3 ################################################################################
6 # A parser for Privoxy log messages. For incomplete documentation run
7 # perldoc privoxy-log-parser(.pl), for fancy screenshots see:
9 # https://www.fabiankeil.de/sourcecode/privoxy-log-parser/
12 # - LOG_LEVEL_CGI, LOG_LEVEL_ERROR, LOG_LEVEL_WRITE content highlighting
13 # - create fancy statistics
14 # - grep through Privoxy sources to find unsupported log messages
15 # - hunt down substitutions that match content from variables which
16 # can contain stuff like ()?'[]
17 # - replace $h{'foo'} with h('foo') where possible
18 # - hunt down XXX comments instead of just creating them
19 # - add example log lines for every regex and mark them up for
21 # - Handle incomplete input without Perl warning about undefined variables.
22 # - Use generic highlighting function that takes a regex and the
24 # - Add --compress and --decompress options.
26 # Copyright (c) 2007-2020 Fabian Keil <fk@fabiankeil.de>
28 # Permission to use, copy, modify, and distribute this software for any
29 # purpose with or without fee is hereby granted, provided that the above
30 # copyright notice and this permission notice appear in all copies.
32 # THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
33 # WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
34 # MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
35 # ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
36 # WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
37 # ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
38 # OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
39 ################################################################################
46 PRIVOXY_LOG_PARSER_VERSION => '0.9.1',
47 # Feel free to mess with these ...
48 DEFAULT_BACKGROUND => 'black', # Choose registered colour (like 'black')
49 DEFAULT_TEXT_COLOUR => 'white', # Choose registered colour (like 'black')
50 HEADER_DEFAULT_COLOUR => 'yellow',
51 REGISTER_HEADERS_WITH_THE_SAME_COLOUR => 1,
53 CLI_OPTION_DEFAULT_TO_HTML_OUTPUT => 0,
54 CLI_OPTION_TITLE => 'Privoxy-Log-Parser in da house',
55 CLI_OPTION_NO_EMBEDDED_CSS => 0,
56 CLI_OPTION_NO_MSECS => 0,
57 CLI_OPTION_NO_SYNTAX_HIGHLIGHTING => 0,
58 CLI_OPTION_SHORTEN_THREAD_IDS => 0,
59 CLI_OPTION_SHOW_INEFFECTIVE_FILTERS => 0,
60 CLI_OPTION_STATISTICS => 0,
61 CLI_OPTION_STRICT_CHECKS => 0,
62 CLI_OPTION_UNBREAK_LINES_ONLY => 0,
63 CLI_OPTION_URL_STATISTICS_THRESHOLD => 0,
64 CLI_OPTION_HOST_STATISTICS_THRESHOLD => 0,
65 CLI_OPTION_SHOW_COMPLETE_REQUEST_DISTRIBUTION => 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_GIF_NOT_CHANGED => 1,
73 SUPPRESS_NEED_TO_DE_CHUNK_FIRST => 1,
75 DEBUG_HEADER_REGISTERING => 0,
76 DEBUG_HEADER_HIGHLIGHTING => 0,
79 DEBUG_SUPPRESS_LOG_MESSAGES => 0,
81 PUNISH_MISSING_LOG_KNOWLEDGE_WITH_DEATH => 0,
82 PUNISH_MISSING_HIGHLIGHT_KNOWLEDGE_WITH_DEATH => 1,
84 LOG_UNPARSED_LINES_TO_EXTRA_FILE => 0,
85 ERROR_LOG_FILE => '/var/log/privoxy-log-parser',
87 # You better leave these alone unless you know what you're doing.
88 COLOUR_RESET => "\033[0;0m",
92 # For performance reasons, these are global.
95 my %req; # request data from previous lines
100 my $thread_colour_index = 0;
101 my $header_colour_index = 0;
102 my $time_colour_index = 0;
104 my $no_special_header_highlighting;
107 my $header_highlight_regex = '';
109 my $html_output_mode;
110 my $no_msecs_mode; # XXX: should probably be removed
111 my $shorten_thread_ids;
114 sub prepare_our_stuff() {
116 # Syntax Higlight hash
118 'red', 'green', 'brown', 'blue', 'purple', 'cyan',
119 'light_gray', 'light_red', 'light_green', 'yellow',
120 'light_blue', 'pink', 'light_cyan', 'white'
127 Filter => 'purple', # XXX: Used?
128 'Re-Filter' => 'purple',
130 Request => 'light_cyan',
131 CGI => 'light_green',
133 Error => 'light_red',
135 'Fatal error' => 'light_red',
136 'Gif-Deanimate' => 'blue',
138 Writing => 'light_green',
139 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 ###############################################################
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 $outro = '</p></body></html>';
467 return cli_option_is_set('html-output') ? "<br>\n" : "\n";
470 sub get_colour_html_markup($) {
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();
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 unless cli_option_is_set('strict-checks');
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(my $errorlog_fd, ">>", ERROR_LOG_FILE) || die "Writing " . ERROR_LOG_FILE . " failed";
575 print $errorlog_fd $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:/
915 or $c =~ m/^Added header: /
916 or $c =~ m/^Enlisting (?:sorted|left-over) header/
917 or $c =~ m/^Multiple Content-Type headers detected. Removing and ignoring: Content-Type:/
920 # XXX: Some of these may need highlighting
922 # Modified: User-Agent: Mozilla/5.0 (X11; U; SunOS i86pc; pl-PL; rv:1.8.1.1) Gecko/20070214 Firefox/2.0.0.1
923 # Accept-Language header crunched and replaced with: Accept-Language: pl-pl
924 # cookie 'Set-Cookie: eZSessionCookie=07bfec287c197440d299f81580593c3d; \
925 # expires=Thursday, 12-Apr-07 15:16:18 GMT; path=/' send by \
926 # http://wirres.net/article/articleview/4265/1/6/ appears to be using time format 1 (XXX: gone with the wind)
927 # Cookie rewritten to a temporary one: Set-Cookie: NSC_gffe-iuuq-mc-wtfswfs=8efb33a53660;path=/
928 # Text mode is already enabled
929 # Denied request with NULL byte(s) turned into line break(s)
930 # Replaced: 'Connection: Yo, home to Bel Air' with 'Connection: close'
931 # addh-unique: Host: people.freebsd.org
932 # Suppressed offer to compress content
933 # Crunched incoming cookie -- yum!
934 # Accepted the client's request to fetch without filtering.
935 # Crunched outgoing cookie: Cookie: PREF=ID=6cf0abd347b30262:TM=1173357617:LM=1173357617:S=jZypyyJ7LPiwFi1_
936 # addh-unique: Host: subkeys.pgp.net:11371
937 # Referer forged to: Referer: http://10.0.0.1/
938 # Downgraded answer to HTTP/1.0
939 # Parameter: +hide-referrer{pille-palle} is a bad idea, but I don't care.
940 # Referer overwritten with: Referer: pille-palle
941 # Referer replaced with: Referer: pille-palle
942 # crunched x-forwarded-for!
944 # modified # XXX: pretty stupid log message
945 # Content filtering is enabled. Crunching: 'Range: 1234-5678' to prevent range-mismatch problems
946 # force-text-mode overruled the client's request to fetch without filtering!
947 # Server time in the future.
948 # content-disposition header crunched and replaced with: content-disposition: filename=baz
949 # Content-Disposition header crunched and replaced with: content-disposition: filename=baz
950 # Reducing white space in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
951 # Ignoring single quote in 'X-LWS-Test: "This is quoted" this is not "this is " but " this again is not'
952 # Converting tab to space in 'X-LWS-Test: "This is quoted" this is not "this is " but "\
954 # A HTTP/1.1 response without Connection header implies keep-alive.
955 # Disabled filter mode on behalf of the client.
956 # Keeping the server header 'Connection: keep-alive' around.
957 # Keeping the client header 'Connection: close' around. The connection will not be kept alive.
958 # Keeping the client header 'Connection: keep-alive' around. The connection will be kept alive if possible.
959 # Content modified with no Content-Length header set. Creating a fake one for adjustment later on.
960 # Appended client IP address to X-Forwarded-For: 10.0.0.2, 10.0.0.1
961 # Removing 'Connection: close' to imply keep-alive.
962 # keep-alive support is disabled. Crunching: Keep-Alive: 300.
963 # Continue hack in da house.
964 # Merged multiple header lines to: 'X-FORWARDED-PROTO: http X-HOST: 127.0.0.1'
965 # Added header: Content-Encoding: deflate
966 # Enlisting sorted header User-Agent: Mozilla/5.0 (X11; SunOS i86pc; rv:10.0.3) Gecko/20100101 Firefox/10.0.3
967 # Enlisting left-over header Connection: close
968 # Multiple Content-Type headers detected. Removing and ignoring: Content-Type: text/html
970 } elsif ($c =~ m/^scanning headers for:/) {
972 return '' unless SHOW_SCAN_INTRO;
974 } elsif ($c =~ m/^[Cc]runch(ing|ed)|crumble crunched:/) {
975 # crunched User-Agent!
976 # Crunching: Content-Encoding: gzip
978 $c =~ s@(Crunching|crunched)@$h{$1}$1$h{'Standard'}@;
980 } elsif ($c =~ m/^Offending request data with NULL bytes turned into \'°\' characters:/) {
982 # Offending request data with NULL bytes turned into '°' characters: °°n°°(°°°
984 $c = h('warning') . $c . h('Standard');
986 } elsif ($c =~ m/^(Transforming \")(.*?)(\" to \")(.*?)(\")/) {
988 # Transforming "Proxy-Authenticate: Basic realm="Correos Proxy Server"" to\
989 # "Proxy-Authenticate: Basic realm="Correos Proxy Server""
991 $c =~ s@(?<=^Transforming \")(.*)(?=\" to)@$h{'Header'}$1$h{'Standard'}@;
992 $c =~ s@(?<=to \")(.*)(?=\")@$h{'Header'}$1$h{'Standard'}@;
994 } elsif ($c =~ m/^Removing empty header/) {
996 # Removing empty header
999 } elsif ($c =~ m/^Content-Type: .* not replaced/) {
1001 # Content-Type: application/octet-stream not replaced. It doesn't look like text.\
1002 # Enable force-text-mode if you know what you're doing.
1003 # XXX: Could highlight more here.
1004 $c =~ s@(?<=^Content-Type: )(.*)(?= not replaced)@$h{'content-type'}$1$h{'Standard'}@;
1006 } elsif ($c =~ m/^(Server|Client) keep-alive timeout is/) {
1008 # Server keep-alive timeout is 5. Sticking with 10.
1009 # Client keep-alive timeout is 20. Sticking with 10.
1011 $c =~ s@(?<=timeout is )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1012 $c =~ s@(?<=Sticking with )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1014 } elsif ($c =~ m/^Reducing keep-alive timeout/) {
1016 # Reducing keep-alive timeout from 60 to 10.
1018 $c =~ s@(?<= from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1019 $c =~ s@(?<= to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1021 } elsif ($c =~ m/^Killed all-caps Host header line: HOST:/) {
1023 # Killed all-caps Host header line: HOST: bestproxydb.com
1024 $c = highlight_matched_host($c, '(?<=HOST: )[^\s]+');
1025 $c = highlight_matched_pattern($c, 'HOST', 'HOST');
1029 found_unknown_content($c);
1033 unless ($c =~ m/^Transforming/) {
1034 $c = highlight_known_headers($c) unless $no_special_header_highlighting;
1040 sub handle_loglevel_re_filter($) {
1042 my $content = shift;
1046 if ($c =~ m/^(?:re_)?filtering ([^\s]+) \(size (\d+)\) with (?:filter )?\'?([^\s]+?)\'? produced (\d+) hits \(new size (\d+)\)/) {
1048 # XXX: only the second version gets highlighted properly.
1049 # re_filtering www.lfk.de/favicon.ico (size 209) with filter untrackable-hulk produced 0 hits (new size 209).
1050 # filtering aci.blogg.de/ (size 37988) with 'blogg.de' produced 3 hits (new size 38057)
1051 $req{$t}{'content_source'} = $1;
1052 $req{$t}{'content_size'} = $2;
1053 $req{$t}{'content_filter'} = $3;
1054 $req{$t}{'content_hits'} = $4;
1055 $req{$t}{'new_content_size'} = $5;
1056 $req{$t}{'content_size_change'} = $req{$t}{'new_content_size'} - $req{$t}{'content_size'};
1057 #return '' if ($req{$t}{'content_hits'} == 0 && !cli_option_is_set('show-ineffective-filters'));
1058 if ($req{$t}{'content_hits'} == 0 and
1059 not (cli_option_is_set('show-ineffective-filters')
1060 or ($req{$t}{'content_filter'} =~ m/^privoxy-filter-test$/))) {
1064 $c =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1065 $c =~ s@(?<=\(new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1066 $c =~ s@(?<=produced )(\d+)(?= hits)@$h{'Number'}$1$h{'Standard'}@;
1068 $c =~ s@([^\s]+?)(\'? produced)@$h{'filter'}$1$h{'Standard'}$2@;
1069 $c = highlight_matched_host($c, '(?<=filtering )[^\s]+');
1072 $c .= "(" . $h{'Number'};
1073 $c .= "+" if ($req{$t}{'content_size_change'} >= 0);
1074 $c .= $req{$t}{'content_size_change'} . $h{'Standard'} . ")";
1077 } elsif ($c =~ /\.{3}$/
1078 and $c =~ m/^(?:re_)?filtering \'?(.*?)\'? \(size (\d*)\) with (?:filter )?\'?([^\s]*?)\'? ?\.{3}$/) {
1080 # Used by Privoxy 3.0.5 and 3.0.6:
1082 # Used by Privoxy 3.0.7:
1083 # filtering 'Connection: close' (size 17) with 'generic-content-ads' ...
1085 $req{$t}{'filtered_header'} = $1;
1086 $req{$t}{'old_header_size'} = $2;
1087 $req{$t}{'header_filter_name'} = $3;
1089 unless (cli_option_is_set('show-ineffective-filters') or
1090 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/) {
1093 $content =~ s@(?<=\(size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1094 $content =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1096 } elsif ($c =~ m/^ ?\.\.\. ?produced (\d*) hits \(new size (\d*)\)\./) {
1098 # ...produced 0 hits (new size 23).
1099 #... produced 1 hits (new size 54).
1101 $req{$t}{'header_filter_hits'} = $1;
1102 $req{$t}{'new_header_size'} = $2;
1104 unless (cli_option_is_set('show-ineffective-filters') or
1105 (defined($req{$t}{'header_filter_name'}) and
1106 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1108 if ($req{$t}{'header_filter_hits'} == 0 and
1109 not (defined($req{$t}{'header_filter_name'}) and
1110 $req{$t}{'header_filter_name'} =~ m/^privoxy-filter-test$/)) {
1113 # Reformat including information from the intro
1114 $c = "'" . h('filter') . $req{$t}{'header_filter_name'} . h('Standard') . "'";
1116 # XXX: Hide behind constant, it may be interesting if LOG_LEVEL_HEADER isn't enabled as well.
1117 # $c .= $req{$t}{'filtered_header'} . " ";
1118 $c .= h('Number') . $req{$t}{'header_filter_hits'}. h('Standard');
1119 $c .= ($req{$t}{'header_filter_hits'} == 1) ? " time, " : " times, ";
1121 if ($req{$t}{'old_header_size'} != $req{$t}{'new_header_size'}) {
1123 $c .= "changing size from ";
1124 $c .= h('Number') . $req{$t}{'old_header_size'} . h('Standard');
1126 $c .= h('Number') . $req{$t}{'new_header_size'} . h('Standard');
1131 $c .= "keeping the size at " . $req{$t}{'old_header_size'};
1135 # Highlight from last line (XXX: What?)
1136 # $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1137 # $c =~ s@($req{$t}{'header_filter_name'})@$h{'filter'}$1$h{'Standard'}@;
1142 $c =~ s@(?<=produced )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1143 $c =~ s@(?<=new size )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1148 } elsif ($c =~ m/^(Tagger|Filter) ([^\s]*) has empty joblist. Nothing to do./) {
1150 # Filter privoxy-filter-test has empty joblist. Nothing to do.
1151 # Tagger variable-test has empty joblist. Nothing to do.
1153 $content =~ s@(?<=$1 )([^\s]*)@$h{'filter'}$1$h{'Standard'}@;
1155 } elsif ($c =~ m/^De-chunking successful. Shrunk from (\d+) to (\d+)/) {
1157 $req{$t}{'chunked-size'} = $1;
1158 $req{$t}{'dechunked-size'} = $2;
1159 $req{$t}{'dechunk-change'} = $req{$t}{'dechunked-size'} - $req{$t}{'chunked-size'};
1161 $content .= " (" . h('Number') . $req{$t}{'dechunk-change'} . h('Standard') . ")";
1163 $content =~ s@(?<=from )($req{$t}{'chunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1164 $content =~ s@(?<=to )($req{$t}{'dechunked-size'})@$h{'Number'}$1$h{'Standard'}@;
1166 } elsif ($c =~ m/^Decompression successful. Old size: (\d+), new size: (\d+)./) {
1168 # Decompression successful. Old size: 670, new size: 1166.
1170 $req{$t}{'size-compressed'} = $1;
1171 $req{$t}{'size-decompressed'} = $2;
1172 $req{$t}{'decompression-gain'} = $req{$t}{'size-decompressed'} - $req{$t}{'size-compressed'};
1174 $content =~ s@(?<=Old size: )($req{$t}{'size-compressed'})@$h{'Number'}$1$h{'Standard'}@;
1175 $content =~ s@(?<=new size: )($req{$t}{'size-decompressed'})@$h{'Number'}$1$h{'Standard'}@;
1177 # XXX: Create sub get_percentage()
1178 if ($req{$t}{'size-decompressed'}) {
1179 $req{$t}{'decompression-gain-percent'} =
1180 $req{$t}{'decompression-gain'} / $req{$t}{'size-decompressed'} * 100;
1182 $content .= " (saved: ";
1183 #$content .= h('Number') . $req{$t}{'decompression-gain'} . h('Standard');
1185 $content .= h('Number') . sprintf("%.2f%%", $req{$t}{'decompression-gain-percent'}) . h('Standard');
1189 } elsif ($c =~ m/^(Need to de-chunk first)/) {
1191 # Need to de-chunk first
1192 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1194 } elsif ($c =~ m/^(Adding (?:dynamic )?re_filter job)/) {
1196 return '' if (SUPPRESS_SUCCEEDED_FILTER_ADDITIONS && m/succeeded/);
1198 # Adding re_filter job ...
1199 # Adding dynamic re_filter job s@^(?:\w*)\s+.*\s+HTTP/\d\.\d\s*@IP-ADDRESS: $origin@D\
1200 # to filter client-ip-address succeeded.
1202 } elsif ($c =~ m/^Compressed content from /) {
1204 # Compressed content from 29258 to 8630 bytes. Compression level: 3
1205 $content =~ s@(?<=from )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1206 $content =~ s@(?<=to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1207 $content =~ s@(?<=level: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1209 } elsif ($c =~ m/^Reading in filter/) {
1211 return '' unless SHOW_FILTER_READIN_IN;
1213 } elsif ($c =~ m/^Decompression didn't result/) {
1215 # Decompression didn't result in any content.
1217 # Nothing to highlight.
1221 found_unknown_content($content);
1228 sub handle_loglevel_redirect($) {
1232 if ($c =~ m/^Decoding "([^""]*)"/) {
1234 $req{$t}{'original-destination'} = $1;
1235 $c = highlight_matched_path($c, '(?<=Decoding ")[^"]*');
1238 } elsif ($c =~ m/^Checking/) {
1240 # Checking /_ylt=A0geu.Z76BRGR9k/**http://search.yahoo.com/search?p=view+odb+presentation+on+freebsd\
1241 # &ei=UTF-8&xargs=0&pstart=1&fr=moz2&b=11 for redirects.
1243 # TODO: Change colour if really url-decoded
1244 $req{$t}{'decoded-original-destination'} = $1;
1245 $c = highlight_matched_path($c, '(?<=Checking ")[^"]*');
1248 } elsif ($c =~ m/^pcrs command "([^""]*)" changed /) {
1250 # pcrs command "s@&from=rss@@" changed \
1251 # "http://it.slashdot.org/article.pl?sid=07/03/02/1657247&from=rss"\
1252 # to "http://it.slashdot.org/article.pl?sid=07/03/02/1657247" (1 hit).
1253 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1254 $c = highlight_matched_url($c, '(?<=changed ")[^""]*');
1255 $c =~ s@(?<=changed )"([^""]*)"@$1@; # Remove quotes
1256 $c = highlight_matched_url($c, '(?<=to ")[^""]*');
1257 $c =~ s@(?<=to )"([^""]*)"@$1@; # Remove quotes
1258 $c =~ s@(\d+)(?= hits?)@$h{'hits'}$1$h{'Standard'}@;
1260 } elsif ($c =~ m/^pcrs command "([^""]*)" didn\'t change/) {
1262 # pcrs command "s@^http://([^.]+?)/?$@http://www.bing.com/search?q=$1@" didn't \
1263 # change "http://www.example.org/".
1264 $c =~ s@(?<=pcrs command )"([^""]*)"@$h{'filter'}$1$h{'Standard'}@;
1265 $c = highlight_matched_url($c, '(?<=change ")[^""]*');
1267 } elsif ($c =~ m/(^New URL is: )(.*)/) {
1269 # New URL is: http://it.slashdot.org/article.pl?sid=07/03/04/1511210
1270 # XXX: Use URL highlighter
1272 $c = $1 . h('rewritten-URL') . $2 . h('Standard');
1274 } elsif ($c =~ m/No pcrs command recognized, assuming that/) {
1275 # No pcrs command recognized, assuming that "http://config.privoxy.org/user-manual/favicon.png"\
1276 # is already properly formatted.
1277 # XXX: assume the same?
1278 $c = highlight_matched_url($c, '(?<=assuming that \")[^"]*');
1280 } elsif ($c =~ m/^Percent-encoding redirect/) {
1282 # Percent-encoding redirect URL: http://www.example.org/\x02
1283 $c = highlight_matched_url($c, '(?<=redirect URL: ).*');
1287 found_unknown_content($c);
1294 sub handle_loglevel_gif_deanimate($) {
1296 my $content = shift;
1298 if ($content =~ m/Success! GIF shrunk from (\d+) bytes to (\d+)\./) {
1300 my $bytes_from = $1;
1302 # Gif-Deanimate: Success! GIF shrunk from 205 bytes to 133.
1303 $content =~ s@$bytes_from@$h{'Number'}$bytes_from$h{'Standard'}@;
1304 # XXX: Do we need g in case of ($1 == $2)?
1305 $content =~ s@$bytes_to@$h{'Number'}$bytes_to$h{'Standard'}@;
1307 } elsif ($content =~ m/GIF (not) changed/) {
1309 # Gif-Deanimate: GIF not changed.
1310 return '' if SUPPRESS_GIF_NOT_CHANGED;
1311 $content =~ s@($1)@$h{'not'}$1$h{'Standard'}@;
1313 } elsif ($content =~ m/^failed! \(gif parsing\)/) {
1315 # failed! (gif parsing)
1316 # XXX: Replace this error message with something less stupid
1317 $content =~ s@(failed!)@$h{'error'}$1$h{'Standard'}@;
1319 } elsif ($content =~ m/^Need to de-chunk first/) {
1321 # Need to de-chunk first
1322 return '' if SUPPRESS_NEED_TO_DE_CHUNK_FIRST;
1324 } elsif ($content =~ m/^(?:No GIF header found|failed while parsing)/) {
1326 # No GIF header found (XXX: Did I ever commit this?)
1327 # failed while parsing 195 134747048 (XXX: never committed)
1329 # Ignore these for now
1333 found_unknown_content($content);
1340 sub handle_loglevel_request($) {
1342 my $content = shift;
1344 if ($content =~ m/crunch! /) {
1346 # config.privoxy.org/send-stylesheet crunch! (CGI Call)
1348 # Highlight crunch reasons
1349 foreach my $reason (keys %reason_colours) {
1350 $content =~ s@\(($reason)\)@$reason_colours{$reason}($1)$h{'Standard'}@g;
1352 # Highlight request URL domain and ditch 'crunch!'
1353 $content = highlight_matched_pattern($content, 'request_', '[^ ]*(?= crunch!)');
1354 $content =~ s@ crunch!@@;
1356 } elsif ($content =~ m/\[too long, truncated\]$/) {
1358 # config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit[...]&filter... [too long, truncated]
1359 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1361 } elsif ($content =~ m/(.*)/) { # XXX: Pretty stupid
1363 # trac.vidalia-project.net/wiki/Volunteer?format=txt
1364 $content = h('request_') . $content . h('Standard');
1368 found_unknown_content($content);
1375 sub handle_loglevel_crunch($) {
1377 my $content = shift;
1379 # Highlight crunch reason
1380 foreach my $reason (keys %reason_colours) {
1381 $content =~ s@($reason)@$reason_colours{$reason}$1$h{'Standard'}@g;
1384 if ($content =~ m/\[too long, truncated\]$/) {
1386 # Blocked: config.privoxy.org/edit-actions-submit?f=3&v=1176116716&s=7&Submit=Submit\
1387 # [...]&filter... [too long, truncated]
1388 $content = highlight_matched_pattern($content, 'request_', '^.*(?=\.\.\. \[too long, truncated\]$)');
1392 # Blocked: http://ads.example.org/
1393 $content = highlight_matched_pattern($content, 'request_', '(?<=: ).*');
1399 sub handle_loglevel_connect($) {
1403 if ($c =~ m/^via [^\s]+ to: [^\s]+/) {
1405 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
1407 $c = highlight_matched_host($c, '(?<=via )[^\s]+');
1408 $c = highlight_matched_host($c, '(?<=to: )[^\s]+');
1410 } elsif ($c =~ m/^connect to: .* failed: .*/) {
1412 # connect to: www.example.org.noconnect failed: Operation not permitted
1414 $c = highlight_matched_host($c, '(?<=connect to: )[^\s]+');
1416 $c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1418 } elsif ($c =~ m/^to ([^\s]*)( successful)?$/) {
1420 # Connect: to www.nzherald.co.nz successful
1421 # Connect: to archiv.radiotux.de
1423 return '' if SUPPRESS_SUCCESSFUL_CONNECTIONS;
1424 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1426 } elsif ($c =~ m/^to ([^\s]*)$/) {
1428 # Connect: to lists.sourceforge.net:443
1430 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1432 } elsif ($c =~ m/^[Aa]ccepted connection from .*/ or
1436 # Accepted connection from 10.0.0.1 on socket 5
1437 # Privoxy between 3.0.20 and 3.0.6:
1438 # accepted connection from 10.0.0.1( on socket 5)?
1439 # Privoxy 3.0.6 and earlier just say:
1441 $c = highlight_matched_host($c, '(?<=connection from )[^ ]*');
1442 $c = highlight_matched_pattern($c, 'Number', '(?<=socket )\d+');
1444 } elsif ($c =~ m/^Closing client socket/) {
1446 # Closing client socket 5. Keep-alive: 0, Socket alive: 1. Data available: 0.
1447 # Privoxy 3.0.20 and later
1448 # Closing client socket 8. Keep-alive: 1. Socket alive: 0. Data available: 0. \
1449 # Configuration file change detected: 0. Requests received: 11.
1451 $c = highlight_matched_pattern($c, 'Number', '(?<=socket )\d+');
1452 $c = highlight_matched_pattern($c, 'Number', '(?<=Keep-alive: )\d+');
1453 $c = highlight_matched_pattern($c, 'Number', '(?<=Socket alive: )\d+');
1454 $c = highlight_matched_pattern($c, 'Number', '(?<=available: )\d+');
1455 $c = highlight_matched_pattern($c, 'Number', '(?<=detected: )\d+');
1456 $c = highlight_matched_pattern($c, 'Number', '(?<=received: )\d+');
1458 } elsif ($c =~ m/^write header to: .* failed:/) {
1460 # write header to: 10.0.0.1 failed: Broken pipe
1462 $c = highlight_matched_host($c, '(?<=write header to: )[^\s]*');
1463 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1465 } elsif ($c =~ m/^write header to client failed:/) {
1467 # write header to client failed: Broken pipe
1469 $c =~ s@(?<=failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1471 } elsif ($c =~ m/^socks4_connect:/) {
1473 # socks4_connect: SOCKS request rejected or failed.
1474 $c =~ s@(?<=socks4_connect: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1476 } elsif ($c =~ m/^Listening for new connections/ or
1477 $c =~ m/^accept connection/) {
1479 # Privoxy versions above 3.0.6 say:
1480 # Listening for new connections ...
1481 # earlier versions say:
1482 # accept connection ...
1485 } elsif ($c =~ m/^accept failed:/) {
1487 $c =~ s@(?<=accept failed: )(.*)@$h{'Error'}$1$h{'Standard'}@;
1489 } elsif ($c =~ m/^Overriding forwarding settings/) {
1491 # Overriding forwarding settings based on 'forward 10.0.0.1:8123'
1492 $c =~ s@(?<=based on \')(.*)(?=\')@$h{'configuration-line'}$1$h{'Standard'}@;
1494 } elsif ($c =~ m/^Denying suspicious CONNECT request from/) {
1496 # Denying suspicious CONNECT request from 10.0.0.1
1497 $c = highlight_matched_host($c, '(?<=from )[^\s]+'); # XXX: not an URL
1499 } elsif ($c =~ m/^socks5_connect:/) {
1501 $c =~ s@(?<=socks5_connect: )(.*)@$h{'error'}$1$h{'Standard'}@;
1503 } elsif ($c =~ m/^Created new connection to/) {
1505 # Created new connection to www.privoxy.org:80 on socket 11.
1506 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1507 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1509 } elsif ($c =~ m/^Found reusable socket/) {
1511 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
1513 # Found reusable socket 8 for www.privoxy.org:80 in slot 2.\
1514 # Timestamp made 0 seconds ago. Timeout: 1. Latency: 0.
1515 $c =~ s@(?<=Found reusable socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1516 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1517 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1518 $c =~ s@(?<=made )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1519 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1520 $c =~ s@(?<=Latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1522 } elsif ($c =~ m/^Marking open socket/) {
1524 # Marking open socket 9 for www.privoxy.org:80 in slot 0 as unused.
1525 $c =~ s@(?<=Marking open socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1526 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1527 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1529 } elsif ($c =~ m/^No reusable/) {
1531 # No reusable socket for addons.mozilla.org:443 found. Opening a new one.
1532 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1534 } elsif ($c =~ m/^(Remembering|Forgetting) socket/) {
1536 # Remembering socket 13 for www.privoxy.org:80 in slot 0.
1537 # Forgetting socket 38 for www.privoxy.org:80 in slot 5.
1539 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1540 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1541 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1543 } elsif ($c =~ m/^Socket/) {
1545 # Socket 16 already forgotten or never remembered.
1546 $c =~ s@(?<=Socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1548 } elsif ($c =~ m/^The connection to/) {
1550 # The connection to www.privoxy.org:80 in slot 6 timed out. Closing socket 19. Timeout is: 61.
1552 # The connection to 1.bp.blogspot.com:80 in slot 0 timed out. Closing socket 5.\
1553 # Timeout is: 1. Assumed latency: 4.
1554 # The connection to 10.0.0.1:80 in slot 0 is no longer usable. Closing socket 4.
1555 $c = highlight_matched_host($c, '(?<=connection to )[^\s]+');
1556 $c =~ s@(?<=in slot )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1557 $c =~ s@(?<=Closing socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1558 $c =~ s@(?<=Timeout is: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1559 $c =~ s@(?<=Assumed latency: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1561 } elsif ($c =~ m/^Stopped waiting for the request line/ or
1562 $c =~ m/^No request line on socket \d received in time/ or
1563 $c =~ m/^The client side of the connection on socket \d/) {
1565 # Stopped waiting for the request line. Timeout: 121.
1566 # Privoxy 3.0.19 and later:
1567 # No request line on socket 5 received in time. Timeout: 1.
1568 # The client side of the connection on socket 5 got closed \
1569 # without sending a complete request line.
1570 $c =~ s@(?<=Timeout: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1571 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1573 } elsif ($c =~ m/^Waiting for \d/) {
1575 # Waiting for 1 connections to timeout.
1576 $c =~ s@(?<=^Waiting for )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1578 } elsif ($c =~ m/^Initialized/) {
1580 # Initialized 20 socket slots.
1581 $c =~ s@(?<=Initialized )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1583 } elsif ($c =~ m/^Done reading from server/) {
1585 # Done reading from server. Expected content length: 24892. \
1586 # Actual content length: 24892. Most recently received: 4412.
1588 # Done reading from server. Expected content length: 24892. \
1589 # Actual content length: 24892. Bytes most recently read: 4412.
1590 # Done reading from server. Content length: 6018 as expected. \
1591 # Bytes most recently read: 294.
1592 $c =~ s@(?<=ontent length: )(\d+)@$h{'Number'}$1$h{'Standard'}@g;
1593 $c =~ s@(?<=received: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1594 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1596 } elsif ($c =~ m/^Continuing buffering (?:server )?headers/) {
1598 # Continuing buffering headers. byte_count: 19. header_offset: 517. len: 536.
1599 $c =~ s@(?<=byte_count: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1600 $c =~ s@(?<=header_offset: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1601 $c =~ s@(?<=len: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1602 # 3.0.15 up to 3.0.19:
1603 # Continuing buffering headers. Bytes most recently read: 498.
1604 $c =~ s@(?<=read: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1606 # Continuing buffering server headers from socket 5. Bytes most recently read: 498.
1607 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1609 } elsif ($c =~ m/^Received \d+ bytes while/) {
1611 # Received 206 bytes while expecting 12103.
1612 $c =~ s@(?<=Received )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1613 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1615 } elsif ($c =~ m/^(Rejecting c|C)onnection from/) {
1617 # Connection from 81.163.28.218 dropped due to ACL
1618 # Rejecting connection from 178.63.152.227. Maximum number of connections reached.
1619 # Connection from 192.168.2.1 on 127.0.1.1:8118 (socket 3) dropped due to ACL
1620 $c = highlight_matched_host($c, '(?<=onnection from )[\d.:]+');
1621 $c = highlight_matched_host($c, '(?<=on )[\d.:]+');
1622 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1624 } elsif ($c =~ m/^(?:Reusing|Closing) server socket / or
1625 $c =~ m/^No additional client request/) {
1627 # Reusing server socket 4. Opened for 10.0.0.1.
1628 # Closing server socket 2. Opened for 10.0.0.1.
1629 # No additional client request received in time. \
1630 # Closing server socket 4, initially opened for 10.0.0.1.
1631 # No additional client request received in time on socket 29.
1632 # Privoxy 3.0.20 and later
1633 # Reusing server socket 7 connected to www.privoxy.org. Total requests: 2.
1634 # Closing server socket 6 connected to d.asset.soup.io. Keep-alive: 0.\
1635 # Tainted: 1. Socket alive: 1. Timeout: 60. Configuration file change detected: 0.
1636 # Reusing server socket 35 connected to nl.wikipedia.org. Requests already sent: 5.
1638 $c =~ s@(?<= socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1639 $c = highlight_matched_host($c, '(?<=for )[^\s]+(?=\.)');
1640 $c = highlight_matched_host($c, '(?<=connected to )[^\s]+(?=\.)');
1641 for my $number_pattern ('requests', 'Keep-alive', 'Tainted', ' alive', 'Timeout', 'detected') {
1642 $c = highlight_matched_pattern($c, 'Number', '(?<='. $number_pattern . ': )\d+');
1644 $c =~ s@(?<=already sent: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1646 } elsif ($c =~ m/^Connected to /) {
1648 # Connected to tor-jail[10.0.0.2]:9050.
1650 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1651 $c = highlight_matched_host($c, '(?<=Connected to )[^\[\s]+');
1652 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1654 } elsif ($c =~ m/^Could not connect to /) {
1656 # Could not connect to [10.0.0.1]:80.
1658 $c = highlight_matched_host($c, '(?<=\[)[^\]]+');
1659 $c =~ s@(?<=\]:)(\d+)@$h{'Number'}$1$h{'Standard'}@;
1661 } elsif ($c =~ m/^Waiting for the next client request/ or
1662 $c =~ m/^The connection on server socket/ or
1663 $c =~ m/^Client request (?:\d+ )?(?:arrived in time|has been pipelined) /) {
1665 # Waiting for the next client request on socket 3. Keeping the server \
1666 # socket 12 to a.fsdn.com open.
1667 # The connection on server socket 6 to upload.wikimedia.org isn't reusable. Closing.
1668 # Privoxy 3.0.20 and later:
1669 # Client request 4 arrived in time on socket 7.
1670 # Used by Privoxy 3.0.18 and 3.0.19:
1671 # Client request arrived in time on socket 21.
1672 # Used by earlier version:
1673 # Client request arrived in time or the client closed the connection on socket 12.
1674 # Client request 8 has been pipelined on socket 7 and the socket is still alive.
1676 $c =~ s@(?<=request )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1677 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1678 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1679 $c = highlight_matched_host($c, '(?<=to )[^\s]+');
1681 } elsif ($c =~ m/^Marking the server socket/) {
1683 # Marking the server socket 7 tainted.
1685 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1687 } elsif ($c =~ m/^Reduced expected bytes to /) {
1689 # Reduced expected bytes to 0 to account for the 1542 ones we already got.
1690 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1691 $c =~ s@(?<=for the )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1693 } elsif ($c =~ m/^The client closed socket /) {
1695 # The client closed socket 2 while the server socket 4 is still open.
1696 $c =~ s@(?<=closed socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1697 $c =~ s@(?<=server socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1699 } elsif ($c =~ m/^Expected client content length set /) {
1701 # Expected client content length set to 667325411 after reading 4999 bytes.
1702 $c =~ s@(?<=set to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1703 $c =~ s@(?<=reading )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1705 } elsif ($c =~ m/^Reducing expected bytes to /) {
1707 # Reducing expected bytes to 0. Marking the server socket tainted after throwing 4 bytes away.
1708 $c =~ s@(?<=bytes to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1709 $c =~ s@(?<=after throwing )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1711 } elsif ($c =~ m/^Waiting for up to /) {
1713 # Waiting for up to 4999 bytes from the client.
1714 $c =~ s@(?<=up to )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1716 } elsif ($c =~ m/^Optimistically sending /) {
1718 # Optimistically sending 318 bytes of client headers intended for www.privoxy.org
1719 $c =~ s@(?<=sending )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1720 $c = highlight_matched_host($c, '(?<=for )[^\s]+');
1722 } elsif ($c =~ m/^Stopping to watch the client socket/) {
1724 # Stopping to watch the client socket. There's already another request waiting.
1725 # Privoxy 3.0.20 and later:
1726 # Stopping to watch the client socket 5. There's already another request waiting.
1727 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1729 } elsif ($c =~ m/^Drained \d+ bytes before closing/) {
1731 # Drained 180 bytes before closing socket 6
1732 $c =~ s@(?<=Drained )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1733 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1735 } elsif ($c =~ m/^Tainting client socket/ or
1736 $c =~ m/^Failed to shutdown socket/) {
1738 # Tainting client socket 7 due to unread data.
1739 # Failed to shutdown socket 11: Connection reset by peer
1741 $c =~ s@(?<=socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1743 } elsif ($c =~ m/^Shifting \d+ pipelined bytes/) {
1745 # Shifting 360 pipelined bytes by 360 bytes
1746 $c =~ s@(?<=Shifting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1747 $c =~ s@(?<=by )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1749 } elsif ($c =~ m/^Flushed (\d+) bytes of request body while expecting (\d+)/) {
1751 # Flushed 30 bytes of request body while expecting 30
1752 $c =~ s@(?<=Flushed )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1753 $c =~ s@(?<=expecting )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1755 } elsif ($c =~ m/^Performing the TLS\/SSL handshake with client. Hash of host:/) {
1757 # Performing the TLS/SSL handshake with client. Hash of host: bab5296b25e256c7b06b92b17b56bcae
1758 $c = highlight_matched_host($c, '(?<=Hash of host: ).+');
1760 } elsif ($c =~ m/^Forwarding \d+ bytes of encrypted POST data/) {
1762 # Forwarding 1954 bytes of encrypted POST data
1763 $c =~ s@(?<=Forwarding )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1765 } elsif ($c =~ m/^Forwarded the last \d+ bytes/) {
1767 # Forwarded the last 1954 bytes
1768 $c =~ s@(?<=the last )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1770 } elsif ($c =~ m/^Waiting for the next client connection. Currently active threads:/) {
1772 # Waiting for the next client connection. Currently active threads: 30
1773 $c =~ s@(?<=threads: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1775 } elsif ($c =~ m/^Data arrived in time on client socket/) {
1777 # Data arrived in time on client socket 6. Requests so far: 3
1778 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1779 $c =~ s@(?<=Requests so far: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1781 } elsif ($c =~ m/^Dropping the client connection on socket/) {
1783 # Dropping the client connection on socket 71. The server connection has not been established yet.
1784 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1786 } elsif ($c =~ m/^Looks like we / or
1787 $c =~ m/^Unsetting keep-alive flag/ or
1788 $c =~ m/^No connections to wait/ or
1789 $c =~ m/^Complete client request received/ or
1790 $c =~ m/^Possible pipeline attempt detected./ or
1791 $c =~ m/^POST request detected. The connection will not be kept alive./ or
1792 $c =~ m/^The server still wants to talk, but the client hung up on us./ or
1793 $c =~ m/^The server didn't specify how long the connection will stay open/ or
1794 $c =~ m/^There might be a request body. The connection will not be kept alive/ or
1795 $c =~ m/^There better be a request body./ or
1796 $c =~ m/^Done reading from the client\.$/) {
1798 # Looks like we reached the end of the last chunk. We better stop reading.
1799 # Looks like we read the end of the last chunk together with the server \
1800 # headers. We better stop reading.
1801 # Looks like we got the last chunk together with the server headers. \
1802 # We better stop reading.
1803 # Unsetting keep-alive flag.
1804 # No connections to wait for left.
1805 # Client request arrived in time or the client closed the connection.
1806 # Complete client request received
1807 # Possible pipeline attempt detected. The connection will not be \
1808 # kept alive and we will only serve the first request.
1809 # POST request detected. The connection will not be kept alive.
1810 # The server still wants to talk, but the client hung up on us.
1811 # The server didn't specify how long the connection will stay open. Assume it's only a second.
1812 # There might be a request body. The connection will not be kept alive.
1813 # Privoxy 3.0.20 and later
1814 # There better be a request body.
1815 # Done reading from the client.
1819 found_unknown_content($c);
1827 sub handle_loglevel_info($) {
1831 if ($c =~ m/^Rewrite detected:/) {
1833 # Rewrite detected: GET http://10.0.0.2:88/blah.txt HTTP/1.1
1834 $c = highlight_matched_request_line($c, '(?<=^Rewrite detected: ).*');
1836 } elsif ($c =~ m/^Decompress(ing deflated|ion didn)/ or
1837 $c =~ m/^Compressed content detected/ or
1838 $c =~ m/^SDCH-compressed content detected/ or
1841 # Decompressing deflated iob: 117
1842 # Decompression didn't result in any content.
1843 # Compressed content detected, content filtering disabled. Consider recompiling Privoxy\
1844 # with zlib support or enable the prevent-compression action.
1845 # SDCH-compressed content detected, content filtering disabled.\
1846 # Consider suppressing SDCH offers made by the client.
1847 # Tagger 'complete-url' created empty tag. Ignored.
1851 } elsif ($c =~ m/^(Re)?loading configuration file /) {
1853 # loading configuration file '/usr/local/etc/privoxy/config':
1854 # Reloading configuration file '/usr/local/etc/privoxy/config'
1855 $c =~ s@(?<=loading configuration file \')([^\']*)@$h{'file'}$1$h{'Standard'}@;
1857 } elsif ($c =~ m/^Loading (actions|filter|trust) file: /) {
1859 # Loading actions file: /usr/local/etc/privoxy/default.action
1860 # Loading filter file: /usr/local/etc/privoxy/default.filter
1861 # Loading trust file: /usr/local/etc/privoxy/trust
1863 $c =~ s@(?<= file: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1865 } elsif ($c =~ m/^exiting by signal/) {
1867 # exiting by signal 15 .. bye
1868 $c =~ s@(?<=exiting by signal )(\d+)@$h{'signal'}$1$h{'Standard'}@;
1870 } elsif ($c =~ m/^Privoxy version/) {
1872 # Privoxy version 3.0.7
1873 $c =~ s@(?<=^Privoxy version )(\d+\.\d+\.\d+)$@$h{'version'}$1$h{'Standard'}@;
1875 } elsif ($c =~ m/^Program name: /) {
1877 # Program name: /usr/local/sbin/privoxy
1878 $c =~ s@(?<=Program name: )(.*)@$h{'program-name'}$1$h{'Standard'}@;
1880 } elsif ($c =~ m/^Listening on port /) {
1882 # Listening on port 8118 on IP address 10.0.0.1
1883 $c =~ s@(?<=Listening on port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1884 $c =~ s@(?<=on IP address )(.*)@$h{'ip-address'}$1$h{'Standard'}@;
1886 } elsif ($c =~ m/^\(Re-\)Open(?:ing)? logfile/) {
1888 # (Re-)Open logfile /var/log/privoxy/privoxy.log
1889 $c =~ s@(?<=Open logfile )(.*)@$h{'file'}$1$h{'Standard'}@;
1891 } elsif ($c =~ m/^(Request from|Malformed server response detected)/) {
1893 # Request from 10.0.0.1 denied. limit-connect{,} doesn't allow CONNECT requests to port 443.
1894 # Request from 10.0.0.1 marked for blocking. limit-connect{,} doesn't allow CONNECT requests to port 443.
1896 # Request from 10.0.0.1 marked for blocking. limit-connect{0} doesn't allow CONNECT requests to www.example.org:443
1897 # Malformed server response detected. Downgrading to HTTP/1.0 impossible.
1899 $c =~ s@(?<=Request from )([^\s]*)@$h{'ip-address'}$1$h{'Standard'}@;
1900 $c =~ s@(denied|blocking)@$h{'warning'}$1$h{'Standard'}@;
1901 $c =~ s@(CONNECT)@$h{'method'}$1$h{'Standard'}@;
1902 $c =~ s@(?<=to port )(\d+)@$h{'port'}$1$h{'Standard'}@;
1903 $c =~ s@(?<=to )([^\s]+)@$h{'request_'}$1$h{'Standard'}@;
1905 } elsif ($c =~ m/^Status code/) {
1907 # Status code 304 implies no body.
1908 $c =~ s@(?<=Status code )(\d+)@$h{'status-code'}$1$h{'Standard'}@;
1910 } elsif ($c =~ m/^Method/) {
1912 # Method HEAD implies no body.
1913 $c =~ s@(?<=Method )([^\s]+)@$h{'method'}$1$h{'Standard'}@;
1915 } elsif ($c =~ m/^Buffer limit reached while extending /) {
1917 # Buffer limit reached while extending the buffer (iob). Needed: 4197470. Limit: 4194304
1918 $c =~ s@(?<=Needed: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1919 $c =~ s@(?<=Limit: )(\d+)@$h{'Number'}$1$h{'Standard'}@;
1921 } elsif ($c =~ m/^File modification detected: /) {
1923 # File modification detected: /usr/local/etc/privoxy/user-agent.action
1924 $c =~ s@(?<= detected: )(.*)$@$h{'file'}$1$h{'Standard'}@;
1926 } elsif ($c =~ m/^No logfile configured/ or
1927 $c =~ m/^Malformerd HTTP headers detected and MS IIS5 hack enabled/ or
1928 $c =~ m/^Invalid \"chunked\" transfer/ or
1929 $c =~ m/^Support for/ or
1930 $c =~ m/^Flushing header and buffers/ or
1931 $c =~ m/^Can not resolve/
1934 # No logfile configured. Please enable it before reporting any problems.
1935 # Malformerd HTTP headers detected and MS IIS5 hack enabled. Expect an invalid \
1936 # response or even no response at all.
1937 # No logfile configured. Logging disabled.
1938 # Invalid "chunked" transfer encoding detected and ignored.
1939 # Support for 'Connection: keep-alive' is experimental, incomplete and\
1940 # known not to work properly in some situations.
1941 # Flushing header and buffers. Stepping back from filtering.
1942 # Can not resolve doesnotexist: hostname nor servname provided, or not known
1946 found_unknown_content($c);
1953 sub handle_loglevel_cgi($) {
1957 if ($c =~ m/^Granting access to/) {
1959 #Granting access to http://config.privoxy.org/send-stylesheet, referrer http://p.p/ is trustworthy.
1961 } elsif ($c =~ m/^Substituting: s(.)/) {
1963 # Substituting: s/@else-not-FEATURE_ZLIB@.*@endif-FEATURE_ZLIB@//sigTU
1964 # XXX: prone to span several lines
1967 #$c =~ s@(?<=failed: )(.*)@$h{'error'}$1$h{'Standard'}@;
1968 $c =~ s@(?!<=\\)($delimiter)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g; # XXX: Too aggressive
1969 #$c =~ s@(?!<=\\)($1)@$h{'pcrs-delimiter'}$1$h{'Standard'}@g;
1975 sub handle_loglevel_force($) {
1979 if ($c =~ m/^Ignored force prefix in request:/) {
1981 # Ignored force prefix in request: "GET http://10.0.0.1/PRIVOXY-FORCE/block HTTP/1.1"
1982 $c =~ s@^(Ignored)@$h{'ignored'}$1$h{'Standard'}@;
1983 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1985 } elsif ($c =~ m/^Enforcing request:/) {
1987 # Enforcing request: "GET http://10.0.0.1/block HTTP/1.1".
1988 $c = highlight_matched_request_line($c, '(?<=request: ")[^"]*');
1992 found_unknown_content($c);
1999 sub handle_loglevel_error($) {
2003 if ($c =~ m/^(?:Empty|No) server or forwarder response received on socket \d+\./) {
2005 # Empty server or forwarder response received on socket 4.
2006 # Empty server or forwarder response received on socket 3. \
2007 # Closing client socket 15 without sending data.
2008 # Used by Privoxy 3.0.18 and later:
2009 # No server or forwarder response received on socket 8. \
2010 # Closing client socket 10 without sending data.
2012 $c =~ s@(?<=on socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
2013 $c =~ s@(?<=client socket )(\d+)@$h{'Number'}$1$h{'Standard'}@;
2015 } elsif ($c =~ m/^Didn't receive data in time:/) {
2017 # Didn't receive data in time: a.fsdn.com:443
2018 $c =~ s@(?<=in time: )(.*)@$h{'destination'}$1$h{'Standard'}@;
2021 # XXX: There are probably more messages that deserve highlighting.
2027 sub handle_loglevel_ignore($) {
2031 sub gather_loglevel_clf_stats($) {
2033 my $content = shift;
2034 my ($method, $resource, $http_version, $status_code, $size);
2038 # +0200] "GET https://www.youtube.com/watch?v=JmcA9LIIXWw HTTP/1.1" 200 68004
2039 # +0200] "VERSION-CONTROL http://p.p/ HTTP/1.1" 200 2787
2040 $content =~ m/^[+-]\d{4}\] "([^ ]+) (.+) (HTTP\/\d\.\d)" (\d+) (\d+)/;
2047 $stats{requests_clf}++;
2049 unless (defined $method) {
2050 # +0200] "Invalid request" 400 0
2051 return if ($content =~ m/^[+-]\d{4}\] "Invalid request"/);
2052 print("Failed to parse: $content\n");
2055 $stats{'method'}{$method}++;
2056 if ($cli_options{'url-statistics-threshold'} != 0) {
2057 $stats{'resource'}{$resource}++;
2059 $stats{'http-version'}{$http_version}++;
2061 if ($cli_options{'host-statistics-threshold'} != 0) {
2062 $resource =~ m@(?:https?://)?([^/]+)/?@;
2063 $stats{'hosts'}{$1}++;
2065 $stats{'content-size-total'} += $size;
2066 $stats{'status-code'}{$status_code}++;
2069 sub gather_loglevel_request_stats($$) {
2077 sub gather_loglevel_crunch_stats($$) {
2084 if ($c =~ m/^Redirected:/) {
2085 # Redirected: http://www.example.org/http://p.p/
2086 $stats{'fast-redirections'}++;
2088 } elsif ($c =~ m/^Blocked:/) {
2089 # Blocked: blogger.googleusercontent.com:443
2090 $stats{'blocked'}++;
2092 } elsif ($c =~ m/^Connection timeout:/) {
2093 # Connection timeout: http://c.tile.openstreetmap.org/18/136116/87842.png
2094 $stats{'connection-timeout'}++;
2096 } elsif ($c =~ m/^Connection failure:/) {
2097 # Connection failure: http://127.0.0.1:8080/
2098 $stats{'connection-failure'}++;
2103 sub gather_loglevel_error_stats($$) {
2110 if ($c =~ m/^Empty server or forwarder response received on socket \d+./) {
2112 # Empty server or forwarder response received on socket 4.
2113 $stats{'empty-responses'}++;
2114 if ($thread_data{$thread}{'new_connection'}) {
2115 $stats{'empty-responses-on-new-connections'}++;
2117 $stats{'empty-responses-on-reused-connections'}++;
2122 sub gather_loglevel_connect_stats($$) {
2124 my ($c, $thread) = @_;
2128 if ($c =~ m/^via ([^\s]+) to: [^\s]+/) {
2130 # Connect: via 10.0.0.1:8123 to: www.example.org.noconnect
2131 $thread_data{$thread}{'forwarder'} = $1; # XXX: is this missue?
2133 } elsif ($c =~ m/^to ([^\s]*)$/) {
2135 # Connect: to lists.sourceforge.net:443
2137 $thread_data{$thread}{'forwarder'} = 'direct connection';
2139 } elsif ($c =~ m/^Created new connection to/) {
2141 # Created new connection to www.privoxy.org:80 on socket 11.
2143 $thread_data{$thread}{'new_connection'} = 1;
2145 } elsif ($c =~ m/^Reusing server socket \d./ or
2146 $c =~ m/^Found reusable socket/) {
2148 # Reusing server socket 4. Opened for 10.0.0.1.
2149 # Found reusable socket 9 for www.privoxy.org:80 in slot 0.
2151 $thread_data{$thread}{'new_connection'} = 0;
2152 $stats{'reused-connections'}++;
2154 } elsif ($c =~ m/^Closing client socket \d+. .* Requests received: (\d+)\.$/) {
2156 # Closing client socket 12. Keep-alive: 1. Socket alive: 1. Data available: 0. \
2157 # Configuration file change detected: 0. Requests received: 14.
2159 $stats{'client-requests-on-connection'}{$1}++;
2160 $stats{'closed-client-connections'}++;
2164 sub gather_loglevel_header_stats($$) {
2166 my ($c, $thread) = @_;
2170 if ($c =~ m/^A HTTP\/1\.1 response without/ or
2171 $c =~ m/^Keeping the server header 'Connection: keep-alive' around./)
2173 # A HTTP/1.1 response without Connection header implies keep-alive.
2174 # Keeping the server header 'Connection: keep-alive' around.
2175 $stats{'server-keep-alive'}++;
2184 'server-keep-alive' => 0,
2185 'reused-connections' => 0,
2186 'empty-responses' => 0,
2187 'empty-responses-on-new-connections' => 0,
2188 'empty-responses-on-reused-connections' => 0,
2189 'fast-redirections' => 0,
2191 'connection-failure' => 0,
2192 'connection-timeout' => 0,
2193 'reused-connections' => 0,
2194 'server-keep-alive' => 0,
2195 'closed-client-connections' => 0,
2196 'content-size-total' => 0,
2198 $stats{'client-requests-on-connection'}{1} = 0;
2201 sub get_percentage($$) {
2205 # If small is 0 the percentage is always 0%.
2206 # Make sure it works even if big is 0 as well.
2207 return "0.00%" if ($small eq 0);
2209 # Prevent division by zero.
2210 # XXX: Is this still supposed to be reachable?
2211 return "NaN" if ($big eq 0);
2213 return sprintf("%.2f%%", $small / $big * 100);
2220 my $new_connections = $stats{requests} - $stats{crunches} - $stats{'reused-connections'};
2221 my $client_requests_checksum = 0;
2224 if ($stats{requests_clf} && $stats{requests}
2225 && $stats{requests_clf} != $stats{requests}) {
2226 print "Inconsistent request counts: " . $stats{requests} . "/" . $stats{requests_clf} . "\n";
2229 # To get the total number of requests we can use either the number
2230 # of Common-Log-Format lines or the number of "Request:" messages.
2231 # We prefer the number of CLF lines if available because using
2232 # it works when analysing old log files from Privoxy versions before 3.0.29.
2233 # In Privoxy 3.0.28 and earlier "Request:" messages excluded
2234 # crunched messages.
2235 $requests_total = $stats{requests_clf} ? $stats{requests_clf} : $stats{requests};
2237 if ($requests_total eq 0) {
2238 print "No requests yet.\n";
2242 print "Client requests total: " . $requests_total . "\n";
2243 if ($stats{crunches}) {
2244 my $outgoing_requests = $requests_total - $stats{crunches};
2245 print "Crunches: " . $stats{crunches} . " (" .
2246 get_percentage($requests_total, $stats{crunches}) . ")\n";
2247 print "Blocks: " . $stats{'blocked'} . " (" .
2248 get_percentage($requests_total, $stats{'blocked'}) . ")\n";
2249 print "Fast redirections: " . $stats{'fast-redirections'} . " (" .
2250 get_percentage($requests_total, $stats{'fast-redirections'}) . ")\n";
2251 print "Connection timeouts: " . $stats{'connection-timeout'} . " (" .
2252 get_percentage($requests_total, $stats{'connection-timeout'}) . ")\n";
2253 print "Connection failures: " . $stats{'connection-failure'} . " (" .
2254 get_percentage($requests_total, $stats{'connection-failure'}) . ")\n";
2255 print "Outgoing requests: " . $outgoing_requests . " (" .
2256 get_percentage($requests_total, $outgoing_requests) . ")\n";
2258 print "No crunches detected. Is 'debug 1024' enabled?\n";
2261 print "Server keep-alive offers: " . $stats{'server-keep-alive'} . " (" .
2262 get_percentage($requests_total, $stats{'server-keep-alive'}) . ")\n";
2263 print "New outgoing connections: " . $new_connections . " (" .
2264 get_percentage($requests_total, $new_connections) . ")\n";
2265 print "Reused connections: " . $stats{'reused-connections'} . " (" .
2266 get_percentage($requests_total, $stats{'reused-connections'}) .
2267 "; server offers accepted: " .
2268 get_percentage($stats{'server-keep-alive'}, $stats{'reused-connections'}) . ")\n";
2269 print "Empty responses: " . $stats{'empty-responses'} . " (" .
2270 get_percentage($requests_total, $stats{'empty-responses'}) . ")\n";
2271 print "Empty responses on new connections: "
2272 . $stats{'empty-responses-on-new-connections'} . " (" .
2273 get_percentage($requests_total, $stats{'empty-responses-on-new-connections'})
2275 print "Empty responses on reused connections: " .
2276 $stats{'empty-responses-on-reused-connections'} . " (" .
2277 get_percentage($requests_total, $stats{'empty-responses-on-reused-connections'}) .
2279 print "Client connections: " . $stats{'closed-client-connections'} . "\n";
2280 if ($stats{'content-size-total'}) {
2281 print "Bytes of content transfered to the client: " . $stats{'content-size-total'} . "\n";
2283 my $lines_printed = 0;
2284 print "Client requests per connection distribution:\n";
2285 foreach my $client_requests (sort {
2286 $stats{'client-requests-on-connection'}{$b} <=> $stats{'client-requests-on-connection'}{$a}}
2287 keys %{$stats{'client-requests-on-connection'}
2290 my $count = $stats{'client-requests-on-connection'}{$client_requests};
2291 $client_requests_checksum += $count * $client_requests;
2292 if ($cli_options{'show-complete-request-distribution'} or ($lines_printed < 10)) {
2293 printf "%8d: %d\n", $count, $client_requests;
2297 unless ($cli_options{'show-complete-request-distribution'}) {
2298 printf "Enable --show-complete-request-distribution to get less common numbers as well.\n";
2300 # Due to log rotation we may not have a complete picture for all the requests
2301 printf "Improperly accounted requests: ~%d\n", abs($requests_total - $client_requests_checksum);
2303 if (exists $stats{method}) {
2304 print "Method distribution:\n";
2305 foreach my $method (sort {$stats{'method'}{$b} <=> $stats{'method'}{$a}} keys %{$stats{'method'}}) {
2306 printf "%8d : %-8s\n", $stats{'method'}{$method}, $method;
2309 print "Method distribution unknown. No CLF message parsed yet. Is 'debug 512' enabled?\n";
2311 if (exists $stats{'http-version'}) {
2312 print "Client HTTP versions:\n";
2313 foreach my $http_version (sort {$stats{'http-version'}{$b} <=> $stats{'http-version'}{$a}} keys %{$stats{'http-version'}}) {
2314 printf "%8d : %-8s\n", $stats{'http-version'}{$http_version}, $http_version;
2317 print "HTTP version distribution unknown. No CLF message parsed yet. Is 'debug 512' enabled?\n";
2319 if (exists $stats{'status-code'}) {
2320 print "HTTP status codes:\n";
2321 foreach my $status_code (sort {$stats{'status-code'}{$b} <=> $stats{'status-code'}{$a}} keys %{$stats{'status-code'}}) {
2322 printf "%8d : %-8d\n", $stats{'status-code'}{$status_code}, $status_code;
2325 print "Status code distribution unknown. No CLF message parsed yet. Is 'debug 512' enabled?\n";
2328 if ($cli_options{'url-statistics-threshold'} == 0) {
2329 print "URL statistics are disabled. Increase --url-statistics-threshold to enable them.\n";
2331 print "Requested URLs:\n";
2332 foreach my $resource (sort {$stats{'resource'}{$b} <=> $stats{'resource'}{$a}} keys %{$stats{'resource'}}) {
2333 if ($stats{'resource'}{$resource} < $cli_options{'url-statistics-threshold'}) {
2334 print "Skipped statistics for URLs below the treshold.\n";
2337 printf "%d : %s\n", $stats{'resource'}{$resource}, $resource;
2341 if ($cli_options{'host-statistics-threshold'} == 0) {
2342 print "Host statistics are disabled. Increase --host-statistics-threshold to enable them.\n";
2344 print "Requested Hosts:\n";
2345 foreach my $host (sort {$stats{'hosts'}{$b} <=> $stats{'hosts'}{$a}} keys %{$stats{'hosts'}}) {
2346 if ($stats{'hosts'}{$host} < $cli_options{'host-statistics-threshold'}) {
2347 print "Skipped statistics for Hosts below the treshold.\n";
2350 printf "%d : %s\n", $stats{'hosts'}{$host}, $host;
2356 ################################################################################
2357 # Functions that actually print stuff
2358 ################################################################################
2360 sub print_clf_message() {
2362 our ($ip, $timestamp, $request_line, $status_code, $size);
2365 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2367 # Rebuild highlighted
2368 $output .= $h{'Number'} . $ip . $h{'Standard'};
2370 $output .= "[" . $h{'Timestamp'} . $timestamp . $h{'Standard'} . "]";
2372 $output .= "\"" . highlight_request_line("$request_line") . "\"";
2374 $output .= $h{'Status'} . $status_code . $h{'Standard'};
2376 $output .= $h{'Number'} . $size . $h{'Standard'};
2377 $output .= $line_end;
2382 sub print_non_clf_message($) {
2384 my $content = shift;
2385 my $msec_string = $no_msecs_mode ? '' : '.' . $req{$t}{'msecs'};
2386 my $line_start = $html_output_mode ? '' : $h{"Standard"};
2388 return if DEBUG_SUPPRESS_LOG_MESSAGES;
2391 . $time_colours[$time_colour_index % 2]
2392 . $req{$t}{'time-stamp'}
2394 . $h{Standard} . " "
2395 . $thread_colours{$t}
2399 . $h{$req{$t}{'log-level'}}
2400 . $req{$t}{'log-level'}
2407 sub shorten_thread_id($) {
2409 my $thread_id = shift;
2411 our %short_thread_ids;
2414 unless (defined $short_thread_ids{$thread_id}) {
2415 $short_thread_ids{$thread_id} = sprintf "%.3d", $max_threadid++;
2418 return $short_thread_ids{$thread_id}
2423 my ($day, $time_stamp, $thread, $log_level, $content, $c, $msecs);
2425 my $last_thread = 0;
2426 my $last_timestamp = 0;
2427 my $filters_that_did_nothing;
2430 $time_colour = paint_it('white');
2432 my %log_level_handlers = (
2433 'Re-Filter' => \&handle_loglevel_re_filter,
2434 'Header' => \&handle_loglevel_header,
2435 'Connect' => \&handle_loglevel_connect,
2436 'Redirect' => \&handle_loglevel_redirect,
2437 'Request' => \&handle_loglevel_request,
2438 'Crunch' => \&handle_loglevel_crunch,
2439 'Gif-Deanimate' => \&handle_loglevel_gif_deanimate,
2440 'Info' => \&handle_loglevel_info,
2441 'CGI' => \&handle_loglevel_cgi,
2442 'Force' => \&handle_loglevel_force,
2443 'Error' => \&handle_loglevel_error,
2444 'Fatal error' => \&handle_loglevel_ignore,
2445 'Writing' => \&handle_loglevel_ignore,
2446 'Received' => \&handle_loglevel_ignore,
2447 'Actions' => \&handle_loglevel_ignore,
2448 'Unknown log level' => \&handle_loglevel_ignore,
2453 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/) {
2454 $thread = $t = ($shorten_thread_ids) ? shorten_thread_id($4) : $4;
2455 $req{$t}{'day'} = $day = $1;
2456 $req{$t}{'time-stamp'} = $time_stamp = $2;
2457 $req{$t}{'msecs'} = $msecs = $3 ? $3 : 0; # Only the cool kids have micro second resolution;
2458 $req{$t}{'log-level'} = $log_level = $5;
2459 $req{$t}{'content'} = $content = $c = $6;
2460 $req{$t}{'log-message'} = $_;
2461 $no_special_header_highlighting = 0;
2463 if (defined($log_level_handlers{$log_level})) {
2465 $content = $log_level_handlers{$log_level}($content);
2469 die "No handler found for log level \"$log_level\"\n";
2472 # Highlight Truncations
2473 if (length($_) > 4000) {
2474 $content =~ s@(too long, truncated)]$@$h{'Truncation'}$1$h{'Standard'}]@g;
2477 next unless $content;
2479 # Register threads to keep the colour constant
2480 if (!defined($thread_colours{$thread})) {
2481 $thread_colours{$thread} = $all_colours[$thread_colour_index % @all_colours];
2482 $thread_colour_index++;
2485 # Switch timestamp colour if timestamps differ
2486 if (($msecs ne $last_msecs) || ($time_stamp ne $last_timestamp)) {
2487 debug_message("Tick tack!") if DEBUG_TICKS;
2488 $time_colour = $time_colours[$time_colour_index % 2];
2489 $time_colour_index++;
2490 $last_msecs = $msecs;
2491 $last_timestamp = $time_stamp;
2494 $last_thread = $thread;
2496 print_non_clf_message($content);
2498 } elsif (m/^((?:\d+\.\d+\.\d+\.\d+|[:\d]+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2500 # LOG_LEVEL_CLF lines look like this
2501 # 61.152.239.32 - - [04/Mar/2007:18:28:23 +0100] "GET \
2502 # http://ad.yieldmanager.com/imp?z=1&Z=120x600&s=109339&u=http%3A%2F%2Fwww.365loan.co.uk%2F&r=1\
2503 # HTTP/1.1" 403 1730
2504 our ($ip, $timestamp, $request_line, $status_code, $size) = ($1, $2, $3, $4, $5);
2506 print_clf_message();
2510 # Some Privoxy log messages span more than one line,
2511 # usually to dump lots of content that doesn't need any syntax highlighting.
2512 # XXX: add mechanism to forward these lines to the right handler anyway.
2514 unless (DEBUG_SUPPRESS_LOG_MESSAGES or (SUPPRESS_EMPTY_LINES and m/^\s+$/)) {
2515 print and print get_line_end(); # unless (SUPPRESS_EMPTY_LINES and m/^\s+$/);
2523 my ($day, $time_stamp, $msecs, $thread, $log_level, $content);
2524 my $strict_checks = cli_option_is_set('strict-checks');
2525 my %log_level_handlers = (
2526 'Connect:' => \&gather_loglevel_connect_stats,
2527 'Crunch:' => \&gather_loglevel_crunch_stats,
2528 'Error:' => \&gather_loglevel_error_stats,
2529 'Header:' => \&gather_loglevel_header_stats,
2530 'Request:' => \&gather_loglevel_request_stats,
2532 my %ignored_log_levels = (
2533 'Actions:' => \&handle_loglevel_ignore,
2534 'CGI:' => \&handle_loglevel_ignore,
2535 'Fatal error:' => \&handle_loglevel_ignore,
2536 'Force:' => \&handle_loglevel_ignore,
2537 'Gif-Deanimate:' => \&handle_loglevel_ignore,
2538 'Info:' => \&handle_loglevel_ignore,
2539 'Re-Filter:' => \&handle_loglevel_ignore,
2540 'Received:' => \&handle_loglevel_ignore,
2541 'Redirect:' => \&handle_loglevel_ignore,
2542 'Unknown log level:' => \&handle_loglevel_ignore,
2543 'Writing:' => \&handle_loglevel_ignore,
2547 (undef, $time_stamp, $thread, $log_level, $content) = split(/ /, $_, 5);
2550 next if (not defined($log_level));
2552 if ($time_stamp eq "-") {
2554 gather_loglevel_clf_stats($content);
2556 } elsif (defined($log_level_handlers{$log_level})) {
2558 $content = $log_level_handlers{$log_level}($content, $thread);
2560 } elsif ($strict_checks and not defined($ignored_log_levels{$log_level})) {
2562 die "No handler found for: $_";
2570 sub unbreak_lines_only_loop() {
2571 my $log_messages_reached = 0;
2575 # Log level other than LOG_LEVEL_CLF?
2576 if (m/^(\d{4}-\d{2}-\d{2}|\w{3} \d{2}) (\d\d:\d\d:\d\d)\.?(\d+)? (?:Privoxy\()?([^\)\s]*)[\)]? ([\w -]*): (.*?)\r?$/ or
2578 m/^((?:\d+\.\d+\.\d+\.\d+)) - - \[(.*)\] "(.*)" (\d+) (\d+)/) {
2579 $log_messages_reached = 1;
2584 $_ = "\n". $_ if /^(?:\d+\.\d+\.\d+\.\d+)/;
2589 print "\n" unless $log_messages_reached;
2594 sub VersionMessage {
2595 my $version_message;
2597 $version_message .= 'Privoxy-Log-Parser ' . PRIVOXY_LOG_PARSER_VERSION . "\n";
2598 $version_message .= 'https://www.fabiankeil.de/sourcecode/privoxy-log-parser/' . "\n";
2600 print $version_message;
2603 sub get_cli_options() {
2605 our %cli_options = (
2606 'html-output' => CLI_OPTION_DEFAULT_TO_HTML_OUTPUT,
2607 'title' => CLI_OPTION_TITLE,
2608 'no-syntax-highlighting' => CLI_OPTION_NO_SYNTAX_HIGHLIGHTING,
2609 'no-embedded-css' => CLI_OPTION_NO_EMBEDDED_CSS,
2610 'no-msecs' => CLI_OPTION_NO_MSECS,
2611 'shorten-thread-ids' => CLI_OPTION_SHORTEN_THREAD_IDS,
2612 'show-ineffective-filters' => CLI_OPTION_SHOW_INEFFECTIVE_FILTERS,
2613 'statistics' => CLI_OPTION_STATISTICS,
2614 'strict-checks' => CLI_OPTION_STRICT_CHECKS,
2615 'url-statistics-threshold' => CLI_OPTION_URL_STATISTICS_THRESHOLD,
2616 'unbreak-lines-only' => CLI_OPTION_UNBREAK_LINES_ONLY,
2617 'host-statistics-threshold'=> CLI_OPTION_HOST_STATISTICS_THRESHOLD,
2618 'show-complete-request-distribution' => CLI_OPTION_SHOW_COMPLETE_REQUEST_DISTRIBUTION,
2622 'html-output' => \$cli_options{'html-output'},
2623 'title' => \$cli_options{'title'},
2624 'no-syntax-highlighting' => \$cli_options{'no-syntax-highlighting'},
2625 'no-embedded-css' => \$cli_options{'no-embedded-css'},
2626 'no-msecs' => \$cli_options{'no-msecs'},
2627 'shorten-thread-ids' => \$cli_options{'shorten-thread-ids'},
2628 'show-ineffective-filters' => \$cli_options{'show-ineffective-filters'},
2629 'statistics' => \$cli_options{'statistics'},
2630 'strict-checks' => \$cli_options{'strict-checks'},
2631 'unbreak-lines-only' => \$cli_options{'unbreak-lines-only'},
2632 'url-statistics-threshold=i'=> \$cli_options{'url-statistics-threshold'},
2633 'host-statistics-threshold=i'=> \$cli_options{'host-statistics-threshold'},
2634 'show-complete-request-distribution' => \$cli_options{'show-complete-request-distribution'},
2635 'version' => sub { VersionMessage && exit(0) },
2639 $html_output_mode = cli_option_is_set('html-output');
2640 $no_msecs_mode = cli_option_is_set('no-msecs');
2641 $shorten_thread_ids = cli_option_is_set('shorten-thread-ids');
2642 $line_end = get_line_end();
2653 Options and their default values if they have any:
2654 [--host-statistics-threshold $cli_options{'host-statistics-threshold'}]
2658 [--no-syntax-highlighting]
2659 [--shorten-thread-ids]
2660 [--show-ineffective-filters]
2661 [--show-complete-request-distribution]
2663 [--unbreak-lines-only]
2664 [--url-statistics-threshold $cli_options{'url-statistics-threshold'}]
2665 [--title $cli_options{'title'}]
2667 see "perldoc $0" for more information
2673 ################################################################################
2675 ################################################################################
2679 set_background(DEFAULT_BACKGROUND);
2680 prepare_our_stuff();
2684 # XXX: should explicitly reject incompatible argument combinations
2685 if (cli_option_is_set('unbreak-lines-only')) {
2686 unbreak_lines_only_loop();
2687 } elsif (cli_option_is_set('statistics')) {
2700 B<privoxy-log-parser> - A parser and syntax-highlighter for Privoxy log messages
2704 B<privoxy-log-parser> [B<--html-output>]
2705 [B<--no-msecs>] [B<--no-syntax-higlighting>] [B<--statistics>]
2706 [B<--shorten-thread-ids>] [B<--show-ineffective-filters>]
2707 [B<--url-statistics-threshold>] [B<--version>]
2711 B<privoxy-log-parser> reads Privoxy log messages and
2713 - syntax-highlights recognized lines,
2715 - reformats some of them for easier comprehension,
2717 - filters out less useful messages, and
2719 - (in some cases) calculates additional information,
2720 like the compression ratio or how a filter affected
2723 With B<privoxy-log-parser> you should be able to increase Privoxy's log level
2724 without getting confused by the resulting amount of output. For example for
2725 "debug 64" B<privoxy-log-parser> will (by default) only show messages that
2726 affect the content. If a filter doesn't cause any hits, B<privoxy-log-parser>
2727 will hide the "filter foo caused 0 hits" message.
2731 [B<--host-statistics-threshold>] Only show the request count for a host
2732 if it's above or equal to the given threshold. If the threshold is 0, host
2733 statistics are disabled.
2735 [B<--html-output>] Use HTML and CSS for the syntax highlighting. If this option is
2736 omitted, ANSI escape sequences are used unless B<--no-syntax-highlighting> is active.
2737 This option is only intended to make embedding log excerpts in web pages easier.
2738 It does not escape any input!
2740 [B<--no-msecs>] Don't expect milisecond resolution
2742 [B<--no-syntax-highlighting>] Disable syntax-highlighting. Useful when
2743 the filtered output is piped into less in which case the ANSI control
2744 codes don't work, or if the terminal itself doesn't support the control
2747 [B<--shorten-thread-ids>] Shorten the thread ids to a three-digit decimal number.
2748 Note that the mapping from thread ids to shortened ids is created at run-time
2749 and thus varies with the input.
2751 [B<--show-ineffective-filters>] Don't suppress log lines for filters
2752 that didn't modify the content.
2754 [B<--show-complete-request-distribution>] Show the complete client request
2755 distribution in the B<--statistics> output. Without this option only the
2756 ten most common numbers are shown.
2758 [B<--statistics>] Gather various statistics instead of syntax highlighting
2759 log messages. This is an experimental feature, if the results look wrong
2760 they very well might be. Also note that the results are pretty much guaranteed
2761 to be incorrect if Privoxy and Privoxy-Log-Parser aren't in sync.
2763 [B<--strict-checks>] When generating statistics, look more careful at the
2764 input data and abort if it is unexpected, even if it doesn't affect the
2765 results. Significantly slows the parsing down and is not expected to catch
2766 any problems that matter.
2767 When highlighting, print warnings in case of unknown messages which can't be
2768 properly highlighted.
2770 [B<--unbreak-lines-only>] Tries to fix lines that got messed up by a broken or
2771 interestingly configured mail client and thus are no longer recognized properly.
2772 Only fixes some breakage, but may be good enough or at least better than nothing.
2773 Doesn't do anything else, so you probably want to pipe the output into
2774 B<privoxy-log-parser> again.
2776 [B<--url-statistics-threshold>] Only show the request count for a resource
2777 if it's above or equal to the given threshold. If the threshold is 0, URL
2778 statistics are disabled.
2780 [B<--version>] Print version and exit.
2784 To monitor a log file:
2786 tail -F /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log | B<privoxy-log-parser>
2788 Replace '-F' with '-f' if your tail implementation lacks '-F' support
2789 or if the log won't get rotated anyway. The log file location depends
2790 on your system (Doh!).
2792 To monitor Privoxy without having it write to a log file:
2794 privoxy --no-daemon /usr/jails/privoxy-jail/usr/local/etc/privoxy/config 2>&1 | B<privoxy-log-parser>
2796 Again, the config file location depends on your system. Output redirection
2797 depends on your shell, the above works with bourne shells.
2799 To read a processed Privoxy log file from top to bottom, letting the content
2800 scroll by slightly faster than you can read:
2802 B<privoxy-log-parser> < /usr/jails/privoxy-jail/var/log/privoxy/privoxy.log
2804 This is probably only useful to fill screens in the background of haxor movies.
2808 Syntax highlighting with ANSI escape sequences will look strange
2809 if your background color isn't black.
2811 Some messages aren't recognized yet and will not be fully highlighted.
2813 B<privoxy-log-parser> is developed with Privoxy 3.0.7 or later in mind,
2814 using earlier Privoxy versions will probably result in an increased amount
2815 of unrecognized log lines.
2817 Privoxy's log files tend to be rather large. If you use HTML
2818 highlighting some browsers can't handle them, get confused and
2819 will eventually crash because of segmentation faults or unexpected
2820 exceptions. This is a problem in the browser and not B<privoxy-log-parser>'s
2825 Many settings can't be controlled through command line options yet.
2833 Fabian Keil <fk@fabiankeil.de>