h{'Standard'}@;
660 } elsif ($rl =~ m/\.\.\. \[too long, truncated\]$/) {
662 $rl =~ s@^(\w+)@$h{'method'}$1$h{'Standard'}@;
663 $rl = highlight_matched_url($rl, '[^\s]*(?=\.\.\.)');
665 } elsif ($rl =~ m/^ $/) {
667 $rl = h('error') . "No request line specified!" . h('Standard');
671 debug_message ("Can't parse request line: $rl");
678 sub highlight_response_line ($) {
681 my ($http_version, $status_code, $status_message);
686 # TODO: Mark different status codes differently
688 if ($rl =~ m/((?:HTTP\/\d\.\d|ICY)) (\d+) (.*)/) {
689 ($http_version, $status_code, $status_message) = ($1, $2, $3);
691 debug_message ("Can't parse response line: $rl") and die 'Fix this';
694 # Rebuild highlighted
696 $rl .= h('http-version') . $http_version . h('Standard');
698 $rl .= h('status-code') . $status_code . h('Standard');
700 $rl .= h('status-message') . $status_message . h('Standard');
705 sub highlight_matched_url ($$) {
707 my $result = shift; # XXX: Stupid name;
710 #print "Got $result, regex ($regex)\n";
712 if ($result =~ m@(.*?)($regex)(.*)@) {
713 $result = $1 . highlight_url($2) . $3;
714 #print "Now the result is $result\n";
720 sub highlight_matched_host ($$) {
722 my ($result, $regex) = @_; # XXX: result ist stupid name;
724 if ($result =~ m@(.*?)($regex)(.*)@) {
725 $result = $1 . $h{host} . $2 . $h{Standard} . $3;
731 sub highlight_matched_pattern ($$$) {
733 my $result = shift; # XXX: Stupid name;
737 die "Unknown key $key" unless defined $h{$key};
739 if ($result =~ m@(.*?)($regex)(.*)@) {
740 $result = $1 . h($key) . $2 . h('Standard') . $3;
746 sub highlight_matched_path ($$) {
748 my $result = shift; # XXX: Stupid name;
751 if ($result =~ m@(.*?)($regex)(.*)@) {
752 $result = $1 . h('path') . $2 . h('Standard') . $3;
758 sub highlight_url ($) {
762 if ($html_output_mode) {
764 $url = '<a href="' . $url . '">' . $url . '</a>';
768 $url = h('URL') . $url . h('Standard');
775 sub update_header_highlight_regex ($) {
778 my $headers = join ('|', keys %header_colours);
780 $header_highlight_regex = qr/$headers/;
781 print "Registering '$header'\n" if DEBUG_HEADER_HIGHLIGHTING;
784 ################################################################################
785 # loglevel-specific highlighter functions
786 ################################################################################
788 sub handle_loglevel_header ($) {
792 if ($c =~ /^scan:/) {
794 if ($c =~ m/^scan: ([^: ]+):/) {
796 # Register new headers
797 # scan: Accept: image/png,image/*;q=0.8,*/*;q=0.5
799 if (!defined($header_colours{$header}) and $header =~ /^[\d\w-]*$/) {
800 debug_message "Registering previously unknown header $1" if DEBUG_HEADER_REGISTERING;
802 if (REGISTER_HEADERS_WITH_THE_SAME_COLOUR) {
803 $header_colours{$header} = $header_colours{'Default'};
805 $header_colours{$header} = $all_colours[$header_colour_index % @all_colours];
806 $header_colour_index++;
808 update_header_highlight_regex($header);
811 } elsif ($c =~ m/^(scan: )(\w+ .+ HTTP\/\d\.\d)/) {
813 # scan: GET http://p.p/ HTTP/1.1
814 $c = $1 . highlight_request_line($2);
816 } elsif ($c =~ m/^(scan: )((?:HTTP\/\d\.\d|ICY) (\d+) (.*))/) {
818 # scan: HTTP/1.1 200 OK
819 $req{$t}{'response_line'} = $2;
820 $req{$t}{'status_code'} = $3;
821 $req{$t}{'status_message'} = $4;
822 $c = $1 . highlight_response_line($req{$t}{'response_line'});
825 } elsif ($c =~ m/^Crunching (?:server|client) header: .* \(contains: ([^\)]*)\)/) {
827 # Crunching server header: Set-Cookie: trac_form_token=d5308c34e16d15e9e301a456; (contains: Cookie:)
828 $c =~ s@(?<=contains: )($1)@$h{'crunch-pattern'}$1$h{'Standard'}@;
829 $c =~ s@(Crunching)@$h{$1}$1$h{'Standard'}@;
831 } elsif ($c =~ m/^New host is: ([^\s]*)\./) {
833 # New host is: trac.vidalia-project.net. Crunching Referer: http://www.vidalia-project.net/!
834 $c = highlight_matched_host($c, '(?<=New host is: )[^\s]+(?=\.)');
835 $c = highlight_matched_url($c, '(?<=Crunching Referer: )[^\s!]+');
837 } elsif ($c =~ m/^Text mode enabled by force. (Take cover)!/) {
839 # Text mode enabled by force. Take cover!
840 $c =~ s@($1)@$h{'warning'}$1$h{'Standard'}@;
842 } elsif ($c =~ m/^(New HTTP Request-Line: )(.*)/) {
844 # New HTTP Request-Line: GET http://www.privoxy.org/ HTTP/1.1
845 $c = $1 . highlight_request_line($2);
847 } elsif ($c =~ m/^Adjust(ed)? Content-Length to \d+/) {
849 # Adjusted Content-Length to 2132
850 # Adjust Content-Length to 33533
851 $c =~ s@(?<=Content-Length to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
852 $c = highlight_known_headers($c);
854 } elsif ($c =~ m/^Destination extracted from "Host:" header. New request URL:/) {
856 # Destination extracted from "Host:" header. New request URL: http://www.cccmz.de/~ridcully/blog/
857 $c = highlight_matched_url($c, '(?<=New request URL: ).*');
859 } elsif ($c =~ m/^Couldn\'t parse:/) {
861 # XXX: These should probable be logged with LOG_LEVEL_ERROR
862 # Couldn't parse: If-Modified-Since: Wed, 21 Mar 2007 16:34:50 GMT (crunching!)
863 # Couldn't parse: at, 24 Mar 2007 13:46:21 GMT in If-Modified-Since: Sat, 24 Mar 2007 13:46:21 GMT (crunching!)
864 $c =~ s@^(Couldn\'t parse)@$h{'error'}$1$h{'Standard'}@;
866 } elsif ($c =~ /^Tagger \'([^\']*)\' added tag \'([^\']*)\'/ or
867 $c =~ m/^Adding tag \'([^\']*)\' created by header tagger \'([^\']*)\'/) {
869 # Adding tag 'GET request' created by header tagger 'method-man' (XXX: no longer used)
870 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. No action bit update necessary.
871 # Tagger 'revalidation' added tag 'REVALIDATION-REQUEST'. Action bits updated accordingly.
873 # XXX: Save tag and tagger
875 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
876 $c =~ s@(?<=added tag \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
877 $c =~ s@(?<=Action bits )(updated)@$h{'action-bits-update'}$1$h{'Standard'}@;
878 $no_special_header_highlighting = 1;
880 } elsif ($c =~ /^Tagger \'([^\']*)\' didn['']t add tag \'([^\']*)\'/) {
882 # Tagger 'revalidation' didn't add tag 'REVALIDATION-REQUEST'. Tag already present
883 # XXX: Save tag and tagger
885 $c =~ s@(?<=^Tagger \')([^\']*)@$h{'tag'}$1$h{'Standard'}@;
886 $c =~ s@(?<=didn['']t add tag \')([^\']*)@$h{'tagger'}$1$h{'Standard'}@;
888 } elsif ($c =~ m/^(?:scan:|Randomiz|addh:|Adding:|Removing:|Referer:|Modified:|Accept-Language header|[Cc]ookie)/
889 or $c =~ m/^(Text mode is already enabled|Denied request with NULL byte|Replaced:|add-unique:)/
890 or $c =~ m/^(Crunched (incoming|outgoing) cookie|Suppressed offer|Accepted the client)/
891 or $c =~ m/^(addh-unique|Referer forged to)/
892 or $c =~ m/^Downgraded answer to HTTP\/1.0/
893 or $c =~ m/^Parameter: \+hide-referrer\{[^\}]*\} is a bad idea, but I don\'t care./
894 or $c =~ m/^Referer (?:overwritten|replaced) with: Referer: / #XXX: should this be highlighted?
895 or $c =~ m/^Referer crunched!/
896 or $c =~ m/^crunched x-forwarded-for!/
897 or $c =~ m/^crunched From!/
898 or $c =~ m/^ modified$/
899 or $c =~ m/^Content filtering is enabled. Crunching:/
900 or $c =~ m/^force-text-mode overruled the client/
901 or $c =~ m/^Server time in the future\./
902 or $c =~ m/^content-disposition header crunched and replaced with:/i
903 or $c =~ m/^Reducing white space in /
904 or $c =~ m/^Ignoring single quote in /
905 or $c =~ m/^Converting tab to space in /
906 or $c =~ m/A HTTP\/1\.1 response without/
907 or $c =~ m/Disabled filter mode on behalf of the client/
908 or $c =~ m/Keeping the (?:server|client) header /
909 or $c =~ m/Content modified with no Content-Length header set/
910 or $c =~ m/^Appended client IP address to/
911 or $c =~ m/^Removing 'Connection: close' to imply keep-alive./
912 or $c =~ m/^keep-alive support is disabled/
913 or $c =~ m/^Continue hack in da house/
914 or $c =~ m/^Merged multiple header lines to:/
917 # XXX: Some of these may need highlighting
919 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
920 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
921 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
922 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
923 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
924 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
925 # Text mode is already enabled
926 # Denied request with NULL byte(s) turned into line break(s)
927 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
928 # addh-unique: Host: people.freebsd.org
929 # Suppressed offer to compress content
930 # Crunched incoming cookie -- yum!
931 # Accepted the client's request to fetch without filtering.
932 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
933 # addh-unique: Host: subkeys.pgp.net:11371
934 # Referer forged to: Referer: http://10.0.0.1/
935 # Downgraded answer to HTTP/1.0
936 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
937 # Referer overwritten with: Referer: pille-palle
938 # Referer replaced with: Referer: pille-palle
939 # crunched x-forwarded-for!
941 # modified # XXX: pretty stupid log message
942 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
943 # force-text-mode overruled the client's request to fetch without filtering!
944 # Server time in the future.
945 # content-disposition header crunched and replaced with: content-disposition: filename=baz
946 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
947 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
948 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
949 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
951 # A HTTP/1.1 response without Connection header implies keep-alive.
952 # Disabled filter mode on behalf of the client.
953 # Keeping the server header 'Connection: keep-alive' around.
954 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
955 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
956 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
957 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
958 # Removing 'Connection: close' to imply keep-alive.
959 # keep-alive support is disabled. Crunching: Keep-Alive: 300.
960 # Continue hack in da house.
961 # Merged multiple header lines to: 'X-FORWARDED-PROTO: http X-HOST: 127.0.0.1'
963 } elsif ($c =~ m/^scanning headers for:/) {
965 return '' unless SHOW_SCAN_INTRO;
967 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
968 # crunched User-Agent!
969 # Crunching: Content-Encoding: gzip
971 $c =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
973 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
975 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
977 $c = h('warning') . $c . h('Standard');
979 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
981 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
982 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
984 $c =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
985 $c =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
987 } elsif ($c =~ m/^Removing empty header/) {
989 # Removing empty header
992 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
994 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
995 # Enable force-text-mode if you know what you're doing.
996 # XXX: Could highlight more here.
997 $c =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
999 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
1001 # Server keep-alive timeout is 5. Sticking with 10.
1002 # Client keep-alive timeout is 20. Sticking with 10.
1004 $c =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1005 $c =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1007 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1009 # Reducing keep-alive timeout from 60 to 10.
1011 $c =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1012 $c =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1014 } elsif ($c =~ m/^Killed all-caps Host header line: HOST:/) {
1016 # Killed all-caps Host header line: HOST: bestproxydb.com
1017 $c = highlight_matched_host($c, '(?<=HOST: )[^\s]+');
1018 $c = highlight_matched_pattern($c, 'HOST', 'HOST');
1022 found_unknown_content($c);
1026 unless ($c =~ m/^Transforming/) {
1027 $c = highlight_known_headers($c) unless $no_special_header_highlighting;
1033 sub handle_loglevel_re_filter ($) {
1035 my $content = shift;
1039 if ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1041 # XXX: only the second version gets highlighted properly.
1042 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1043 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1044 $req{$t}{'content_source'} = $1;
1045 $req{$t}{'content_size'} = $2;
1046 $req{$t}{'content_filter'} = $3;
1047 $req{$t}{'content_hits'} = $4;
1048 $req{$t}{'new_content_size'} = $5;
1049 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1050 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1051 if ($req{$t}{'content_hits'} == 0 and
1052 not (cli_option_is_set('show-ineffective-filters')
1053 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1057 $c =~ s@(?<=\(size )(\d+)\)(?= with)@$h{'Number'}$1$h{'Standard'}@;
1058 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1059 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1061 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1062 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1065 $c .= "(" . $h{'Number'};
1066 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1067 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1070 } elsif ($c =~ /\.{3}$/
1071 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1073 # Used by Privoxy 3.0.5 and 3.0.6:
1075 # Used by Privoxy 3.0.7:
1076 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1078 $req{$t}{'filtered_header'} = $1;
1079 $req{$t}{'old_header_size'} = $2;
1080 $req{$t}{'header_filter_name'} = $3;
1082 unless (cli_option_is_set('show-ineffective-filters') or
1083 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1086 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1087 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1089 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1091 # ...produced 0 hits (new size 23).
1092 #... produced 1 hits (new size 54).
1094 $req{$t}{'header_filter_hits'} = $1;
1095 $req{$t}{'new_header_size'} = $2;
1097 unless (cli_option_is_set('show-ineffective-filters') or
1098 (defined($req{$t}{'header_filter_name'}) and
1099 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1101 if ($req{$t}{'header_filter_hits'} == 0 and
1102 not (defined($req{$t}{'header_filter_name'}) and
1103 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1106 # Reformat including information from the intro
1107 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1109 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1110 # $c .= $req{$t}{'filtered_header'} . " ";
1111 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1112 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1114 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1116 $c .= "changing size from ";
1117 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1119 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1124 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1128 # Highlight from last line (XXX: What?)
1129 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1130 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1135 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1136 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1141 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1143 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1144 # Tagger variable-test has empty joblist. Nothing to do.
1146 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1148 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1150 $req{$t}{'chunked-size'} = $1;
1151 $req{$t}{'dechunked-size'} = $2;
1152 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1154 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1156 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1157 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1159 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1161 # Decompression successful. Old size: 670, new size: 1166.
1163 $req{$t}{'size-compressed'} = $1;
1164 $req{$t}{'size-decompressed'} = $2;
1165 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1167 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1168 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1170 # XXX: Create sub get_percentage()
1171 if ($req{$t}{'size-decompressed'}) {
1172 $req{$t}{'decompression-gain-percent'} =
1173 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1175 $content .= " (saved: ";
1176 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1178 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1182 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1184 # Need to de-chunk first
1185 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1187 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1189 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1191 # Adding re_filter job ...
1192 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1193 # to filter client-ip-address succeeded.
1195 } elsif ($c =~ m/^Reading in filter/) {
1197 return '' unless SHOW_FILTER_READIN_IN;
1201 found_unknown_content($content);
1208 sub handle_loglevel_redirect ($) {
1212 if ($c =~ m/^Decoding "([^""]*)"/) {
1214 $req{$t}{'original-destination'} = $1;
1215 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1218 } elsif ($c =~ m/^Checking/) {
1220 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1221 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1223 # TODO: Change colour if really url-decoded
1224 $req{$t}{'decoded-original-destination'} = $1;
1225 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1228 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1230 # pcrs command "s@&from=rss@@" changed \
1231 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1232 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1233 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1234 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1235 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1236 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1237 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1238 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1240 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1242 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1243 # change "http://www.example.org/".
1244 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1245 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1247 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1249 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1250 # XXX: Use URL highlighter
1252 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1254 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1255 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1256 # is already properly formatted.
1257 # XXX: assume the same?
1258 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1262 found_unknown_content($c);
1269 sub handle_loglevel_gif_deanimate ($) {
1271 my $content = shift;
1273 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1275 my $bytes_from = $1;
1277 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1278 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1279 # XXX: Do we need g in case of ($1 == $2)?
1280 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1282 } elsif ($content =~ m/GIF (not) changed/) {
1284 # Gif-Deanimate: GIF not changed.
1285 return '' if SUPPRESS_GIF_NOT_CHANGED;
1286 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1288 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1290 # failed! (gif parsing)
1291 # XXX: Replace this error message with something less stupid
1292 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1294 } elsif ($content =~ m/^Need to de-chunk first/) {
1296 # Need to de-chunk first
1297 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1299 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1301 # No GIF header found (XXX: Did I ever commit this?)
1302 # failed while parsing 195 134747048 (XXX: never commited)
1304 # Ignore these for now
1308 found_unknown_content($content);
1315 sub handle_loglevel_request ($) {
1317 my $content = shift;
1319 if ($content =~ m/crunch! /) {
1321 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1323 # Highlight crunch reasons
1324 foreach my $reason (keys %reason_colours) {
1325 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1327 # Highlight request URL domain and ditch 'crunch!'
1328 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1329 $content =~ s@ crunch!@@;
1331 } elsif ($content =~ m/\[too long, truncated\]$/) {
1333 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1334 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1336 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1338 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1339 $content = h('request_') . $content . h('Standard');
1343 found_unknown_content($content);
1350 sub handle_loglevel_crunch ($) {
1352 my $content = shift;
1354 # Highlight crunch reason
1355 foreach my $reason (keys %reason_colours) {
1356 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1359 if ($content =~ m/\[too long, truncated\]$/) {
1361 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1362 # [...]&filter... [too long, truncated]
1363 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1367 # Blocked: http://ads.example.org/
1368 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1374 sub handle_loglevel_connect ($) {
1378 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1380 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1382 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1383 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1385 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1387 # connect to: www.example.org.noconnect failed: Operation not permitted
1389 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1391 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1393 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1395 # Connect: to www.nzherald.co.nz successful
1396 # Connect: to archiv.radiotux.de
1398 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1399 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1401 } elsif ($c =~ m/^to ([^\s]*)$/) {
1403 # Connect: to lists.sourceforge.net:443
1405 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1407 } elsif ($c =~ m/^accepted connection from .*/ or
1410 # accepted connection from 10.0.0.1
1411 # Privoxy 3.0.6 and earlier just say:
1413 return '' if SUPPRESS_ACCEPTED_CONNECTIONS;
1414 $c = highlight_matched_host($c, '(?<=connection from ).*');
1416 } elsif ($c =~ m/^write header to: .* failed:/) {
1418 # write header to: 10.0.0.1 failed: Broken pipe
1420 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1421 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1423 } elsif ($c =~ m/^write header to client failed:/) {
1425 # write header to client failed: Broken pipe
1427 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1429 } elsif ($c =~ m/^socks4_connect:/) {
1431 # socks4_connect: SOCKS request rejected or failed.
1432 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1434 } elsif ($c =~ m/^Listening for new connections/ or
1435 $c =~ m/^accept connection/) {
1437 # Privoxy versions above 3.0.6 say:
1438 # Listening for new connections ...
1439 # earlier versions say:
1440 # accept connection ...
1443 } elsif ($c =~ m/^accept failed:/) {
1445 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1447 } elsif ($c =~ m/^Overriding forwarding settings/) {
1449 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1450 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1452 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1454 # Denying suspicious CONNECT request from 10.0.0.1
1455 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1457 } elsif ($c =~ m/^socks5_connect:/) {
1459 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1461 } elsif ($c =~ m/^Created new connection to/) {
1463 # Created new connection to www.privoxy.org:80 on socket 11.
1464 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1465 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1467 } elsif ($c =~ m/^Found reusable socket/) {
1469 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1471 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1472 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1473 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1474 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1475 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1476 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1477 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1478 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1480 } elsif ($c =~ m/^Marking open socket/) {
1482 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1483 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1484 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1485 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1487 } elsif ($c =~ m/^No reusable/) {
1489 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1490 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1492 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1494 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1495 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1497 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1498 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1499 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1501 } elsif ($c =~ m/^Socket/) {
1503 # Socket 16 already forgotten or never remembered.
1504 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1506 } elsif ($c =~ m/^The connection to/) {
1508 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1510 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1511 # Timeout is: 1. Assumed latency: 4.
1512 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1513 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1514 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1515 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1516 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1517 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1519 } elsif ($c =~ m/^Stopped waiting for the request line./) {
1521 # Stopped waiting for the request line. Timeout: 121.
1522 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1524 } elsif ($c =~ m/^Waiting for \d/) {
1526 # Waiting for 1 connections to timeout.
1527 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1529 } elsif ($c =~ m/^Initialized/) {
1531 # Initialized 20 socket slots.
1532 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1534 } elsif ($c =~ m/^Done reading from server/) {
1536 # Done reading from server. Expected content length: 24892. \
1537 # Actual content length: 24892. Most recently received: 4412.
1539 # Done reading from server. Expected content length: 24892. \
1540 # Actual content length: 24892. Bytes most recently read: 4412.
1541 # Done reading from server. Content length: 6018 as expected. \
1542 # Bytes most recently read: 294.
1543 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1544 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1545 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1547 } elsif ($c =~ m/^Continuing buffering headers/) {
1549 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1550 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1551 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1552 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1554 # Continuing buffering headers. Bytes most recently read: %d.
1555 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1557 } elsif ($c =~ m/^Received \d+ bytes while/) {
1559 # Received 206 bytes while expecting 12103.
1560 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1561 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1563 } elsif ($c =~ m/^(Rejecting c|C)onnection from/) {
1565 # Connection from 81.163.28.218 dropped due to ACL
1566 # Rejecting connection from 178.63.152.227. Maximum number of connections reached.
1567 $c =~ s@(?<=onnection from )((?:\d+\.?){3}\d+)@$h{'Number'}$1$h{'Standard'}@;
1569 } elsif ($c =~ m/^(?:Reusing|Closing) server socket \d./ or
1570 $c =~ m/^No additional client request/) {
1572 # Reusing server socket 4. Opened for 10.0.0.1.
1573 # Closing server socket 2. Opened for 10.0.0.1.
1574 # No additional client request received in time. \
1575 # Closing server socket 4, initially opened for 10.0.0.1.
1576 # No additional client request received in time on socket 29.
1578 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1579 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.$)');
1581 } elsif ($c =~ m/^Connected to /) {
1583 # Connected to tor-jail[10.0.0.2]:9050.
1585 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1586 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1587 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1589 } elsif ($c =~ m/^Could not connect to /) {
1591 # Could not connect to [10.0.0.1]:80.
1593 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1594 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1596 } elsif ($c =~ m/^Waiting for the next client request/ or
1597 $c =~ m/^The connection on server socket/ or
1598 $c =~ m/^Client request arrived in time or the client closed the connection/) {
1600 # Waiting for the next client request on socket 3. Keeping the server \
1601 # socket 12 to a.fsdn.com open.
1602 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1603 # Client request arrived in time or the client closed the connection on socket 12.
1605 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1606 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1607 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1609 } elsif ($c =~ m/^Marking the server socket/) {
1611 # Marking the server socket 7 tainted.
1613 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1615 } elsif ($c =~ m/^Reduced expected bytes to /) {
1617 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1618 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1619 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1621 } elsif ($c =~ m/^The client closed socket /) {
1623 # The client closed socket 2 while the server socket 4 is still open.
1624 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1625 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1627 } elsif ($c =~ m/^Expected client content length set /) {
1629 # Expected client content length set to 667325411 after reading 4999 bytes.
1630 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1631 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1633 } elsif ($c =~ m/^Waiting for up to /) {
1635 # Waiting for up to 4999 bytes from the client.
1636 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1638 } elsif ($c =~ m/^Looks like we / or
1639 $c =~ m/^Unsetting keep-alive flag/ or
1640 $c =~ m/^No connections to wait/ or
1641 $c =~ m/^Complete client request received/ or
1642 $c =~ m/^Possible pipeline attempt detected./ or
1643 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1644 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1645 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1646 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1647 $c =~ m/^Stopping to watch the client socket. There's already another request waiting./ or
1648 $c =~ m/^Done reading from the client\.$/) {
1650 # Looks like we reached the end of the last chunk. We better stop reading.
1651 # Looks like we read the end of the last chunk together with the server \
1652 # headers. We better stop reading.
1653 # Looks like we got the last chunk together with the server headers. \
1654 # We better stop reading.
1655 # Unsetting keep-alive flag.
1656 # No connections to wait for left.
1657 # Client request arrived in time or the client closed the connection.
1658 # Complete client request received
1659 # Possible pipeline attempt detected. The connection will not be \
1660 # kept alive and we will only serve the first request.
1661 # POST request detected. The connection will not be kept alive.
1662 # The server still wants to talk, but the client hung up on us.
1663 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1664 # There might be a request body. The connection will not be kept alive.
1665 # Stopping to watch the client socket. There's already another request waiting.
1666 # Done reading from the client\.
1670 found_unknown_content($c);
1678 sub handle_loglevel_info ($) {
1682 if ($c =~ m/^Rewrite detected:/) {
1684 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1685 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1687 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1688 $c =~ m/^Compressed content detected/ or
1691 # Decompressing deflated iob: 117
1692 # Decompression didn't result in any content.
1693 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1694 # with zlib support or enable the prevent-compression action.
1695 # Tagger 'complete-url' created empty tag. Ignored.
1699 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1701 # loading configuration file '/usr/local/etc/privoxy/config':
1702 # Reloading configuration file '/usr/local/etc/privoxy/config'
1703 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1705 } elsif ($c =~ m/^Loading (actions|filter) file: /) {
1707 # Loading actions file: /usr/local/etc/privoxy/default.action
1708 # Loading filter file: /usr/local/etc/privoxy/default.filter
1709 $c =~ s@(?<= file: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1711 } elsif ($c =~ m/^exiting by signal/) {
1713 # exiting by signal 15 .. bye
1714 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1716 } elsif ($c =~ m/^Privoxy version/) {
1718 # Privoxy version 3.0.7
1719 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1721 } elsif ($c =~ m/^Program name: /) {
1723 # Program name: /usr/local/sbin/privoxy
1724 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1726 } elsif ($c =~ m/^Listening on port /) {
1728 # Listening on port 8118 on IP address 10.0.0.1
1729 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1730 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1732 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1734 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1735 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1737 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1739 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1740 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1741 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1743 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1744 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1745 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1746 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1748 } elsif ($c =~ m/^Status code/) {
1750 # Status code 304 implies no body.
1751 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1753 } elsif ($c =~ m/^Method/) {
1755 # Method HEAD implies no body.
1756 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1758 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1760 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1761 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1762 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1764 } elsif ($c =~ m/^No logfile configured/ or
1765 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1766 $c =~ m/^Invalid \"chunked\" transfer/ or
1767 $c =~ m/^Support for/ or
1768 $c =~ m/^Flushing header and buffers/ or
1769 $c =~ m/^Can not resolve/
1772 # No logfile configured. Please enable it before reporting any problems.
1773 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1774 # response or even no response at all.
1775 # No logfile configured. Logging disabled.
1776 # Invalid "chunked" transfer encoding detected and ignored.
1777 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1778 # known not to work properly in some situations.
1779 # Flushing header and buffers. Stepping back from filtering.
1780 # Can not resolve doesnotexist: hostname nor servname provided, or not known
1784 found_unknown_content($c);
1791 sub handle_loglevel_cgi ($) {
1795 if ($c =~ m/^Granting access to/) {
1797 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1799 } elsif ($c =~ m/^Substituting: s(.)/) {
1801 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1802 # XXX: prone to span several lines
1805 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1806 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1807 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1813 sub handle_loglevel_force ($) {
1817 if ($c =~ m/^Ignored force prefix in request:/) {
1819 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1820 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1821 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1823 } elsif ($c =~ m/^Enforcing request:/) {
1825 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1826 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1830 found_unknown_content($c);
1837 sub handle_loglevel_error ($) {
1841 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1843 # Empty server or forwarder response received on socket 4.
1844 # Empty server or forwarder response received on socket 3. \
1845 # Closing client socket 15 without sending data.
1846 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1847 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1849 # XXX: There are probably more messages that deserve highlighting.
1855 sub handle_loglevel_ignore ($) {
1859 sub gather_loglevel_request_stats ($$) {
1867 sub gather_loglevel_crunch_stats ($$) {
1875 if ($c =~ m/^Redirected:/) {
1876 # Redirected: http://www.example.org/http://p.p/
1877 $stats{'fast-redirections'}++;
1879 } elsif ($c =~ m/^Blocked:/) {
1880 # Blocked: blogger.googleusercontent.com:443
1881 $stats{'blocked'}++;
1886 sub gather_loglevel_error_stats ($$) {
1893 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
1895 # Empty server or forwarder response received on socket 4.
1896 $stats{'empty-responses'}++;
1897 if ($thread_data{$thread}{'new_connection'}) {
1898 $stats{'empty-responses-on-new-connections'}++;
1900 $stats{'empty-responses-on-reused-connections'}++;
1905 sub gather_loglevel_connect_stats ($$) {
1907 my ($c, $thread) = @_;
1911 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
1913 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1914 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
1916 } elsif ($c =~ m/^to ([^\s]*)$/) {
1918 # Connect: to lists.sourceforge.net:443
1920 $thread_data{$thread}{'forwarder'} = 'direct connection';
1922 } elsif ($c =~ m/^Created new connection to/) {
1924 # Created new connection to www.privoxy.org:80 on socket 11.
1926 $thread_data{$thread}{'new_connection'} = 1;
1928 } elsif ($c =~ m/^Reusing server socket \d./ or
1929 $c =~ m/^Found reusable socket/) {
1931 # Reusing server socket 4. Opened for 10.0.0.1.
1932 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1934 $thread_data{$thread}{'new_connection'} = 0;
1935 $stats{'reused-connections'}++;
1939 sub gather_loglevel_header_stats ($$) {
1941 my ($c, $thread) = @_;
1944 if ($c =~ m/^A HTTP\/1\.1 response without/ or
1945 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
1947 # A HTTP/1.1 response without Connection header implies keep-alive.
1948 # Keeping the server header 'Connection: keep-alive' around.
1949 $stats{'server-keep-alive'}++;
1951 } elsif ($c =~ m/^scan: ((\w+) (.+) (HTTP\/\d\.\d))/) {
1953 # scan: HTTP/1.1 200 OK
1954 $stats{'method'}{$2}++;
1955 $stats{'ressource'}{$3}++;
1956 $stats{'http-version'}{$4}++;
1958 } elsif ($c =~ m/^scan: Host: ([^\s]+)/) {
1961 $stats{'hosts'}{$1}++;
1969 'server-keep-alive' => 0,
1970 'reused-connections' => 0,
1971 'empty-responses' => 0,
1972 'empty-responses-on-new-connections' => 0,
1973 'empty-responses-on-reused-connections' => 0,
1974 'fast-redirections' => 0,
1976 'reused-connections' => 0,
1977 'server-keep-alive' => 0,
1981 sub get_percentage ($$) {
1984 return "NaN" if ($big eq 0);
1985 return sprintf("%.2f%%", $small / $big * 100);
1988 sub print_stats () {
1992 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
1993 my $outgoing_requests = $stats{requests} - $stats{crunches};
1995 if ($stats{requests} eq 0) {
1996 print "No requests yet.\n";
2000 print "Client requests total: " . $stats{requests} . "\n";
2001 print "Crunches: " . $stats{crunches} . " (" .
2002 get_percentage($stats{requests}, $stats{crunches}) . ")\n";
2003 print "Blocks: " . $stats{'blocked'} . " (" .
2004 get_percentage($stats{requests}, $stats{'blocked'}) . ")\n";
2005 print "Fast redirections: " . $stats{'fast-redirections'} . " (" .
2006 get_percentage($stats{requests}, $stats{'fast-redirections'}) . ")\n";
2007 print "Outgoing requests: " . $outgoing_requests . " (" .
2008 get_percentage($stats{requests}, $outgoing_requests) . ")\n";
2009 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
2010 get_percentage($stats{requests}, $stats{'server-keep-alive'}) . ")\n";
2011 print "New outgoing connections: " . $new_connections . " (" .
2012 get_percentage($stats{requests}, $new_connections) . ")\n";
2013 print "Reused connections: " . $stats{'reused-connections'} . " (" .
2014 get_percentage($stats{requests}, $stats{'reused-connections'}) .
2015 "; server offers accepted: " .
2016 get_percentage($stats{'server-keep-alive'}, $stats{'reused-connections'}) . ")\n";
2017 print "Empty responses: " . $stats{'empty-responses'} . " (" .
2018 get_percentage($stats{requests}, $stats{'empty-responses'}) . ")\n";
2019 print "Empty responses on new connections: "
2020 . $stats{'empty-responses-on-new-connections'} . " (" .
2021 get_percentage($stats{requests}, $stats{'empty-responses-on-new-connections'})
2023 print "Empty responses on reused connections: " .
2024 $stats{'empty-responses-on-reused-connections'} . " (" .
2025 get_percentage($stats{requests}, $stats{'empty-responses-on-reused-connections'}) .
2028 if ($stats{method} eq 0) {
2029 print "No response lines parsed yet yet.\n";
2032 print "Method distribution:\n";
2033 foreach my $method (sort {$stats{'method'}{$b} <=> $stats{'method'}{$a}} keys %{$stats{'method'}}) {
2034 printf "%8d : %-8s\n", $stats{'method'}{$method}, $method;
2036 print "Client HTTP versions:\n";
2037 foreach my $http_version (sort {$stats{'http-version'}{$b} <=> $stats{'http-version'}{$a}} keys %{$stats{'http-version'}}) {
2038 printf "%d : %s\n", $stats{'http-version'}{$http_version}, $http_version;
2041 if ($cli_options{'url-statistics-threshold'} == 0) {
2042 print "URL statistics are disabled. Increase --url-statistics-threshold to enable them.\n";
2044 print "Requested URLs:\n";
2045 foreach my $ressource (sort {$stats{'ressource'}{$b} <=> $stats{'ressource'}{$a}} keys %{$stats{'ressource'}}) {
2046 if ($stats{'ressource'}{$ressource} < $cli_options{'url-statistics-threshold'}) {
2047 print "Skipped statistics for URLs below the treshold.\n";
2050 printf "%d : %s\n", $stats{'ressource'}{$ressource}, $ressource;
2054 if ($cli_options{'host-statistics-threshold'} == 0) {
2055 print "Host statistics are disabled. Increase --host-statistics-threshold to enable them.\n";
2057 print "Requested Hosts:\n";
2058 foreach my $host (sort {$stats{'hosts'}{$b} <=> $stats{'hosts'}{$a}} keys %{$stats{'hosts'}}) {
2059 if ($stats{'hosts'}{$host} < $cli_options{'host-statistics-threshold'}) {
2060 print "Skipped statistics for Hosts below the treshold.\n";
2063 printf "%d : %s\n", $stats{'hosts'}{$host}, $host;
2069 ################################################################################
2070 # Functions that actually print stuff
2071 ################################################################################
2073 sub print_clf_message () {
2075 our ($ip, $timestamp, $request_line, $status_code, $size);
2078 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2080 # Rebuild highlighted
2081 $output .= $h{'Number'} . $ip . $h{'Standard'};
2083 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2085 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2087 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2089 $output .= $h{'Number'} . $size . $h{'Standard'};
2090 $output .= $line_end;
2095 sub print_non_clf_message ($) {
2097 my $content = shift;
2098 my $msec_string = "." . $req{$t}{'msecs'} unless $no_msecs_mode;
2099 my $line_start = $html_output_mode ? '' : $h{"Standard"};
2101 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2104 . $time_colours[$time_colour_index % 2]
2105 . $req{$t}{'time-stamp'}
2107 . $h{Standard} . " "
2108 . $thread_colours{$t}
2112 . $h{$req{$t}{'log-level'}}
2113 . $req{$t}{'log-level'}
2120 sub shorten_thread_id ($) {
2122 my $thread_id = shift;
2124 our %short_thread_ids;
2127 unless (defined $short_thread_ids{$thread_id}) {
2128 $short_thread_ids{$thread_id} = sprintf "%.3d", $max_threadid++;
2131 return $short_thread_ids{$thread_id}
2136 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2138 my $last_thread = 0;
2139 my $last_timestamp = 0;
2140 my $filters_that_did_nothing;
2143 $time_colour = paint_it('white');
2145 my %log_level_handlers = (
2146 'Re-Filter' => \&handle_loglevel_re_filter,
2147 'Header' => \&handle_loglevel_header,
2148 'Connect' => \&handle_loglevel_connect,
2149 'Redirect' => \&handle_loglevel_redirect,
2150 'Request' => \&handle_loglevel_request,
2151 'Crunch' => \&handle_loglevel_crunch,
2152 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2153 'Info' => \&handle_loglevel_info,
2154 'CGI' => \&handle_loglevel_cgi,
2155 'Force' => \&handle_loglevel_force,
2156 'Error' => \&handle_loglevel_error,
2157 'Fatal error' => \&handle_loglevel_ignore,
2158 'Writing' => \&handle_loglevel_ignore,
2159 'Received' => \&handle_loglevel_ignore,
2160 'Unknown log level' => \&handle_loglevel_ignore,
2165 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2166 $thread = $t = ($shorten_thread_ids) ? shorten_thread_id($4) : $4;
2167 $req{$t}{'day'} = $day = $1;
2168 $req{$t}{'time-stamp'} = $time_stamp = $2;
2169 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2170 $req{$t}{'log-level'} = $log_level = $5;
2171 $req{$t}{'content'} = $content = $c = $6;
2172 $req{$t}{'log-message'} = $_;
2173 $no_special_header_highlighting = 0;
2175 if (defined($log_level_handlers{$log_level})) {
2177 $content = $log_level_handlers{$log_level}($content);
2181 die "No handler found for log level \"$log_level\"\n";
2184 # Highlight Truncations
2185 if (length($_) > 4000) {
2186 $content =~ s@(too long, truncated)]$@$h{'Truncation'}$1$h{'Standard'}]@g;
2189 next unless $content;
2191 # Register threads to keep the colour constant
2192 if (!defined($thread_colours{$thread})) {
2193 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2194 $thread_colour_index++;
2197 # Switch timestamp colour if timestamps differ
2198 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2199 debug_message("Tick tack!") if DEBUG_TICKS;
2200 $time_colour = $time_colours[$time_colour_index % 2];
2201 $time_colour_index++;
2202 $last_msecs = $msecs;
2203 $last_timestamp = $time_stamp;
2206 $last_thread = $thread;
2208 print_non_clf_message($content);
2210 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2212 # LOG_LEVEL_CLF lines look like this
2213 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2214 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2215 # HTTP/1.1" 403 1730
2216 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2218 print_clf_message();
2222 # Some Privoxy log messages span more than one line,
2223 # usually to dump lots of content that doesn't need any syntax highlighting.
2224 # XXX: add mechanism to forward these lines to the right handler anyway.
2226 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2227 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2235 my ($day, $time_stamp, $msecs, $thread, $log_level, $content);
2236 my %log_level_handlers = (
2237 'Re-Filter' => \&handle_loglevel_ignore,
2238 'Header' => \&gather_loglevel_header_stats,
2239 'Connect' => \&gather_loglevel_connect_stats,
2240 'Redirect' => \&handle_loglevel_ignore,
2241 'Request' => \&gather_loglevel_request_stats,
2242 'Crunch' => \&gather_loglevel_crunch_stats,
2243 'Gif-Deanimate' => \&handle_loglevel_ignore,
2244 'Info' => \&handle_loglevel_ignore,
2245 'CGI' => \&handle_loglevel_ignore,
2246 'Force' => \&handle_loglevel_ignore,
2247 'Error' => \&gather_loglevel_error_stats,
2248 'Fatal error' => \&handle_loglevel_ignore,
2249 'Writing' => \&handle_loglevel_ignore,
2250 'Received' => \&handle_loglevel_ignore,
2251 'Unknown log level' => \&handle_loglevel_ignore
2255 if (m/^(\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2258 $msecs = $3 ? $3 : 0;
2263 if (defined($log_level_handlers{$log_level})) {
2265 $content = $log_level_handlers{$log_level}($content, $thread);
2269 die "No handler found for log level \"$log_level\"\n";
2279 sub VersionMessage {
2280 my $version_message;
2282 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2283 $version_message .= 'Copyright (C) 2007-2010 Fabian Keil <fk@fabiankeil.de>' . "\n";
2284 $version_message .= 'http://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2286 print $version_message;
2289 sub get_cli_options () {
2291 our %cli_options = (
2292 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2293 'title' => CLI_OPTION_TITLE,
2294 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2295 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2296 'no-msecs' => CLI_OPTION_NO_MSECS,
2297 'shorten-thread-ids' => CLI_OPTION_SHORTEN_THREAD_IDS,
2298 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2299 'accept-unknown-messages' => CLI_OPTION_ACCEPT_UNKNOWN_MESSAGES,
2300 'statistics' => CLI_OPTION_STATISTICS,
2301 'url-statistics-threshold' => CLI_OPTION_URL_STATISTICS_THRESHOLD,
2302 'host-statistics-threshold'=> CLI_OPTION_HOST_STATISTICS_THRESHOLD,
2306 'html-output' => \$cli_options{'html-output'},
2307 'title' => \$cli_options{'title'},
2308 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2309 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2310 'no-msecs' => \$cli_options{'no-msecs'},
2311 'shorten-thread-ids' => \$cli_options{'shorten-thread-ids'},
2312 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2313 'accept-unknown-messages' => \$cli_options{'accept-unknown-messages'},
2314 'statistics' => \$cli_options{'statistics'},
2315 'url-statistics-threshold=s'=> \$cli_options{'url-statistics-threshold'},
2316 'host-statistics-threshold=s'=> \$cli_options{'host-statistics-threshold'},
2317 'version' => sub { VersionMessage && exit(0) },
2321 $html_output_mode = cli_option_is_set('html-output');
2322 $no_msecs_mode = cli_option_is_set('no-msecs');
2323 $shorten_thread_ids = cli_option_is_set('shorten-thread-ids');
2324 $line_end = get_line_end();
2335 Options and their default values if they have any:
2336 [--accept-unknown-messages]
2337 [--host-statistics-threshold $cli_options{'host-statistics-threshold'}]
2341 [--no-syntax-highlighting]
2342 [--shorten-thread-ids]
2343 [--show-ineffective-filters]
2345 [--url-statistics-threshold $cli_options{'url-statistics-threshold'}]
2346 [--title $cli_options{'title'}]
2348 see "perldoc $0" for more information
2354 ################################################################################
2356 ################################################################################
2360 set_background(DEFAULT_BACKGROUND);
2361 prepare_our_stuff();
2365 if (cli_option_is_set('statistics')) {
2378 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2382 B<privoxy-log-parser> [B<--accept-unknown-messages>] [B<--html-output>]
2383 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--statistics>]
2384 [B<--shorten-thread-ids>] [B<--show-ineffective-filters>]
2385 [B<--url-statistics-threshold>] [B<--version>]
2389 B<privoxy-log-parser> reads Privoxy log messages and
2391 - syntax-highlights recognized lines,
2393 - reformats some of them for easier comprehension,
2395 - filters out less useful messages, and
2397 - (in some cases) calculates additional information,
2398 like the compression ratio or how a filter affected
2401 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2402 without getting confused by the resulting amount of output. For example for
2403 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2404 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2405 will hide the "filter foo caused 0 hits" message.
2409 [B<--accept-unknown-messages>] Don't print warnings in case of unknown messages,
2410 just don't highlight them.
2412 [B<--host-statistics-threshold>] Only show the request count for a host
2413 if it's above or equal to the given threshold. If the threshold is 0, host
2414 statistics are disabled.
2416 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2417 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2418 This option is only intended to make embedding log excerpts in web pages easier.
2419 It does not escape any input!
2421 [B<--no-msecs>] Don't expect milisecond resolution
2423 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2424 the filtered output is piped into less in which case the ANSI control
2425 codes don't work, or if the terminal itself doesn't support the control
2428 [B<--shorten-thread-ids>] Shorten the thread ids to a three-digit decimal number.
2429 Note that the mapping from thread ids to shortened ids is created at run-time
2430 and thus varies with the input.
2432 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2433 that didn't modify the content.
2435 [B<--statistics>] Gather various statistics instead of syntax highlighting
2436 log messages. This is an experimental feature, if the results look wrong
2437 they very well might be. Also note that the results are pretty much guaranteed
2438 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2440 [B<--url-statistics-threshold>] Only show the request count for a ressource
2441 if it's above or equal to the given threshold. If the threshold is 0, URL
2442 statistics are disabled.
2444 [B<--version>] Print version and exit.
2448 To monitor a log file:
2450 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2452 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2453 or if the log won't get rotated anyway. The log file location depends
2454 on your system (Doh!).
2456 To monitor Privoxy without having it write to a log file:
2458 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2460 Again, the config file location depends on your system. Output redirection
2461 depends on your shell, the above works with bourne shells.
2463 To read a processed Privoxy log file from top to bottom, letting the content
2464 scroll by slightly faster than you can read:
2466 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2468 This is probably only useful to fill screens in the background of haxor movies.
2472 Syntax highlighting with ANSI escape sequences will look strange
2473 if your background color isn't black.
2475 Some messages aren't recognized yet and will not be fully highlighted.
2477 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2478 using earlier Privoxy versions will probably result in an increased amount
2479 of unrecognized log lines.
2481 Privoxy's log files tend to be rather large. If you use HTML
2482 highlighting some browsers can't handle them, get confused and
2483 will eventually crash because of segmentation faults or unexpected
2484 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2489 Many settings can't be controlled through command line options yet.
2497 Fabian Keil <fk@fabiankeil.de>